{
  "schema_version": 1,
  "report_sha256": "5161cfd64291a4772e08ef4620fdbff23d827611dd7d972614a4ba5c1e91e501",
  "measurements_sha256": "24e9ac46790824768124d498e02a4d66aa77b8f2e48cb0f40ddce2f94fc90027",
  "models": [
    {
      "id": "descript-dac-44khz",
      "name": "Descript DAC",
      "version": "44.1 kHz \u00b7 8 kbps \u00b7 0.0.1",
      "family": "Descript",
      "adapter": "backend.codecs:DescriptDACAdapter",
      "description": "Descript's universal residual vector quantized audio codec, using all nine codebooks from the original 44.1 kHz checkpoint.",
      "sample_rate": 44100,
      "channels": 1,
      "codebooks": 9,
      "codebook_sizes": [
        1024,
        1024,
        1024,
        1024,
        1024,
        1024,
        1024,
        1024,
        1024
      ],
      "hop_length": 512,
      "frame_rate": 86.1328125,
      "bitrate_kbps": 7.751953125,
      "advertised_bitrate_kbps": 8,
      "source_url": "https://github.com/descriptinc/descript-audio-codec",
      "source_revision": "c7cfc5d2647e26471dc394f95846a0830e7bec34",
      "license": "MIT",
      "checkpoint": "models/descript-dac/weights_44khz_8kbps_0.0.1.pth",
      "checkpoint_url": "https://github.com/descriptinc/descript-audio-codec/releases/download/0.0.1/weights.pth",
      "checkpoint_version": "0.0.1",
      "dependencies": [
        "torch",
        "torchaudio",
        "dac"
      ],
      "max_chunk_seconds": 30,
      "options": {
        "model_type": "44khz",
        "model_bitrate": "8kbps",
        "n_quantizers": 9
      },
      "measured_checkpoint_sha256": "a88eed82a7024ccc1facdb1e605c4c2f99281c8118c22c9895ffa846d8fb61aa",
      "label": "Descript DAC"
    },
    {
      "id": "fish-400m-morepitchloss1-step3802000",
      "name": "Fish 400M MorePitchLoss1",
      "version": "Step 3,802,000 \u00b7 targeted high-pitch loss \u00b7 44.1 kHz \u00b7 2026-09-15",
      "family": "Custom Fish-style DAC",
      "adapter": "backend.bundled_dac:BundledDACAdapter",
      "description": "Custom Fish-style DAC generator trained through step 3,802,000 with targeted extra pitch-loss capacity above 250 Hz. Uses its exact bundled architecture and all ten discrete codebooks. No loudness normalization or direct-forward fallback; not the official S2-Pro checkpoint.",
      "sample_rate": 44100,
      "channels": 1,
      "codebooks": 10,
      "codebook_sizes": [
        4096,
        1024,
        1024,
        1024,
        1024,
        1024,
        1024,
        1024,
        1024,
        1024
      ],
      "hop_length": 2048,
      "frame_rate": 21.533203125,
      "bitrate_kbps": 2.19638671875,
      "source_url": "",
      "source_kind": "r2_bundle",
      "source_revision": "archive-sha256:7cdbab0d27c259c3048e985f0fc4624d330de25b596edc3dfa711f20e1cf0075",
      "license": "Bundled code: MIT; checkpoint terms unspecified",
      "checkpoint": "models/fish-400m-morepitchloss1-step3802000/checkpoint/dac/weights.pth",
      "checkpoint_sha256": "a97c5ebe1cbfcd0a341e268de5a31b42b1daac4d64741a25687ae271c3c94321",
      "checkpoint_revision": "step3802000-generator-20260915",
      "training_step": 3802000,
      "artifact_bucket": "artifacts",
      "artifact_key": "fishdac_400m_targeted_pitchloss_step3802000_G_20260915.tar.zst",
      "artifact_sha256": "7cdbab0d27c259c3048e985f0fc4624d330de25b596edc3dfa711f20e1cf0075",
      "bundle_directory": "models/fish-400m-morepitchloss1-step3802000",
      "required_files": [
        "models/fish-400m-morepitchloss1-step3802000/checkpoint/dac/package.pth"
      ],
      "dependencies": [
        "torch",
        "torchaudio",
        "dac",
        "audiotools",
        "einops",
        "vector_quantize_pytorch"
      ],
      "max_chunk_seconds": 30,
      "options": {
        "n_quantizers": 10,
        "source_package": "models/fish-400m-morepitchloss1-step3802000/checkpoint/dac/package.pth",
        "source_package_sha256": "0d819f48b78a775ffc8a8d52cdbfe35f253e4fa1ca1030c8a550c726c81e9791"
      },
      "measured_checkpoint_sha256": "a97c5ebe1cbfcd0a341e268de5a31b42b1daac4d64741a25687ae271c3c94321",
      "label": "Fish 400M MorePitchLoss1"
    },
    {
      "id": "fish-dac-s2",
      "name": "Fish DAC",
      "version": "S2-Pro codec \u00b7 44.1 kHz",
      "family": "Fish Audio",
      "adapter": "backend.codecs:FishDACAdapter",
      "description": "The actual modded DAC codec from Fish Speech S2-Pro: one semantic codebook plus nine residual codebooks. Only the codec is loaded; no text-to-speech generation is involved.",
      "sample_rate": 44100,
      "channels": 1,
      "codebooks": 10,
      "codebook_sizes": [
        4096,
        1024,
        1024,
        1024,
        1024,
        1024,
        1024,
        1024,
        1024,
        1024
      ],
      "hop_length": 2048,
      "frame_rate": 21.533203125,
      "bitrate_kbps": 2.19638671875,
      "source_url": "https://github.com/fishaudio/fish-speech",
      "source_revision": "befe4001745417f8c42131739d862b8a6fdbd15a",
      "source_directory": "vendor/fish-speech",
      "license": "Fish Audio Research License",
      "checkpoint": "models/fish-dac-s2/codec.pth",
      "checkpoint_url": "https://huggingface.co/fishaudio/s2-pro/blob/1de9996b6be38b745688de084d87a5633f714e4e/codec.pth",
      "checkpoint_repository": "fishaudio/s2-pro",
      "checkpoint_filename": "codec.pth",
      "checkpoint_revision": "1de9996b6be38b745688de084d87a5633f714e4e",
      "dependencies": [
        "torch",
        "torchaudio",
        "dac",
        "hydra",
        "omegaconf"
      ],
      "max_chunk_seconds": 30,
      "options": {
        "config": "fish_speech/configs/modded_dac_vq.yaml"
      },
      "measured_checkpoint_sha256": "74fc41c5a7151c6f350af8bd7e5d6e3accfcc7f3dfbfac23afd35af07052bb2f",
      "label": "Fish DAC S2-Pro"
    },
    {
      "id": "qwen3-tts-tokenizer-12hz",
      "name": "Qwen3-TTS Tokenizer 12Hz",
      "version": "24 kHz \u00b7 12.5 frames/s \u00b7 16 codebooks",
      "family": "Qwen",
      "adapter": "backend.qwen_tokenizer:QwenTokenizerAdapter",
      "description": "Qwen's discrete audio tokenizer only, not its speech generation model. Uses all sixteen 2,048-entry codebooks. Despite the 12Hz product name, the pinned checkpoint's native token frame rate is 12.5 Hz (1,920 samples at 24 kHz). Reconstruction is decoded directly from discrete indices without loudness normalization.",
      "sample_rate": 24000,
      "channels": 1,
      "codebooks": 16,
      "codebook_sizes": [
        2048,
        2048,
        2048,
        2048,
        2048,
        2048,
        2048,
        2048,
        2048,
        2048,
        2048,
        2048,
        2048,
        2048,
        2048,
        2048
      ],
      "hop_length": 1920,
      "frame_rate": 12.5,
      "bitrate_kbps": 2.2,
      "source_url": "https://github.com/QwenLM/Qwen3-TTS",
      "source_revision": "qwen-tts==0.1.1",
      "source_kind": "hf_snapshot",
      "license": "Apache-2.0",
      "checkpoint": "models/qwen3-tts-tokenizer-12hz/model.safetensors",
      "checkpoint_url": "https://huggingface.co/Qwen/Qwen3-TTS-Tokenizer-12Hz/blob/7dd38ad4e9bad454aae9cd937d0cd577604fe229/model.safetensors",
      "checkpoint_repository": "Qwen/Qwen3-TTS-Tokenizer-12Hz",
      "checkpoint_filename": "model.safetensors",
      "checkpoint_revision": "7dd38ad4e9bad454aae9cd937d0cd577604fe229",
      "checkpoint_sha256": "836b7b357f5ea43e889936a3709af68dfe3751881acefe4ecf0dbd30ba571258",
      "required_files": [
        "models/qwen3-tts-tokenizer-12hz/config.json",
        "models/qwen3-tts-tokenizer-12hz/preprocessor_config.json"
      ],
      "dependencies": [
        "torch",
        "torchaudio",
        "qwen_tts",
        "transformers",
        "accelerate",
        "onnxruntime",
        "sox",
        "safetensors"
      ],
      "max_chunk_seconds": 30,
      "options": {
        "n_quantizers": 16,
        "dtype": "float32",
        "decoder_policy": "direct discrete decode; no wrapper chunking"
      },
      "measured_checkpoint_sha256": "836b7b357f5ea43e889936a3709af68dfe3751881acefe4ecf0dbd30ba571258",
      "label": "Qwen3-TTS 12Hz"
    }
  ],
  "bands": [
    {
      "id": "0-5",
      "label": "0\u20135%",
      "low": 0,
      "high": 5,
      "target": 2.5
    },
    {
      "id": "5-10",
      "label": ">5\u201310%",
      "low": 5,
      "high": 10,
      "target": 7.5
    },
    {
      "id": "10-20",
      "label": ">10\u201320%",
      "low": 10,
      "high": 20,
      "target": 15
    },
    {
      "id": "20-30",
      "label": ">20\u201330%",
      "low": 20,
      "high": 30,
      "target": 25
    },
    {
      "id": "30-40",
      "label": ">30\u201340%",
      "low": 30,
      "high": 40,
      "target": 35
    },
    {
      "id": "40-50",
      "label": ">40\u201350%",
      "low": 40,
      "high": 50,
      "target": 45
    },
    {
      "id": "over-50",
      "label": ">50%",
      "low": 50,
      "high": null,
      "target": 75
    }
  ],
  "metrics": [
    {
      "id": "mr_stft",
      "label": "MR-STFT"
    },
    {
      "id": "log_mel",
      "label": "Log-mel L1"
    },
    {
      "id": "log1p_mel",
      "label": "Log1p-mel L1"
    }
  ],
  "generated_at": "2026-09-30T14:09:59.173449+00:00",
  "sources": [
    {
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "audio_sha256": "6daab40b44b596802e4ca509696b069474e543da2fe7aa59c6299b74f8f2d05e",
      "label": "Cartoon dialogue",
      "duration_seconds": 5.718458333333333,
      "eligible_complete_anchors": 1,
      "manifest_text": "",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "original_sample_rate": 44100
    },
    {
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "audio_sha256": "ee4f6e29286c050620350704bb6927bfd96a58dec185898f0553f6ef1f68c4fb",
      "label": "LibriSpeech Other",
      "duration_seconds": 5.13,
      "eligible_complete_anchors": 6,
      "manifest_text": "THEY ARE REGARDED AS GOOD THAT IS TO SAY AS MORAL",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "original_sample_rate": 16000
    },
    {
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "audio_sha256": "bd0ea4ced877804fa43ccb8b96fdb68b16c93b38137ede65ac60bf535a3296a0",
      "label": "Podcast speech",
      "duration_seconds": 10.563,
      "eligible_complete_anchors": 1,
      "manifest_text": "We can deal in many coins in our realm, but for physical goods, physical cash generally. Unless you find something real juicy for me, but I feel like the information may be flowing your way, so I'll get different things from you over time.",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "original_sample_rate": 48000
    }
  ],
  "examples": [
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "0-5",
      "audio": "listening-audio/cartoon-voices-descript-dac-44khz-window-134.wav",
      "baseline": {
        "log_mel": 0.3338501751422882,
        "log1p_mel": 0.0026197368279099464,
        "mr_stft": 0.9093503952026367
      },
      "treatment": {
        "log_mel": 0.3460738956928253,
        "log1p_mel": 0.002793185878545046,
        "mr_stft": 0.9323517084121704
      },
      "decoder_drift": {
        "log_mel": 0.025301355868577957,
        "log1p_mel": 0.000250395416514948,
        "mr_stft": 0.11527948081493378
      },
      "relative_percent": {
        "mr_stft": 2.5294224680474375,
        "log_mel": 3.6614390108758554,
        "log1p_mel": 6.620857820038317
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 134,
      "window_ms": 1555.736961451247,
      "scalar_tokens": 1206,
      "codebooks": 9,
      "seam_count": 3,
      "id": "descript-dac-44khz--cartoon-voices--mr_stft--0-5",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 2.5294224680474375,
        "log_mel": 3.6614390108758554,
        "log1p_mel": 6.620857820038317
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "5-10",
      "audio": "listening-audio/cartoon-voices-descript-dac-44khz-window-59.wav",
      "baseline": {
        "log_mel": 0.3338501751422882,
        "log1p_mel": 0.0026197368279099464,
        "mr_stft": 0.9093503952026367
      },
      "treatment": {
        "log_mel": 0.35594743490219116,
        "log1p_mel": 0.0030640324112027884,
        "mr_stft": 0.9774693846702576
      },
      "decoder_drift": {
        "log_mel": 0.05163843557238579,
        "log1p_mel": 0.0007157908985391259,
        "mr_stft": 0.21719682216644287
      },
      "relative_percent": {
        "mr_stft": 7.490950664011242,
        "log_mel": 6.6189151317608275,
        "log1p_mel": 16.959550232658515
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 59,
      "window_ms": 684.9886621315193,
      "scalar_tokens": 531,
      "codebooks": 9,
      "seam_count": 8,
      "id": "descript-dac-44khz--cartoon-voices--mr_stft--5-10",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 7.490950664011242,
        "log_mel": 6.6189151317608275,
        "log1p_mel": 16.959550232658515
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "10-20",
      "audio": "listening-audio/cartoon-voices-descript-dac-44khz-window-21.wav",
      "baseline": {
        "log_mel": 0.3338501751422882,
        "log1p_mel": 0.0026197368279099464,
        "mr_stft": 0.9093503952026367
      },
      "treatment": {
        "log_mel": 0.450326144695282,
        "log1p_mel": 0.0037931061815470457,
        "mr_stft": 1.0482585430145264
      },
      "decoder_drift": {
        "log_mel": 0.20218178629875183,
        "log1p_mel": 0.0018406793242320418,
        "mr_stft": 0.4348534941673279
      },
      "relative_percent": {
        "mr_stft": 15.275536090896601,
        "log_mel": 34.88869505710197,
        "log1p_mel": 44.78958882954765
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 21,
      "window_ms": 243.8095238095238,
      "scalar_tokens": 189,
      "codebooks": 9,
      "seam_count": 23,
      "id": "descript-dac-44khz--cartoon-voices--mr_stft--10-20",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 15.275536090896601,
        "log_mel": 34.88869505710197,
        "log1p_mel": 44.78958882954765
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "20-30",
      "audio": "listening-audio/cartoon-voices-descript-dac-44khz-window-10.wav",
      "baseline": {
        "log_mel": 0.3338501751422882,
        "log1p_mel": 0.0026197368279099464,
        "mr_stft": 0.9093503952026367
      },
      "treatment": {
        "log_mel": 0.514198362827301,
        "log1p_mel": 0.004839731380343437,
        "mr_stft": 1.1390408277511597
      },
      "decoder_drift": {
        "log_mel": 0.3511083722114563,
        "log1p_mel": 0.0037460278254002333,
        "mr_stft": 0.7443992495536804
      },
      "relative_percent": {
        "mr_stft": 25.258737859495795,
        "log_mel": 54.02069584302232,
        "log1p_mel": 84.74112852796078
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 10,
      "window_ms": 116.09977324263039,
      "scalar_tokens": 90,
      "codebooks": 9,
      "seam_count": 49,
      "id": "descript-dac-44khz--cartoon-voices--mr_stft--20-30",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 25.258737859495795,
        "log_mel": 54.02069584302232,
        "log1p_mel": 84.74112852796078
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "30-40",
      "audio": "listening-audio/cartoon-voices-descript-dac-44khz-window-8.wav",
      "baseline": {
        "log_mel": 0.3338501751422882,
        "log1p_mel": 0.0026197368279099464,
        "mr_stft": 0.9093503952026367
      },
      "treatment": {
        "log_mel": 0.5645103454589844,
        "log1p_mel": 0.005966950207948685,
        "mr_stft": 1.2240171432495117
      },
      "decoder_drift": {
        "log_mel": 0.4319925010204315,
        "log1p_mel": 0.005039328243583441,
        "mr_stft": 0.8991166949272156
      },
      "relative_percent": {
        "mr_stft": 34.60346525463991,
        "log_mel": 69.09092386079712,
        "log1p_mel": 127.76906994544106
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 8,
      "window_ms": 92.87981859410431,
      "scalar_tokens": 72,
      "codebooks": 9,
      "seam_count": 61,
      "id": "descript-dac-44khz--cartoon-voices--mr_stft--30-40",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 34.60346525463991,
        "log_mel": 69.09092386079712,
        "log1p_mel": 127.76906994544106
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "40-50",
      "audio": "listening-audio/cartoon-voices-descript-dac-44khz-window-6.wav",
      "baseline": {
        "log_mel": 0.3338501751422882,
        "log1p_mel": 0.0026197368279099464,
        "mr_stft": 0.9093503952026367
      },
      "treatment": {
        "log_mel": 0.6534324288368225,
        "log1p_mel": 0.007164353039115667,
        "mr_stft": 1.3265814781188965
      },
      "decoder_drift": {
        "log_mel": 0.5623480677604675,
        "log1p_mel": 0.006562489550560713,
        "mr_stft": 1.0968327522277832
      },
      "relative_percent": {
        "mr_stft": 45.88232271271904,
        "log_mel": 95.72625012352536,
        "log1p_mel": 173.47605922811198
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 6,
      "window_ms": 69.65986394557822,
      "scalar_tokens": 54,
      "codebooks": 9,
      "seam_count": 82,
      "id": "descript-dac-44khz--cartoon-voices--mr_stft--40-50",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 45.88232271271904,
        "log_mel": 95.72625012352536,
        "log1p_mel": 173.47605922811198
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "over-50",
      "audio": "listening-audio/cartoon-voices-descript-dac-44khz-window-4.wav",
      "baseline": {
        "log_mel": 0.3338501751422882,
        "log1p_mel": 0.0026197368279099464,
        "mr_stft": 0.9093503952026367
      },
      "treatment": {
        "log_mel": 0.8385322093963623,
        "log1p_mel": 0.009735261090099812,
        "mr_stft": 1.5278514623641968
      },
      "decoder_drift": {
        "log_mel": 0.7930729389190674,
        "log1p_mel": 0.009430600330233574,
        "mr_stft": 1.395703673362732
      },
      "relative_percent": {
        "mr_stft": 68.01570334433464,
        "log_mel": 151.17021701096206,
        "log1p_mel": 271.61217823038754
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-07,
      "window_frames": 4,
      "window_ms": 46.439909297052154,
      "scalar_tokens": 36,
      "codebooks": 9,
      "seam_count": 123,
      "id": "descript-dac-44khz--cartoon-voices--mr_stft--over-50",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 68.01571645361464,
        "log_mel": 151.17021701096206,
        "log1p_mel": 271.61217823038754
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "0-5",
      "audio": "listening-audio/cartoon-voices-descript-dac-44khz-window-142.wav",
      "baseline": {
        "log_mel": 0.3338501751422882,
        "log1p_mel": 0.0026197368279099464,
        "mr_stft": 0.9093503952026367
      },
      "treatment": {
        "log_mel": 0.342407763004303,
        "log1p_mel": 0.002806874457746744,
        "mr_stft": 0.9301456212997437
      },
      "decoder_drift": {
        "log_mel": 0.01989361084997654,
        "log1p_mel": 0.0003135719452984631,
        "mr_stft": 0.1059214174747467
      },
      "relative_percent": {
        "mr_stft": 2.2868221322401183,
        "log_mel": 2.5633018938413077,
        "log1p_mel": 7.143375160553744
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 142,
      "window_ms": 1648.6167800453516,
      "scalar_tokens": 1278,
      "codebooks": 9,
      "seam_count": 3,
      "id": "descript-dac-44khz--cartoon-voices--log_mel--0-5",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 2.2868221322401183,
        "log_mel": 2.5633018938413077,
        "log1p_mel": 7.143375160553744
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "5-10",
      "audio": "listening-audio/cartoon-voices-descript-dac-44khz-window-77.wav",
      "baseline": {
        "log_mel": 0.3338501751422882,
        "log1p_mel": 0.0026197368279099464,
        "mr_stft": 0.9093503952026367
      },
      "treatment": {
        "log_mel": 0.35891181230545044,
        "log1p_mel": 0.0030343495309352875,
        "mr_stft": 0.9536435604095459
      },
      "decoder_drift": {
        "log_mel": 0.04420437291264534,
        "log1p_mel": 0.0006212840671651065,
        "mr_stft": 0.16683706641197205
      },
      "relative_percent": {
        "mr_stft": 4.870857860796218,
        "log_mel": 7.50685158469091,
        "log1p_mel": 15.82650205960281
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 77,
      "window_ms": 893.968253968254,
      "scalar_tokens": 693,
      "codebooks": 9,
      "seam_count": 6,
      "id": "descript-dac-44khz--cartoon-voices--log_mel--5-10",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 4.870857860796218,
        "log_mel": 7.50685158469091,
        "log1p_mel": 15.82650205960281
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "10-20",
      "audio": "listening-audio/cartoon-voices-descript-dac-44khz-window-30.wav",
      "baseline": {
        "log_mel": 0.3338501751422882,
        "log1p_mel": 0.0026197368279099464,
        "mr_stft": 0.9093503952026367
      },
      "treatment": {
        "log_mel": 0.38396286964416504,
        "log1p_mel": 0.003289861371740699,
        "mr_stft": 0.981374979019165
      },
      "decoder_drift": {
        "log_mel": 0.10767170786857605,
        "log1p_mel": 0.0012629308039322495,
        "mr_stft": 0.30116018652915955
      },
      "relative_percent": {
        "mr_stft": 7.920443450236658,
        "log_mel": 15.010534135714803,
        "log1p_mel": 25.57984209297026
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 30,
      "window_ms": 348.29931972789115,
      "scalar_tokens": 270,
      "codebooks": 9,
      "seam_count": 16,
      "id": "descript-dac-44khz--cartoon-voices--log_mel--10-20",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 7.920443450236658,
        "log_mel": 15.010534135714803,
        "log1p_mel": 25.57984209297026
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "20-30",
      "audio": "listening-audio/cartoon-voices-descript-dac-44khz-window-20.wav",
      "baseline": {
        "log_mel": 0.3338501751422882,
        "log1p_mel": 0.0026197368279099464,
        "mr_stft": 0.9093503952026367
      },
      "treatment": {
        "log_mel": 0.41470077633857727,
        "log1p_mel": 0.0037167295813560486,
        "mr_stft": 1.0243782997131348
      },
      "decoder_drift": {
        "log_mel": 0.16933271288871765,
        "log1p_mel": 0.0018759395461529493,
        "mr_stft": 0.4288777709007263
      },
      "relative_percent": {
        "mr_stft": 12.649458901358447,
        "log_mel": 24.217630307316806,
        "log1p_mel": 41.874158570397114
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 20,
      "window_ms": 232.19954648526078,
      "scalar_tokens": 180,
      "codebooks": 9,
      "seam_count": 24,
      "id": "descript-dac-44khz--cartoon-voices--log_mel--20-30",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 12.649458901358447,
        "log_mel": 24.217630307316806,
        "log1p_mel": 41.874158570397114
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "30-40",
      "audio": "listening-audio/cartoon-voices-descript-dac-44khz-window-21.wav",
      "baseline": {
        "log_mel": 0.3338501751422882,
        "log1p_mel": 0.0026197368279099464,
        "mr_stft": 0.9093503952026367
      },
      "treatment": {
        "log_mel": 0.450326144695282,
        "log1p_mel": 0.0037931061815470457,
        "mr_stft": 1.0482585430145264
      },
      "decoder_drift": {
        "log_mel": 0.20218178629875183,
        "log1p_mel": 0.0018406793242320418,
        "mr_stft": 0.4348534941673279
      },
      "relative_percent": {
        "mr_stft": 15.275536090896601,
        "log_mel": 34.88869505710197,
        "log1p_mel": 44.78958882954765
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 21,
      "window_ms": 243.8095238095238,
      "scalar_tokens": 189,
      "codebooks": 9,
      "seam_count": 23,
      "id": "descript-dac-44khz--cartoon-voices--log_mel--30-40",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 15.275536090896601,
        "log_mel": 34.88869505710197,
        "log1p_mel": 44.78958882954765
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "40-50",
      "audio": "listening-audio/cartoon-voices-descript-dac-44khz-window-13.wav",
      "baseline": {
        "log_mel": 0.3338501751422882,
        "log1p_mel": 0.0026197368279099464,
        "mr_stft": 0.9093503952026367
      },
      "treatment": {
        "log_mel": 0.4749208092689514,
        "log1p_mel": 0.0044457498006522655,
        "mr_stft": 1.09793221950531
      },
      "decoder_drift": {
        "log_mel": 0.27771490812301636,
        "log1p_mel": 0.0029422836378216743,
        "mr_stft": 0.6115362644195557
      },
      "relative_percent": {
        "mr_stft": 20.738081304803345,
        "log_mel": 42.25567174452982,
        "log1p_mel": 69.70215302882661
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 13,
      "window_ms": 150.92970521541952,
      "scalar_tokens": 117,
      "codebooks": 9,
      "seam_count": 37,
      "id": "descript-dac-44khz--cartoon-voices--log_mel--40-50",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 20.738081304803345,
        "log_mel": 42.25567174452982,
        "log1p_mel": 69.70215302882661
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "over-50",
      "audio": "listening-audio/cartoon-voices-descript-dac-44khz-window-8.wav",
      "baseline": {
        "log_mel": 0.3338501751422882,
        "log1p_mel": 0.0026197368279099464,
        "mr_stft": 0.9093503952026367
      },
      "treatment": {
        "log_mel": 0.5645103454589844,
        "log1p_mel": 0.005966950207948685,
        "mr_stft": 1.2240171432495117
      },
      "decoder_drift": {
        "log_mel": 0.4319925010204315,
        "log1p_mel": 0.005039328243583441,
        "mr_stft": 0.8991166949272156
      },
      "relative_percent": {
        "mr_stft": 34.60346525463991,
        "log_mel": 69.09092386079712,
        "log1p_mel": 127.76906994544106
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 8,
      "window_ms": 92.87981859410431,
      "scalar_tokens": 72,
      "codebooks": 9,
      "seam_count": 61,
      "id": "descript-dac-44khz--cartoon-voices--log_mel--over-50",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 34.60346525463991,
        "log_mel": 69.09092386079712,
        "log1p_mel": 127.76906994544106
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "0-5",
      "audio": "listening-audio/cartoon-voices-descript-dac-44khz-window-160.wav",
      "baseline": {
        "log_mel": 0.3338501751422882,
        "log1p_mel": 0.0026197368279099464,
        "mr_stft": 0.9093503952026367
      },
      "treatment": {
        "log_mel": 0.3414854407310486,
        "log1p_mel": 0.0026937988586723804,
        "mr_stft": 0.9161485433578491
      },
      "decoder_drift": {
        "log_mel": 0.01895088516175747,
        "log1p_mel": 0.0001619661197764799,
        "mr_stft": 0.06869953870773315
      },
      "relative_percent": {
        "mr_stft": 0.747582910952338,
        "log_mel": 2.2870335729212155,
        "log1p_mel": 2.8270790399019363
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 160,
      "window_ms": 1857.5963718820863,
      "scalar_tokens": 1440,
      "codebooks": 9,
      "seam_count": 3,
      "id": "descript-dac-44khz--cartoon-voices--log1p_mel--0-5",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 0.747582910952338,
        "log_mel": 2.2870335729212155,
        "log1p_mel": 2.8270790399019363
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "5-10",
      "audio": "listening-audio/cartoon-voices-descript-dac-44khz-window-125.wav",
      "baseline": {
        "log_mel": 0.3338501751422882,
        "log1p_mel": 0.0026197368279099464,
        "mr_stft": 0.9093503952026367
      },
      "treatment": {
        "log_mel": 0.3480883240699768,
        "log1p_mel": 0.0028191397432237864,
        "mr_stft": 0.9291256666183472
      },
      "decoder_drift": {
        "log_mel": 0.024449365213513374,
        "log1p_mel": 0.0002737299946602434,
        "mr_stft": 0.0893414244055748
      },
      "relative_percent": {
        "mr_stft": 2.1746591325012554,
        "log_mel": 4.2648319479300145,
        "log1p_mel": 7.6115628558356185
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 125,
      "window_ms": 1451.2471655328798,
      "scalar_tokens": 1125,
      "codebooks": 9,
      "seam_count": 3,
      "id": "descript-dac-44khz--cartoon-voices--log1p_mel--5-10",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 2.1746591325012554,
        "log_mel": 4.2648319479300145,
        "log1p_mel": 7.6115628558356185
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "10-20",
      "audio": "listening-audio/cartoon-voices-descript-dac-44khz-window-72.wav",
      "baseline": {
        "log_mel": 0.3338501751422882,
        "log1p_mel": 0.0026197368279099464,
        "mr_stft": 0.9093503952026367
      },
      "treatment": {
        "log_mel": 0.35574862360954285,
        "log1p_mel": 0.00301732262596488,
        "mr_stft": 0.9507513046264648
      },
      "decoder_drift": {
        "log_mel": 0.04319530725479126,
        "log1p_mel": 0.0006160733755677938,
        "mr_stft": 0.16534413397312164
      },
      "relative_percent": {
        "mr_stft": 4.552800509269317,
        "log_mel": 6.559364079387242,
        "log1p_mel": 15.176554905025764
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 72,
      "window_ms": 835.9183673469388,
      "scalar_tokens": 648,
      "codebooks": 9,
      "seam_count": 6,
      "id": "descript-dac-44khz--cartoon-voices--log1p_mel--10-20",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 4.552800509269317,
        "log_mel": 6.559364079387242,
        "log1p_mel": 15.176554905025764
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "20-30",
      "audio": "listening-audio/cartoon-voices-descript-dac-44khz-window-38.wav",
      "baseline": {
        "log_mel": 0.3338501751422882,
        "log1p_mel": 0.0026197368279099464,
        "mr_stft": 0.9093503952026367
      },
      "treatment": {
        "log_mel": 0.39039796590805054,
        "log1p_mel": 0.0032837181352078915,
        "mr_stft": 0.9879237413406372
      },
      "decoder_drift": {
        "log_mel": 0.10030942410230637,
        "log1p_mel": 0.001024187309667468,
        "mr_stft": 0.27319714426994324
      },
      "relative_percent": {
        "mr_stft": 8.640601747414589,
        "log_mel": 16.93807431482144,
        "log1p_mel": 25.345343861416648
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 38,
      "window_ms": 441.1791383219955,
      "scalar_tokens": 342,
      "codebooks": 9,
      "seam_count": 12,
      "id": "descript-dac-44khz--cartoon-voices--log1p_mel--20-30",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 8.640601747414589,
        "log_mel": 16.93807431482144,
        "log1p_mel": 25.345343861416648
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "30-40",
      "audio": "listening-audio/cartoon-voices-descript-dac-44khz-window-26.wav",
      "baseline": {
        "log_mel": 0.3338501751422882,
        "log1p_mel": 0.0026197368279099464,
        "mr_stft": 0.9093503952026367
      },
      "treatment": {
        "log_mel": 0.406360924243927,
        "log1p_mel": 0.0035263923928141594,
        "mr_stft": 1.012105941772461
      },
      "decoder_drift": {
        "log_mel": 0.1386127918958664,
        "log1p_mel": 0.0013806024799123406,
        "mr_stft": 0.35128170251846313
      },
      "relative_percent": {
        "mr_stft": 11.299884743210175,
        "log_mel": 21.719548019028135,
        "log1p_mel": 34.60865058065974
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 26,
      "window_ms": 301.85941043083903,
      "scalar_tokens": 234,
      "codebooks": 9,
      "seam_count": 18,
      "id": "descript-dac-44khz--cartoon-voices--log1p_mel--30-40",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 11.299884743210175,
        "log_mel": 21.719548019028135,
        "log1p_mel": 34.60865058065974
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "40-50",
      "audio": "listening-audio/cartoon-voices-descript-dac-44khz-window-21.wav",
      "baseline": {
        "log_mel": 0.3338501751422882,
        "log1p_mel": 0.0026197368279099464,
        "mr_stft": 0.9093503952026367
      },
      "treatment": {
        "log_mel": 0.450326144695282,
        "log1p_mel": 0.0037931061815470457,
        "mr_stft": 1.0482585430145264
      },
      "decoder_drift": {
        "log_mel": 0.20218178629875183,
        "log1p_mel": 0.0018406793242320418,
        "mr_stft": 0.4348534941673279
      },
      "relative_percent": {
        "mr_stft": 15.275536090896601,
        "log_mel": 34.88869505710197,
        "log1p_mel": 44.78958882954765
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 21,
      "window_ms": 243.8095238095238,
      "scalar_tokens": 189,
      "codebooks": 9,
      "seam_count": 23,
      "id": "descript-dac-44khz--cartoon-voices--log1p_mel--40-50",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 15.275536090896601,
        "log_mel": 34.88869505710197,
        "log1p_mel": 44.78958882954765
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "over-50",
      "audio": "listening-audio/cartoon-voices-descript-dac-44khz-window-12.wav",
      "baseline": {
        "log_mel": 0.3338501751422882,
        "log1p_mel": 0.0026197368279099464,
        "mr_stft": 0.9093503952026367
      },
      "treatment": {
        "log_mel": 0.4680282175540924,
        "log1p_mel": 0.00453445129096508,
        "mr_stft": 1.0936191082000732
      },
      "decoder_drift": {
        "log_mel": 0.27997082471847534,
        "log1p_mel": 0.0031358671840280294,
        "mr_stft": 0.6342747211456299
      },
      "relative_percent": {
        "mr_stft": 20.26377444487444,
        "log_mel": 40.19109540817733,
        "log1p_mel": 73.08804619824019
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 12,
      "window_ms": 139.31972789115645,
      "scalar_tokens": 108,
      "codebooks": 9,
      "seam_count": 41,
      "id": "descript-dac-44khz--cartoon-voices--log1p_mel--over-50",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 20.26377444487444,
        "log_mel": 40.19109540817733,
        "log1p_mel": 73.08804619824019
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "0-5",
      "audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-38.wav",
      "baseline": {
        "log_mel": 0.45535868406295776,
        "log1p_mel": 0.004118511453270912,
        "mr_stft": 1.0745127201080322
      },
      "treatment": {
        "log_mel": 0.4636480510234833,
        "log1p_mel": 0.004297120496630669,
        "mr_stft": 1.1007288694381714
      },
      "decoder_drift": {
        "log_mel": 0.08566039800643921,
        "log1p_mel": 0.000996531918644905,
        "mr_stft": 0.3180096447467804
      },
      "relative_percent": {
        "mr_stft": 2.4398174949016305,
        "log_mel": 1.8204038378192955,
        "log1p_mel": 4.3367378089456565
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-07,
      "window_frames": 38,
      "window_ms": 1764.716553287982,
      "scalar_tokens": 380,
      "codebooks": 10,
      "seam_count": 3,
      "id": "fish-400m-morepitchloss1-step3802000--cartoon-voices--mr_stft--0-5",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 2.439828589166715,
        "log_mel": 1.8204234722240642,
        "log1p_mel": 4.3367378089456565
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "5-10",
      "audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-30.wav",
      "baseline": {
        "log_mel": 0.45535868406295776,
        "log1p_mel": 0.004118511453270912,
        "mr_stft": 1.0745127201080322
      },
      "treatment": {
        "log_mel": 0.4737754762172699,
        "log1p_mel": 0.0046226345002651215,
        "mr_stft": 1.1547068357467651
      },
      "decoder_drift": {
        "log_mel": 0.11302269995212555,
        "log1p_mel": 0.0015046156477183104,
        "mr_stft": 0.45715194940567017
      },
      "relative_percent": {
        "mr_stft": 7.463300725809008,
        "log_mel": 4.044458313606203,
        "log1p_mel": 12.240418722007828
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-07,
      "window_frames": 30,
      "window_ms": 1393.1972789115646,
      "scalar_tokens": 300,
      "codebooks": 10,
      "seam_count": 4,
      "id": "fish-400m-morepitchloss1-step3802000--cartoon-voices--mr_stft--5-10",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 7.463289631543923,
        "log_mel": 4.044458313606203,
        "log1p_mel": 12.240418722007828
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "10-20",
      "audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-16.wav",
      "baseline": {
        "log_mel": 0.45535868406295776,
        "log1p_mel": 0.004118511453270912,
        "mr_stft": 1.0745127201080322
      },
      "treatment": {
        "log_mel": 0.5001124143600464,
        "log1p_mel": 0.00491088442504406,
        "mr_stft": 1.2355040311813354
      },
      "decoder_drift": {
        "log_mel": 0.18310463428497314,
        "log1p_mel": 0.0023247674107551575,
        "mr_stft": 0.6534328460693359
      },
      "relative_percent": {
        "mr_stft": 14.982727338687729,
        "log_mel": 9.828236918152422,
        "log1p_mel": 19.2393048013463
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 16,
      "window_ms": 743.0385487528345,
      "scalar_tokens": 160,
      "codebooks": 10,
      "seam_count": 7,
      "id": "fish-400m-morepitchloss1-step3802000--cartoon-voices--mr_stft--10-20",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 14.98270515015756,
        "log_mel": 9.828236918152422,
        "log1p_mel": 19.2393048013463
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "20-30",
      "audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-8.wav",
      "baseline": {
        "log_mel": 0.45535868406295776,
        "log1p_mel": 0.004118511453270912,
        "mr_stft": 1.0745127201080322
      },
      "treatment": {
        "log_mel": 0.5402026772499084,
        "log1p_mel": 0.0059037101455032825,
        "mr_stft": 1.337961196899414
      },
      "decoder_drift": {
        "log_mel": 0.2953680157661438,
        "log1p_mel": 0.004123294726014137,
        "mr_stft": 0.9359821081161499
      },
      "relative_percent": {
        "mr_stft": 24.517948634883968,
        "log_mel": 18.632343283744234,
        "log1p_mel": 43.34572605873341
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-07,
      "window_frames": 8,
      "window_ms": 371.51927437641723,
      "scalar_tokens": 80,
      "codebooks": 10,
      "seam_count": 15,
      "id": "fish-400m-morepitchloss1-step3802000--cartoon-voices--mr_stft--20-30",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 24.517937540618885,
        "log_mel": 18.632330194141076,
        "log1p_mel": 43.34572605873341
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "30-40",
      "audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-4.wav",
      "baseline": {
        "log_mel": 0.45535868406295776,
        "log1p_mel": 0.004118511453270912,
        "mr_stft": 1.0745127201080322
      },
      "treatment": {
        "log_mel": 0.6082643866539001,
        "log1p_mel": 0.0071660117246210575,
        "mr_stft": 1.4419114589691162
      },
      "decoder_drift": {
        "log_mel": 0.41535520553588867,
        "log1p_mel": 0.005648731254041195,
        "mr_stft": 1.1230965852737427
      },
      "relative_percent": {
        "mr_stft": 34.19212560128144,
        "log_mel": 33.579177897000804,
        "log1p_mel": 73.99518748284233
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 4,
      "window_ms": 185.75963718820861,
      "scalar_tokens": 40,
      "codebooks": 10,
      "seam_count": 30,
      "id": "fish-400m-morepitchloss1-step3802000--cartoon-voices--mr_stft--30-40",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 34.192147789811614,
        "log_mel": 33.57921716581031,
        "log1p_mel": 73.99518748284233
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "40-50",
      "audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-3.wav",
      "baseline": {
        "log_mel": 0.45535868406295776,
        "log1p_mel": 0.004118511453270912,
        "mr_stft": 1.0745127201080322
      },
      "treatment": {
        "log_mel": 0.6692796349525452,
        "log1p_mel": 0.008835365995764732,
        "mr_stft": 1.5791540145874023
      },
      "decoder_drift": {
        "log_mel": 0.5031737089157104,
        "log1p_mel": 0.00759125268086791,
        "mr_stft": 1.3165324926376343
      },
      "relative_percent": {
        "mr_stft": 46.96466454381603,
        "log_mel": 46.97855962268434,
        "log1p_mel": 114.52813949922867
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 3,
      "window_ms": 139.31972789115645,
      "scalar_tokens": 30,
      "codebooks": 10,
      "seam_count": 41,
      "id": "fish-400m-morepitchloss1-step3802000--cartoon-voices--mr_stft--40-50",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 46.96464235528586,
        "log_mel": 46.978572712287516,
        "log1p_mel": 114.52813949922867
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "over-50",
      "audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-1.wav",
      "baseline": {
        "log_mel": 0.45535868406295776,
        "log1p_mel": 0.004118511453270912,
        "mr_stft": 1.0745127201080322
      },
      "treatment": {
        "log_mel": 0.8963348865509033,
        "log1p_mel": 0.013486160896718502,
        "mr_stft": 1.9584412574768066
      },
      "decoder_drift": {
        "log_mel": 0.8308484554290771,
        "log1p_mel": 0.013109906576573849,
        "mr_stft": 1.8471240997314453
      },
      "relative_percent": {
        "mr_stft": 82.2631990135867,
        "log_mel": 96.84150493262061,
        "log1p_mel": 227.45231013034638
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-07,
      "window_frames": 1,
      "window_ms": 46.439909297052154,
      "scalar_tokens": 10,
      "codebooks": 10,
      "seam_count": 123,
      "id": "fish-400m-morepitchloss1-step3802000--cartoon-voices--mr_stft--over-50",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 82.26318791932161,
        "log_mel": 96.84150493262061,
        "log1p_mel": 227.45231013034638
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "0-5",
      "audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-32.wav",
      "baseline": {
        "log_mel": 0.45535868406295776,
        "log1p_mel": 0.004118511453270912,
        "mr_stft": 1.0745127201080322
      },
      "treatment": {
        "log_mel": 0.4660554528236389,
        "log1p_mel": 0.004282723646610975,
        "mr_stft": 1.1055259704589844
      },
      "decoder_drift": {
        "log_mel": 0.08999848365783691,
        "log1p_mel": 0.000986166181974113,
        "mr_stft": 0.3250672221183777
      },
      "relative_percent": {
        "mr_stft": 2.886261816224378,
        "log_mel": 2.3490863653326644,
        "log1p_mel": 3.9871734048389396
      },
      "recorded_metric_max_abs_difference": 8.940696716308594e-08,
      "window_frames": 32,
      "window_ms": 1486.077097505669,
      "scalar_tokens": 320,
      "codebooks": 10,
      "seam_count": 3,
      "id": "fish-400m-morepitchloss1-step3802000--cartoon-voices--log_mel--0-5",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 2.886261816224378,
        "log_mel": 2.3490667309278956,
        "log1p_mel": 3.9871734048389396
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "5-10",
      "audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-15.wav",
      "baseline": {
        "log_mel": 0.45535868406295776,
        "log1p_mel": 0.004118511453270912,
        "mr_stft": 1.0745127201080322
      },
      "treatment": {
        "log_mel": 0.4890357255935669,
        "log1p_mel": 0.005058096721768379,
        "mr_stft": 1.1885583400726318
      },
      "decoder_drift": {
        "log_mel": 0.1815240979194641,
        "log1p_mel": 0.0023396110627800226,
        "mr_stft": 0.6228591799736023
      },
      "relative_percent": {
        "mr_stft": 10.613705899464222,
        "log_mel": 7.3957174221702005,
        "log1p_mel": 22.813710224145446
      },
      "recorded_metric_max_abs_difference": 8.940696716308594e-08,
      "window_frames": 15,
      "window_ms": 696.5986394557823,
      "scalar_tokens": 150,
      "codebooks": 10,
      "seam_count": 8,
      "id": "fish-400m-morepitchloss1-step3802000--cartoon-voices--log_mel--5-10",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 10.613705899464222,
        "log_mel": 7.395697787765432,
        "log1p_mel": 22.813732837232358
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "10-20",
      "audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-10.wav",
      "baseline": {
        "log_mel": 0.45535868406295776,
        "log1p_mel": 0.004118511453270912,
        "mr_stft": 1.0745127201080322
      },
      "treatment": {
        "log_mel": 0.5161362886428833,
        "log1p_mel": 0.00519574573263526,
        "mr_stft": 1.2278010845184326
      },
      "decoder_drift": {
        "log_mel": 0.2545762062072754,
        "log1p_mel": 0.003024084260687232,
        "mr_stft": 0.780166506767273
      },
      "relative_percent": {
        "mr_stft": 14.265849211630432,
        "log_mel": 13.347193477817253,
        "log1p_mel": 26.155913163937193
      },
      "recorded_metric_max_abs_difference": 4.76837158203125e-07,
      "window_frames": 10,
      "window_ms": 464.39909297052156,
      "scalar_tokens": 100,
      "codebooks": 10,
      "seam_count": 12,
      "id": "fish-400m-morepitchloss1-step3802000--cartoon-voices--log_mel--10-20",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 14.265804834570094,
        "log_mel": 13.347154209007716,
        "log1p_mel": 26.155924470480652
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "20-30",
      "audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-5.wav",
      "baseline": {
        "log_mel": 0.45535868406295776,
        "log1p_mel": 0.004118511453270912,
        "mr_stft": 1.0745127201080322
      },
      "treatment": {
        "log_mel": 0.5705623626708984,
        "log1p_mel": 0.006501206662505865,
        "mr_stft": 1.3537538051605225
      },
      "decoder_drift": {
        "log_mel": 0.3587104082107544,
        "log1p_mel": 0.004704788327217102,
        "mr_stft": 1.0112167596817017
      },
      "relative_percent": {
        "mr_stft": 25.98769468493729,
        "log_mel": 25.29954575150095,
        "log1p_mel": 57.853310261954526
      },
      "recorded_metric_max_abs_difference": 9.313225746154785e-10,
      "window_frames": 5,
      "window_ms": 232.19954648526078,
      "scalar_tokens": 50,
      "codebooks": 10,
      "seam_count": 24,
      "id": "fish-400m-morepitchloss1-step3802000--cartoon-voices--log_mel--20-30",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 25.98769468493729,
        "log_mel": 25.29954575150095,
        "log1p_mel": 57.85333287504144
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "30-40",
      "audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-4.wav",
      "baseline": {
        "log_mel": 0.45535868406295776,
        "log1p_mel": 0.004118511453270912,
        "mr_stft": 1.0745127201080322
      },
      "treatment": {
        "log_mel": 0.6082643866539001,
        "log1p_mel": 0.0071660117246210575,
        "mr_stft": 1.4419114589691162
      },
      "decoder_drift": {
        "log_mel": 0.41535520553588867,
        "log1p_mel": 0.005648731254041195,
        "mr_stft": 1.1230965852737427
      },
      "relative_percent": {
        "mr_stft": 34.19212560128144,
        "log_mel": 33.579177897000804,
        "log1p_mel": 73.99518748284233
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 4,
      "window_ms": 185.75963718820861,
      "scalar_tokens": 40,
      "codebooks": 10,
      "seam_count": 30,
      "id": "fish-400m-morepitchloss1-step3802000--cartoon-voices--log_mel--30-40",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 34.192147789811614,
        "log_mel": 33.57921716581031,
        "log1p_mel": 73.99518748284233
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "40-50",
      "audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-3.wav",
      "baseline": {
        "log_mel": 0.45535868406295776,
        "log1p_mel": 0.004118511453270912,
        "mr_stft": 1.0745127201080322
      },
      "treatment": {
        "log_mel": 0.6692796349525452,
        "log1p_mel": 0.008835365995764732,
        "mr_stft": 1.5791540145874023
      },
      "decoder_drift": {
        "log_mel": 0.5031737089157104,
        "log1p_mel": 0.00759125268086791,
        "mr_stft": 1.3165324926376343
      },
      "relative_percent": {
        "mr_stft": 46.96466454381603,
        "log_mel": 46.97855962268434,
        "log1p_mel": 114.52813949922867
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 3,
      "window_ms": 139.31972789115645,
      "scalar_tokens": 30,
      "codebooks": 10,
      "seam_count": 41,
      "id": "fish-400m-morepitchloss1-step3802000--cartoon-voices--log_mel--40-50",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 46.96464235528586,
        "log_mel": 46.978572712287516,
        "log1p_mel": 114.52813949922867
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "over-50",
      "audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-2.wav",
      "baseline": {
        "log_mel": 0.45535868406295776,
        "log1p_mel": 0.004118511453270912,
        "mr_stft": 1.0745127201080322
      },
      "treatment": {
        "log_mel": 0.705790102481842,
        "log1p_mel": 0.009349692612886429,
        "mr_stft": 1.6179420948028564
      },
      "decoder_drift": {
        "log_mel": 0.5769746899604797,
        "log1p_mel": 0.0083378991112113,
        "mr_stft": 1.4092063903808594
      },
      "relative_percent": {
        "mr_stft": 50.57449432894452,
        "log_mel": 54.99651751107479,
        "log1p_mel": 127.01630720149933
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 2,
      "window_ms": 92.87981859410431,
      "scalar_tokens": 20,
      "codebooks": 10,
      "seam_count": 61,
      "id": "fish-400m-morepitchloss1-step3802000--cartoon-voices--log_mel--over-50",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 50.57451651747469,
        "log_mel": 54.996491331868434,
        "log1p_mel": 127.01630720149933
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "0-5",
      "audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-40.wav",
      "baseline": {
        "log_mel": 0.45535868406295776,
        "log1p_mel": 0.004118511453270912,
        "mr_stft": 1.0745127201080322
      },
      "treatment": {
        "log_mel": 0.4631478488445282,
        "log1p_mel": 0.0042311884462833405,
        "mr_stft": 1.0946147441864014
      },
      "decoder_drift": {
        "log_mel": 0.07864557951688766,
        "log1p_mel": 0.0008301229681819677,
        "mr_stft": 0.2816340923309326
      },
      "relative_percent": {
        "mr_stft": 1.870803732909554,
        "log_mel": 1.7105558879587646,
        "log1p_mel": 2.7358669337423036
      },
      "recorded_metric_max_abs_difference": 2.9802322387695312e-08,
      "window_frames": 40,
      "window_ms": 1857.5963718820863,
      "scalar_tokens": 400,
      "codebooks": 10,
      "seam_count": 3,
      "id": "fish-400m-morepitchloss1-step3802000--cartoon-voices--log1p_mel--0-5",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 1.870803732909554,
        "log_mel": 1.710549343157175,
        "log1p_mel": 2.7358556271988466
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "5-10",
      "audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-36.wav",
      "baseline": {
        "log_mel": 0.45535868406295776,
        "log1p_mel": 0.004118511453270912,
        "mr_stft": 1.0745127201080322
      },
      "treatment": {
        "log_mel": 0.47029584646224976,
        "log1p_mel": 0.004454261623322964,
        "mr_stft": 1.125699520111084
      },
      "decoder_drift": {
        "log_mel": 0.10058180242776871,
        "log1p_mel": 0.0012957011349499226,
        "mr_stft": 0.3778277337551117
      },
      "relative_percent": {
        "mr_stft": 4.7637221081854975,
        "log_mel": 3.280306914543618,
        "log1p_mel": 8.152221351367128
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 36,
      "window_ms": 1671.8367346938776,
      "scalar_tokens": 360,
      "codebooks": 10,
      "seam_count": 3,
      "id": "fish-400m-morepitchloss1-step3802000--cartoon-voices--log1p_mel--5-10",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 4.763744296715666,
        "log_mel": 3.2803200041467973,
        "log1p_mel": 8.152232657910584
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "10-20",
      "audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-14.wav",
      "baseline": {
        "log_mel": 0.45535868406295776,
        "log1p_mel": 0.004118511453270912,
        "mr_stft": 1.0745127201080322
      },
      "treatment": {
        "log_mel": 0.4879765510559082,
        "log1p_mel": 0.004733889363706112,
        "mr_stft": 1.1985557079315186
      },
      "decoder_drift": {
        "log_mel": 0.1767263561487198,
        "log1p_mel": 0.0021881544962525368,
        "mr_stft": 0.6339682936668396
      },
      "relative_percent": {
        "mr_stft": 11.544115346630335,
        "log_mel": 7.163115173716705,
        "log1p_mel": 14.941755472027829
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 14,
      "window_ms": 650.1587301587301,
      "scalar_tokens": 140,
      "codebooks": 10,
      "seam_count": 8,
      "id": "fish-400m-morepitchloss1-step3802000--cartoon-voices--log1p_mel--10-20",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 11.544093158100166,
        "log_mel": 7.163095539311937,
        "log1p_mel": 14.941755472027829
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "20-30",
      "audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-12.wav",
      "baseline": {
        "log_mel": 0.45535868406295776,
        "log1p_mel": 0.004118511453270912,
        "mr_stft": 1.0745127201080322
      },
      "treatment": {
        "log_mel": 0.5101870894432068,
        "log1p_mel": 0.00516648730263114,
        "mr_stft": 1.2205626964569092
      },
      "decoder_drift": {
        "log_mel": 0.226170152425766,
        "log1p_mel": 0.002889861585572362,
        "mr_stft": 0.7093726396560669
      },
      "relative_percent": {
        "mr_stft": 13.592205435612993,
        "log_mel": 12.040707095127768,
        "log1p_mel": 25.445500425352165
      },
      "recorded_metric_max_abs_difference": 4.76837158203125e-07,
      "window_frames": 12,
      "window_ms": 557.2789115646258,
      "scalar_tokens": 120,
      "codebooks": 10,
      "seam_count": 10,
      "id": "fish-400m-morepitchloss1-step3802000--cartoon-voices--log1p_mel--20-30",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 13.59224981267333,
        "log_mel": 12.040746363937306,
        "log1p_mel": 25.445500425352165
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "30-40",
      "audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-9.wav",
      "baseline": {
        "log_mel": 0.45535868406295776,
        "log1p_mel": 0.004118511453270912,
        "mr_stft": 1.0745127201080322
      },
      "treatment": {
        "log_mel": 0.5158568024635315,
        "log1p_mel": 0.005552873481065035,
        "mr_stft": 1.2778257131576538
      },
      "decoder_drift": {
        "log_mel": 0.24887490272521973,
        "log1p_mel": 0.0033205461222678423,
        "mr_stft": 0.8220294713973999
      },
      "relative_percent": {
        "mr_stft": 18.92141332949324,
        "log_mel": 13.285816328520772,
        "log1p_mel": 34.82719531239753
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 9,
      "window_ms": 417.9591836734694,
      "scalar_tokens": 90,
      "codebooks": 10,
      "seam_count": 13,
      "id": "fish-400m-morepitchloss1-step3802000--cartoon-voices--log1p_mel--30-40",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 18.92143551802341,
        "log_mel": 13.285816328520772,
        "log1p_mel": 34.82719531239753
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "40-50",
      "audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-7.wav",
      "baseline": {
        "log_mel": 0.45535868406295776,
        "log1p_mel": 0.004118511453270912,
        "mr_stft": 1.0745127201080322
      },
      "treatment": {
        "log_mel": 0.5430534482002258,
        "log1p_mel": 0.006032158620655537,
        "mr_stft": 1.3280375003814697
      },
      "decoder_drift": {
        "log_mel": 0.3008230924606323,
        "log1p_mel": 0.004115909803658724,
        "mr_stft": 0.9213342666625977
      },
      "relative_percent": {
        "mr_stft": 23.59439544354096,
        "log_mel": 19.25839282448898,
        "log1p_mel": 46.46453431287196
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-07,
      "window_frames": 7,
      "window_ms": 325.07936507936506,
      "scalar_tokens": 70,
      "codebooks": 10,
      "seam_count": 17,
      "id": "fish-400m-morepitchloss1-step3802000--cartoon-voices--log1p_mel--40-50",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 23.594406537806044,
        "log_mel": 19.258405914092158,
        "log1p_mel": 46.46453431287196
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "over-50",
      "audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-4.wav",
      "baseline": {
        "log_mel": 0.45535868406295776,
        "log1p_mel": 0.004118511453270912,
        "mr_stft": 1.0745127201080322
      },
      "treatment": {
        "log_mel": 0.6082643866539001,
        "log1p_mel": 0.0071660117246210575,
        "mr_stft": 1.4419114589691162
      },
      "decoder_drift": {
        "log_mel": 0.41535520553588867,
        "log1p_mel": 0.005648731254041195,
        "mr_stft": 1.1230965852737427
      },
      "relative_percent": {
        "mr_stft": 34.19212560128144,
        "log_mel": 33.579177897000804,
        "log1p_mel": 73.99518748284233
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 4,
      "window_ms": 185.75963718820861,
      "scalar_tokens": 40,
      "codebooks": 10,
      "seam_count": 30,
      "id": "fish-400m-morepitchloss1-step3802000--cartoon-voices--log1p_mel--over-50",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 34.192147789811614,
        "log_mel": 33.57921716581031,
        "log1p_mel": 73.99518748284233
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "0-5",
      "audio": "listening-audio/cartoon-voices-fish-dac-s2-window-118.wav",
      "baseline": {
        "log_mel": 0.3437022268772125,
        "log1p_mel": 0.0027478106785565615,
        "mr_stft": 0.8997645378112793
      },
      "treatment": {
        "log_mel": 0.3564625382423401,
        "log1p_mel": 0.002993264002725482,
        "mr_stft": 0.9294652342796326
      },
      "decoder_drift": {
        "log_mel": 0.02667841874063015,
        "log1p_mel": 0.000348048924934119,
        "mr_stft": 0.1327459216117859
      },
      "relative_percent": {
        "mr_stft": 3.3009409929181777,
        "log_mel": 3.7126065434793354,
        "log1p_mel": 8.93268688721518
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 118,
      "window_ms": 5479.909297052154,
      "scalar_tokens": 1180,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--cartoon-voices--mr_stft--0-5",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 3.3009409929181777,
        "log_mel": 3.7126065434793354,
        "log1p_mel": 8.93268688721518
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "5-10",
      "audio": "listening-audio/cartoon-voices-fish-dac-s2-window-117.wav",
      "baseline": {
        "log_mel": 0.3437022268772125,
        "log1p_mel": 0.0027478106785565615,
        "mr_stft": 0.8997645378112793
      },
      "treatment": {
        "log_mel": 0.35766270756721497,
        "log1p_mel": 0.0031525169033557177,
        "mr_stft": 0.9557125568389893
      },
      "decoder_drift": {
        "log_mel": 0.032835401594638824,
        "log1p_mel": 0.0005498357932083309,
        "mr_stft": 0.18377912044525146
      },
      "relative_percent": {
        "mr_stft": 6.2180733599266125,
        "log_mel": 4.061795239688637,
        "log1p_mel": 14.728315453368502
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 117,
      "window_ms": 5433.469387755103,
      "scalar_tokens": 1170,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--cartoon-voices--mr_stft--5-10",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 6.2180733599266125,
        "log_mel": 4.061795239688637,
        "log1p_mel": 14.728315453368502
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "10-20",
      "audio": "listening-audio/cartoon-voices-fish-dac-s2-window-78.wav",
      "baseline": {
        "log_mel": 0.3437022268772125,
        "log1p_mel": 0.0027478106785565615,
        "mr_stft": 0.8997645378112793
      },
      "treatment": {
        "log_mel": 0.3768247067928314,
        "log1p_mel": 0.0037109116092324257,
        "mr_stft": 1.0400002002716064
      },
      "decoder_drift": {
        "log_mel": 0.11456084996461868,
        "log1p_mel": 0.0016049162950366735,
        "mr_stft": 0.404829204082489
      },
      "relative_percent": {
        "mr_stft": 15.58581790758915,
        "log_mel": 9.636969831868992,
        "log1p_mel": 35.0497557270498
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 78,
      "window_ms": 3622.312925170068,
      "scalar_tokens": 780,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--cartoon-voices--mr_stft--10-20",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 15.58581790758915,
        "log_mel": 9.636969831868992,
        "log1p_mel": 35.0497557270498
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "20-30",
      "audio": "listening-audio/cartoon-voices-fish-dac-s2-window-88.wav",
      "baseline": {
        "log_mel": 0.3437022268772125,
        "log1p_mel": 0.0027478106785565615,
        "mr_stft": 0.8997645378112793
      },
      "treatment": {
        "log_mel": 0.3863738477230072,
        "log1p_mel": 0.004519135225564241,
        "mr_stft": 1.128000259399414
      },
      "decoder_drift": {
        "log_mel": 0.11575476825237274,
        "log1p_mel": 0.002466872800141573,
        "mr_stft": 0.48441872000694275
      },
      "relative_percent": {
        "mr_stft": 25.366161034011103,
        "log_mel": 12.415287859347824,
        "log1p_mel": 64.46312188939325
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 88,
      "window_ms": 4086.71201814059,
      "scalar_tokens": 880,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--cartoon-voices--mr_stft--20-30",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 25.366161034011103,
        "log_mel": 12.415287859347824,
        "log1p_mel": 64.46312188939325
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "30-40",
      "audio": "listening-audio/cartoon-voices-fish-dac-s2-window-71.wav",
      "baseline": {
        "log_mel": 0.3437022268772125,
        "log1p_mel": 0.0027478106785565615,
        "mr_stft": 0.8997645378112793
      },
      "treatment": {
        "log_mel": 0.38752496242523193,
        "log1p_mel": 0.0049530575051903725,
        "mr_stft": 1.2168292999267578
      },
      "decoder_drift": {
        "log_mel": 0.1440238207578659,
        "log1p_mel": 0.0030335006304085255,
        "mr_stft": 0.6109956502914429
      },
      "relative_percent": {
        "mr_stft": 35.2386373091291,
        "log_mel": 12.750204136348264,
        "log1p_mel": 80.25468580652864
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 71,
      "window_ms": 3297.233560090703,
      "scalar_tokens": 710,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--cartoon-voices--mr_stft--30-40",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 35.2386373091291,
        "log_mel": 12.750204136348264,
        "log1p_mel": 80.25468580652864
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "40-50",
      "audio": "listening-audio/cartoon-voices-fish-dac-s2-window-31.wav",
      "baseline": {
        "log_mel": 0.3437022268772125,
        "log1p_mel": 0.0027478106785565615,
        "mr_stft": 0.8997645378112793
      },
      "treatment": {
        "log_mel": 0.4366110563278198,
        "log1p_mel": 0.006303395610302687,
        "mr_stft": 1.3047292232513428
      },
      "decoder_drift": {
        "log_mel": 0.24959714710712433,
        "log1p_mel": 0.005052767228335142,
        "mr_stft": 0.8344498872756958
      },
      "relative_percent": {
        "mr_stft": 45.0078513235429,
        "log_mel": 27.031779891201847,
        "log1p_mel": 129.39701266514808
      },
      "recorded_metric_max_abs_difference": 1.3113021850585938e-06,
      "window_frames": 31,
      "window_ms": 1439.6371882086166,
      "scalar_tokens": 310,
      "codebooks": 10,
      "seam_count": 3,
      "id": "fish-dac-s2--cartoon-voices--mr_stft--40-50",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 45.00799706191443,
        "log_mel": 27.031788562173432,
        "log1p_mel": 129.39701266514808
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "over-50",
      "audio": "listening-audio/cartoon-voices-fish-dac-s2-window-34.wav",
      "baseline": {
        "log_mel": 0.3437022268772125,
        "log1p_mel": 0.0027478106785565615,
        "mr_stft": 0.8997645378112793
      },
      "treatment": {
        "log_mel": 0.47349977493286133,
        "log1p_mel": 0.008339455351233482,
        "mr_stft": 1.5626857280731201
      },
      "decoder_drift": {
        "log_mel": 0.3184358477592468,
        "log1p_mel": 0.007270817179232836,
        "mr_stft": 1.1147527694702148
      },
      "relative_percent": {
        "mr_stft": 73.67718579734523,
        "log_mel": 37.76453508461523,
        "log1p_mel": 203.49453899110105
      },
      "recorded_metric_max_abs_difference": 2.9802322387695312e-08,
      "window_frames": 34,
      "window_ms": 1578.9569160997733,
      "scalar_tokens": 340,
      "codebooks": 10,
      "seam_count": 3,
      "id": "fish-dac-s2--cartoon-voices--mr_stft--over-50",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 73.67718579734523,
        "log_mel": 37.76452641364367,
        "log1p_mel": 203.4945728843595
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "0-5",
      "audio": "listening-audio/cartoon-voices-fish-dac-s2-window-121.wav",
      "baseline": {
        "log_mel": 0.3437022268772125,
        "log1p_mel": 0.0027478106785565615,
        "mr_stft": 0.8997645378112793
      },
      "treatment": {
        "log_mel": 0.35386818647384644,
        "log1p_mel": 0.0027816100046038628,
        "mr_stft": 0.9077519178390503
      },
      "decoder_drift": {
        "log_mel": 0.016298383474349976,
        "log1p_mel": 5.374338070396334e-05,
        "mr_stft": 0.0416983962059021
      },
      "relative_percent": {
        "mr_stft": 0.8877189189074652,
        "log_mel": 2.957781126121617,
        "log1p_mel": 1.2300456618451028
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 121,
      "window_ms": 5619.22902494331,
      "scalar_tokens": 1210,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--cartoon-voices--log_mel--0-5",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 0.8877189189074652,
        "log_mel": 2.957781126121617,
        "log1p_mel": 1.2300456618451028
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "5-10",
      "audio": "listening-audio/cartoon-voices-fish-dac-s2-window-110.wav",
      "baseline": {
        "log_mel": 0.3437022268772125,
        "log1p_mel": 0.0027478106785565615,
        "mr_stft": 0.8997645378112793
      },
      "treatment": {
        "log_mel": 0.3696087896823883,
        "log1p_mel": 0.0039429557509720325,
        "mr_stft": 1.1300413608551025
      },
      "decoder_drift": {
        "log_mel": 0.06216747313737869,
        "log1p_mel": 0.0015248670242726803,
        "mr_stft": 0.42821070551872253
      },
      "relative_percent": {
        "mr_stft": 25.59300943377727,
        "log_mel": 7.53750216882676,
        "log1p_mel": 43.49444747930329
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 110,
      "window_ms": 5108.390022675737,
      "scalar_tokens": 1100,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--cartoon-voices--log_mel--5-10",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 25.59300943377727,
        "log_mel": 7.53750216882676,
        "log1p_mel": 43.49444747930329
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "10-20",
      "audio": "listening-audio/cartoon-voices-fish-dac-s2-window-70.wav",
      "baseline": {
        "log_mel": 0.3437022268772125,
        "log1p_mel": 0.0027478106785565615,
        "mr_stft": 0.8997645378112793
      },
      "treatment": {
        "log_mel": 0.39568617939949036,
        "log1p_mel": 0.004600330255925655,
        "mr_stft": 1.1654642820358276
      },
      "decoder_drift": {
        "log_mel": 0.16282883286476135,
        "log1p_mel": 0.002695328090339899,
        "mr_stft": 0.5747238397598267
      },
      "relative_percent": {
        "mr_stft": 29.529919557718486,
        "log_mel": 15.124706346708972,
        "log1p_mel": 67.41802089299075
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 70,
      "window_ms": 3250.7936507936506,
      "scalar_tokens": 700,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--cartoon-voices--log_mel--10-20",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 29.529919557718486,
        "log_mel": 15.124706346708972,
        "log1p_mel": 67.41802089299075
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "20-30",
      "audio": "listening-audio/cartoon-voices-fish-dac-s2-window-52.wav",
      "baseline": {
        "log_mel": 0.3437022268772125,
        "log1p_mel": 0.0027478106785565615,
        "mr_stft": 0.8997645378112793
      },
      "treatment": {
        "log_mel": 0.4297525882720947,
        "log1p_mel": 0.006634743884205818,
        "mr_stft": 1.3280307054519653
      },
      "decoder_drift": {
        "log_mel": 0.232290118932724,
        "log1p_mel": 0.0052491347305476665,
        "mr_stft": 0.7991837859153748
      },
      "relative_percent": {
        "mr_stft": 47.5975824388972,
        "log_mel": 25.036311861204098,
        "log1p_mel": 141.45564088465812
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 52,
      "window_ms": 2414.8752834467123,
      "scalar_tokens": 520,
      "codebooks": 10,
      "seam_count": 2,
      "id": "fish-dac-s2--cartoon-voices--log_mel--20-30",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 47.59760893678293,
        "log_mel": 25.03627717731782,
        "log1p_mel": 141.45564088465812
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "30-40",
      "audio": "listening-audio/cartoon-voices-fish-dac-s2-window-36.wav",
      "baseline": {
        "log_mel": 0.3437022268772125,
        "log1p_mel": 0.0027478106785565615,
        "mr_stft": 0.8997645378112793
      },
      "treatment": {
        "log_mel": 0.4638552665710449,
        "log1p_mel": 0.007180857006460428,
        "mr_stft": 1.4334895610809326
      },
      "decoder_drift": {
        "log_mel": 0.3009440004825592,
        "log1p_mel": 0.006014987826347351,
        "mr_stft": 0.9800841212272644
      },
      "relative_percent": {
        "mr_stft": 59.318299492883455,
        "log_mel": 34.958469948103364,
        "log1p_mel": 161.3301222860291
      },
      "recorded_metric_max_abs_difference": 3.5762786865234375e-07,
      "window_frames": 36,
      "window_ms": 1671.8367346938776,
      "scalar_tokens": 360,
      "codebooks": 10,
      "seam_count": 3,
      "id": "fish-dac-s2--cartoon-voices--log_mel--30-40",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 59.31825974605487,
        "log_mel": 34.95839190935921,
        "log1p_mel": 161.3301222860291
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "40-50",
      "audio": "listening-audio/cartoon-voices-fish-dac-s2-window-20.wav",
      "baseline": {
        "log_mel": 0.3437022268772125,
        "log1p_mel": 0.0027478106785565615,
        "mr_stft": 0.8997645378112793
      },
      "treatment": {
        "log_mel": 0.5029767751693726,
        "log1p_mel": 0.008916974999010563,
        "mr_stft": 1.6046950817108154
      },
      "decoder_drift": {
        "log_mel": 0.38990193605422974,
        "log1p_mel": 0.008278998546302319,
        "mr_stft": 1.2452908754348755
      },
      "relative_percent": {
        "mr_stft": 78.34611326362268,
        "log_mel": 46.34085433175292,
        "log1p_mel": 224.51198580008045
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-07,
      "window_frames": 20,
      "window_ms": 928.7981859410431,
      "scalar_tokens": 200,
      "codebooks": 10,
      "seam_count": 6,
      "id": "fish-dac-s2--cartoon-voices--log_mel--40-50",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 78.34612651256556,
        "log_mel": 46.34088901563922,
        "log1p_mel": 224.51201969333886
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "over-50",
      "audio": "listening-audio/cartoon-voices-fish-dac-s2-window-8.wav",
      "baseline": {
        "log_mel": 0.3437022268772125,
        "log1p_mel": 0.0027478106785565615,
        "mr_stft": 0.8997645378112793
      },
      "treatment": {
        "log_mel": 0.5993815660476685,
        "log1p_mel": 0.011280094273388386,
        "mr_stft": 1.7889409065246582
      },
      "decoder_drift": {
        "log_mel": 0.5205504298210144,
        "log1p_mel": 0.010858474299311638,
        "mr_stft": 1.5434869527816772
      },
      "relative_percent": {
        "mr_stft": 98.8232288945665,
        "log_mel": 74.38978254330522,
        "log1p_mel": 310.5120618904456
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 8,
      "window_ms": 371.51927437641723,
      "scalar_tokens": 80,
      "codebooks": 10,
      "seam_count": 15,
      "id": "fish-dac-s2--cartoon-voices--log_mel--over-50",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 98.82325539245225,
        "log_mel": 74.38983456913466,
        "log1p_mel": 310.5120618904456
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "0-5",
      "audio": "listening-audio/cartoon-voices-fish-dac-s2-window-120.wav",
      "baseline": {
        "log_mel": 0.3437022268772125,
        "log1p_mel": 0.0027478106785565615,
        "mr_stft": 0.8997645378112793
      },
      "treatment": {
        "log_mel": 0.34659576416015625,
        "log1p_mel": 0.0027916408143937588,
        "mr_stft": 0.9046617746353149
      },
      "decoder_drift": {
        "log_mel": 0.013186821714043617,
        "log1p_mel": 9.982252959161997e-05,
        "mr_stft": 0.06660713255405426
      },
      "relative_percent": {
        "mr_stft": 0.5442798219130074,
        "log_mel": 0.8418733009773138,
        "log1p_mel": 1.5950930018301612
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 120,
      "window_ms": 5572.789115646258,
      "scalar_tokens": 1200,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--cartoon-voices--log1p_mel--0-5",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 0.5442798219130074,
        "log_mel": 0.8418733009773138,
        "log1p_mel": 1.5950930018301612
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "5-10",
      "audio": "listening-audio/cartoon-voices-fish-dac-s2-window-118.wav",
      "baseline": {
        "log_mel": 0.3437022268772125,
        "log1p_mel": 0.0027478106785565615,
        "mr_stft": 0.8997645378112793
      },
      "treatment": {
        "log_mel": 0.3564625382423401,
        "log1p_mel": 0.002993264002725482,
        "mr_stft": 0.9294652342796326
      },
      "decoder_drift": {
        "log_mel": 0.02667841874063015,
        "log1p_mel": 0.000348048924934119,
        "mr_stft": 0.1327459216117859
      },
      "relative_percent": {
        "mr_stft": 3.3009409929181777,
        "log_mel": 3.7126065434793354,
        "log1p_mel": 8.93268688721518
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 118,
      "window_ms": 5479.909297052154,
      "scalar_tokens": 1180,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--cartoon-voices--log1p_mel--5-10",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 3.3009409929181777,
        "log_mel": 3.7126065434793354,
        "log1p_mel": 8.93268688721518
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "10-20",
      "audio": "listening-audio/cartoon-voices-fish-dac-s2-window-117.wav",
      "baseline": {
        "log_mel": 0.3437022268772125,
        "log1p_mel": 0.0027478106785565615,
        "mr_stft": 0.8997645378112793
      },
      "treatment": {
        "log_mel": 0.35766270756721497,
        "log1p_mel": 0.0031525169033557177,
        "mr_stft": 0.9557125568389893
      },
      "decoder_drift": {
        "log_mel": 0.032835401594638824,
        "log1p_mel": 0.0005498357932083309,
        "mr_stft": 0.18377912044525146
      },
      "relative_percent": {
        "mr_stft": 6.2180733599266125,
        "log_mel": 4.061795239688637,
        "log1p_mel": 14.728315453368502
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 117,
      "window_ms": 5433.469387755103,
      "scalar_tokens": 1170,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--cartoon-voices--log1p_mel--10-20",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 6.2180733599266125,
        "log_mel": 4.061795239688637,
        "log1p_mel": 14.728315453368502
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "20-30",
      "audio": "listening-audio/cartoon-voices-fish-dac-s2-window-62.wav",
      "baseline": {
        "log_mel": 0.3437022268772125,
        "log1p_mel": 0.0027478106785565615,
        "mr_stft": 0.8997645378112793
      },
      "treatment": {
        "log_mel": 0.36653637886047363,
        "log1p_mel": 0.0034516453742980957,
        "mr_stft": 0.9799065589904785
      },
      "decoder_drift": {
        "log_mel": 0.09710627049207687,
        "log1p_mel": 0.001360422233119607,
        "mr_stft": 0.31226593255996704
      },
      "relative_percent": {
        "mr_stft": 8.90699931052501,
        "log_mel": 6.643585696469345,
        "log1p_mel": 25.61438097734092
      },
      "recorded_metric_max_abs_difference": 4.76837158203125e-07,
      "window_frames": 62,
      "window_ms": 2879.2743764172333,
      "scalar_tokens": 620,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--cartoon-voices--log1p_mel--20-30",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 8.907052306296471,
        "log_mel": 6.643542341611486,
        "log1p_mel": 25.61436403071169
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "30-40",
      "audio": "listening-audio/cartoon-voices-fish-dac-s2-window-78.wav",
      "baseline": {
        "log_mel": 0.3437022268772125,
        "log1p_mel": 0.0027478106785565615,
        "mr_stft": 0.8997645378112793
      },
      "treatment": {
        "log_mel": 0.3768247067928314,
        "log1p_mel": 0.0037109116092324257,
        "mr_stft": 1.0400002002716064
      },
      "decoder_drift": {
        "log_mel": 0.11456084996461868,
        "log1p_mel": 0.0016049162950366735,
        "mr_stft": 0.404829204082489
      },
      "relative_percent": {
        "mr_stft": 15.58581790758915,
        "log_mel": 9.636969831868992,
        "log1p_mel": 35.0497557270498
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 78,
      "window_ms": 3622.312925170068,
      "scalar_tokens": 780,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--cartoon-voices--log1p_mel--30-40",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 15.58581790758915,
        "log_mel": 9.636969831868992,
        "log1p_mel": 35.0497557270498
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "40-50",
      "audio": "listening-audio/cartoon-voices-fish-dac-s2-window-98.wav",
      "baseline": {
        "log_mel": 0.3437022268772125,
        "log1p_mel": 0.0027478106785565615,
        "mr_stft": 0.8997645378112793
      },
      "treatment": {
        "log_mel": 0.36655765771865845,
        "log1p_mel": 0.003972452599555254,
        "mr_stft": 1.0873693227767944
      },
      "decoder_drift": {
        "log_mel": 0.07301639020442963,
        "log1p_mel": 0.0016746091423556209,
        "mr_stft": 0.3954096734523773
      },
      "relative_percent": {
        "mr_stft": 20.850431094103005,
        "log_mel": 6.649776770172333,
        "log1p_mel": 44.56791476048863
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 98,
      "window_ms": 4551.111111111111,
      "scalar_tokens": 980,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--cartoon-voices--log1p_mel--40-50",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 20.850431094103005,
        "log_mel": 6.649776770172333,
        "log1p_mel": 44.56791476048863
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "over-50",
      "audio": "listening-audio/cartoon-voices-fish-dac-s2-window-60.wav",
      "baseline": {
        "log_mel": 0.3437022268772125,
        "log1p_mel": 0.0027478106785565615,
        "mr_stft": 0.8997645378112793
      },
      "treatment": {
        "log_mel": 0.38748699426651,
        "log1p_mel": 0.004807527642697096,
        "mr_stft": 1.2827085256576538
      },
      "decoder_drift": {
        "log_mel": 0.1455337107181549,
        "log1p_mel": 0.0029924476984888315,
        "mr_stft": 0.7020615935325623
      },
      "relative_percent": {
        "mr_stft": 42.56046685035002,
        "log_mel": 12.739157318564477,
        "log1p_mel": 74.95847440342993
      },
      "recorded_metric_max_abs_difference": 5.960464477539062e-07,
      "window_frames": 60,
      "window_ms": 2786.394557823129,
      "scalar_tokens": 600,
      "codebooks": 10,
      "seam_count": 2,
      "id": "fish-dac-s2--cartoon-voices--log1p_mel--over-50",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 42.560533095064336,
        "log_mel": 12.739183331479186,
        "log1p_mel": 74.95849135005915
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "0-5",
      "audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-46.wav",
      "baseline": {
        "log_mel": 0.35760822892189026,
        "log1p_mel": 0.0033701565116643906,
        "mr_stft": 1.0018222332000732
      },
      "treatment": {
        "log_mel": 0.38111546635627747,
        "log1p_mel": 0.0034674652852118015,
        "mr_stft": 1.0257624387741089
      },
      "decoder_drift": {
        "log_mel": 0.08966473489999771,
        "log1p_mel": 0.000629258225671947,
        "mr_stft": 0.2660871744155884
      },
      "relative_percent": {
        "mr_stft": 2.3896660286291116,
        "log_mel": 6.573460992566171,
        "log1p_mel": 2.887366601836372
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 46,
      "window_ms": 3680.0,
      "scalar_tokens": 736,
      "codebooks": 16,
      "seam_count": 1,
      "id": "qwen3-tts-tokenizer-12hz--cartoon-voices--mr_stft--0-5",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 2.3896660286291116,
        "log_mel": 6.573460992566171,
        "log1p_mel": 2.887366601836372
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "5-10",
      "audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-32.wav",
      "baseline": {
        "log_mel": 0.35760822892189026,
        "log1p_mel": 0.0033701565116643906,
        "mr_stft": 1.0018222332000732
      },
      "treatment": {
        "log_mel": 0.38344472646713257,
        "log1p_mel": 0.003875700756907463,
        "mr_stft": 1.0786893367767334
      },
      "decoder_drift": {
        "log_mel": 0.10885705053806305,
        "log1p_mel": 0.001262948731891811,
        "mr_stft": 0.39866000413894653
      },
      "relative_percent": {
        "mr_stft": 7.672728856408706,
        "log_mel": 7.224805095546505,
        "log1p_mel": 15.000616247148812
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 32,
      "window_ms": 2560.0,
      "scalar_tokens": 512,
      "codebooks": 16,
      "seam_count": 2,
      "id": "qwen3-tts-tokenizer-12hz--cartoon-voices--mr_stft--5-10",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 7.672728856408706,
        "log_mel": 7.224805095546505,
        "log1p_mel": 15.000616247148812
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "10-20",
      "audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-60.wav",
      "baseline": {
        "log_mel": 0.35760822892189026,
        "log1p_mel": 0.0033701565116643906,
        "mr_stft": 1.0018222332000732
      },
      "treatment": {
        "log_mel": 0.3937692642211914,
        "log1p_mel": 0.004459210205823183,
        "mr_stft": 1.1436285972595215
      },
      "decoder_drift": {
        "log_mel": 0.07585228979587555,
        "log1p_mel": 0.0015247149858623743,
        "mr_stft": 0.3836081922054291
      },
      "relative_percent": {
        "mr_stft": 14.154842981122794,
        "log_mel": 10.111913645924385,
        "log1p_mel": 32.314632581290745
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 60,
      "window_ms": 4800.0,
      "scalar_tokens": 960,
      "codebooks": 16,
      "seam_count": 1,
      "id": "qwen3-tts-tokenizer-12hz--cartoon-voices--mr_stft--10-20",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 14.154842981122794,
        "log_mel": 10.111913645924385,
        "log1p_mel": 32.314632581290745
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "20-30",
      "audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-58.wav",
      "baseline": {
        "log_mel": 0.35760822892189026,
        "log1p_mel": 0.0033701565116643906,
        "mr_stft": 1.0018222332000732
      },
      "treatment": {
        "log_mel": 0.4129151403903961,
        "log1p_mel": 0.005124110262840986,
        "mr_stft": 1.2270891666412354
      },
      "decoder_drift": {
        "log_mel": 0.10367392003536224,
        "log1p_mel": 0.0022877766750752926,
        "mr_stft": 0.5205024480819702
      },
      "relative_percent": {
        "mr_stft": 22.485719120207847,
        "log_mel": 15.465782662564553,
        "log1p_mel": 52.043688330379226
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 58,
      "window_ms": 4640.0,
      "scalar_tokens": 928,
      "codebooks": 16,
      "seam_count": 1,
      "id": "qwen3-tts-tokenizer-12hz--cartoon-voices--mr_stft--20-30",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 22.485719120207847,
        "log_mel": 15.465782662564553,
        "log1p_mel": 52.043688330379226
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "30-40",
      "audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-10.wav",
      "baseline": {
        "log_mel": 0.35760822892189026,
        "log1p_mel": 0.0033701565116643906,
        "mr_stft": 1.0018222332000732
      },
      "treatment": {
        "log_mel": 0.48399704694747925,
        "log1p_mel": 0.006297344341874123,
        "mr_stft": 1.3555781841278076
      },
      "decoder_drift": {
        "log_mel": 0.3165271282196045,
        "log1p_mel": 0.0049058496952056885,
        "mr_stft": 0.975512683391571
      },
      "relative_percent": {
        "mr_stft": 35.31124976112263,
        "log_mel": 35.342815909640365,
        "log1p_mel": 86.8561391756879
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-07,
      "window_frames": 10,
      "window_ms": 800.0,
      "scalar_tokens": 160,
      "codebooks": 16,
      "seam_count": 7,
      "id": "qwen3-tts-tokenizer-12hz--cartoon-voices--mr_stft--30-40",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 35.3112616603684,
        "log_mel": 35.342832577222374,
        "log1p_mel": 86.85611154128662
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "40-50",
      "audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-14.wav",
      "baseline": {
        "log_mel": 0.35760822892189026,
        "log1p_mel": 0.0033701565116643906,
        "mr_stft": 1.0018222332000732
      },
      "treatment": {
        "log_mel": 0.49788182973861694,
        "log1p_mel": 0.007388490252196789,
        "mr_stft": 1.4296282529830933
      },
      "decoder_drift": {
        "log_mel": 0.3251117467880249,
        "log1p_mel": 0.005922285839915276,
        "mr_stft": 1.0469763278961182
      },
      "relative_percent": {
        "mr_stft": 42.70278754110892,
        "log_mel": 39.22549580014436,
        "log1p_mel": 119.23285243948203
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-07,
      "window_frames": 14,
      "window_ms": 1120.0,
      "scalar_tokens": 224,
      "codebooks": 16,
      "seam_count": 5,
      "id": "qwen3-tts-tokenizer-12hz--cartoon-voices--mr_stft--40-50",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 42.70279944035467,
        "log_mel": 39.22549580014436,
        "log1p_mel": 119.23285243948203
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "mr_stft",
      "band_id": "over-50",
      "audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-1.wav",
      "baseline": {
        "log_mel": 0.35760822892189026,
        "log1p_mel": 0.0033701565116643906,
        "mr_stft": 1.0018222332000732
      },
      "treatment": {
        "log_mel": 0.7989727258682251,
        "log1p_mel": 0.012088381685316563,
        "mr_stft": 1.819854736328125
      },
      "decoder_drift": {
        "log_mel": 0.7650249004364014,
        "log1p_mel": 0.011803830042481422,
        "mr_stft": 1.7678649425506592
      },
      "relative_percent": {
        "mr_stft": 81.65445685059807,
        "log_mel": 123.42123621622218,
        "log1p_mel": 258.68902952956853
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 1,
      "window_ms": 80.0,
      "scalar_tokens": 16,
      "codebooks": 16,
      "seam_count": 71,
      "id": "qwen3-tts-tokenizer-12hz--cartoon-voices--mr_stft--over-50",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 81.65445685059807,
        "log_mel": 123.42123621622218,
        "log1p_mel": 258.68902952956853
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "0-5",
      "audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-64.wav",
      "baseline": {
        "log_mel": 0.35760822892189026,
        "log1p_mel": 0.0033701565116643906,
        "mr_stft": 1.0018222332000732
      },
      "treatment": {
        "log_mel": 0.3667219877243042,
        "log1p_mel": 0.0035118188243359327,
        "mr_stft": 1.0225368738174438
      },
      "decoder_drift": {
        "log_mel": 0.03334556892514229,
        "log1p_mel": 0.0003627531696110964,
        "mr_stft": 0.14271603524684906
      },
      "relative_percent": {
        "mr_stft": 2.067696237006311,
        "log_mel": 2.5485316235283273,
        "log1p_mel": 4.203434237586223
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 64,
      "window_ms": 5120.0,
      "scalar_tokens": 1024,
      "codebooks": 16,
      "seam_count": 1,
      "id": "qwen3-tts-tokenizer-12hz--cartoon-voices--log_mel--0-5",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 2.067696237006311,
        "log_mel": 2.5485316235283273,
        "log1p_mel": 4.203434237586223
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "5-10",
      "audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-40.wav",
      "baseline": {
        "log_mel": 0.35760822892189026,
        "log1p_mel": 0.0033701565116643906,
        "mr_stft": 1.0018222332000732
      },
      "treatment": {
        "log_mel": 0.3848930895328522,
        "log1p_mel": 0.0035349244717508554,
        "mr_stft": 1.030065894126892
      },
      "decoder_drift": {
        "log_mel": 0.10525720566511154,
        "log1p_mel": 0.0007422871422022581,
        "mr_stft": 0.28558212518692017
      },
      "relative_percent": {
        "mr_stft": 2.8192288003632493,
        "log_mel": 7.62981900422699,
        "log1p_mel": 4.889029916450149
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 40,
      "window_ms": 3200.0,
      "scalar_tokens": 640,
      "codebooks": 16,
      "seam_count": 1,
      "id": "qwen3-tts-tokenizer-12hz--cartoon-voices--log_mel--5-10",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 2.8192288003632493,
        "log_mel": 7.62981900422699,
        "log1p_mel": 4.889029916450149
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "10-20",
      "audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-16.wav",
      "baseline": {
        "log_mel": 0.35760822892189026,
        "log1p_mel": 0.0033701565116643906,
        "mr_stft": 1.0018222332000732
      },
      "treatment": {
        "log_mel": 0.4098943769931793,
        "log1p_mel": 0.004058558493852615,
        "mr_stft": 1.1102581024169922
      },
      "decoder_drift": {
        "log_mel": 0.19293257594108582,
        "log1p_mel": 0.002149249892681837,
        "mr_stft": 0.5906268358230591
      },
      "relative_percent": {
        "mr_stft": 10.823863318599681,
        "log_mel": 14.62106960707259,
        "log1p_mel": 20.426409865702343
      },
      "recorded_metric_max_abs_difference": 3.5762786865234375e-07,
      "window_frames": 16,
      "window_ms": 1280.0,
      "scalar_tokens": 256,
      "codebooks": 16,
      "seam_count": 4,
      "id": "qwen3-tts-tokenizer-12hz--cartoon-voices--log_mel--10-20",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 10.823899016336958,
        "log_mel": 14.621086274654594,
        "log1p_mel": 20.426409865702343
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "20-30",
      "audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-15.wav",
      "baseline": {
        "log_mel": 0.35760822892189026,
        "log1p_mel": 0.0033701565116643906,
        "mr_stft": 1.0018222332000732
      },
      "treatment": {
        "log_mel": 0.4539038836956024,
        "log1p_mel": 0.006097804289311171,
        "mr_stft": 1.329467535018921
      },
      "decoder_drift": {
        "log_mel": 0.2558121681213379,
        "log1p_mel": 0.004393856972455978,
        "mr_stft": 0.8867181539535522
      },
      "relative_percent": {
        "mr_stft": 32.70493416504301,
        "log_mel": 26.92769544594158,
        "log1p_mel": 80.9353443439842
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-07,
      "window_frames": 15,
      "window_ms": 1200.0,
      "scalar_tokens": 240,
      "codebooks": 16,
      "seam_count": 4,
      "id": "qwen3-tts-tokenizer-12hz--cartoon-voices--log_mel--20-30",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 32.704922265797265,
        "log_mel": 26.92769544594158,
        "log1p_mel": 80.93531670958289
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "30-40",
      "audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-10.wav",
      "baseline": {
        "log_mel": 0.35760822892189026,
        "log1p_mel": 0.0033701565116643906,
        "mr_stft": 1.0018222332000732
      },
      "treatment": {
        "log_mel": 0.48399704694747925,
        "log1p_mel": 0.006297344341874123,
        "mr_stft": 1.3555781841278076
      },
      "decoder_drift": {
        "log_mel": 0.3165271282196045,
        "log1p_mel": 0.0049058496952056885,
        "mr_stft": 0.975512683391571
      },
      "relative_percent": {
        "mr_stft": 35.31124976112263,
        "log_mel": 35.342815909640365,
        "log1p_mel": 86.8561391756879
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-07,
      "window_frames": 10,
      "window_ms": 800.0,
      "scalar_tokens": 160,
      "codebooks": 16,
      "seam_count": 7,
      "id": "qwen3-tts-tokenizer-12hz--cartoon-voices--log_mel--30-40",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 35.3112616603684,
        "log_mel": 35.342832577222374,
        "log1p_mel": 86.85611154128662
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "40-50",
      "audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-9.wav",
      "baseline": {
        "log_mel": 0.35760822892189026,
        "log1p_mel": 0.0033701565116643906,
        "mr_stft": 1.0018222332000732
      },
      "treatment": {
        "log_mel": 0.5019466876983643,
        "log1p_mel": 0.006904057692736387,
        "mr_stft": 1.420159101486206
      },
      "decoder_drift": {
        "log_mel": 0.3468869924545288,
        "log1p_mel": 0.0056712571531534195,
        "mr_stft": 1.0874910354614258
      },
      "relative_percent": {
        "mr_stft": 41.75759475309897,
        "log_mel": 40.362174889437675,
        "log1p_mel": 104.85866661803014
      },
      "recorded_metric_max_abs_difference": 1.0728836059570312e-06,
      "window_frames": 9,
      "window_ms": 720.0,
      "scalar_tokens": 144,
      "codebooks": 16,
      "seam_count": 7,
      "id": "qwen3-tts-tokenizer-12hz--cartoon-voices--log_mel--40-50",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 41.75770184631078,
        "log_mel": 40.36222489218366,
        "log1p_mel": 104.85884624163853
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log_mel",
      "band_id": "over-50",
      "audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-3.wav",
      "baseline": {
        "log_mel": 0.35760822892189026,
        "log1p_mel": 0.0033701565116643906,
        "mr_stft": 1.0018222332000732
      },
      "treatment": {
        "log_mel": 0.6338729858398438,
        "log1p_mel": 0.009329487569630146,
        "mr_stft": 1.6265119314193726
      },
      "decoder_drift": {
        "log_mel": 0.5417962074279785,
        "log1p_mel": 0.008759819902479649,
        "mr_stft": 1.4338970184326172
      },
      "relative_percent": {
        "mr_stft": 62.355343844175096,
        "log_mel": 77.2534675029237,
        "log1p_mel": 176.8265372050235
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-07,
      "window_frames": 3,
      "window_ms": 240.0,
      "scalar_tokens": 48,
      "codebooks": 16,
      "seam_count": 23,
      "id": "qwen3-tts-tokenizer-12hz--cartoon-voices--log_mel--over-50",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 62.35533194492935,
        "log_mel": 77.2534675029237,
        "log1p_mel": 176.82662010822736
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "0-5",
      "audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-47.wav",
      "baseline": {
        "log_mel": 0.35760822892189026,
        "log1p_mel": 0.0033701565116643906,
        "mr_stft": 1.0018222332000732
      },
      "treatment": {
        "log_mel": 0.376310795545578,
        "log1p_mel": 0.0034512102138251066,
        "mr_stft": 1.02200186252594
      },
      "decoder_drift": {
        "log_mel": 0.07546132057905197,
        "log1p_mel": 0.0006186891696415842,
        "mr_stft": 0.23848006129264832
      },
      "relative_percent": {
        "mr_stft": 2.0142924220605263,
        "log_mel": 5.229903875554509,
        "log1p_mel": 2.405042670279034
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 47,
      "window_ms": 3760.0,
      "scalar_tokens": 752,
      "codebooks": 16,
      "seam_count": 1,
      "id": "qwen3-tts-tokenizer-12hz--cartoon-voices--log1p_mel--0-5",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 2.0142924220605263,
        "log_mel": 5.229903875554509,
        "log1p_mel": 2.405042670279034
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "5-10",
      "audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-42.wav",
      "baseline": {
        "log_mel": 0.35760822892189026,
        "log1p_mel": 0.0033701565116643906,
        "mr_stft": 1.0018222332000732
      },
      "treatment": {
        "log_mel": 0.38150081038475037,
        "log1p_mel": 0.0036363545805215836,
        "mr_stft": 1.0335429906845093
      },
      "decoder_drift": {
        "log_mel": 0.10217133164405823,
        "log1p_mel": 0.000983192352578044,
        "mr_stft": 0.33034101128578186
      },
      "relative_percent": {
        "mr_stft": 3.166306000527852,
        "log_mel": 6.681216910161991,
        "log1p_mel": 7.898685652605741
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 42,
      "window_ms": 3360.0,
      "scalar_tokens": 672,
      "codebooks": 16,
      "seam_count": 1,
      "id": "qwen3-tts-tokenizer-12hz--cartoon-voices--log1p_mel--5-10",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 3.166306000527852,
        "log_mel": 6.681216910161991,
        "log1p_mel": 7.898685652605741
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "10-20",
      "audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-32.wav",
      "baseline": {
        "log_mel": 0.35760822892189026,
        "log1p_mel": 0.0033701565116643906,
        "mr_stft": 1.0018222332000732
      },
      "treatment": {
        "log_mel": 0.38344472646713257,
        "log1p_mel": 0.003875700756907463,
        "mr_stft": 1.0786893367767334
      },
      "decoder_drift": {
        "log_mel": 0.10885705053806305,
        "log1p_mel": 0.001262948731891811,
        "mr_stft": 0.39866000413894653
      },
      "relative_percent": {
        "mr_stft": 7.672728856408706,
        "log_mel": 7.224805095546505,
        "log1p_mel": 15.000616247148812
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 32,
      "window_ms": 2560.0,
      "scalar_tokens": 512,
      "codebooks": 16,
      "seam_count": 2,
      "id": "qwen3-tts-tokenizer-12hz--cartoon-voices--log1p_mel--10-20",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 7.672728856408706,
        "log_mel": 7.224805095546505,
        "log1p_mel": 15.000616247148812
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "20-30",
      "audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-66.wav",
      "baseline": {
        "log_mel": 0.35760822892189026,
        "log1p_mel": 0.0033701565116643906,
        "mr_stft": 1.0018222332000732
      },
      "treatment": {
        "log_mel": 0.378473162651062,
        "log1p_mel": 0.0041479854844510555,
        "mr_stft": 1.1002113819122314
      },
      "decoder_drift": {
        "log_mel": 0.04626914858818054,
        "log1p_mel": 0.0010976181365549564,
        "mr_stft": 0.30534106492996216
      },
      "relative_percent": {
        "mr_stft": 9.821018684909632,
        "log_mel": 5.834578748949637,
        "log1p_mel": 23.07990652940106
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 66,
      "window_ms": 5280.0,
      "scalar_tokens": 1056,
      "codebooks": 16,
      "seam_count": 1,
      "id": "qwen3-tts-tokenizer-12hz--cartoon-voices--log1p_mel--20-30",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 9.821018684909632,
        "log_mel": 5.834578748949637,
        "log1p_mel": 23.07990652940106
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "30-40",
      "audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-31.wav",
      "baseline": {
        "log_mel": 0.35760822892189026,
        "log1p_mel": 0.0033701565116643906,
        "mr_stft": 1.0018222332000732
      },
      "treatment": {
        "log_mel": 0.4041663110256195,
        "log1p_mel": 0.004619194660335779,
        "mr_stft": 1.1614590883255005
      },
      "decoder_drift": {
        "log_mel": 0.1405613124370575,
        "log1p_mel": 0.0021746342536062002,
        "mr_stft": 0.5354441404342651
      },
      "relative_percent": {
        "mr_stft": 15.934648866346969,
        "log_mel": 13.019298309798844,
        "log1p_mel": 37.061725304102765
      },
      "recorded_metric_max_abs_difference": 3.5762786865234375e-07,
      "window_frames": 31,
      "window_ms": 2480.0,
      "scalar_tokens": 496,
      "codebooks": 16,
      "seam_count": 2,
      "id": "qwen3-tts-tokenizer-12hz--cartoon-voices--log1p_mel--30-40",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 15.93461316860969,
        "log_mel": 13.019306643589834,
        "log1p_mel": 37.061725304102765
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "40-50",
      "audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-56.wav",
      "baseline": {
        "log_mel": 0.35760822892189026,
        "log1p_mel": 0.0033701565116643906,
        "mr_stft": 1.0018222332000732
      },
      "treatment": {
        "log_mel": 0.4045254588127136,
        "log1p_mel": 0.004892849363386631,
        "mr_stft": 1.2074813842773438
      },
      "decoder_drift": {
        "log_mel": 0.09721845388412476,
        "log1p_mel": 0.0020068970043212175,
        "mr_stft": 0.4964035153388977
      },
      "relative_percent": {
        "mr_stft": 20.5285073800312,
        "log_mel": 13.119728825107968,
        "log1p_mel": 45.18166579065615
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 56,
      "window_ms": 4480.0,
      "scalar_tokens": 896,
      "codebooks": 16,
      "seam_count": 1,
      "id": "qwen3-tts-tokenizer-12hz--cartoon-voices--log1p_mel--40-50",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 20.5285073800312,
        "log_mel": 13.119728825107968,
        "log1p_mel": 45.18166579065615
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "metric": "log1p_mel",
      "band_id": "over-50",
      "audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-8.wav",
      "baseline": {
        "log_mel": 0.35760822892189026,
        "log1p_mel": 0.0033701565116643906,
        "mr_stft": 1.0018222332000732
      },
      "treatment": {
        "log_mel": 0.4803314805030823,
        "log1p_mel": 0.005908168386667967,
        "mr_stft": 1.3221604824066162
      },
      "decoder_drift": {
        "log_mel": 0.3165564239025116,
        "log1p_mel": 0.0044966707937419415,
        "mr_stft": 0.9596074819564819
      },
      "relative_percent": {
        "mr_stft": 31.97555799728078,
        "log_mel": 34.31779295212962,
        "log1p_mel": 75.30842755282454
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 8,
      "window_ms": 640.0,
      "scalar_tokens": 128,
      "codebooks": 16,
      "seam_count": 8,
      "id": "qwen3-tts-tokenizer-12hz--cartoon-voices--log1p_mel--over-50",
      "original_audio": "listening-audio/cartoon-voices-original.wav",
      "full_audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 31.97553419878927,
        "log_mel": 34.31779295212962,
        "log1p_mel": 75.30842755282454
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "0-5",
      "audio": "listening-audio/librispeech-other-500-descript-dac-44khz-window-158.wav",
      "baseline": {
        "log_mel": 0.2750609219074249,
        "log1p_mel": 0.001427017035894096,
        "mr_stft": 0.9531030654907227
      },
      "treatment": {
        "log_mel": 0.2848808169364929,
        "log1p_mel": 0.0014770738780498505,
        "mr_stft": 0.9766303300857544
      },
      "decoder_drift": {
        "log_mel": 0.01855829730629921,
        "log1p_mel": 8.445665298495442e-05,
        "mr_stft": 0.09519176185131073
      },
      "relative_percent": {
        "mr_stft": 2.468491126184591,
        "log_mel": 3.570080024807387,
        "log1p_mel": 3.507795695262428
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 158,
      "window_ms": 1834.3764172335602,
      "scalar_tokens": 1422,
      "codebooks": 9,
      "seam_count": 2,
      "id": "descript-dac-44khz--librispeech-other-500--mr_stft--0-5",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 2.468491126184591,
        "log_mel": 3.570080024807387,
        "log1p_mel": 3.507795695262428
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "5-10",
      "audio": "listening-audio/librispeech-other-500-descript-dac-44khz-window-73.wav",
      "baseline": {
        "log_mel": 0.2750609219074249,
        "log1p_mel": 0.001427017035894096,
        "mr_stft": 0.9531030654907227
      },
      "treatment": {
        "log_mel": 0.31231653690338135,
        "log1p_mel": 0.0016612253384664655,
        "mr_stft": 1.0257751941680908
      },
      "decoder_drift": {
        "log_mel": 0.06812459230422974,
        "log1p_mel": 0.00037242911639623344,
        "mr_stft": 0.24168968200683594
      },
      "relative_percent": {
        "mr_stft": 7.624792250558077,
        "log_mel": 13.544495793006629,
        "log1p_mel": 16.412439142720302
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 73,
      "window_ms": 847.5283446712018,
      "scalar_tokens": 657,
      "codebooks": 9,
      "seam_count": 6,
      "id": "descript-dac-44khz--librispeech-other-500--mr_stft--5-10",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 7.624792250558077,
        "log_mel": 13.544495793006629,
        "log1p_mel": 16.412439142720302
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "10-20",
      "audio": "listening-audio/librispeech-other-500-descript-dac-44khz-window-29.wav",
      "baseline": {
        "log_mel": 0.2750609219074249,
        "log1p_mel": 0.001427017035894096,
        "mr_stft": 0.9531030654907227
      },
      "treatment": {
        "log_mel": 0.3546628952026367,
        "log1p_mel": 0.001979151740670204,
        "mr_stft": 1.0954874753952026
      },
      "decoder_drift": {
        "log_mel": 0.14993102848529816,
        "log1p_mel": 0.00085022347047925,
        "mr_stft": 0.46409034729003906
      },
      "relative_percent": {
        "mr_stft": 14.93903598255355,
        "log_mel": 28.939760960302017,
        "log1p_mel": 38.691528614454754
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 29,
      "window_ms": 336.68934240362813,
      "scalar_tokens": 261,
      "codebooks": 9,
      "seam_count": 15,
      "id": "descript-dac-44khz--librispeech-other-500--mr_stft--10-20",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 14.93903598255355,
        "log_mel": 28.939760960302017,
        "log1p_mel": 38.691528614454754
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "20-30",
      "audio": "listening-audio/librispeech-other-500-descript-dac-44khz-window-20.wav",
      "baseline": {
        "log_mel": 0.2750609219074249,
        "log1p_mel": 0.001427017035894096,
        "mr_stft": 0.9531030654907227
      },
      "treatment": {
        "log_mel": 0.3880966901779175,
        "log1p_mel": 0.0023798064794391394,
        "mr_stft": 1.1868233680725098
      },
      "decoder_drift": {
        "log_mel": 0.21299518644809723,
        "log1p_mel": 0.0014436064520850778,
        "mr_stft": 0.674298882484436
      },
      "relative_percent": {
        "mr_stft": 24.522038701182012,
        "log_mel": 41.09481182809971,
        "log1p_mel": 66.76790953291419
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 20,
      "window_ms": 232.19954648526078,
      "scalar_tokens": 180,
      "codebooks": 9,
      "seam_count": 22,
      "id": "descript-dac-44khz--librispeech-other-500--mr_stft--20-30",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 24.522038701182012,
        "log_mel": 41.09481182809971,
        "log1p_mel": 66.76790953291419
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "30-40",
      "audio": "listening-audio/librispeech-other-500-descript-dac-44khz-window-14.wav",
      "baseline": {
        "log_mel": 0.2750609219074249,
        "log1p_mel": 0.001427017035894096,
        "mr_stft": 0.9531030654907227
      },
      "treatment": {
        "log_mel": 0.4333249628543854,
        "log1p_mel": 0.002648234600201249,
        "mr_stft": 1.2831380367279053
      },
      "decoder_drift": {
        "log_mel": 0.30428963899612427,
        "log1p_mel": 0.0019343754975125194,
        "mr_stft": 0.9332152605056763
      },
      "relative_percent": {
        "mr_stft": 34.62741682267678,
        "log_mel": 57.53781375030296,
        "log1p_mel": 85.57834514862681
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 14,
      "window_ms": 162.53968253968253,
      "scalar_tokens": 126,
      "codebooks": 9,
      "seam_count": 31,
      "id": "descript-dac-44khz--librispeech-other-500--mr_stft--30-40",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 34.62741682267678,
        "log_mel": 57.53781375030296,
        "log1p_mel": 85.57834514862681
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "40-50",
      "audio": "listening-audio/librispeech-other-500-descript-dac-44khz-window-10.wav",
      "baseline": {
        "log_mel": 0.2750609219074249,
        "log1p_mel": 0.001427017035894096,
        "mr_stft": 0.9531030654907227
      },
      "treatment": {
        "log_mel": 0.5085204839706421,
        "log1p_mel": 0.0032874313183128834,
        "mr_stft": 1.3914259672164917
      },
      "decoder_drift": {
        "log_mel": 0.4267071783542633,
        "log1p_mel": 0.0027924063615500927,
        "mr_stft": 1.1983904838562012
      },
      "relative_percent": {
        "mr_stft": 45.98903493192423,
        "log_mel": 84.8755833595987,
        "log1p_mel": 130.37085301880413
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 10,
      "window_ms": 116.09977324263039,
      "scalar_tokens": 90,
      "codebooks": 9,
      "seam_count": 44,
      "id": "descript-dac-44khz--librispeech-other-500--mr_stft--40-50",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 45.98903493192423,
        "log_mel": 84.8755833595987,
        "log1p_mel": 130.37085301880413
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "over-50",
      "audio": "listening-audio/librispeech-other-500-descript-dac-44khz-window-7.wav",
      "baseline": {
        "log_mel": 0.2750609219074249,
        "log1p_mel": 0.001427017035894096,
        "mr_stft": 0.9531030654907227
      },
      "treatment": {
        "log_mel": 0.6299577951431274,
        "log1p_mel": 0.0040318043902516365,
        "mr_stft": 1.623096227645874
      },
      "decoder_drift": {
        "log_mel": 0.6064090132713318,
        "log1p_mel": 0.003765730420127511,
        "mr_stft": 1.6535224914550781
      },
      "relative_percent": {
        "mr_stft": 70.29598229339373,
        "log_mel": 129.02482503681395,
        "log1p_mel": 182.53372516506184
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 7,
      "window_ms": 81.26984126984127,
      "scalar_tokens": 63,
      "codebooks": 9,
      "seam_count": 63,
      "id": "descript-dac-44khz--librispeech-other-500--mr_stft--over-50",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 70.29598229339373,
        "log_mel": 129.02482503681395,
        "log1p_mel": 182.53372516506184
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "0-5",
      "audio": "listening-audio/librispeech-other-500-descript-dac-44khz-window-320.wav",
      "baseline": {
        "log_mel": 0.2750609219074249,
        "log1p_mel": 0.001427017035894096,
        "mr_stft": 0.9531030654907227
      },
      "treatment": {
        "log_mel": 0.28181537985801697,
        "log1p_mel": 0.001434879843145609,
        "mr_stft": 0.9601745009422302
      },
      "decoder_drift": {
        "log_mel": 0.011716928333044052,
        "log1p_mel": 9.887937267194502e-06,
        "mr_stft": 0.02719348669052124
      },
      "relative_percent": {
        "mr_stft": 0.7419381709644179,
        "log_mel": 2.4556225230952133,
        "log1p_mel": 0.5509960325446706
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 320,
      "window_ms": 3715.1927437641725,
      "scalar_tokens": 2880,
      "codebooks": 9,
      "seam_count": 1,
      "id": "descript-dac-44khz--librispeech-other-500--log_mel--0-5",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 0.7419381709644179,
        "log_mel": 2.4556225230952133,
        "log1p_mel": 0.5509960325446706
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "5-10",
      "audio": "listening-audio/librispeech-other-500-descript-dac-44khz-window-126.wav",
      "baseline": {
        "log_mel": 0.2750609219074249,
        "log1p_mel": 0.001427017035894096,
        "mr_stft": 0.9531030654907227
      },
      "treatment": {
        "log_mel": 0.29526516795158386,
        "log1p_mel": 0.001724258647300303,
        "mr_stft": 1.0314221382141113
      },
      "decoder_drift": {
        "log_mel": 0.034809619188308716,
        "log1p_mel": 0.0003679404908325523,
        "mr_stft": 0.2148275375366211
      },
      "relative_percent": {
        "mr_stft": 8.217272146015464,
        "log_mel": 7.3453713104906,
        "log1p_mel": 20.829576937739258
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 126,
      "window_ms": 1462.857142857143,
      "scalar_tokens": 1134,
      "codebooks": 9,
      "seam_count": 3,
      "id": "descript-dac-44khz--librispeech-other-500--log_mel--5-10",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 8.217272146015464,
        "log_mel": 7.3453713104906,
        "log1p_mel": 20.829576937739258
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "10-20",
      "audio": "listening-audio/librispeech-other-500-descript-dac-44khz-window-41.wav",
      "baseline": {
        "log_mel": 0.2750609219074249,
        "log1p_mel": 0.001427017035894096,
        "mr_stft": 0.9531030654907227
      },
      "treatment": {
        "log_mel": 0.3165615200996399,
        "log1p_mel": 0.0017753904685378075,
        "mr_stft": 1.0627696514129639
      },
      "decoder_drift": {
        "log_mel": 0.08861570060253143,
        "log1p_mel": 0.0006041586748324335,
        "mr_stft": 0.34007224440574646
      },
      "relative_percent": {
        "mr_stft": 11.50626725408519,
        "log_mel": 15.087784155025297,
        "log1p_mel": 24.41270313394952
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 41,
      "window_ms": 476.0090702947846,
      "scalar_tokens": 369,
      "codebooks": 9,
      "seam_count": 10,
      "id": "descript-dac-44khz--librispeech-other-500--log_mel--10-20",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 11.50626725408519,
        "log_mel": 15.087784155025297,
        "log1p_mel": 24.41270313394952
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "20-30",
      "audio": "listening-audio/librispeech-other-500-descript-dac-44khz-window-32.wav",
      "baseline": {
        "log_mel": 0.2750609219074249,
        "log1p_mel": 0.001427017035894096,
        "mr_stft": 0.9531030654907227
      },
      "treatment": {
        "log_mel": 0.34446561336517334,
        "log1p_mel": 0.0018919677240774035,
        "mr_stft": 1.099223017692566
      },
      "decoder_drift": {
        "log_mel": 0.12981925904750824,
        "log1p_mel": 0.0007837152807042003,
        "mr_stft": 0.4376610219478607
      },
      "relative_percent": {
        "mr_stft": 15.330970751480134,
        "log_mel": 25.23247976355849,
        "log1p_mel": 32.58199982819359
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 32,
      "window_ms": 371.51927437641723,
      "scalar_tokens": 288,
      "codebooks": 9,
      "seam_count": 13,
      "id": "descript-dac-44khz--librispeech-other-500--log_mel--20-30",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 15.330970751480134,
        "log_mel": 25.23247976355849,
        "log1p_mel": 32.58199982819359
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "30-40",
      "audio": "listening-audio/librispeech-other-500-descript-dac-44khz-window-24.wav",
      "baseline": {
        "log_mel": 0.2750609219074249,
        "log1p_mel": 0.001427017035894096,
        "mr_stft": 0.9531030654907227
      },
      "treatment": {
        "log_mel": 0.3694794774055481,
        "log1p_mel": 0.002005386631935835,
        "mr_stft": 1.1254897117614746
      },
      "decoder_drift": {
        "log_mel": 0.17730411887168884,
        "log1p_mel": 0.0010236691450700164,
        "mr_stft": 0.5598599910736084
      },
      "relative_percent": {
        "mr_stft": 18.086884043542085,
        "log_mel": 34.32641570579076,
        "log1p_mel": 40.529971366414856
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 24,
      "window_ms": 278.6394557823129,
      "scalar_tokens": 216,
      "codebooks": 9,
      "seam_count": 18,
      "id": "descript-dac-44khz--librispeech-other-500--log_mel--30-40",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 18.086884043542085,
        "log_mel": 34.32641570579076,
        "log1p_mel": 40.529971366414856
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "40-50",
      "audio": "listening-audio/librispeech-other-500-descript-dac-44khz-window-19.wav",
      "baseline": {
        "log_mel": 0.2750609219074249,
        "log1p_mel": 0.001427017035894096,
        "mr_stft": 0.9531030654907227
      },
      "treatment": {
        "log_mel": 0.3966156244277954,
        "log1p_mel": 0.0022442908957600594,
        "mr_stft": 1.183593511581421
      },
      "decoder_drift": {
        "log_mel": 0.22724705934524536,
        "log1p_mel": 0.0013546556001529098,
        "mr_stft": 0.6901254057884216
      },
      "relative_percent": {
        "mr_stft": 24.18316071326725,
        "log_mel": 44.191919985377346,
        "log1p_mel": 57.27148585538093
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 19,
      "window_ms": 220.58956916099774,
      "scalar_tokens": 171,
      "codebooks": 9,
      "seam_count": 23,
      "id": "descript-dac-44khz--librispeech-other-500--log_mel--40-50",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 24.18316071326725,
        "log_mel": 44.191919985377346,
        "log1p_mel": 57.27148585538093
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "over-50",
      "audio": "listening-audio/librispeech-other-500-descript-dac-44khz-window-11.wav",
      "baseline": {
        "log_mel": 0.2750609219074249,
        "log1p_mel": 0.001427017035894096,
        "mr_stft": 0.9531030654907227
      },
      "treatment": {
        "log_mel": 0.4872071444988251,
        "log1p_mel": 0.0029306686483323574,
        "mr_stft": 1.3611743450164795
      },
      "decoder_drift": {
        "log_mel": 0.3911093473434448,
        "log1p_mel": 0.0023469694424420595,
        "mr_stft": 1.1206802129745483
      },
      "relative_percent": {
        "mr_stft": 42.81502119769742,
        "log_mel": 77.1269946745836,
        "log1p_mel": 105.37026360698985
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 11,
      "window_ms": 127.70975056689343,
      "scalar_tokens": 99,
      "codebooks": 9,
      "seam_count": 40,
      "id": "descript-dac-44khz--librispeech-other-500--log_mel--over-50",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 42.81502119769742,
        "log_mel": 77.1269946745836,
        "log1p_mel": 105.37026360698985
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "0-5",
      "audio": "listening-audio/librispeech-other-500-descript-dac-44khz-window-151.wav",
      "baseline": {
        "log_mel": 0.2750609219074249,
        "log1p_mel": 0.001427017035894096,
        "mr_stft": 0.9531030654907227
      },
      "treatment": {
        "log_mel": 0.2840215861797333,
        "log1p_mel": 0.0014636075356975198,
        "mr_stft": 0.9648658633232117
      },
      "decoder_drift": {
        "log_mel": 0.01900932379066944,
        "log1p_mel": 5.368295751395635e-05,
        "mr_stft": 0.058541350066661835
      },
      "relative_percent": {
        "mr_stft": 1.2341580106483674,
        "log_mel": 3.2577016793843727,
        "log1p_mel": 2.564124946167734
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 151,
      "window_ms": 1753.1065759637188,
      "scalar_tokens": 1359,
      "codebooks": 9,
      "seam_count": 2,
      "id": "descript-dac-44khz--librispeech-other-500--log1p_mel--0-5",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 1.2341580106483674,
        "log_mel": 3.2577016793843727,
        "log1p_mel": 2.564124946167734
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "5-10",
      "audio": "listening-audio/librispeech-other-500-descript-dac-44khz-window-125.wav",
      "baseline": {
        "log_mel": 0.2750609219074249,
        "log1p_mel": 0.001427017035894096,
        "mr_stft": 0.9531030654907227
      },
      "treatment": {
        "log_mel": 0.29306986927986145,
        "log1p_mel": 0.0015336633659899235,
        "mr_stft": 0.9901172518730164
      },
      "decoder_drift": {
        "log_mel": 0.03181895613670349,
        "log1p_mel": 0.00016408468945883214,
        "mr_stft": 0.1332034170627594
      },
      "relative_percent": {
        "mr_stft": 3.8835449934510713,
        "log_mel": 6.547257693878317,
        "log1p_mel": 7.473374697941737
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 125,
      "window_ms": 1451.2471655328798,
      "scalar_tokens": 1125,
      "codebooks": 9,
      "seam_count": 3,
      "id": "descript-dac-44khz--librispeech-other-500--log1p_mel--5-10",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 3.8835449934510713,
        "log_mel": 6.547257693878317,
        "log1p_mel": 7.473374697941737
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "10-20",
      "audio": "listening-audio/librispeech-other-500-descript-dac-44khz-window-60.wav",
      "baseline": {
        "log_mel": 0.2750609219074249,
        "log1p_mel": 0.001427017035894096,
        "mr_stft": 0.9531030654907227
      },
      "treatment": {
        "log_mel": 0.3157922923564911,
        "log1p_mel": 0.0016476847231388092,
        "mr_stft": 1.0142887830734253
      },
      "decoder_drift": {
        "log_mel": 0.07525686919689178,
        "log1p_mel": 0.0003849150671157986,
        "mr_stft": 0.2404162585735321
      },
      "relative_percent": {
        "mr_stft": 6.41963286008318,
        "log_mel": 14.80812692934068,
        "log1p_mel": 15.463563622171762
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 60,
      "window_ms": 696.5986394557823,
      "scalar_tokens": 540,
      "codebooks": 9,
      "seam_count": 7,
      "id": "descript-dac-44khz--librispeech-other-500--log1p_mel--10-20",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 6.41963286008318,
        "log_mel": 14.80812692934068,
        "log1p_mel": 15.463563622171762
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "20-30",
      "audio": "listening-audio/librispeech-other-500-descript-dac-44khz-window-43.wav",
      "baseline": {
        "log_mel": 0.2750609219074249,
        "log1p_mel": 0.001427017035894096,
        "mr_stft": 0.9531030654907227
      },
      "treatment": {
        "log_mel": 0.32705211639404297,
        "log1p_mel": 0.0017849330324679613,
        "mr_stft": 1.0529364347457886
      },
      "decoder_drift": {
        "log_mel": 0.09762489795684814,
        "log1p_mel": 0.0005428153672255576,
        "mr_stft": 0.3299129605293274
      },
      "relative_percent": {
        "mr_stft": 10.474561762495727,
        "log_mel": 18.90170153073083,
        "log1p_mel": 25.081410212430555
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 43,
      "window_ms": 499.22902494331066,
      "scalar_tokens": 387,
      "codebooks": 9,
      "seam_count": 10,
      "id": "descript-dac-44khz--librispeech-other-500--log1p_mel--20-30",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 10.474561762495727,
        "log_mel": 18.90170153073083,
        "log1p_mel": 25.081410212430555
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "30-40",
      "audio": "listening-audio/librispeech-other-500-descript-dac-44khz-window-35.wav",
      "baseline": {
        "log_mel": 0.2750609219074249,
        "log1p_mel": 0.001427017035894096,
        "mr_stft": 0.9531030654907227
      },
      "treatment": {
        "log_mel": 0.33294549584388733,
        "log1p_mel": 0.0019179624505341053,
        "mr_stft": 1.0764377117156982
      },
      "decoder_drift": {
        "log_mel": 0.1138540729880333,
        "log1p_mel": 0.0008175605325959623,
        "mr_stft": 0.3965660333633423
      },
      "relative_percent": {
        "mr_stft": 12.940326255427003,
        "log_mel": 21.044273950315695,
        "log1p_mel": 34.403612731393075
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 35,
      "window_ms": 406.3492063492063,
      "scalar_tokens": 315,
      "codebooks": 9,
      "seam_count": 12,
      "id": "descript-dac-44khz--librispeech-other-500--log1p_mel--30-40",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 12.940326255427003,
        "log_mel": 21.044273950315695,
        "log1p_mel": 34.403612731393075
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "40-50",
      "audio": "listening-audio/librispeech-other-500-descript-dac-44khz-window-31.wav",
      "baseline": {
        "log_mel": 0.2750609219074249,
        "log1p_mel": 0.001427017035894096,
        "mr_stft": 0.9531030654907227
      },
      "treatment": {
        "log_mel": 0.35399800539016724,
        "log1p_mel": 0.0020587248727679253,
        "mr_stft": 1.1295511722564697
      },
      "decoder_drift": {
        "log_mel": 0.14194467663764954,
        "log1p_mel": 0.0009247882990166545,
        "mr_stft": 0.4879429340362549
      },
      "relative_percent": {
        "mr_stft": 18.513014295563046,
        "log_mel": 28.698036396936644,
        "log1p_mel": 44.267715169779564
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 31,
      "window_ms": 359.90929705215416,
      "scalar_tokens": 279,
      "codebooks": 9,
      "seam_count": 14,
      "id": "descript-dac-44khz--librispeech-other-500--log1p_mel--40-50",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 18.513014295563046,
        "log_mel": 28.698036396936644,
        "log1p_mel": 44.267715169779564
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "over-50",
      "audio": "listening-audio/librispeech-other-500-descript-dac-44khz-window-21.wav",
      "baseline": {
        "log_mel": 0.2750609219074249,
        "log1p_mel": 0.001427017035894096,
        "mr_stft": 0.9531030654907227
      },
      "treatment": {
        "log_mel": 0.3925490379333496,
        "log1p_mel": 0.0024966008495539427,
        "mr_stft": 1.2333118915557861
      },
      "decoder_drift": {
        "log_mel": 0.21306130290031433,
        "log1p_mel": 0.0015619989717379212,
        "mr_stft": 0.7021669149398804
      },
      "relative_percent": {
        "mr_stft": 29.39963538159358,
        "log_mel": 42.713488783211,
        "log1p_mel": 74.95242080201938
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 21,
      "window_ms": 243.8095238095238,
      "scalar_tokens": 189,
      "codebooks": 9,
      "seam_count": 21,
      "id": "descript-dac-44khz--librispeech-other-500--log1p_mel--over-50",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 29.39963538159358,
        "log_mel": 42.713488783211,
        "log1p_mel": 74.95242080201938
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "0-5",
      "audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-96.wav",
      "baseline": {
        "log_mel": 0.40510469675064087,
        "log1p_mel": 0.002791645936667919,
        "mr_stft": 1.1514151096343994
      },
      "treatment": {
        "log_mel": 0.4141429364681244,
        "log1p_mel": 0.0028523432556539774,
        "mr_stft": 1.180903673171997
      },
      "decoder_drift": {
        "log_mel": 0.028687817975878716,
        "log1p_mel": 0.00011697001900756732,
        "mr_stft": 0.1066138744354248
      },
      "relative_percent": {
        "mr_stft": 2.5610714407735102,
        "log_mel": 2.231087368272844,
        "log1p_mel": 2.1742484671428697
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 96,
      "window_ms": 4458.231292517006,
      "scalar_tokens": 960,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-400m-morepitchloss1-step3802000--librispeech-other-500--mr_stft--0-5",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 2.5610714407735102,
        "log_mel": 2.231087368272844,
        "log1p_mel": 2.1742484671428697
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "5-10",
      "audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-38.wav",
      "baseline": {
        "log_mel": 0.40510469675064087,
        "log1p_mel": 0.002791645936667919,
        "mr_stft": 1.1514151096343994
      },
      "treatment": {
        "log_mel": 0.42126649618148804,
        "log1p_mel": 0.0030476097017526627,
        "mr_stft": 1.2358253002166748
      },
      "decoder_drift": {
        "log_mel": 0.0704689472913742,
        "log1p_mel": 0.0006581613561138511,
        "mr_stft": 0.3613162934780121
      },
      "relative_percent": {
        "mr_stft": 7.33099556154666,
        "log_mel": 3.9895364236656627,
        "log1p_mel": 9.168919371998129
      },
      "recorded_metric_max_abs_difference": 3.337860107421875e-06,
      "window_frames": 38,
      "window_ms": 1764.716553287982,
      "scalar_tokens": 380,
      "codebooks": 10,
      "seam_count": 2,
      "id": "fish-400m-morepitchloss1-step3802000--librispeech-other-500--mr_stft--5-10",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 7.331285453530834,
        "log_mel": 3.989492283487639,
        "log1p_mel": 9.168919371998129
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "10-20",
      "audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-19.wav",
      "baseline": {
        "log_mel": 0.40510469675064087,
        "log1p_mel": 0.002791645936667919,
        "mr_stft": 1.1514151096343994
      },
      "treatment": {
        "log_mel": 0.4431701600551605,
        "log1p_mel": 0.0031610713340342045,
        "mr_stft": 1.3149535655975342
      },
      "decoder_drift": {
        "log_mel": 0.15371789038181305,
        "log1p_mel": 0.001106564886868,
        "mr_stft": 0.585857093334198
      },
      "relative_percent": {
        "mr_stft": 14.203257764705036,
        "log_mel": 9.39645074713873,
        "log1p_mel": 13.233246828114154
      },
      "recorded_metric_max_abs_difference": 2.7418136596679688e-06,
      "window_frames": 19,
      "window_ms": 882.358276643991,
      "scalar_tokens": 190,
      "codebooks": 10,
      "seam_count": 5,
      "id": "fish-400m-morepitchloss1-step3802000--librispeech-other-500--mr_stft--10-20",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 14.203495890263461,
        "log_mel": 9.396465460531388,
        "log1p_mel": 13.233255168378255
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "20-30",
      "audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-14.wav",
      "baseline": {
        "log_mel": 0.40510469675064087,
        "log1p_mel": 0.002791645936667919,
        "mr_stft": 1.1514151096343994
      },
      "treatment": {
        "log_mel": 0.47964707016944885,
        "log1p_mel": 0.0036992698442190886,
        "mr_stft": 1.47435462474823
      },
      "decoder_drift": {
        "log_mel": 0.23303094506263733,
        "log1p_mel": 0.00205229758284986,
        "mr_stft": 0.8954468965530396
      },
      "relative_percent": {
        "mr_stft": 28.04718406173872,
        "log_mel": 18.40076750941546,
        "log1p_mel": 32.51214259049271
      },
      "recorded_metric_max_abs_difference": 1.5497207641601562e-06,
      "window_frames": 14,
      "window_ms": 650.1587301587301,
      "scalar_tokens": 140,
      "codebooks": 10,
      "seam_count": 7,
      "id": "fish-400m-morepitchloss1-step3802000--librispeech-other-500--mr_stft--20-30",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 28.047318654445654,
        "log_mel": 18.400789579504483,
        "log1p_mel": 32.51214259049271
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "30-40",
      "audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-6.wav",
      "baseline": {
        "log_mel": 0.40510469675064087,
        "log1p_mel": 0.002791645936667919,
        "mr_stft": 1.1514151096343994
      },
      "treatment": {
        "log_mel": 0.5131045579910278,
        "log1p_mel": 0.003771210554987192,
        "mr_stft": 1.5390914678573608
      },
      "decoder_drift": {
        "log_mel": 0.357206791639328,
        "log1p_mel": 0.002598223276436329,
        "mr_stft": 1.1409868001937866
      },
      "relative_percent": {
        "mr_stft": 33.669556268551794,
        "log_mel": 26.659740582288393,
        "log1p_mel": 35.0891424106766
      },
      "recorded_metric_max_abs_difference": 5.960464477539062e-07,
      "window_frames": 6,
      "window_ms": 278.6394557823129,
      "scalar_tokens": 60,
      "codebooks": 10,
      "seam_count": 18,
      "id": "fish-400m-morepitchloss1-step3802000--librispeech-other-500--mr_stft--30-40",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 33.669608034977536,
        "log_mel": 26.659755295681077,
        "log1p_mel": 35.0891424106766
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "40-50",
      "audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-4.wav",
      "baseline": {
        "log_mel": 0.40510469675064087,
        "log1p_mel": 0.002791645936667919,
        "mr_stft": 1.1514151096343994
      },
      "treatment": {
        "log_mel": 0.5777700543403625,
        "log1p_mel": 0.0043848915956914425,
        "mr_stft": 1.6879606246948242
      },
      "decoder_drift": {
        "log_mel": 0.45666971802711487,
        "log1p_mel": 0.003454972989857197,
        "mr_stft": 1.3759009838104248
      },
      "relative_percent": {
        "mr_stft": 46.598790529228886,
        "log_mel": 42.622403288502106,
        "log1p_mel": 57.07191009061863
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 4,
      "window_ms": 185.75963718820861,
      "scalar_tokens": 40,
      "codebooks": 10,
      "seam_count": 27,
      "id": "fish-400m-morepitchloss1-step3802000--librispeech-other-500--mr_stft--40-50",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 46.59881123579916,
        "log_mel": 42.62244742868013,
        "log1p_mel": 57.07191009061863
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "over-50",
      "audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-2.wav",
      "baseline": {
        "log_mel": 0.40510469675064087,
        "log1p_mel": 0.002791645936667919,
        "mr_stft": 1.1514151096343994
      },
      "treatment": {
        "log_mel": 0.6810618042945862,
        "log1p_mel": 0.006322666071355343,
        "mr_stft": 2.0669620037078857
      },
      "decoder_drift": {
        "log_mel": 0.6198489665985107,
        "log1p_mel": 0.005612446926534176,
        "mr_stft": 1.838541030883789
      },
      "relative_percent": {
        "mr_stft": 79.5149278841923,
        "log_mel": 68.11994769683172,
        "log1p_mel": 126.48524256990892
      },
      "recorded_metric_max_abs_difference": 4.76837158203125e-07,
      "window_frames": 2,
      "window_ms": 92.87981859410431,
      "scalar_tokens": 20,
      "codebooks": 10,
      "seam_count": 55,
      "id": "fish-400m-morepitchloss1-step3802000--librispeech-other-500--mr_stft--over-50",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 79.5148864710517,
        "log_mel": 68.11988884326104,
        "log1p_mel": 126.48524256990892
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "0-5",
      "audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-39.wav",
      "baseline": {
        "log_mel": 0.40510469675064087,
        "log1p_mel": 0.002791645936667919,
        "mr_stft": 1.1514151096343994
      },
      "treatment": {
        "log_mel": 0.41474828124046326,
        "log1p_mel": 0.0028901880141347647,
        "mr_stft": 1.1976348161697388
      },
      "decoder_drift": {
        "log_mel": 0.059605300426483154,
        "log1p_mel": 0.00045172663521952927,
        "mr_stft": 0.2810191512107849
      },
      "relative_percent": {
        "mr_stft": 4.0141653647410624,
        "log_mel": 2.380516584274117,
        "log1p_mel": 3.5298916733138563
      },
      "recorded_metric_max_abs_difference": 3.5762786865234375e-07,
      "window_frames": 39,
      "window_ms": 1811.156462585034,
      "scalar_tokens": 390,
      "codebooks": 10,
      "seam_count": 2,
      "id": "fish-400m-morepitchloss1-step3802000--librispeech-other-500--log_mel--0-5",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 4.014134304885619,
        "log_mel": 2.380516584274117,
        "log1p_mel": 3.5298916733138563
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "5-10",
      "audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-18.wav",
      "baseline": {
        "log_mel": 0.40510469675064087,
        "log1p_mel": 0.002791645936667919,
        "mr_stft": 1.1514151096343994
      },
      "treatment": {
        "log_mel": 0.43541085720062256,
        "log1p_mel": 0.003192150965332985,
        "mr_stft": 1.3030869960784912
      },
      "decoder_drift": {
        "log_mel": 0.14879655838012695,
        "log1p_mel": 0.0011750549310818315,
        "mr_stft": 0.555861234664917
      },
      "relative_percent": {
        "mr_stft": 13.172650347818603,
        "log_mel": 7.48106864548066,
        "log1p_mel": 14.346555320804931
      },
      "recorded_metric_max_abs_difference": 1.5497207641601562e-06,
      "window_frames": 18,
      "window_ms": 835.9183673469388,
      "scalar_tokens": 180,
      "codebooks": 10,
      "seam_count": 6,
      "id": "fish-400m-morepitchloss1-step3802000--librispeech-other-500--log_mel--5-10",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 13.172784940525561,
        "log_mel": 7.481134855747684,
        "log1p_mel": 14.346572001333113
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "10-20",
      "audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-11.wav",
      "baseline": {
        "log_mel": 0.40510469675064087,
        "log1p_mel": 0.002791645936667919,
        "mr_stft": 1.1514151096343994
      },
      "treatment": {
        "log_mel": 0.4656517207622528,
        "log1p_mel": 0.0031750500202178955,
        "mr_stft": 1.375402808189392
      },
      "decoder_drift": {
        "log_mel": 0.22366055846214294,
        "log1p_mel": 0.0016553692985326052,
        "mr_stft": 0.8089880347251892
      },
      "relative_percent": {
        "mr_stft": 19.453253364558833,
        "log_mel": 14.946018769286496,
        "log1p_mel": 13.733979603717362
      },
      "recorded_metric_max_abs_difference": 3.5762786865234375e-07,
      "window_frames": 11,
      "window_ms": 510.83900226757373,
      "scalar_tokens": 110,
      "codebooks": 10,
      "seam_count": 10,
      "id": "fish-400m-morepitchloss1-step3802000--librispeech-other-500--log_mel--10-20",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 19.453222304703388,
        "log_mel": 14.94607026616086,
        "log1p_mel": 13.733979603717362
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "20-30",
      "audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-6.wav",
      "baseline": {
        "log_mel": 0.40510469675064087,
        "log1p_mel": 0.002791645936667919,
        "mr_stft": 1.1514151096343994
      },
      "treatment": {
        "log_mel": 0.5131045579910278,
        "log1p_mel": 0.003771210554987192,
        "mr_stft": 1.5390914678573608
      },
      "decoder_drift": {
        "log_mel": 0.357206791639328,
        "log1p_mel": 0.002598223276436329,
        "mr_stft": 1.1409868001937866
      },
      "relative_percent": {
        "mr_stft": 33.669556268551794,
        "log_mel": 26.659740582288393,
        "log1p_mel": 35.0891424106766
      },
      "recorded_metric_max_abs_difference": 5.960464477539062e-07,
      "window_frames": 6,
      "window_ms": 278.6394557823129,
      "scalar_tokens": 60,
      "codebooks": 10,
      "seam_count": 18,
      "id": "fish-400m-morepitchloss1-step3802000--librispeech-other-500--log_mel--20-30",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 33.669608034977536,
        "log_mel": 26.659755295681077,
        "log1p_mel": 35.0891424106766
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "30-40",
      "audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-5.wav",
      "baseline": {
        "log_mel": 0.40510469675064087,
        "log1p_mel": 0.002791645936667919,
        "mr_stft": 1.1514151096343994
      },
      "treatment": {
        "log_mel": 0.5277677774429321,
        "log1p_mel": 0.004265657160431147,
        "mr_stft": 1.707873821258545
      },
      "decoder_drift": {
        "log_mel": 0.3963123857975006,
        "log1p_mel": 0.0033061110880225897,
        "mr_stft": 1.3770450353622437
      },
      "relative_percent": {
        "mr_stft": 48.328244693682535,
        "log_mel": 30.2793528873341,
        "log1p_mel": 52.80079412658587
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 5,
      "window_ms": 232.19954648526078,
      "scalar_tokens": 50,
      "codebooks": 10,
      "seam_count": 22,
      "id": "fish-400m-morepitchloss1-step3802000--librispeech-other-500--log_mel--30-40",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 48.32826540025283,
        "log_mel": 30.2793528873341,
        "log1p_mel": 52.800810807114054
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "40-50",
      "audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-3.wav",
      "baseline": {
        "log_mel": 0.40510469675064087,
        "log1p_mel": 0.002791645936667919,
        "mr_stft": 1.1514151096343994
      },
      "treatment": {
        "log_mel": 0.5786410570144653,
        "log1p_mel": 0.004568345379084349,
        "mr_stft": 1.745819091796875
      },
      "decoder_drift": {
        "log_mel": 0.4892440736293793,
        "log1p_mel": 0.0038249765057116747,
        "mr_stft": 1.4805587530136108
      },
      "relative_percent": {
        "mr_stft": 51.623778182936334,
        "log_mel": 42.837410095653226,
        "log1p_mel": 63.64343769672598
      },
      "recorded_metric_max_abs_difference": 4.76837158203125e-07,
      "window_frames": 3,
      "window_ms": 139.31972789115645,
      "scalar_tokens": 30,
      "codebooks": 10,
      "seam_count": 36,
      "id": "fish-400m-morepitchloss1-step3802000--librispeech-other-500--log_mel--40-50",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 51.62373676979575,
        "log_mel": 42.83738066886789,
        "log1p_mel": 63.643404335669615
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "over-50",
      "audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-2.wav",
      "baseline": {
        "log_mel": 0.40510469675064087,
        "log1p_mel": 0.002791645936667919,
        "mr_stft": 1.1514151096343994
      },
      "treatment": {
        "log_mel": 0.6810618042945862,
        "log1p_mel": 0.006322666071355343,
        "mr_stft": 2.0669620037078857
      },
      "decoder_drift": {
        "log_mel": 0.6198489665985107,
        "log1p_mel": 0.005612446926534176,
        "mr_stft": 1.838541030883789
      },
      "relative_percent": {
        "mr_stft": 79.5149278841923,
        "log_mel": 68.11994769683172,
        "log1p_mel": 126.48524256990892
      },
      "recorded_metric_max_abs_difference": 4.76837158203125e-07,
      "window_frames": 2,
      "window_ms": 92.87981859410431,
      "scalar_tokens": 20,
      "codebooks": 10,
      "seam_count": 55,
      "id": "fish-400m-morepitchloss1-step3802000--librispeech-other-500--log_mel--over-50",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 79.5148864710517,
        "log_mel": 68.11988884326104,
        "log1p_mel": 126.48524256990892
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "0-5",
      "audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-96.wav",
      "baseline": {
        "log_mel": 0.40510469675064087,
        "log1p_mel": 0.002791645936667919,
        "mr_stft": 1.1514151096343994
      },
      "treatment": {
        "log_mel": 0.4141429364681244,
        "log1p_mel": 0.0028523432556539774,
        "mr_stft": 1.180903673171997
      },
      "decoder_drift": {
        "log_mel": 0.028687817975878716,
        "log1p_mel": 0.00011697001900756732,
        "mr_stft": 0.1066138744354248
      },
      "relative_percent": {
        "mr_stft": 2.5610714407735102,
        "log_mel": 2.231087368272844,
        "log1p_mel": 2.1742484671428697
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 96,
      "window_ms": 4458.231292517006,
      "scalar_tokens": 960,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-400m-morepitchloss1-step3802000--librispeech-other-500--log1p_mel--0-5",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 2.5610714407735102,
        "log_mel": 2.231087368272844,
        "log1p_mel": 2.1742484671428697
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "5-10",
      "audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-17.wav",
      "baseline": {
        "log_mel": 0.40510469675064087,
        "log1p_mel": 0.002791645936667919,
        "mr_stft": 1.1514151096343994
      },
      "treatment": {
        "log_mel": 0.4458019733428955,
        "log1p_mel": 0.0030055835377424955,
        "mr_stft": 1.291111946105957
      },
      "decoder_drift": {
        "log_mel": 0.17134414613246918,
        "log1p_mel": 0.0010392674012109637,
        "mr_stft": 0.6019343733787537
      },
      "relative_percent": {
        "mr_stft": 12.132621441446467,
        "log_mel": 10.046113243980859,
        "log1p_mel": 7.663493362984641
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-06,
      "window_frames": 17,
      "window_ms": 789.4784580498866,
      "scalar_tokens": 170,
      "codebooks": 10,
      "seam_count": 6,
      "id": "fish-400m-morepitchloss1-step3802000--librispeech-other-500--log1p_mel--5-10",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 12.13272497429796,
        "log_mel": 10.046186810944224,
        "log1p_mel": 7.663493362984641
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "10-20",
      "audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-16.wav",
      "baseline": {
        "log_mel": 0.40510469675064087,
        "log1p_mel": 0.002791645936667919,
        "mr_stft": 1.1514151096343994
      },
      "treatment": {
        "log_mel": 0.4541194438934326,
        "log1p_mel": 0.003203360131010413,
        "mr_stft": 1.347142219543457
      },
      "decoder_drift": {
        "log_mel": 0.17713259160518646,
        "log1p_mel": 0.0012772391783073545,
        "mr_stft": 0.6699364185333252
      },
      "relative_percent": {
        "mr_stft": 16.99883111410667,
        "log_mel": 12.099278911338418,
        "log1p_mel": 14.748080655024332
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-06,
      "window_frames": 16,
      "window_ms": 743.0385487528345,
      "scalar_tokens": 160,
      "codebooks": 10,
      "seam_count": 6,
      "id": "fish-400m-morepitchloss1-step3802000--librispeech-other-500--log1p_mel--10-20",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 16.998934646958162,
        "log_mel": 12.099278911338418,
        "log1p_mel": 14.748080655024332
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "20-30",
      "audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-9.wav",
      "baseline": {
        "log_mel": 0.40510469675064087,
        "log1p_mel": 0.002791645936667919,
        "mr_stft": 1.1514151096343994
      },
      "treatment": {
        "log_mel": 0.47021111845970154,
        "log1p_mel": 0.003466957248747349,
        "mr_stft": 1.404052495956421
      },
      "decoder_drift": {
        "log_mel": 0.26262393593788147,
        "log1p_mel": 0.002138136886060238,
        "mr_stft": 0.891067624092102
      },
      "relative_percent": {
        "mr_stft": 21.941468737737836,
        "log_mel": 16.071505028522637,
        "log1p_mel": 24.190435585304716
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-07,
      "window_frames": 9,
      "window_ms": 417.9591836734694,
      "scalar_tokens": 90,
      "codebooks": 10,
      "seam_count": 12,
      "id": "fish-400m-morepitchloss1-step3802000--librispeech-other-500--log1p_mel--20-30",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 21.941479091022977,
        "log_mel": 16.071505028522637,
        "log1p_mel": 24.190435585304716
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "30-40",
      "audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-6.wav",
      "baseline": {
        "log_mel": 0.40510469675064087,
        "log1p_mel": 0.002791645936667919,
        "mr_stft": 1.1514151096343994
      },
      "treatment": {
        "log_mel": 0.5131045579910278,
        "log1p_mel": 0.003771210554987192,
        "mr_stft": 1.5390914678573608
      },
      "decoder_drift": {
        "log_mel": 0.357206791639328,
        "log1p_mel": 0.002598223276436329,
        "mr_stft": 1.1409868001937866
      },
      "relative_percent": {
        "mr_stft": 33.669556268551794,
        "log_mel": 26.659740582288393,
        "log1p_mel": 35.0891424106766
      },
      "recorded_metric_max_abs_difference": 5.960464477539062e-07,
      "window_frames": 6,
      "window_ms": 278.6394557823129,
      "scalar_tokens": 60,
      "codebooks": 10,
      "seam_count": 18,
      "id": "fish-400m-morepitchloss1-step3802000--librispeech-other-500--log1p_mel--30-40",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 33.669608034977536,
        "log_mel": 26.659755295681077,
        "log1p_mel": 35.0891424106766
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "40-50",
      "audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-8.wav",
      "baseline": {
        "log_mel": 0.40510469675064087,
        "log1p_mel": 0.002791645936667919,
        "mr_stft": 1.1514151096343994
      },
      "treatment": {
        "log_mel": 0.5197157263755798,
        "log1p_mel": 0.003920335788279772,
        "mr_stft": 1.5236324071884155
      },
      "decoder_drift": {
        "log_mel": 0.3291643559932709,
        "log1p_mel": 0.002570197219029069,
        "mr_stft": 1.067570686340332
      },
      "relative_percent": {
        "mr_stft": 32.326942250410774,
        "log_mel": 28.291705957555678,
        "log1p_mel": 40.43098147894233
      },
      "recorded_metric_max_abs_difference": 8.344650268554688e-07,
      "window_frames": 8,
      "window_ms": 371.51927437641723,
      "scalar_tokens": 80,
      "codebooks": 10,
      "seam_count": 13,
      "id": "fish-400m-morepitchloss1-step3802000--librispeech-other-500--log1p_mel--40-50",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 32.32701472340682,
        "log_mel": 28.291750097733704,
        "log1p_mel": 40.43098147894233
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "over-50",
      "audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-3.wav",
      "baseline": {
        "log_mel": 0.40510469675064087,
        "log1p_mel": 0.002791645936667919,
        "mr_stft": 1.1514151096343994
      },
      "treatment": {
        "log_mel": 0.5786410570144653,
        "log1p_mel": 0.004568345379084349,
        "mr_stft": 1.745819091796875
      },
      "decoder_drift": {
        "log_mel": 0.4892440736293793,
        "log1p_mel": 0.0038249765057116747,
        "mr_stft": 1.4805587530136108
      },
      "relative_percent": {
        "mr_stft": 51.623778182936334,
        "log_mel": 42.837410095653226,
        "log1p_mel": 63.64343769672598
      },
      "recorded_metric_max_abs_difference": 4.76837158203125e-07,
      "window_frames": 3,
      "window_ms": 139.31972789115645,
      "scalar_tokens": 30,
      "codebooks": 10,
      "seam_count": 36,
      "id": "fish-400m-morepitchloss1-step3802000--librispeech-other-500--log1p_mel--over-50",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 51.62373676979575,
        "log_mel": 42.83738066886789,
        "log1p_mel": 63.643404335669615
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "0-5",
      "audio": "listening-audio/librispeech-other-500-fish-dac-s2-window-104.wav",
      "baseline": {
        "log_mel": 0.312560111284256,
        "log1p_mel": 0.0020922308322042227,
        "mr_stft": 0.953180193901062
      },
      "treatment": {
        "log_mel": 0.32658255100250244,
        "log1p_mel": 0.002095166128128767,
        "mr_stft": 0.9771348237991333
      },
      "decoder_drift": {
        "log_mel": 0.027214888483285904,
        "log1p_mel": 9.203323315887246e-06,
        "mr_stft": 0.06590276956558228
      },
      "relative_percent": {
        "mr_stft": 2.5131271139859246,
        "log_mel": 4.486317739212042,
        "log1p_mel": 0.14029503242967145
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 104,
      "window_ms": 4829.750566893425,
      "scalar_tokens": 1040,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--librispeech-other-500--mr_stft--0-5",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 2.5131271139859246,
        "log_mel": 4.486317739212042,
        "log1p_mel": 0.14029503242967145
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "5-10",
      "audio": "listening-audio/librispeech-other-500-fish-dac-s2-window-94.wav",
      "baseline": {
        "log_mel": 0.312560111284256,
        "log1p_mel": 0.0020922308322042227,
        "mr_stft": 0.953180193901062
      },
      "treatment": {
        "log_mel": 0.34312930703163147,
        "log1p_mel": 0.002204885007813573,
        "mr_stft": 1.0319013595581055
      },
      "decoder_drift": {
        "log_mel": 0.05956544354557991,
        "log1p_mel": 0.00020297878654673696,
        "mr_stft": 0.2037234753370285
      },
      "relative_percent": {
        "mr_stft": 8.25879158639069,
        "log_mel": 9.780261346136566,
        "log1p_mel": 5.384404716503766
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 94,
      "window_ms": 4365.351473922902,
      "scalar_tokens": 940,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--librispeech-other-500--mr_stft--5-10",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 8.25879158639069,
        "log_mel": 9.780261346136566,
        "log1p_mel": 5.384404716503766
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "10-20",
      "audio": "listening-audio/librispeech-other-500-fish-dac-s2-window-84.wav",
      "baseline": {
        "log_mel": 0.312560111284256,
        "log1p_mel": 0.0020922308322042227,
        "mr_stft": 0.953180193901062
      },
      "treatment": {
        "log_mel": 0.3590467572212219,
        "log1p_mel": 0.002290960866957903,
        "mr_stft": 1.104896068572998
      },
      "decoder_drift": {
        "log_mel": 0.096099853515625,
        "log1p_mel": 0.0004135071358177811,
        "mr_stft": 0.37474164366722107
      },
      "relative_percent": {
        "mr_stft": 15.916809396868747,
        "log_mel": 14.87286581322238,
        "log1p_mel": 9.49847558380128
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 84,
      "window_ms": 3900.9523809523807,
      "scalar_tokens": 840,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--librispeech-other-500--mr_stft--10-20",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 15.916809396868747,
        "log_mel": 14.87286581322238,
        "log1p_mel": 9.49847558380128
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "20-30",
      "audio": "listening-audio/librispeech-other-500-fish-dac-s2-window-61.wav",
      "baseline": {
        "log_mel": 0.312560111284256,
        "log1p_mel": 0.0020922308322042227,
        "mr_stft": 0.953180193901062
      },
      "treatment": {
        "log_mel": 0.3859747350215912,
        "log1p_mel": 0.002372331917285919,
        "mr_stft": 1.1936979293823242
      },
      "decoder_drift": {
        "log_mel": 0.16705222427845,
        "log1p_mel": 0.0006669689901173115,
        "mr_stft": 0.6239050626754761
      },
      "relative_percent": {
        "mr_stft": 25.233186444726673,
        "log_mel": 23.48816150457813,
        "log1p_mel": 13.387676004497195
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 61,
      "window_ms": 2832.834467120181,
      "scalar_tokens": 610,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--librispeech-other-500--mr_stft--20-30",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 25.233186444726673,
        "log_mel": 23.48816150457813,
        "log1p_mel": 13.387676004497195
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "30-40",
      "audio": "listening-audio/librispeech-other-500-fish-dac-s2-window-69.wav",
      "baseline": {
        "log_mel": 0.312560111284256,
        "log1p_mel": 0.0020922308322042227,
        "mr_stft": 0.953180193901062
      },
      "treatment": {
        "log_mel": 0.39344918727874756,
        "log1p_mel": 0.0029086782597005367,
        "mr_stft": 1.285857081413269
      },
      "decoder_drift": {
        "log_mel": 0.15968763828277588,
        "log1p_mel": 0.001208504312671721,
        "mr_stft": 0.7058143615722656
      },
      "relative_percent": {
        "mr_stft": 34.90178348656898,
        "log_mel": 25.879526233252292,
        "log1p_mel": 39.02281789032638
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 69,
      "window_ms": 3204.3537414965986,
      "scalar_tokens": 690,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--librispeech-other-500--mr_stft--30-40",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 34.90178348656898,
        "log_mel": 25.879526233252292,
        "log1p_mel": 39.02281789032638
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "40-50",
      "audio": "listening-audio/librispeech-other-500-fish-dac-s2-window-26.wav",
      "baseline": {
        "log_mel": 0.312560111284256,
        "log1p_mel": 0.0020922308322042227,
        "mr_stft": 0.953180193901062
      },
      "treatment": {
        "log_mel": 0.4644201397895813,
        "log1p_mel": 0.002920235740020871,
        "mr_stft": 1.3855721950531006
      },
      "decoder_drift": {
        "log_mel": 0.32600221037864685,
        "log1p_mel": 0.0017677006544545293,
        "mr_stft": 1.1267317533493042
      },
      "relative_percent": {
        "mr_stft": 45.363091251654765,
        "log_mel": 48.585863334050686,
        "log1p_mel": 39.57521775665271
      },
      "recorded_metric_max_abs_difference": 9.059906005859375e-06,
      "window_frames": 26,
      "window_ms": 1207.4376417233561,
      "scalar_tokens": 260,
      "codebooks": 10,
      "seam_count": 4,
      "id": "fish-dac-s2--librispeech-other-500--mr_stft--40-50",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 45.36214075918086,
        "log_mel": 48.58581565950535,
        "log1p_mel": 39.575206628308756
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "over-50",
      "audio": "listening-audio/librispeech-other-500-fish-dac-s2-window-13.wav",
      "baseline": {
        "log_mel": 0.312560111284256,
        "log1p_mel": 0.0020922308322042227,
        "mr_stft": 0.953180193901062
      },
      "treatment": {
        "log_mel": 0.5366503596305847,
        "log1p_mel": 0.004136400762945414,
        "mr_stft": 1.6885923147201538
      },
      "decoder_drift": {
        "log_mel": 0.44999808073043823,
        "log1p_mel": 0.0035101952962577343,
        "mr_stft": 1.5938925743103027
      },
      "relative_percent": {
        "mr_stft": 77.15352517022882,
        "log_mel": 71.6950884825259,
        "log1p_mel": 97.70288723771468
      },
      "recorded_metric_max_abs_difference": 3.2186508178710938e-06,
      "window_frames": 13,
      "window_ms": 603.7188208616781,
      "scalar_tokens": 130,
      "codebooks": 10,
      "seam_count": 8,
      "id": "fish-dac-s2--librispeech-other-500--mr_stft--over-50",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 77.15386284518667,
        "log_mel": 71.69499313343523,
        "log1p_mel": 97.70288723771468
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "0-5",
      "audio": "listening-audio/librispeech-other-500-fish-dac-s2-window-107.wav",
      "baseline": {
        "log_mel": 0.312560111284256,
        "log1p_mel": 0.0020922308322042227,
        "mr_stft": 0.953180193901062
      },
      "treatment": {
        "log_mel": 0.3215210437774658,
        "log1p_mel": 0.002094396622851491,
        "mr_stft": 0.9677411913871765
      },
      "decoder_drift": {
        "log_mel": 0.015774572268128395,
        "log1p_mel": 5.366084224078804e-06,
        "mr_stft": 0.038296088576316833
      },
      "relative_percent": {
        "mr_stft": 1.5276227495371053,
        "log_mel": 2.8669469230705458,
        "log1p_mel": 0.1035158556088378
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 107,
      "window_ms": 4969.07029478458,
      "scalar_tokens": 1070,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--librispeech-other-500--log_mel--0-5",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 1.5276227495371053,
        "log_mel": 2.8669469230705458,
        "log1p_mel": 0.1035158556088378
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "5-10",
      "audio": "listening-audio/librispeech-other-500-fish-dac-s2-window-102.wav",
      "baseline": {
        "log_mel": 0.312560111284256,
        "log1p_mel": 0.0020922308322042227,
        "mr_stft": 0.953180193901062
      },
      "treatment": {
        "log_mel": 0.3353276252746582,
        "log1p_mel": 0.0020980979315936565,
        "mr_stft": 0.986586332321167
      },
      "decoder_drift": {
        "log_mel": 0.03932806849479675,
        "log1p_mel": 1.3819048035657033e-05,
        "mr_stft": 0.08703748881816864
      },
      "relative_percent": {
        "mr_stft": 3.504703374435869,
        "log_mel": 7.284203315917193,
        "log1p_mel": 0.2804231396997814
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 102,
      "window_ms": 4736.8707482993195,
      "scalar_tokens": 1020,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--librispeech-other-500--log_mel--5-10",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 3.504703374435869,
        "log_mel": 7.284203315917193,
        "log1p_mel": 0.2804231396997814
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "10-20",
      "audio": "listening-audio/librispeech-other-500-fish-dac-s2-window-84.wav",
      "baseline": {
        "log_mel": 0.312560111284256,
        "log1p_mel": 0.0020922308322042227,
        "mr_stft": 0.953180193901062
      },
      "treatment": {
        "log_mel": 0.3590467572212219,
        "log1p_mel": 0.002290960866957903,
        "mr_stft": 1.104896068572998
      },
      "decoder_drift": {
        "log_mel": 0.096099853515625,
        "log1p_mel": 0.0004135071358177811,
        "mr_stft": 0.37474164366722107
      },
      "relative_percent": {
        "mr_stft": 15.916809396868747,
        "log_mel": 14.87286581322238,
        "log1p_mel": 9.49847558380128
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 84,
      "window_ms": 3900.9523809523807,
      "scalar_tokens": 840,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--librispeech-other-500--log_mel--10-20",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 15.916809396868747,
        "log_mel": 14.87286581322238,
        "log1p_mel": 9.49847558380128
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "20-30",
      "audio": "listening-audio/librispeech-other-500-fish-dac-s2-window-65.wav",
      "baseline": {
        "log_mel": 0.312560111284256,
        "log1p_mel": 0.0020922308322042227,
        "mr_stft": 0.953180193901062
      },
      "treatment": {
        "log_mel": 0.39176562428474426,
        "log1p_mel": 0.0026066184509545565,
        "mr_stft": 1.241534948348999
      },
      "decoder_drift": {
        "log_mel": 0.16832250356674194,
        "log1p_mel": 0.0009394927765242755,
        "mr_stft": 0.6721748113632202
      },
      "relative_percent": {
        "mr_stft": 30.25186174586707,
        "log_mel": 25.340889685202118,
        "log1p_mel": 24.585605509331508
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 65,
      "window_ms": 3018.5941043083903,
      "scalar_tokens": 650,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--librispeech-other-500--log_mel--20-30",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 30.25186174586707,
        "log_mel": 25.340889685202118,
        "log1p_mel": 24.585605509331508
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "30-40",
      "audio": "listening-audio/librispeech-other-500-fish-dac-s2-window-35.wav",
      "baseline": {
        "log_mel": 0.312560111284256,
        "log1p_mel": 0.0020922308322042227,
        "mr_stft": 0.953180193901062
      },
      "treatment": {
        "log_mel": 0.4264054596424103,
        "log1p_mel": 0.002906706416979432,
        "mr_stft": 1.4450769424438477
      },
      "decoder_drift": {
        "log_mel": 0.26021015644073486,
        "log1p_mel": 0.0014569963095709682,
        "mr_stft": 1.1278598308563232
      },
      "relative_percent": {
        "mr_stft": 51.60585078143613,
        "log_mel": 36.42350519085218,
        "log1p_mel": 38.92857194524455
      },
      "recorded_metric_max_abs_difference": 3.814697265625e-06,
      "window_frames": 35,
      "window_ms": 1625.3968253968253,
      "scalar_tokens": 350,
      "codebooks": 10,
      "seam_count": 3,
      "id": "fish-dac-s2--librispeech-other-500--log_mel--30-40",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 51.60625098879357,
        "log_mel": 36.42340984176151,
        "log1p_mel": 38.928583073588506
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "40-50",
      "audio": "listening-audio/librispeech-other-500-fish-dac-s2-window-33.wav",
      "baseline": {
        "log_mel": 0.312560111284256,
        "log1p_mel": 0.0020922308322042227,
        "mr_stft": 0.953180193901062
      },
      "treatment": {
        "log_mel": 0.45321404933929443,
        "log1p_mel": 0.0034228013828396797,
        "mr_stft": 1.577036738395691
      },
      "decoder_drift": {
        "log_mel": 0.30246472358703613,
        "log1p_mel": 0.0021401375997811556,
        "mr_stft": 1.2985560894012451
      },
      "relative_percent": {
        "mr_stft": 65.45001128709814,
        "log_mel": 45.000604036489335,
        "log1p_mel": 63.595781600907976
      },
      "recorded_metric_max_abs_difference": 2.086162567138672e-07,
      "window_frames": 33,
      "window_ms": 1532.5170068027212,
      "scalar_tokens": 330,
      "codebooks": 10,
      "seam_count": 3,
      "id": "fish-dac-s2--librispeech-other-500--log_mel--40-50",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 65.44999878061822,
        "log_mel": 45.00053729212588,
        "log1p_mel": 63.59580385759593
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "over-50",
      "audio": "listening-audio/librispeech-other-500-fish-dac-s2-window-14.wav",
      "baseline": {
        "log_mel": 0.312560111284256,
        "log1p_mel": 0.0020922308322042227,
        "mr_stft": 0.953180193901062
      },
      "treatment": {
        "log_mel": 0.5446971654891968,
        "log1p_mel": 0.00419022748246789,
        "mr_stft": 1.711046576499939
      },
      "decoder_drift": {
        "log_mel": 0.4588513970375061,
        "log1p_mel": 0.0035359421744942665,
        "mr_stft": 1.6021203994750977
      },
      "relative_percent": {
        "mr_stft": 79.5092457279428,
        "log_mel": 74.26957113981352,
        "log1p_mel": 100.27558231007285
      },
      "recorded_metric_max_abs_difference": 3.4570693969726562e-06,
      "window_frames": 14,
      "window_ms": 650.1587301587301,
      "scalar_tokens": 140,
      "codebooks": 10,
      "seam_count": 7,
      "id": "fish-dac-s2--librispeech-other-500--log_mel--over-50",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 79.5088830400251,
        "log_mel": 74.26959020963164,
        "log1p_mel": 100.27558231007285
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "0-5",
      "audio": "listening-audio/librispeech-other-500-fish-dac-s2-window-97.wav",
      "baseline": {
        "log_mel": 0.312560111284256,
        "log1p_mel": 0.0020922308322042227,
        "mr_stft": 0.953180193901062
      },
      "treatment": {
        "log_mel": 0.3389350175857544,
        "log1p_mel": 0.0021330846939235926,
        "mr_stft": 1.0063025951385498
      },
      "decoder_drift": {
        "log_mel": 0.053214751183986664,
        "log1p_mel": 7.821156032150611e-05,
        "mr_stft": 0.15406779944896698
      },
      "relative_percent": {
        "mr_stft": 5.573175101349381,
        "log_mel": 8.43834684890803,
        "log1p_mel": 1.9526460030382609
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 97,
      "window_ms": 4504.671201814059,
      "scalar_tokens": 970,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--librispeech-other-500--log1p_mel--0-5",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 5.573175101349381,
        "log_mel": 8.43834684890803,
        "log1p_mel": 1.9526460030382609
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "5-10",
      "audio": "listening-audio/librispeech-other-500-fish-dac-s2-window-89.wav",
      "baseline": {
        "log_mel": 0.312560111284256,
        "log1p_mel": 0.0020922308322042227,
        "mr_stft": 0.953180193901062
      },
      "treatment": {
        "log_mel": 0.3603515625,
        "log1p_mel": 0.002230653539299965,
        "mr_stft": 1.0628594160079956
      },
      "decoder_drift": {
        "log_mel": 0.08478328585624695,
        "log1p_mel": 0.0003113337734248489,
        "mr_stft": 0.2742619216442108
      },
      "relative_percent": {
        "mr_stft": 11.506661889191339,
        "log_mel": 15.290323201952138,
        "log1p_mel": 6.616034185382413
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 89,
      "window_ms": 4133.151927437642,
      "scalar_tokens": 890,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--librispeech-other-500--log1p_mel--5-10",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 11.506661889191339,
        "log_mel": 15.290323201952138,
        "log1p_mel": 6.616034185382413
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "10-20",
      "audio": "listening-audio/librispeech-other-500-fish-dac-s2-window-79.wav",
      "baseline": {
        "log_mel": 0.312560111284256,
        "log1p_mel": 0.0020922308322042227,
        "mr_stft": 0.953180193901062
      },
      "treatment": {
        "log_mel": 0.36783355474472046,
        "log1p_mel": 0.0024058709386736155,
        "mr_stft": 1.1421151161193848
      },
      "decoder_drift": {
        "log_mel": 0.11277566105127335,
        "log1p_mel": 0.0005701177287846804,
        "mr_stft": 0.44846010208129883
      },
      "relative_percent": {
        "mr_stft": 19.821532531542907,
        "log_mel": 17.684100262619996,
        "log1p_mel": 14.990702825030278
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 79,
      "window_ms": 3668.7528344671205,
      "scalar_tokens": 790,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--librispeech-other-500--log1p_mel--10-20",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 19.821532531542907,
        "log_mel": 17.684100262619996,
        "log1p_mel": 14.990702825030278
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "20-30",
      "audio": "listening-audio/librispeech-other-500-fish-dac-s2-window-81.wav",
      "baseline": {
        "log_mel": 0.312560111284256,
        "log1p_mel": 0.0020922308322042227,
        "mr_stft": 0.953180193901062
      },
      "treatment": {
        "log_mel": 0.3729909658432007,
        "log1p_mel": 0.002619432285428047,
        "mr_stft": 1.185491919517517
      },
      "decoder_drift": {
        "log_mel": 0.11825831979513168,
        "log1p_mel": 0.0008418477955274284,
        "mr_stft": 0.5056176781654358
      },
      "relative_percent": {
        "mr_stft": 24.37227788647993,
        "log_mel": 19.33415441613604,
        "log1p_mel": 25.198053919719896
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 81,
      "window_ms": 3761.632653061224,
      "scalar_tokens": 810,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--librispeech-other-500--log1p_mel--20-30",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 24.37227788647993,
        "log_mel": 19.33415441613604,
        "log1p_mel": 25.198053919719896
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "30-40",
      "audio": "listening-audio/librispeech-other-500-fish-dac-s2-window-39.wav",
      "baseline": {
        "log_mel": 0.312560111284256,
        "log1p_mel": 0.0020922308322042227,
        "mr_stft": 0.953180193901062
      },
      "treatment": {
        "log_mel": 0.4274560809135437,
        "log1p_mel": 0.0028408889193087816,
        "mr_stft": 1.3959033489227295
      },
      "decoder_drift": {
        "log_mel": 0.2497773915529251,
        "log1p_mel": 0.0013887936947867274,
        "mr_stft": 1.0374021530151367
      },
      "relative_percent": {
        "mr_stft": 46.44695282743372,
        "log_mel": 36.759639340157605,
        "log1p_mel": 35.782767158431895
      },
      "recorded_metric_max_abs_difference": 5.245208740234375e-06,
      "window_frames": 39,
      "window_ms": 1811.156462585034,
      "scalar_tokens": 390,
      "codebooks": 10,
      "seam_count": 2,
      "id": "fish-dac-s2--librispeech-other-500--log1p_mel--30-40",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 46.447503112550194,
        "log_mel": 36.75975375906639,
        "log1p_mel": 35.78278941511983
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "40-50",
      "audio": "listening-audio/librispeech-other-500-fish-dac-s2-window-30.wav",
      "baseline": {
        "log_mel": 0.312560111284256,
        "log1p_mel": 0.0020922308322042227,
        "mr_stft": 0.953180193901062
      },
      "treatment": {
        "log_mel": 0.4618765711784363,
        "log1p_mel": 0.003048299578949809,
        "mr_stft": 1.470780372619629
      },
      "decoder_drift": {
        "log_mel": 0.3185935616493225,
        "log1p_mel": 0.0018729129806160927,
        "mr_stft": 1.1979916095733643
      },
      "relative_percent": {
        "mr_stft": 54.302447955847136,
        "log_mel": 47.77207791508153,
        "log1p_mel": 45.69614079046631
      },
      "recorded_metric_max_abs_difference": 1.430511474609375e-06,
      "window_frames": 30,
      "window_ms": 1393.1972789115646,
      "scalar_tokens": 300,
      "codebooks": 10,
      "seam_count": 3,
      "id": "fish-dac-s2--librispeech-other-500--log1p_mel--40-50",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 54.30229787808809,
        "log_mel": 47.77212558962687,
        "log1p_mel": 45.69612966212233
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "over-50",
      "audio": "listening-audio/librispeech-other-500-fish-dac-s2-window-24.wav",
      "baseline": {
        "log_mel": 0.312560111284256,
        "log1p_mel": 0.0020922308322042227,
        "mr_stft": 0.953180193901062
      },
      "treatment": {
        "log_mel": 0.5099600553512573,
        "log1p_mel": 0.0036294942256063223,
        "mr_stft": 1.5692453384399414
      },
      "decoder_drift": {
        "log_mel": 0.39107659459114075,
        "log1p_mel": 0.002567119197919965,
        "mr_stft": 1.3840492963790894
      },
      "relative_percent": {
        "mr_stft": 64.63260026601283,
        "log_mel": 63.15583369097189,
        "log1p_mel": 73.4748465484829
      },
      "recorded_metric_max_abs_difference": 7.152557373046875e-07,
      "window_frames": 24,
      "window_ms": 1114.5578231292516,
      "scalar_tokens": 240,
      "codebooks": 10,
      "seam_count": 4,
      "id": "fish-dac-s2--librispeech-other-500--log1p_mel--over-50",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 64.63252522713331,
        "log_mel": 63.155871830608135,
        "log1p_mel": 73.4748465484829
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "0-5",
      "audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-56.wav",
      "baseline": {
        "log_mel": 0.3318670094013214,
        "log1p_mel": 0.0022429267410188913,
        "mr_stft": 1.1133737564086914
      },
      "treatment": {
        "log_mel": 0.3457292318344116,
        "log1p_mel": 0.002273749327287078,
        "mr_stft": 1.1379976272583008
      },
      "decoder_drift": {
        "log_mel": 0.03111836314201355,
        "log1p_mel": 7.367459329543635e-05,
        "mr_stft": 0.08889728039503098
      },
      "relative_percent": {
        "mr_stft": 2.2116446258834355,
        "log_mel": 4.177041417312699,
        "log1p_mel": 1.3742127954738725
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 56,
      "window_ms": 4480.0,
      "scalar_tokens": 896,
      "codebooks": 16,
      "seam_count": 1,
      "id": "qwen3-tts-tokenizer-12hz--librispeech-other-500--mr_stft--0-5",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 2.2116446258834355,
        "log_mel": 4.177041417312699,
        "log1p_mel": 1.3742127954738725
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "5-10",
      "audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-48.wav",
      "baseline": {
        "log_mel": 0.3318670094013214,
        "log1p_mel": 0.0022429267410188913,
        "mr_stft": 1.1133737564086914
      },
      "treatment": {
        "log_mel": 0.3521125912666321,
        "log1p_mel": 0.0024463487789034843,
        "mr_stft": 1.1980257034301758
      },
      "decoder_drift": {
        "log_mel": 0.07104022055864334,
        "log1p_mel": 0.0004766416677739471,
        "mr_stft": 0.3203342854976654
      },
      "relative_percent": {
        "mr_stft": 7.603192237487111,
        "log_mel": 6.100510533370929,
        "log1p_mel": 9.069490954135428
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 48,
      "window_ms": 3840.0,
      "scalar_tokens": 768,
      "codebooks": 16,
      "seam_count": 1,
      "id": "qwen3-tts-tokenizer-12hz--librispeech-other-500--mr_stft--5-10",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 7.603192237487111,
        "log_mel": 6.100510533370929,
        "log1p_mel": 9.069490954135428
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "10-20",
      "audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-23.wav",
      "baseline": {
        "log_mel": 0.3318670094013214,
        "log1p_mel": 0.0022429267410188913,
        "mr_stft": 1.1133737564086914
      },
      "treatment": {
        "log_mel": 0.3699466586112976,
        "log1p_mel": 0.0025186650454998016,
        "mr_stft": 1.2729909420013428
      },
      "decoder_drift": {
        "log_mel": 0.13787542283535004,
        "log1p_mel": 0.0008082955027930439,
        "mr_stft": 0.5490803122520447
      },
      "relative_percent": {
        "mr_stft": 14.33635243096747,
        "log_mel": 11.47437019385289,
        "log1p_mel": 12.293683045378966
      },
      "recorded_metric_max_abs_difference": 4.76837158203125e-07,
      "window_frames": 23,
      "window_ms": 1840.0,
      "scalar_tokens": 368,
      "codebooks": 16,
      "seam_count": 2,
      "id": "qwen3-tts-tokenizer-12hz--librispeech-other-500--mr_stft--10-20",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 14.336395259097333,
        "log_mel": 11.47437019385289,
        "log1p_mel": 12.293683045378966
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "20-30",
      "audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-10.wav",
      "baseline": {
        "log_mel": 0.3318670094013214,
        "log1p_mel": 0.0022429267410188913,
        "mr_stft": 1.1133737564086914
      },
      "treatment": {
        "log_mel": 0.41093602776527405,
        "log1p_mel": 0.003106367774307728,
        "mr_stft": 1.3877489566802979
      },
      "decoder_drift": {
        "log_mel": 0.2475889027118683,
        "log1p_mel": 0.0020219136495143175,
        "mr_stft": 0.8451565504074097
      },
      "relative_percent": {
        "mr_stft": 24.6435843033191,
        "log_mel": 23.825513270086972,
        "log1p_mel": 38.49617633505953
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 10,
      "window_ms": 800.0,
      "scalar_tokens": 160,
      "codebooks": 16,
      "seam_count": 6,
      "id": "qwen3-tts-tokenizer-12hz--librispeech-other-500--mr_stft--20-30",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 24.643562889254156,
        "log_mel": 23.825522250287335,
        "log1p_mel": 38.496165954397554
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "30-40",
      "audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-5.wav",
      "baseline": {
        "log_mel": 0.3318670094013214,
        "log1p_mel": 0.0022429267410188913,
        "mr_stft": 1.1133737564086914
      },
      "treatment": {
        "log_mel": 0.4483236074447632,
        "log1p_mel": 0.0033980992157012224,
        "mr_stft": 1.44966459274292
      },
      "decoder_drift": {
        "log_mel": 0.31688255071640015,
        "log1p_mel": 0.002509737852960825,
        "mr_stft": 0.9830319881439209
      },
      "relative_percent": {
        "mr_stft": 30.204667066966916,
        "log_mel": 35.09134525107698,
        "log1p_mel": 51.50290705248681
      },
      "recorded_metric_max_abs_difference": 5.960464477539062e-07,
      "window_frames": 5,
      "window_ms": 400.0,
      "scalar_tokens": 80,
      "codebooks": 16,
      "seam_count": 12,
      "id": "qwen3-tts-tokenizer-12hz--librispeech-other-500--mr_stft--30-40",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 30.20461353180459,
        "log_mel": 35.09133627087662,
        "log1p_mel": 51.50290705248681
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "40-50",
      "audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-2.wav",
      "baseline": {
        "log_mel": 0.3318670094013214,
        "log1p_mel": 0.0022429267410188913,
        "mr_stft": 1.1133737564086914
      },
      "treatment": {
        "log_mel": 0.5057512521743774,
        "log1p_mel": 0.003984206356108189,
        "mr_stft": 1.63109290599823
      },
      "decoder_drift": {
        "log_mel": 0.4229949712753296,
        "log1p_mel": 0.003455677768215537,
        "mr_stft": 1.3035614490509033
      },
      "relative_percent": {
        "mr_stft": 46.500031692816094,
        "log_mel": 52.39576030372475,
        "log1p_mel": 77.63426166555438
      },
      "recorded_metric_max_abs_difference": 5.960464477539062e-07,
      "window_frames": 2,
      "window_ms": 160.0,
      "scalar_tokens": 32,
      "codebooks": 16,
      "seam_count": 32,
      "id": "qwen3-tts-tokenizer-12hz--librispeech-other-500--mr_stft--40-50",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 46.50008522797842,
        "log_mel": 52.395796224526194,
        "log1p_mel": 77.63426166555438
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "mr_stft",
      "band_id": "over-50",
      "audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-1.wav",
      "baseline": {
        "log_mel": 0.3318670094013214,
        "log1p_mel": 0.0022429267410188913,
        "mr_stft": 1.1133737564086914
      },
      "treatment": {
        "log_mel": 0.6031593084335327,
        "log1p_mel": 0.005139727611094713,
        "mr_stft": 1.906052589416504
      },
      "decoder_drift": {
        "log_mel": 0.5544953346252441,
        "log1p_mel": 0.00471937982365489,
        "mr_stft": 1.6634891033172607
      },
      "relative_percent": {
        "mr_stft": 71.19611257630902,
        "log_mel": 81.74729374926868,
        "log1p_mel": 129.1527189497012
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 1,
      "window_ms": 80.0,
      "scalar_tokens": 16,
      "codebooks": 16,
      "seam_count": 64,
      "id": "qwen3-tts-tokenizer-12hz--librispeech-other-500--mr_stft--over-50",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 71.19611257630902,
        "log_mel": 81.74729374926868,
        "log1p_mel": 129.1527189497012
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "0-5",
      "audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-58.wav",
      "baseline": {
        "log_mel": 0.3318670094013214,
        "log1p_mel": 0.0022429267410188913,
        "mr_stft": 1.1133737564086914
      },
      "treatment": {
        "log_mel": 0.339013010263443,
        "log1p_mel": 0.002244869014248252,
        "mr_stft": 1.1253836154937744
      },
      "decoder_drift": {
        "log_mel": 0.02751513198018074,
        "log1p_mel": 1.7290167306782678e-05,
        "mr_stft": 0.06589779257774353
      },
      "relative_percent": {
        "mr_stft": 1.078690692676476,
        "log_mel": 2.153272443384102,
        "log1p_mel": 0.0865954823151327
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 58,
      "window_ms": 4640.0,
      "scalar_tokens": 928,
      "codebooks": 16,
      "seam_count": 1,
      "id": "qwen3-tts-tokenizer-12hz--librispeech-other-500--log_mel--0-5",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 1.078690692676476,
        "log_mel": 2.153272443384102,
        "log1p_mel": 0.0865954823151327
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "5-10",
      "audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-42.wav",
      "baseline": {
        "log_mel": 0.3318670094013214,
        "log1p_mel": 0.0022429267410188913,
        "mr_stft": 1.1133737564086914
      },
      "treatment": {
        "log_mel": 0.35656052827835083,
        "log1p_mel": 0.0023167552426457405,
        "mr_stft": 1.159103512763977
      },
      "decoder_drift": {
        "log_mel": 0.08168972283601761,
        "log1p_mel": 0.0003050703089684248,
        "mr_stft": 0.2330283522605896
      },
      "relative_percent": {
        "mr_stft": 4.107314016704677,
        "log_mel": 7.440787477362032,
        "log1p_mel": 3.2916144908643474
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 42,
      "window_ms": 3360.0,
      "scalar_tokens": 672,
      "codebooks": 16,
      "seam_count": 1,
      "id": "qwen3-tts-tokenizer-12hz--librispeech-other-500--log_mel--5-10",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 4.107314016704677,
        "log_mel": 7.440787477362032,
        "log1p_mel": 3.2916144908643474
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "10-20",
      "audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-12.wav",
      "baseline": {
        "log_mel": 0.3318670094013214,
        "log1p_mel": 0.0022429267410188913,
        "mr_stft": 1.1133737564086914
      },
      "treatment": {
        "log_mel": 0.38318315148353577,
        "log1p_mel": 0.0026256260462105274,
        "mr_stft": 1.2684404850006104
      },
      "decoder_drift": {
        "log_mel": 0.19493219256401062,
        "log1p_mel": 0.0012100950116291642,
        "mr_stft": 0.6261050701141357
      },
      "relative_percent": {
        "mr_stft": 13.927643587729577,
        "log_mel": 15.462863324313926,
        "log1p_mel": 17.062496879313493
      },
      "recorded_metric_max_abs_difference": 9.5367431640625e-07,
      "window_frames": 12,
      "window_ms": 960.0,
      "scalar_tokens": 192,
      "codebooks": 16,
      "seam_count": 5,
      "id": "qwen3-tts-tokenizer-12hz--librispeech-other-500--log_mel--10-20",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 13.927729243989283,
        "log_mel": 15.46289924511537,
        "log1p_mel": 17.062496879313493
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "20-30",
      "audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-10.wav",
      "baseline": {
        "log_mel": 0.3318670094013214,
        "log1p_mel": 0.0022429267410188913,
        "mr_stft": 1.1133737564086914
      },
      "treatment": {
        "log_mel": 0.41093602776527405,
        "log1p_mel": 0.003106367774307728,
        "mr_stft": 1.3877489566802979
      },
      "decoder_drift": {
        "log_mel": 0.2475889027118683,
        "log1p_mel": 0.0020219136495143175,
        "mr_stft": 0.8451565504074097
      },
      "relative_percent": {
        "mr_stft": 24.6435843033191,
        "log_mel": 23.825513270086972,
        "log1p_mel": 38.49617633505953
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 10,
      "window_ms": 800.0,
      "scalar_tokens": 160,
      "codebooks": 16,
      "seam_count": 6,
      "id": "qwen3-tts-tokenizer-12hz--librispeech-other-500--log_mel--20-30",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 24.643562889254156,
        "log_mel": 23.825522250287335,
        "log1p_mel": 38.496165954397554
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "30-40",
      "audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-5.wav",
      "baseline": {
        "log_mel": 0.3318670094013214,
        "log1p_mel": 0.0022429267410188913,
        "mr_stft": 1.1133737564086914
      },
      "treatment": {
        "log_mel": 0.4483236074447632,
        "log1p_mel": 0.0033980992157012224,
        "mr_stft": 1.44966459274292
      },
      "decoder_drift": {
        "log_mel": 0.31688255071640015,
        "log1p_mel": 0.002509737852960825,
        "mr_stft": 0.9830319881439209
      },
      "relative_percent": {
        "mr_stft": 30.204667066966916,
        "log_mel": 35.09134525107698,
        "log1p_mel": 51.50290705248681
      },
      "recorded_metric_max_abs_difference": 5.960464477539062e-07,
      "window_frames": 5,
      "window_ms": 400.0,
      "scalar_tokens": 80,
      "codebooks": 16,
      "seam_count": 12,
      "id": "qwen3-tts-tokenizer-12hz--librispeech-other-500--log_mel--30-40",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 30.20461353180459,
        "log_mel": 35.09133627087662,
        "log1p_mel": 51.50290705248681
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "40-50",
      "audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-3.wav",
      "baseline": {
        "log_mel": 0.3318670094013214,
        "log1p_mel": 0.0022429267410188913,
        "mr_stft": 1.1133737564086914
      },
      "treatment": {
        "log_mel": 0.46678003668785095,
        "log1p_mel": 0.003827260108664632,
        "mr_stft": 1.5797224044799805
      },
      "decoder_drift": {
        "log_mel": 0.36323878169059753,
        "log1p_mel": 0.003102920949459076,
        "mr_stft": 1.1924428939819336
      },
      "relative_percent": {
        "mr_stft": 41.88608231395246,
        "log_mel": 40.65273843576944,
        "log1p_mel": 70.63687541243668
      },
      "recorded_metric_max_abs_difference": 5.960464477539063e-08,
      "window_frames": 3,
      "window_ms": 240.0,
      "scalar_tokens": 48,
      "codebooks": 16,
      "seam_count": 21,
      "id": "qwen3-tts-tokenizer-12hz--librispeech-other-500--log_mel--40-50",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 41.88608231395246,
        "log_mel": 40.652720475368696,
        "log1p_mel": 70.63687541243668
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log_mel",
      "band_id": "over-50",
      "audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-1.wav",
      "baseline": {
        "log_mel": 0.3318670094013214,
        "log1p_mel": 0.0022429267410188913,
        "mr_stft": 1.1133737564086914
      },
      "treatment": {
        "log_mel": 0.6031593084335327,
        "log1p_mel": 0.005139727611094713,
        "mr_stft": 1.906052589416504
      },
      "decoder_drift": {
        "log_mel": 0.5544953346252441,
        "log1p_mel": 0.00471937982365489,
        "mr_stft": 1.6634891033172607
      },
      "relative_percent": {
        "mr_stft": 71.19611257630902,
        "log_mel": 81.74729374926868,
        "log1p_mel": 129.1527189497012
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 1,
      "window_ms": 80.0,
      "scalar_tokens": 16,
      "codebooks": 16,
      "seam_count": 64,
      "id": "qwen3-tts-tokenizer-12hz--librispeech-other-500--log_mel--over-50",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 71.19611257630902,
        "log_mel": 81.74729374926868,
        "log1p_mel": 129.1527189497012
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "0-5",
      "audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-47.wav",
      "baseline": {
        "log_mel": 0.3318670094013214,
        "log1p_mel": 0.0022429267410188913,
        "mr_stft": 1.1133737564086914
      },
      "treatment": {
        "log_mel": 0.3570224940776825,
        "log1p_mel": 0.002297451952472329,
        "mr_stft": 1.1672046184539795
      },
      "decoder_drift": {
        "log_mel": 0.080325186252594,
        "log1p_mel": 0.00029899674700573087,
        "mr_stft": 0.23693129420280457
      },
      "relative_percent": {
        "mr_stft": 4.834931821900068,
        "log_mel": 7.579989563211131,
        "log1p_mel": 2.4309849473134637
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 47,
      "window_ms": 3760.0,
      "scalar_tokens": 752,
      "codebooks": 16,
      "seam_count": 1,
      "id": "qwen3-tts-tokenizer-12hz--librispeech-other-500--log1p_mel--0-5",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 4.834931821900068,
        "log_mel": 7.579989563211131,
        "log1p_mel": 2.4309849473134637
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "5-10",
      "audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-48.wav",
      "baseline": {
        "log_mel": 0.3318670094013214,
        "log1p_mel": 0.0022429267410188913,
        "mr_stft": 1.1133737564086914
      },
      "treatment": {
        "log_mel": 0.3521125912666321,
        "log1p_mel": 0.0024463487789034843,
        "mr_stft": 1.1980257034301758
      },
      "decoder_drift": {
        "log_mel": 0.07104022055864334,
        "log1p_mel": 0.0004766416677739471,
        "mr_stft": 0.3203342854976654
      },
      "relative_percent": {
        "mr_stft": 7.603192237487111,
        "log_mel": 6.100510533370929,
        "log1p_mel": 9.069490954135428
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 48,
      "window_ms": 3840.0,
      "scalar_tokens": 768,
      "codebooks": 16,
      "seam_count": 1,
      "id": "qwen3-tts-tokenizer-12hz--librispeech-other-500--log1p_mel--5-10",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 7.603192237487111,
        "log_mel": 6.100510533370929,
        "log1p_mel": 9.069490954135428
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "10-20",
      "audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-15.wav",
      "baseline": {
        "log_mel": 0.3318670094013214,
        "log1p_mel": 0.0022429267410188913,
        "mr_stft": 1.1133737564086914
      },
      "treatment": {
        "log_mel": 0.37490904331207275,
        "log1p_mel": 0.0025636828504502773,
        "mr_stft": 1.2188093662261963
      },
      "decoder_drift": {
        "log_mel": 0.17799849808216095,
        "log1p_mel": 0.0010050327982753515,
        "mr_stft": 0.5394338369369507
      },
      "relative_percent": {
        "mr_stft": 9.469920519556618,
        "log_mel": 12.969663356534888,
        "log1p_mel": 14.300784041019398
      },
      "recorded_metric_max_abs_difference": 3.5762786865234375e-07,
      "window_frames": 15,
      "window_ms": 1200.0,
      "scalar_tokens": 240,
      "codebooks": 16,
      "seam_count": 4,
      "id": "qwen3-tts-tokenizer-12hz--librispeech-other-500--log1p_mel--10-20",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 9.469888398459215,
        "log_mel": 12.96967233673525,
        "log1p_mel": 14.300784041019398
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "20-30",
      "audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-7.wav",
      "baseline": {
        "log_mel": 0.3318670094013214,
        "log1p_mel": 0.0022429267410188913,
        "mr_stft": 1.1133737564086914
      },
      "treatment": {
        "log_mel": 0.42705652117729187,
        "log1p_mel": 0.0028320015408098698,
        "mr_stft": 1.3481407165527344
      },
      "decoder_drift": {
        "log_mel": 0.273343950510025,
        "log1p_mel": 0.00170517445076257,
        "mr_stft": 0.7845331430435181
      },
      "relative_percent": {
        "mr_stft": 21.086087110703012,
        "log_mel": 28.6830293700147,
        "log1p_mel": 26.2636665307838
      },
      "recorded_metric_max_abs_difference": 5.960464477539063e-08,
      "window_frames": 7,
      "window_ms": 560.0,
      "scalar_tokens": 112,
      "codebooks": 16,
      "seam_count": 9,
      "id": "qwen3-tts-tokenizer-12hz--librispeech-other-500--log1p_mel--20-30",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 21.086087110703012,
        "log_mel": 28.683047330415423,
        "log1p_mel": 26.263687292107775
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "30-40",
      "audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-6.wav",
      "baseline": {
        "log_mel": 0.3318670094013214,
        "log1p_mel": 0.0022429267410188913,
        "mr_stft": 1.1133737564086914
      },
      "treatment": {
        "log_mel": 0.42036429047584534,
        "log1p_mel": 0.003058828879147768,
        "mr_stft": 1.4175219535827637
      },
      "decoder_drift": {
        "log_mel": 0.27947092056274414,
        "log1p_mel": 0.002072487026453018,
        "mr_stft": 0.9350509643554688
      },
      "relative_percent": {
        "mr_stft": 27.317708489477567,
        "log_mel": 26.666489457379484,
        "log1p_mel": 36.37667353139844
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 6,
      "window_ms": 480.0,
      "scalar_tokens": 96,
      "codebooks": 16,
      "seam_count": 10,
      "id": "qwen3-tts-tokenizer-12hz--librispeech-other-500--log1p_mel--30-40",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 27.317729903542485,
        "log_mel": 26.666498437579843,
        "log1p_mel": 36.37666315073644
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "40-50",
      "audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-4.wav",
      "baseline": {
        "log_mel": 0.3318670094013214,
        "log1p_mel": 0.0022429267410188913,
        "mr_stft": 1.1133737564086914
      },
      "treatment": {
        "log_mel": 0.44054147601127625,
        "log1p_mel": 0.003272973233833909,
        "mr_stft": 1.4137325286865234
      },
      "decoder_drift": {
        "log_mel": 0.31930941343307495,
        "log1p_mel": 0.0025423087645322084,
        "mr_stft": 0.9857778549194336
      },
      "relative_percent": {
        "mr_stft": 26.977353341493515,
        "log_mel": 32.746390430914005,
        "log1p_mel": 45.92421473146733
      },
      "recorded_metric_max_abs_difference": 7.152557373046875e-07,
      "window_frames": 4,
      "window_ms": 320.0,
      "scalar_tokens": 64,
      "codebooks": 16,
      "seam_count": 16,
      "id": "qwen3-tts-tokenizer-12hz--librispeech-other-500--log1p_mel--40-50",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 26.9774175836883,
        "log_mel": 32.746417371515115,
        "log1p_mel": 45.9242354927913
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "metric": "log1p_mel",
      "band_id": "over-50",
      "audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-2.wav",
      "baseline": {
        "log_mel": 0.3318670094013214,
        "log1p_mel": 0.0022429267410188913,
        "mr_stft": 1.1133737564086914
      },
      "treatment": {
        "log_mel": 0.5057512521743774,
        "log1p_mel": 0.003984206356108189,
        "mr_stft": 1.63109290599823
      },
      "decoder_drift": {
        "log_mel": 0.4229949712753296,
        "log1p_mel": 0.003455677768215537,
        "mr_stft": 1.3035614490509033
      },
      "relative_percent": {
        "mr_stft": 46.500031692816094,
        "log_mel": 52.39576030372475,
        "log1p_mel": 77.63426166555438
      },
      "recorded_metric_max_abs_difference": 5.960464477539062e-07,
      "window_frames": 2,
      "window_ms": 160.0,
      "scalar_tokens": 32,
      "codebooks": 16,
      "seam_count": 32,
      "id": "qwen3-tts-tokenizer-12hz--librispeech-other-500--log1p_mel--over-50",
      "original_audio": "listening-audio/librispeech-other-500-original.wav",
      "full_audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 46.50008522797842,
        "log_mel": 52.395796224526194,
        "log1p_mel": 77.63426166555438
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "0-5",
      "audio": "listening-audio/podcast-800-descript-dac-44khz-window-86.wav",
      "baseline": {
        "log_mel": 0.28345760703086853,
        "log1p_mel": 0.007421256974339485,
        "mr_stft": 0.8638533353805542
      },
      "treatment": {
        "log_mel": 0.29972562193870544,
        "log1p_mel": 0.007923826575279236,
        "mr_stft": 0.8854405283927917
      },
      "decoder_drift": {
        "log_mel": 0.033553335815668106,
        "log1p_mel": 0.0010346815688535571,
        "mr_stft": 0.12041231244802475
      },
      "relative_percent": {
        "mr_stft": 2.49894190693003,
        "log_mel": 5.739135060878908,
        "log1p_mel": 6.772028009237352
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 86,
      "window_ms": 998.4580498866213,
      "scalar_tokens": 774,
      "codebooks": 9,
      "seam_count": 10,
      "id": "descript-dac-44khz--podcast-800--mr_stft--0-5",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 2.49894190693003,
        "log_mel": 5.739135060878908,
        "log1p_mel": 6.772028009237352
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "5-10",
      "audio": "listening-audio/podcast-800-descript-dac-44khz-window-33.wav",
      "baseline": {
        "log_mel": 0.28345760703086853,
        "log1p_mel": 0.007421256974339485,
        "mr_stft": 0.8638533353805542
      },
      "treatment": {
        "log_mel": 0.32057592272758484,
        "log1p_mel": 0.009024896658957005,
        "mr_stft": 0.9319227933883667
      },
      "decoder_drift": {
        "log_mel": 0.08461333066225052,
        "log1p_mel": 0.0028745317831635475,
        "mr_stft": 0.2720930576324463
      },
      "relative_percent": {
        "mr_stft": 7.879747084361921,
        "log_mel": 13.09483844357513,
        "log1p_mel": 21.608734075136216
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 33,
      "window_ms": 383.1292517006803,
      "scalar_tokens": 297,
      "codebooks": 9,
      "seam_count": 27,
      "id": "descript-dac-44khz--podcast-800--mr_stft--5-10",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 7.879747084361921,
        "log_mel": 13.09483844357513,
        "log1p_mel": 21.608734075136216
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "10-20",
      "audio": "listening-audio/podcast-800-descript-dac-44khz-window-14.wav",
      "baseline": {
        "log_mel": 0.28345760703086853,
        "log1p_mel": 0.007421256974339485,
        "mr_stft": 0.8638533353805542
      },
      "treatment": {
        "log_mel": 0.37790539860725403,
        "log1p_mel": 0.010366993956267834,
        "mr_stft": 0.9911428689956665
      },
      "decoder_drift": {
        "log_mel": 0.2086612582206726,
        "log1p_mel": 0.005783242639154196,
        "mr_stft": 0.5066355466842651
      },
      "relative_percent": {
        "mr_stft": 14.735086200603398,
        "log_mel": 33.31990013099213,
        "log1p_mel": 39.693235150242565
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 14,
      "window_ms": 162.53968253968253,
      "scalar_tokens": 126,
      "codebooks": 9,
      "seam_count": 64,
      "id": "descript-dac-44khz--podcast-800--mr_stft--10-20",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 14.735086200603398,
        "log_mel": 33.31990013099213,
        "log1p_mel": 39.693235150242565
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "20-30",
      "audio": "listening-audio/podcast-800-descript-dac-44khz-window-9.wav",
      "baseline": {
        "log_mel": 0.28345760703086853,
        "log1p_mel": 0.007421256974339485,
        "mr_stft": 0.8638533353805542
      },
      "treatment": {
        "log_mel": 0.4402306377887726,
        "log1p_mel": 0.012929127551615238,
        "mr_stft": 1.0847861766815186
      },
      "decoder_drift": {
        "log_mel": 0.32549455761909485,
        "log1p_mel": 0.009715450927615166,
        "mr_stft": 0.7379852533340454
      },
      "relative_percent": {
        "mr_stft": 25.575272126909887,
        "log_mel": 55.30739936742339,
        "log1p_mel": 74.21748898226195
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 9,
      "window_ms": 104.48979591836735,
      "scalar_tokens": 81,
      "codebooks": 9,
      "seam_count": 101,
      "id": "descript-dac-44khz--podcast-800--mr_stft--20-30",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 25.575272126909887,
        "log_mel": 55.30739936742339,
        "log1p_mel": 74.21748898226195
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "30-40",
      "audio": "listening-audio/podcast-800-descript-dac-44khz-window-6.wav",
      "baseline": {
        "log_mel": 0.28345760703086853,
        "log1p_mel": 0.007421256974339485,
        "mr_stft": 0.8638533353805542
      },
      "treatment": {
        "log_mel": 0.5198338627815247,
        "log1p_mel": 0.015457912348210812,
        "mr_stft": 1.179431438446045
      },
      "decoder_drift": {
        "log_mel": 0.4462812840938568,
        "log1p_mel": 0.013212451711297035,
        "mr_stft": 0.9500772356987
      },
      "relative_percent": {
        "mr_stft": 36.5314446492782,
        "log_mel": 83.39033770397799,
        "log1p_mel": 108.29237421180409
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 6,
      "window_ms": 69.65986394557822,
      "scalar_tokens": 54,
      "codebooks": 9,
      "seam_count": 151,
      "id": "descript-dac-44khz--podcast-800--mr_stft--30-40",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 36.5314446492782,
        "log_mel": 83.39033770397799,
        "log1p_mel": 108.29237421180409
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "40-50",
      "audio": "listening-audio/podcast-800-descript-dac-44khz-window-5.wav",
      "baseline": {
        "log_mel": 0.28345760703086853,
        "log1p_mel": 0.007421256974339485,
        "mr_stft": 0.8638533353805542
      },
      "treatment": {
        "log_mel": 0.5637060403823853,
        "log1p_mel": 0.016916116699576378,
        "mr_stft": 1.231036901473999
      },
      "decoder_drift": {
        "log_mel": 0.5112473368644714,
        "log1p_mel": 0.015083868987858295,
        "mr_stft": 1.0541346073150635
      },
      "relative_percent": {
        "mr_stft": 42.505313234877896,
        "log_mel": 98.86784704317273,
        "log1p_mel": 127.94139534673592
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 5,
      "window_ms": 58.049886621315196,
      "scalar_tokens": 45,
      "codebooks": 9,
      "seam_count": 181,
      "id": "descript-dac-44khz--podcast-800--mr_stft--40-50",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 42.505313234877896,
        "log_mel": 98.86784704317273,
        "log1p_mel": 127.94139534673592
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "over-50",
      "audio": "listening-audio/podcast-800-descript-dac-44khz-window-3.wav",
      "baseline": {
        "log_mel": 0.28345760703086853,
        "log1p_mel": 0.007421256974339485,
        "mr_stft": 0.8638533353805542
      },
      "treatment": {
        "log_mel": 0.7512640953063965,
        "log1p_mel": 0.022852040827274323,
        "mr_stft": 1.4684619903564453
      },
      "decoder_drift": {
        "log_mel": 0.7377479076385498,
        "log1p_mel": 0.021755147725343704,
        "mr_stft": 1.3902009725570679
      },
      "relative_percent": {
        "mr_stft": 69.98973439276497,
        "log_mel": 165.0357854832888,
        "log1p_mel": 207.92682299359706
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 3,
      "window_ms": 34.82993197278911,
      "scalar_tokens": 27,
      "codebooks": 9,
      "seam_count": 303,
      "id": "descript-dac-44khz--podcast-800--mr_stft--over-50",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 69.98973439276497,
        "log_mel": 165.0357854832888,
        "log1p_mel": 207.92682299359706
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "0-5",
      "audio": "listening-audio/podcast-800-descript-dac-44khz-window-135.wav",
      "baseline": {
        "log_mel": 0.28345760703086853,
        "log1p_mel": 0.007421256974339485,
        "mr_stft": 0.8638533353805542
      },
      "treatment": {
        "log_mel": 0.2907416820526123,
        "log1p_mel": 0.0076684546656906605,
        "mr_stft": 0.8729104399681091
      },
      "decoder_drift": {
        "log_mel": 0.018184104934334755,
        "log1p_mel": 0.00045170856174081564,
        "mr_stft": 0.06772308051586151
      },
      "relative_percent": {
        "mr_stft": 1.048453969743024,
        "log_mel": 2.569722893677895,
        "log1p_mel": 3.3309410010448515
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 135,
      "window_ms": 1567.3469387755101,
      "scalar_tokens": 1215,
      "codebooks": 9,
      "seam_count": 6,
      "id": "descript-dac-44khz--podcast-800--log_mel--0-5",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 1.048453969743024,
        "log_mel": 2.569722893677895,
        "log1p_mel": 3.3309410010448515
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "5-10",
      "audio": "listening-audio/podcast-800-descript-dac-44khz-window-64.wav",
      "baseline": {
        "log_mel": 0.28345760703086853,
        "log1p_mel": 0.007421256974339485,
        "mr_stft": 0.8638533353805542
      },
      "treatment": {
        "log_mel": 0.3047131896018982,
        "log1p_mel": 0.008023688569664955,
        "mr_stft": 0.8920907974243164
      },
      "decoder_drift": {
        "log_mel": 0.04716638848185539,
        "log1p_mel": 0.0012193956645205617,
        "mr_stft": 0.14179927110671997
      },
      "relative_percent": {
        "mr_stft": 3.268779651272946,
        "log_mel": 7.498681299710164,
        "log1p_mel": 8.11764903719816
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 64,
      "window_ms": 743.0385487528345,
      "scalar_tokens": 576,
      "codebooks": 9,
      "seam_count": 14,
      "id": "descript-dac-44khz--podcast-800--log_mel--5-10",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 3.268779651272946,
        "log_mel": 7.498681299710164,
        "log1p_mel": 8.11764903719816
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "10-20",
      "audio": "listening-audio/podcast-800-descript-dac-44khz-window-38.wav",
      "baseline": {
        "log_mel": 0.28345760703086853,
        "log1p_mel": 0.007421256974339485,
        "mr_stft": 0.8638533353805542
      },
      "treatment": {
        "log_mel": 0.32604357600212097,
        "log1p_mel": 0.008726288564503193,
        "mr_stft": 0.9225262403488159
      },
      "decoder_drift": {
        "log_mel": 0.0841037854552269,
        "log1p_mel": 0.0023222730960696936,
        "mr_stft": 0.23284664750099182
      },
      "relative_percent": {
        "mr_stft": 6.791998429040569,
        "log_mel": 15.023752375999845,
        "log1p_mel": 17.585047852083836
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 38,
      "window_ms": 441.1791383219955,
      "scalar_tokens": 342,
      "codebooks": 9,
      "seam_count": 23,
      "id": "descript-dac-44khz--podcast-800--log_mel--10-20",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 6.791998429040569,
        "log_mel": 15.023752375999845,
        "log1p_mel": 17.585047852083836
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "20-30",
      "audio": "listening-audio/podcast-800-descript-dac-44khz-window-20.wav",
      "baseline": {
        "log_mel": 0.28345760703086853,
        "log1p_mel": 0.007421256974339485,
        "mr_stft": 0.8638533353805542
      },
      "treatment": {
        "log_mel": 0.3556617498397827,
        "log1p_mel": 0.009705289267003536,
        "mr_stft": 0.9655972719192505
      },
      "decoder_drift": {
        "log_mel": 0.15377257764339447,
        "log1p_mel": 0.004320144187659025,
        "mr_stft": 0.3900473117828369
      },
      "relative_percent": {
        "mr_stft": 11.777917890873791,
        "log_mel": 25.47264247561758,
        "log1p_mel": 30.77689265526797
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 20,
      "window_ms": 232.19954648526078,
      "scalar_tokens": 180,
      "codebooks": 9,
      "seam_count": 45,
      "id": "descript-dac-44khz--podcast-800--log_mel--20-30",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 11.777917890873791,
        "log_mel": 25.47264247561758,
        "log1p_mel": 30.77689265526797
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "30-40",
      "audio": "listening-audio/podcast-800-descript-dac-44khz-window-14.wav",
      "baseline": {
        "log_mel": 0.28345760703086853,
        "log1p_mel": 0.007421256974339485,
        "mr_stft": 0.8638533353805542
      },
      "treatment": {
        "log_mel": 0.37790539860725403,
        "log1p_mel": 0.010366993956267834,
        "mr_stft": 0.9911428689956665
      },
      "decoder_drift": {
        "log_mel": 0.2086612582206726,
        "log1p_mel": 0.005783242639154196,
        "mr_stft": 0.5066355466842651
      },
      "relative_percent": {
        "mr_stft": 14.735086200603398,
        "log_mel": 33.31990013099213,
        "log1p_mel": 39.693235150242565
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 14,
      "window_ms": 162.53968253968253,
      "scalar_tokens": 126,
      "codebooks": 9,
      "seam_count": 64,
      "id": "descript-dac-44khz--podcast-800--log_mel--30-40",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 14.735086200603398,
        "log_mel": 33.31990013099213,
        "log1p_mel": 39.693235150242565
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "40-50",
      "audio": "listening-audio/podcast-800-descript-dac-44khz-window-10.wav",
      "baseline": {
        "log_mel": 0.28345760703086853,
        "log1p_mel": 0.007421256974339485,
        "mr_stft": 0.8638533353805542
      },
      "treatment": {
        "log_mel": 0.4184426963329315,
        "log1p_mel": 0.01198651548475027,
        "mr_stft": 1.051201343536377
      },
      "decoder_drift": {
        "log_mel": 0.28649768233299255,
        "log1p_mel": 0.00845711212605238,
        "mr_stft": 0.6669464111328125
      },
      "relative_percent": {
        "mr_stft": 21.687478705316355,
        "log_mel": 47.62090907207974,
        "log1p_mel": 61.51597399464943
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 10,
      "window_ms": 116.09977324263039,
      "scalar_tokens": 90,
      "codebooks": 9,
      "seam_count": 90,
      "id": "descript-dac-44khz--podcast-800--log_mel--40-50",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 21.687478705316355,
        "log_mel": 47.62090907207974,
        "log1p_mel": 61.51597399464943
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "over-50",
      "audio": "listening-audio/podcast-800-descript-dac-44khz-window-7.wav",
      "baseline": {
        "log_mel": 0.28345760703086853,
        "log1p_mel": 0.007421256974339485,
        "mr_stft": 0.8638533353805542
      },
      "treatment": {
        "log_mel": 0.482290118932724,
        "log1p_mel": 0.014044269919395447,
        "mr_stft": 1.133995532989502
      },
      "decoder_drift": {
        "log_mel": 0.3957536220550537,
        "log1p_mel": 0.011466650292277336,
        "mr_stft": 0.866928219795227
      },
      "relative_percent": {
        "mr_stft": 31.271766461368312,
        "log_mel": 70.1454139772663,
        "log1p_mel": 89.2438163502003
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 7,
      "window_ms": 81.26984126984127,
      "scalar_tokens": 63,
      "codebooks": 9,
      "seam_count": 129,
      "id": "descript-dac-44khz--podcast-800--log_mel--over-50",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 31.271766461368312,
        "log_mel": 70.1454139772663,
        "log1p_mel": 89.2438163502003
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "0-5",
      "audio": "listening-audio/podcast-800-descript-dac-44khz-window-142.wav",
      "baseline": {
        "log_mel": 0.28345760703086853,
        "log1p_mel": 0.007421256974339485,
        "mr_stft": 0.8638533353805542
      },
      "treatment": {
        "log_mel": 0.2936452031135559,
        "log1p_mel": 0.007609981577843428,
        "mr_stft": 0.874319314956665
      },
      "decoder_drift": {
        "log_mel": 0.022408373653888702,
        "log1p_mel": 0.00044966311543248594,
        "mr_stft": 0.07222670316696167
      },
      "relative_percent": {
        "mr_stft": 1.2115458894998987,
        "log_mel": 3.5940457514614943,
        "log1p_mel": 2.543027470366499
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 142,
      "window_ms": 1648.6167800453516,
      "scalar_tokens": 1278,
      "codebooks": 9,
      "seam_count": 6,
      "id": "descript-dac-44khz--podcast-800--log1p_mel--0-5",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 1.2115458894998987,
        "log_mel": 3.5940457514614943,
        "log1p_mel": 2.543027470366499
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "5-10",
      "audio": "listening-audio/podcast-800-descript-dac-44khz-window-88.wav",
      "baseline": {
        "log_mel": 0.28345760703086853,
        "log1p_mel": 0.007421256974339485,
        "mr_stft": 0.8638533353805542
      },
      "treatment": {
        "log_mel": 0.29863375425338745,
        "log1p_mel": 0.007976552471518517,
        "mr_stft": 0.8885992765426636
      },
      "decoder_drift": {
        "log_mel": 0.03391796350479126,
        "log1p_mel": 0.000987869338132441,
        "mr_stft": 0.12191741168498993
      },
      "relative_percent": {
        "mr_stft": 2.8645998283039598,
        "log_mel": 5.353938947514725,
        "log1p_mel": 7.482499246409047
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 88,
      "window_ms": 1021.6780045351475,
      "scalar_tokens": 792,
      "codebooks": 9,
      "seam_count": 10,
      "id": "descript-dac-44khz--podcast-800--log1p_mel--5-10",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 2.8645998283039598,
        "log_mel": 5.353938947514725,
        "log1p_mel": 7.482499246409047
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "10-20",
      "audio": "listening-audio/podcast-800-descript-dac-44khz-window-44.wav",
      "baseline": {
        "log_mel": 0.28345760703086853,
        "log1p_mel": 0.007421256974339485,
        "mr_stft": 0.8638533353805542
      },
      "treatment": {
        "log_mel": 0.309611052274704,
        "log1p_mel": 0.008489003404974937,
        "mr_stft": 0.9051465392112732
      },
      "decoder_drift": {
        "log_mel": 0.06362567096948624,
        "log1p_mel": 0.0019436792936176062,
        "mr_stft": 0.19881632924079895
      },
      "relative_percent": {
        "mr_stft": 4.780117427286212,
        "log_mel": 9.226580834356412,
        "log1p_mel": 14.387676296985852
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 44,
      "window_ms": 510.83900226757373,
      "scalar_tokens": 396,
      "codebooks": 9,
      "seam_count": 20,
      "id": "descript-dac-44khz--podcast-800--log1p_mel--10-20",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 4.780117427286212,
        "log_mel": 9.226580834356412,
        "log1p_mel": 14.387676296985852
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "20-30",
      "audio": "listening-audio/podcast-800-descript-dac-44khz-window-27.wav",
      "baseline": {
        "log_mel": 0.28345760703086853,
        "log1p_mel": 0.007421256974339485,
        "mr_stft": 0.8638533353805542
      },
      "treatment": {
        "log_mel": 0.33131515979766846,
        "log1p_mel": 0.009233959950506687,
        "mr_stft": 0.944516122341156
      },
      "decoder_drift": {
        "log_mel": 0.10597731173038483,
        "log1p_mel": 0.003242011647671461,
        "mr_stft": 0.30627137422561646
      },
      "relative_percent": {
        "mr_stft": 9.337555769819117,
        "log_mel": 16.88349565499163,
        "log1p_mel": 24.42582142668006
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 27,
      "window_ms": 313.46938775510205,
      "scalar_tokens": 243,
      "codebooks": 9,
      "seam_count": 33,
      "id": "descript-dac-44khz--podcast-800--log1p_mel--20-30",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 9.337555769819117,
        "log_mel": 16.88349565499163,
        "log1p_mel": 24.42582142668006
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "30-40",
      "audio": "listening-audio/podcast-800-descript-dac-44khz-window-19.wav",
      "baseline": {
        "log_mel": 0.28345760703086853,
        "log1p_mel": 0.007421256974339485,
        "mr_stft": 0.8638533353805542
      },
      "treatment": {
        "log_mel": 0.35918113589286804,
        "log1p_mel": 0.009921503253281116,
        "mr_stft": 0.9670728445053101
      },
      "decoder_drift": {
        "log_mel": 0.16042502224445343,
        "log1p_mel": 0.004695218987762928,
        "mr_stft": 0.40139421820640564
      },
      "relative_percent": {
        "mr_stft": 11.948730750606472,
        "log_mel": 26.714234151335802,
        "log1p_mel": 33.690334232957355
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 19,
      "window_ms": 220.58956916099774,
      "scalar_tokens": 171,
      "codebooks": 9,
      "seam_count": 47,
      "id": "descript-dac-44khz--podcast-800--log1p_mel--30-40",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 11.948730750606472,
        "log_mel": 26.714234151335802,
        "log1p_mel": 33.690334232957355
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "40-50",
      "audio": "listening-audio/podcast-800-descript-dac-44khz-window-15.wav",
      "baseline": {
        "log_mel": 0.28345760703086853,
        "log1p_mel": 0.007421256974339485,
        "mr_stft": 0.8638533353805542
      },
      "treatment": {
        "log_mel": 0.3709120452404022,
        "log1p_mel": 0.010472075082361698,
        "mr_stft": 0.9884132146835327
      },
      "decoder_drift": {
        "log_mel": 0.19452273845672607,
        "log1p_mel": 0.005759991239756346,
        "mr_stft": 0.47891396284103394
      },
      "relative_percent": {
        "mr_stft": 14.419100349726154,
        "log_mel": 30.852739894897184,
        "log1p_mel": 41.10918296686184
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 15,
      "window_ms": 174.14965986394557,
      "scalar_tokens": 135,
      "codebooks": 9,
      "seam_count": 60,
      "id": "descript-dac-44khz--podcast-800--log1p_mel--40-50",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 14.419100349726154,
        "log_mel": 30.852739894897184,
        "log1p_mel": 41.10918296686184
      }
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "over-50",
      "audio": "listening-audio/podcast-800-descript-dac-44khz-window-9.wav",
      "baseline": {
        "log_mel": 0.28345760703086853,
        "log1p_mel": 0.007421256974339485,
        "mr_stft": 0.8638533353805542
      },
      "treatment": {
        "log_mel": 0.4402306377887726,
        "log1p_mel": 0.012929127551615238,
        "mr_stft": 1.0847861766815186
      },
      "decoder_drift": {
        "log_mel": 0.32549455761909485,
        "log1p_mel": 0.009715450927615166,
        "mr_stft": 0.7379852533340454
      },
      "relative_percent": {
        "mr_stft": 25.575272126909887,
        "log_mel": 55.30739936742339,
        "log1p_mel": 74.21748898226195
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 9,
      "window_ms": 104.48979591836735,
      "scalar_tokens": 81,
      "codebooks": 9,
      "seam_count": 101,
      "id": "descript-dac-44khz--podcast-800--log1p_mel--over-50",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-descript-dac-44khz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 25.575272126909887,
        "log_mel": 55.30739936742339,
        "log1p_mel": 74.21748898226195
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "0-5",
      "audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-38.wav",
      "baseline": {
        "log_mel": 0.3965880870819092,
        "log1p_mel": 0.012173594906926155,
        "mr_stft": 1.0366172790527344
      },
      "treatment": {
        "log_mel": 0.40996888279914856,
        "log1p_mel": 0.012912478297948837,
        "mr_stft": 1.0603504180908203
      },
      "decoder_drift": {
        "log_mel": 0.09319067746400833,
        "log1p_mel": 0.003631187602877617,
        "mr_stft": 0.3409973382949829
      },
      "relative_percent": {
        "mr_stft": 2.2894793978133654,
        "log_mel": 3.3739782290726605,
        "log1p_mel": 6.069557897004563
      },
      "recorded_metric_max_abs_difference": 2.9802322387695312e-08,
      "window_frames": 38,
      "window_ms": 1764.716553287982,
      "scalar_tokens": 380,
      "codebooks": 10,
      "seam_count": 5,
      "id": "fish-400m-morepitchloss1-step3802000--podcast-800--mr_stft--0-5",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 2.2894793978133654,
        "log_mel": 3.3739857437518506,
        "log1p_mel": 6.069557897004563
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "5-10",
      "audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-29.wav",
      "baseline": {
        "log_mel": 0.3965880870819092,
        "log1p_mel": 0.012173594906926155,
        "mr_stft": 1.0366172790527344
      },
      "treatment": {
        "log_mel": 0.4199870824813843,
        "log1p_mel": 0.013762764632701874,
        "mr_stft": 1.113668441772461
      },
      "decoder_drift": {
        "log_mel": 0.119534932076931,
        "log1p_mel": 0.005164720583707094,
        "mr_stft": 0.4545419216156006
      },
      "relative_percent": {
        "mr_stft": 7.43294215490371,
        "log_mel": 5.900075206909183,
        "log1p_mel": 13.054235317716723
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 29,
      "window_ms": 1346.7573696145125,
      "scalar_tokens": 290,
      "codebooks": 10,
      "seam_count": 7,
      "id": "fish-400m-morepitchloss1-step3802000--podcast-800--mr_stft--5-10",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 7.432919155231232,
        "log_mel": 5.9000601775508255,
        "log1p_mel": 13.054220017017437
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "10-20",
      "audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-10.wav",
      "baseline": {
        "log_mel": 0.3965880870819092,
        "log1p_mel": 0.012173594906926155,
        "mr_stft": 1.0366172790527344
      },
      "treatment": {
        "log_mel": 0.46692079305648804,
        "log1p_mel": 0.016009189188480377,
        "mr_stft": 1.1935734748840332
      },
      "decoder_drift": {
        "log_mel": 0.2558513283729553,
        "log1p_mel": 0.010452142916619778,
        "mr_stft": 0.7729124426841736
      },
      "relative_percent": {
        "mr_stft": 15.141190389448855,
        "log_mel": 17.734447469687286,
        "log1p_mel": 31.50749068684686
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 10,
      "window_ms": 464.39909297052156,
      "scalar_tokens": 100,
      "codebooks": 10,
      "seam_count": 22,
      "id": "fish-400m-morepitchloss1-step3802000--podcast-800--mr_stft--10-20",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 15.141167389776378,
        "log_mel": 17.734424925649762,
        "log1p_mel": 31.50749068684686
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "20-30",
      "audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-6.wav",
      "baseline": {
        "log_mel": 0.3965880870819092,
        "log1p_mel": 0.012173594906926155,
        "mr_stft": 1.0366172790527344
      },
      "treatment": {
        "log_mel": 0.5196914076805115,
        "log1p_mel": 0.01973857916891575,
        "mr_stft": 1.3204081058502197
      },
      "decoder_drift": {
        "log_mel": 0.36210933327674866,
        "log1p_mel": 0.015599893406033516,
        "mr_stft": 1.0062998533248901
      },
      "relative_percent": {
        "mr_stft": 27.37662515685777,
        "log_mel": 31.040599707468573,
        "log1p_mel": 62.14256610169855
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-07,
      "window_frames": 6,
      "window_ms": 278.6394557823129,
      "scalar_tokens": 60,
      "codebooks": 10,
      "seam_count": 37,
      "id": "fish-400m-morepitchloss1-step3802000--podcast-800--mr_stft--20-30",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 27.37663665669401,
        "log_mel": 31.040599707468573,
        "log1p_mel": 62.14258140239784
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "30-40",
      "audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-4.wav",
      "baseline": {
        "log_mel": 0.3965880870819092,
        "log1p_mel": 0.012173594906926155,
        "mr_stft": 1.0366172790527344
      },
      "treatment": {
        "log_mel": 0.576855480670929,
        "log1p_mel": 0.023150241002440453,
        "mr_stft": 1.4318804740905762
      },
      "decoder_drift": {
        "log_mel": 0.45717689394950867,
        "log1p_mel": 0.019900046288967133,
        "mr_stft": 1.1894524097442627
      },
      "relative_percent": {
        "mr_stft": 38.13009902738984,
        "log_mel": 45.454565949125026,
        "log1p_mel": 90.16766353272642
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 4,
      "window_ms": 185.75963718820861,
      "scalar_tokens": 40,
      "codebooks": 10,
      "seam_count": 56,
      "id": "fish-400m-morepitchloss1-step3802000--podcast-800--mr_stft--30-40",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 38.13012202706234,
        "log_mel": 45.45452086104997,
        "log1p_mel": 90.16764823202715
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "40-50",
      "audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-3.wav",
      "baseline": {
        "log_mel": 0.3965880870819092,
        "log1p_mel": 0.012173594906926155,
        "mr_stft": 1.0366172790527344
      },
      "treatment": {
        "log_mel": 0.6119422912597656,
        "log1p_mel": 0.025416988879442215,
        "mr_stft": 1.4801963567733765
      },
      "decoder_drift": {
        "log_mel": 0.5210055708885193,
        "log1p_mel": 0.022862466052174568,
        "mr_stft": 1.27909255027771
      },
      "relative_percent": {
        "mr_stft": 42.79101715591569,
        "log_mel": 54.30173300525245,
        "log1p_mel": 108.78786483178642
      },
      "recorded_metric_max_abs_difference": 5.960464477539062e-07,
      "window_frames": 3,
      "window_ms": 139.31972789115645,
      "scalar_tokens": 30,
      "codebooks": 10,
      "seam_count": 75,
      "id": "fish-400m-morepitchloss1-step3802000--podcast-800--mr_stft--40-50",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 42.79107465509691,
        "log_mel": 54.30173300525245,
        "log1p_mel": 108.78786483178642
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "over-50",
      "audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-1.wav",
      "baseline": {
        "log_mel": 0.3965880870819092,
        "log1p_mel": 0.012173594906926155,
        "mr_stft": 1.0366172790527344
      },
      "treatment": {
        "log_mel": 0.8519214987754822,
        "log1p_mel": 0.038309305906295776,
        "mr_stft": 1.8492192029953003
      },
      "decoder_drift": {
        "log_mel": 0.8483490943908691,
        "log1p_mel": 0.03783388435840607,
        "mr_stft": 1.8036091327667236
      },
      "relative_percent": {
        "mr_stft": 78.38977222964346,
        "log_mel": 114.8126800892864,
        "log1p_mel": 214.6918079596992
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-07,
      "window_frames": 1,
      "window_ms": 46.439909297052154,
      "scalar_tokens": 10,
      "codebooks": 10,
      "seam_count": 227,
      "id": "fish-400m-morepitchloss1-step3802000--podcast-800--mr_stft--over-50",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 78.38978372947969,
        "log_mel": 114.81265003056973,
        "log1p_mel": 214.6918079596992
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "0-5",
      "audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-49.wav",
      "baseline": {
        "log_mel": 0.3965880870819092,
        "log1p_mel": 0.012173594906926155,
        "mr_stft": 1.0366172790527344
      },
      "treatment": {
        "log_mel": 0.4066387712955475,
        "log1p_mel": 0.012574386782944202,
        "mr_stft": 1.0688836574554443
      },
      "decoder_drift": {
        "log_mel": 0.07787753641605377,
        "log1p_mel": 0.0026459048967808485,
        "mr_stft": 0.31774911284446716
      },
      "relative_percent": {
        "mr_stft": 3.112660675711987,
        "log_mel": 2.5342879781364847,
        "log1p_mel": 3.292305018216246
      },
      "recorded_metric_max_abs_difference": 2.9802322387695312e-08,
      "window_frames": 49,
      "window_ms": 2275.5555555555557,
      "scalar_tokens": 490,
      "codebooks": 10,
      "seam_count": 4,
      "id": "fish-400m-morepitchloss1-step3802000--podcast-800--log_mel--0-5",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 3.112660675711987,
        "log_mel": 2.5342954928156747,
        "log1p_mel": 3.292305018216246
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "5-10",
      "audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-32.wav",
      "baseline": {
        "log_mel": 0.3965880870819092,
        "log1p_mel": 0.012173594906926155,
        "mr_stft": 1.0366172790527344
      },
      "treatment": {
        "log_mel": 0.42568883299827576,
        "log1p_mel": 0.014108247123658657,
        "mr_stft": 1.1197254657745361
      },
      "decoder_drift": {
        "log_mel": 0.12742358446121216,
        "log1p_mel": 0.0049926782958209515,
        "mr_stft": 0.46009358763694763
      },
      "relative_percent": {
        "mr_stft": 8.0172488343766,
        "log_mel": 7.3377761118569085,
        "log1p_mel": 15.892201371279269
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 32,
      "window_ms": 1486.077097505669,
      "scalar_tokens": 320,
      "codebooks": 10,
      "seam_count": 7,
      "id": "fish-400m-morepitchloss1-step3802000--podcast-800--log_mel--5-10",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 8.0172488343766,
        "log_mel": 7.3377761118569085,
        "log1p_mel": 15.892201371279269
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "10-20",
      "audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-11.wav",
      "baseline": {
        "log_mel": 0.3965880870819092,
        "log1p_mel": 0.012173594906926155,
        "mr_stft": 1.0366172790527344
      },
      "treatment": {
        "log_mel": 0.45547211170196533,
        "log1p_mel": 0.015545672737061977,
        "mr_stft": 1.159714698791504
      },
      "decoder_drift": {
        "log_mel": 0.24277088046073914,
        "log1p_mel": 0.0095629021525383,
        "mr_stft": 0.7235388159751892
      },
      "relative_percent": {
        "mr_stft": 11.874914901211797,
        "log_mel": 14.847653406163609,
        "log1p_mel": 27.699934620111932
      },
      "recorded_metric_max_abs_difference": 8.940696716308594e-08,
      "window_frames": 11,
      "window_ms": 510.83900226757373,
      "scalar_tokens": 110,
      "codebooks": 10,
      "seam_count": 20,
      "id": "fish-400m-morepitchloss1-step3802000--podcast-800--log_mel--10-20",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 11.874914901211797,
        "log_mel": 14.847675950201133,
        "log1p_mel": 27.699934620111932
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "20-30",
      "audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-8.wav",
      "baseline": {
        "log_mel": 0.3965880870819092,
        "log1p_mel": 0.012173594906926155,
        "mr_stft": 1.0366172790527344
      },
      "treatment": {
        "log_mel": 0.4964759945869446,
        "log1p_mel": 0.01846100576221943,
        "mr_stft": 1.2601650953292847
      },
      "decoder_drift": {
        "log_mel": 0.3157673478126526,
        "log1p_mel": 0.013399356976151466,
        "mr_stft": 0.898218035697937
      },
      "relative_percent": {
        "mr_stft": 21.56512541261413,
        "log_mel": 25.186814924273083,
        "log1p_mel": 51.647938865749985
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-07,
      "window_frames": 8,
      "window_ms": 371.51927437641723,
      "scalar_tokens": 80,
      "codebooks": 10,
      "seam_count": 28,
      "id": "fish-400m-morepitchloss1-step3802000--podcast-800--log_mel--20-30",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 21.56511391277789,
        "log_mel": 25.186807409593893,
        "log1p_mel": 51.647923565050704
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "30-40",
      "audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-5.wav",
      "baseline": {
        "log_mel": 0.3965880870819092,
        "log1p_mel": 0.012173594906926155,
        "mr_stft": 1.0366172790527344
      },
      "treatment": {
        "log_mel": 0.5372368097305298,
        "log1p_mel": 0.02001219429075718,
        "mr_stft": 1.3239917755126953
      },
      "decoder_drift": {
        "log_mel": 0.3905653655529022,
        "log1p_mel": 0.01608518697321415,
        "mr_stft": 1.0324091911315918
      },
      "relative_percent": {
        "mr_stft": 27.72233323397475,
        "log_mel": 35.464686719036976,
        "log1p_mel": 64.39017762428796
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 5,
      "window_ms": 232.19954648526078,
      "scalar_tokens": 50,
      "codebooks": 10,
      "seam_count": 45,
      "id": "fish-400m-morepitchloss1-step3802000--podcast-800--log_mel--30-40",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 27.72233323397475,
        "log_mel": 35.464686719036976,
        "log1p_mel": 64.39017762428796
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "40-50",
      "audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-4.wav",
      "baseline": {
        "log_mel": 0.3965880870819092,
        "log1p_mel": 0.012173594906926155,
        "mr_stft": 1.0366172790527344
      },
      "treatment": {
        "log_mel": 0.576855480670929,
        "log1p_mel": 0.023150241002440453,
        "mr_stft": 1.4318804740905762
      },
      "decoder_drift": {
        "log_mel": 0.45717689394950867,
        "log1p_mel": 0.019900046288967133,
        "mr_stft": 1.1894524097442627
      },
      "relative_percent": {
        "mr_stft": 38.13009902738984,
        "log_mel": 45.454565949125026,
        "log1p_mel": 90.16766353272642
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 4,
      "window_ms": 185.75963718820861,
      "scalar_tokens": 40,
      "codebooks": 10,
      "seam_count": 56,
      "id": "fish-400m-morepitchloss1-step3802000--podcast-800--log_mel--40-50",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 38.13012202706234,
        "log_mel": 45.45452086104997,
        "log1p_mel": 90.16764823202715
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "over-50",
      "audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-2.wav",
      "baseline": {
        "log_mel": 0.3965880870819092,
        "log1p_mel": 0.012173594906926155,
        "mr_stft": 1.0366172790527344
      },
      "treatment": {
        "log_mel": 0.6839240193367004,
        "log1p_mel": 0.02944009006023407,
        "mr_stft": 1.6023468971252441
      },
      "decoder_drift": {
        "log_mel": 0.6297867298126221,
        "log1p_mel": 0.0280220378190279,
        "mr_stft": 1.4664924144744873
      },
      "relative_percent": {
        "mr_stft": 54.574588857854664,
        "log_mel": 72.45198270301208,
        "log1p_mel": 141.83563101384422
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 2,
      "window_ms": 92.87981859410431,
      "scalar_tokens": 20,
      "codebooks": 10,
      "seam_count": 113,
      "id": "fish-400m-morepitchloss1-step3802000--podcast-800--log_mel--over-50",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 54.574611857527145,
        "log_mel": 72.45199773237042,
        "log1p_mel": 141.83563101384422
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "0-5",
      "audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-56.wav",
      "baseline": {
        "log_mel": 0.3965880870819092,
        "log1p_mel": 0.012173594906926155,
        "mr_stft": 1.0366172790527344
      },
      "treatment": {
        "log_mel": 0.4050644636154175,
        "log1p_mel": 0.012415897101163864,
        "mr_stft": 1.048486590385437
      },
      "decoder_drift": {
        "log_mel": 0.06785204261541367,
        "log1p_mel": 0.002271012868732214,
        "mr_stft": 0.2511332631111145
      },
      "relative_percent": {
        "mr_stft": 1.1450041951402534,
        "log_mel": 2.1373250507541464,
        "log1p_mel": 1.9903914668612188
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-07,
      "window_frames": 56,
      "window_ms": 2600.6349206349205,
      "scalar_tokens": 560,
      "codebooks": 10,
      "seam_count": 4,
      "id": "fish-400m-morepitchloss1-step3802000--podcast-800--log1p_mel--0-5",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 1.1449926953040146,
        "log_mel": 2.1373175360749785,
        "log1p_mel": 1.9903914668612188
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "5-10",
      "audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-35.wav",
      "baseline": {
        "log_mel": 0.3965880870819092,
        "log1p_mel": 0.012173594906926155,
        "mr_stft": 1.0366172790527344
      },
      "treatment": {
        "log_mel": 0.4179551601409912,
        "log1p_mel": 0.013073308393359184,
        "mr_stft": 1.0891975164413452
      },
      "decoder_drift": {
        "log_mel": 0.10108020901679993,
        "log1p_mel": 0.004061372950673103,
        "mr_stft": 0.385699599981308
      },
      "relative_percent": {
        "mr_stft": 5.072290270586555,
        "log_mel": 5.387724380805459,
        "log1p_mel": 7.390696777014805
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 35,
      "window_ms": 1625.3968253968253,
      "scalar_tokens": 350,
      "codebooks": 10,
      "seam_count": 6,
      "id": "fish-400m-morepitchloss1-step3802000--podcast-800--log1p_mel--5-10",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 5.072290270586555,
        "log_mel": 5.387724380805459,
        "log1p_mel": 7.390696777014805
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "10-20",
      "audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-17.wav",
      "baseline": {
        "log_mel": 0.3965880870819092,
        "log1p_mel": 0.012173594906926155,
        "mr_stft": 1.0366172790527344
      },
      "treatment": {
        "log_mel": 0.4325087070465088,
        "log1p_mel": 0.013998251408338547,
        "mr_stft": 1.121659517288208
      },
      "decoder_drift": {
        "log_mel": 0.17235513031482697,
        "log1p_mel": 0.0068208989687263966,
        "mr_stft": 0.5679370760917664
      },
      "relative_percent": {
        "mr_stft": 8.203822177572185,
        "log_mel": 9.057412750065975,
        "log1p_mel": 14.988641525883661
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 17,
      "window_ms": 789.4784580498866,
      "scalar_tokens": 170,
      "codebooks": 10,
      "seam_count": 13,
      "id": "fish-400m-morepitchloss1-step3802000--podcast-800--log1p_mel--10-20",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 8.203845177244684,
        "log_mel": 9.057420264745165,
        "log1p_mel": 14.988641525883661
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "20-30",
      "audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-13.wav",
      "baseline": {
        "log_mel": 0.3965880870819092,
        "log1p_mel": 0.012173594906926155,
        "mr_stft": 1.0366172790527344
      },
      "treatment": {
        "log_mel": 0.45442044734954834,
        "log1p_mel": 0.01491579320281744,
        "mr_stft": 1.1419875621795654
      },
      "decoder_drift": {
        "log_mel": 0.22296757996082306,
        "log1p_mel": 0.008279894478619099,
        "mr_stft": 0.6486579775810242
      },
      "relative_percent": {
        "mr_stft": 10.164820252959593,
        "log_mel": 14.582475407460937,
        "log1p_mel": 22.525788946132195
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-07,
      "window_frames": 13,
      "window_ms": 603.7188208616781,
      "scalar_tokens": 130,
      "codebooks": 10,
      "seam_count": 17,
      "id": "fish-400m-morepitchloss1-step3802000--podcast-800--log1p_mel--20-30",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 10.164808753123356,
        "log_mel": 14.58246789278177,
        "log1p_mel": 22.525788946132195
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "30-40",
      "audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-12.wav",
      "baseline": {
        "log_mel": 0.3965880870819092,
        "log1p_mel": 0.012173594906926155,
        "mr_stft": 1.0366172790527344
      },
      "treatment": {
        "log_mel": 0.46232420206069946,
        "log1p_mel": 0.01616702973842621,
        "mr_stft": 1.2139065265655518
      },
      "decoder_drift": {
        "log_mel": 0.23399992287158966,
        "log1p_mel": 0.009928571060299873,
        "mr_stft": 0.7540895342826843
      },
      "relative_percent": {
        "mr_stft": 17.10267145795843,
        "log_mel": 16.575413412560103,
        "log1p_mel": 32.80407194450008
      },
      "recorded_metric_max_abs_difference": 3.5762786865234375e-07,
      "window_frames": 12,
      "window_ms": 557.2789115646258,
      "scalar_tokens": 120,
      "codebooks": 10,
      "seam_count": 18,
      "id": "fish-400m-morepitchloss1-step3802000--podcast-800--log1p_mel--30-40",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 17.102636958449715,
        "log_mel": 16.575405897880913,
        "log1p_mel": 32.80407194450008
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "40-50",
      "audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-7.wav",
      "baseline": {
        "log_mel": 0.3965880870819092,
        "log1p_mel": 0.012173594906926155,
        "mr_stft": 1.0366172790527344
      },
      "treatment": {
        "log_mel": 0.4987375736236572,
        "log1p_mel": 0.0179872028529644,
        "mr_stft": 1.257556438446045
      },
      "decoder_drift": {
        "log_mel": 0.3209703266620636,
        "log1p_mel": 0.013210255652666092,
        "mr_stft": 0.9059342741966248
      },
      "relative_percent": {
        "mr_stft": 21.313474496123174,
        "log_mel": 25.75707386809394,
        "log1p_mel": 47.75588468719787
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-07,
      "window_frames": 7,
      "window_ms": 325.07936507936506,
      "scalar_tokens": 70,
      "codebooks": 10,
      "seam_count": 32,
      "id": "fish-400m-morepitchloss1-step3802000--podcast-800--log1p_mel--40-50",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 21.313462996286937,
        "log_mel": 25.757088897452274,
        "log1p_mel": 47.75588468719787
      }
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "over-50",
      "audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-5.wav",
      "baseline": {
        "log_mel": 0.3965880870819092,
        "log1p_mel": 0.012173594906926155,
        "mr_stft": 1.0366172790527344
      },
      "treatment": {
        "log_mel": 0.5372368097305298,
        "log1p_mel": 0.02001219429075718,
        "mr_stft": 1.3239917755126953
      },
      "decoder_drift": {
        "log_mel": 0.3905653655529022,
        "log1p_mel": 0.01608518697321415,
        "mr_stft": 1.0324091911315918
      },
      "relative_percent": {
        "mr_stft": 27.72233323397475,
        "log_mel": 35.464686719036976,
        "log1p_mel": 64.39017762428796
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 5,
      "window_ms": 232.19954648526078,
      "scalar_tokens": 50,
      "codebooks": 10,
      "seam_count": 45,
      "id": "fish-400m-morepitchloss1-step3802000--podcast-800--log1p_mel--over-50",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 27.72233323397475,
        "log_mel": 35.464686719036976,
        "log1p_mel": 64.39017762428796
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "0-5",
      "audio": "listening-audio/podcast-800-fish-dac-s2-window-193.wav",
      "baseline": {
        "log_mel": 0.3086870610713959,
        "log1p_mel": 0.00981848780065775,
        "mr_stft": 0.9192757606506348
      },
      "treatment": {
        "log_mel": 0.32794904708862305,
        "log1p_mel": 0.01050086971372366,
        "mr_stft": 0.942152202129364
      },
      "decoder_drift": {
        "log_mel": 0.04182243347167969,
        "log1p_mel": 0.0013610806781798601,
        "mr_stft": 0.13912421464920044
      },
      "relative_percent": {
        "mr_stft": 2.4885287372896636,
        "log_mel": 6.239971947762357,
        "log1p_mel": 6.949969556617441
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 193,
      "window_ms": 8962.902494331067,
      "scalar_tokens": 1930,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--podcast-800--mr_stft--0-5",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 2.4885287372896636,
        "log_mel": 6.239971947762357,
        "log1p_mel": 6.949969556617441
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "5-10",
      "audio": "listening-audio/podcast-800-fish-dac-s2-window-196.wav",
      "baseline": {
        "log_mel": 0.3086870610713959,
        "log1p_mel": 0.00981848780065775,
        "mr_stft": 0.9192757606506348
      },
      "treatment": {
        "log_mel": 0.3277676999568939,
        "log1p_mel": 0.011105572804808617,
        "mr_stft": 0.9884568452835083
      },
      "decoder_drift": {
        "log_mel": 0.04371250048279762,
        "log1p_mel": 0.0020882568787783384,
        "mr_stft": 0.23683147132396698
      },
      "relative_percent": {
        "mr_stft": 7.525607395968903,
        "log_mel": 6.181224058848689,
        "log1p_mel": 13.108790582442275
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 196,
      "window_ms": 9102.222222222223,
      "scalar_tokens": 1960,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--podcast-800--mr_stft--5-10",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 7.525607395968903,
        "log_mel": 6.181224058848689,
        "log1p_mel": 13.108790582442275
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "10-20",
      "audio": "listening-audio/podcast-800-fish-dac-s2-window-76.wav",
      "baseline": {
        "log_mel": 0.3086870610713959,
        "log1p_mel": 0.00981848780065775,
        "mr_stft": 0.9192757606506348
      },
      "treatment": {
        "log_mel": 0.37866511940956116,
        "log1p_mel": 0.014163925312459469,
        "mr_stft": 1.0565775632858276
      },
      "decoder_drift": {
        "log_mel": 0.17357012629508972,
        "log1p_mel": 0.00795772671699524,
        "mr_stft": 0.49210843443870544
      },
      "relative_percent": {
        "mr_stft": 14.935866745579695,
        "log_mel": 22.669579377666295,
        "log1p_mel": 44.25770648215925
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-07,
      "window_frames": 76,
      "window_ms": 3529.433106575964,
      "scalar_tokens": 760,
      "codebooks": 10,
      "seam_count": 2,
      "id": "fish-dac-s2--podcast-800--mr_stft--10-20",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 14.935853777839792,
        "log_mel": 22.66956006858201,
        "log1p_mel": 44.25770648215925
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "20-30",
      "audio": "listening-audio/podcast-800-fish-dac-s2-window-95.wav",
      "baseline": {
        "log_mel": 0.3086870610713959,
        "log1p_mel": 0.00981848780065775,
        "mr_stft": 0.9192757606506348
      },
      "treatment": {
        "log_mel": 0.39566367864608765,
        "log1p_mel": 0.015937188640236855,
        "mr_stft": 1.1463160514831543
      },
      "decoder_drift": {
        "log_mel": 0.18554463982582092,
        "log1p_mel": 0.009492713026702404,
        "mr_stft": 0.6070528626441956
      },
      "relative_percent": {
        "mr_stft": 24.697734950808183,
        "log_mel": 28.17630815908252,
        "log1p_mel": 62.318159005801355
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 95,
      "window_ms": 4411.791383219955,
      "scalar_tokens": 950,
      "codebooks": 10,
      "seam_count": 2,
      "id": "fish-dac-s2--podcast-800--mr_stft--20-30",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 24.697734950808183,
        "log_mel": 28.17630815908252,
        "log1p_mel": 62.318159005801355
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "30-40",
      "audio": "listening-audio/podcast-800-fish-dac-s2-window-36.wav",
      "baseline": {
        "log_mel": 0.3086870610713959,
        "log1p_mel": 0.00981848780065775,
        "mr_stft": 0.9192757606506348
      },
      "treatment": {
        "log_mel": 0.43346431851387024,
        "log1p_mel": 0.019383709877729416,
        "mr_stft": 1.246915340423584
      },
      "decoder_drift": {
        "log_mel": 0.2904182970523834,
        "log1p_mel": 0.01509387418627739,
        "mr_stft": 0.8413507342338562
      },
      "relative_percent": {
        "mr_stft": 35.641055034569405,
        "log_mel": 40.42192666236011,
        "log1p_mel": 97.42052209333991
      },
      "recorded_metric_max_abs_difference": 5.960464477539063e-08,
      "window_frames": 36,
      "window_ms": 1671.8367346938776,
      "scalar_tokens": 360,
      "codebooks": 10,
      "seam_count": 6,
      "id": "fish-dac-s2--podcast-800--mr_stft--30-40",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 35.641055034569405,
        "log_mel": 40.42194597144437,
        "log1p_mel": 97.42056003492903
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "40-50",
      "audio": "listening-audio/podcast-800-fish-dac-s2-window-24.wav",
      "baseline": {
        "log_mel": 0.3086870610713959,
        "log1p_mel": 0.00981848780065775,
        "mr_stft": 0.9192757606506348
      },
      "treatment": {
        "log_mel": 0.4910171627998352,
        "log1p_mel": 0.022872354835271835,
        "mr_stft": 1.328324317932129
      },
      "decoder_drift": {
        "log_mel": 0.3535505533218384,
        "log1p_mel": 0.019022610038518906,
        "mr_stft": 0.9662134051322937
      },
      "relative_percent": {
        "mr_stft": 44.4968283501767,
        "log_mel": 59.06632467703867,
        "log1p_mel": 132.95190969977676
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 24,
      "window_ms": 1114.5578231292516,
      "scalar_tokens": 240,
      "codebooks": 10,
      "seam_count": 9,
      "id": "fish-dac-s2--podcast-800--mr_stft--40-50",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 44.49685428565648,
        "log_mel": 59.066353640665085,
        "log1p_mel": 132.95190969977676
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "over-50",
      "audio": "listening-audio/podcast-800-fish-dac-s2-window-3.wav",
      "baseline": {
        "log_mel": 0.3086870610713959,
        "log1p_mel": 0.00981848780065775,
        "mr_stft": 0.9192757606506348
      },
      "treatment": {
        "log_mel": 0.7142701148986816,
        "log1p_mel": 0.031217757612466812,
        "mr_stft": 1.6160378456115723
      },
      "decoder_drift": {
        "log_mel": 0.634157121181488,
        "log1p_mel": 0.02912200428545475,
        "mr_stft": 1.4589993953704834
      },
      "relative_percent": {
        "mr_stft": 75.79467606845098,
        "log_mel": 131.38971630996835,
        "log1p_mel": 217.94873351449812
      },
      "recorded_metric_max_abs_difference": 4.76837158203125e-07,
      "window_frames": 3,
      "window_ms": 139.31972789115645,
      "scalar_tokens": 30,
      "codebooks": 10,
      "seam_count": 75,
      "id": "fish-dac-s2--podcast-800--mr_stft--over-50",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 75.79462419749142,
        "log_mel": 131.38971630996835,
        "log1p_mel": 217.94871454370352
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "0-5",
      "audio": "listening-audio/podcast-800-fish-dac-s2-window-207.wav",
      "baseline": {
        "log_mel": 0.3086870610713959,
        "log1p_mel": 0.00981848780065775,
        "mr_stft": 0.9192757606506348
      },
      "treatment": {
        "log_mel": 0.31635943055152893,
        "log1p_mel": 0.0104245999827981,
        "mr_stft": 0.9363062381744385
      },
      "decoder_drift": {
        "log_mel": 0.026497425511479378,
        "log1p_mel": 0.0011284789070487022,
        "mr_stft": 0.12067978829145432
      },
      "relative_percent": {
        "mr_stft": 1.8525972567524418,
        "log_mel": 2.485484637258084,
        "log1p_mel": 6.173172431906937
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 207,
      "window_ms": 9613.061224489797,
      "scalar_tokens": 2070,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--podcast-800--log_mel--0-5",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 1.8525972567524418,
        "log_mel": 2.485484637258084,
        "log1p_mel": 6.173172431906937
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "5-10",
      "audio": "listening-audio/podcast-800-fish-dac-s2-window-179.wav",
      "baseline": {
        "log_mel": 0.3086870610713959,
        "log1p_mel": 0.00981848780065775,
        "mr_stft": 0.9192757606506348
      },
      "treatment": {
        "log_mel": 0.3318104147911072,
        "log1p_mel": 0.011661222204566002,
        "mr_stft": 0.9842714071273804
      },
      "decoder_drift": {
        "log_mel": 0.05396362021565437,
        "log1p_mel": 0.0029708240181207657,
        "mr_stft": 0.23844027519226074
      },
      "relative_percent": {
        "mr_stft": 7.070310048286665,
        "log_mel": 7.490872354498568,
        "log1p_mel": 18.768006248220903
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 179,
      "window_ms": 8312.743764172335,
      "scalar_tokens": 1790,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--podcast-800--log_mel--5-10",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 7.070310048286665,
        "log_mel": 7.490872354498568,
        "log1p_mel": 18.768006248220903
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "10-20",
      "audio": "listening-audio/podcast-800-fish-dac-s2-window-70.wav",
      "baseline": {
        "log_mel": 0.3086870610713959,
        "log1p_mel": 0.00981848780065775,
        "mr_stft": 0.9192757606506348
      },
      "treatment": {
        "log_mel": 0.3549274206161499,
        "log1p_mel": 0.01354705635458231,
        "mr_stft": 1.0346225500106812
      },
      "decoder_drift": {
        "log_mel": 0.16285519301891327,
        "log1p_mel": 0.007723126094788313,
        "mr_stft": 0.4788864254951477
      },
      "relative_percent": {
        "mr_stft": 12.547572153801534,
        "log_mel": 14.979688291521608,
        "log1p_mel": 37.97497771168774
      },
      "recorded_metric_max_abs_difference": 3.5762786865234375e-07,
      "window_frames": 70,
      "window_ms": 3250.7936507936506,
      "scalar_tokens": 700,
      "codebooks": 10,
      "seam_count": 3,
      "id": "fish-dac-s2--podcast-800--log_mel--10-20",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 12.547611057021225,
        "log_mel": 14.979688291521608,
        "log1p_mel": 37.9749966824823
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "20-30",
      "audio": "listening-audio/podcast-800-fish-dac-s2-window-93.wav",
      "baseline": {
        "log_mel": 0.3086870610713959,
        "log1p_mel": 0.00981848780065775,
        "mr_stft": 0.9192757606506348
      },
      "treatment": {
        "log_mel": 0.38633430004119873,
        "log1p_mel": 0.015319411642849445,
        "mr_stft": 1.123876690864563
      },
      "decoder_drift": {
        "log_mel": 0.18061284720897675,
        "log1p_mel": 0.008787409402430058,
        "mr_stft": 0.5877952575683594
      },
      "relative_percent": {
        "mr_stft": 22.256752431840265,
        "log_mel": 25.154030978915532,
        "log1p_mel": 56.0261819729835
      },
      "recorded_metric_max_abs_difference": 3.5762786865234375e-07,
      "window_frames": 93,
      "window_ms": 4318.91156462585,
      "scalar_tokens": 930,
      "codebooks": 10,
      "seam_count": 2,
      "id": "fish-dac-s2--podcast-800--log_mel--20-30",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 22.256791335059955,
        "log_mel": 25.15405028799982,
        "log1p_mel": 56.0261819729835
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "30-40",
      "audio": "listening-audio/podcast-800-fish-dac-s2-window-26.wav",
      "baseline": {
        "log_mel": 0.3086870610713959,
        "log1p_mel": 0.00981848780065775,
        "mr_stft": 0.9192757606506348
      },
      "treatment": {
        "log_mel": 0.41859540343284607,
        "log1p_mel": 0.01569465547800064,
        "mr_stft": 1.1078605651855469
      },
      "decoder_drift": {
        "log_mel": 0.2642696797847748,
        "log1p_mel": 0.01118028536438942,
        "mr_stft": 0.6617715954780579
      },
      "relative_percent": {
        "mr_stft": 20.51449767384681,
        "log_mel": 35.605101807629566,
        "log1p_mel": 59.8479908173766
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-07,
      "window_frames": 26,
      "window_ms": 1207.4376417233561,
      "scalar_tokens": 260,
      "codebooks": 10,
      "seam_count": 8,
      "id": "fish-dac-s2--podcast-800--log_mel--30-40",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 20.514510641586714,
        "log_mel": 35.60506318946102,
        "log1p_mel": 59.8479908173766
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "40-50",
      "audio": "listening-audio/podcast-800-fish-dac-s2-window-38.wav",
      "baseline": {
        "log_mel": 0.3086870610713959,
        "log1p_mel": 0.00981848780065775,
        "mr_stft": 0.9192757606506348
      },
      "treatment": {
        "log_mel": 0.44920819997787476,
        "log1p_mel": 0.019845223054289818,
        "mr_stft": 1.2320841550827026
      },
      "decoder_drift": {
        "log_mel": 0.291960746049881,
        "log1p_mel": 0.01543481182307005,
        "mr_stft": 0.8177951574325562
      },
      "relative_percent": {
        "mr_stft": 34.02769961112342,
        "log_mel": 45.522199219739214,
        "log1p_mel": 102.1209727730208
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 38,
      "window_ms": 1764.716553287982,
      "scalar_tokens": 380,
      "codebooks": 10,
      "seam_count": 5,
      "id": "fish-dac-s2--podcast-800--log_mel--40-50",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 34.02772554660321,
        "log_mel": 45.5222185288235,
        "log1p_mel": 102.1209727730208
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "over-50",
      "audio": "listening-audio/podcast-800-fish-dac-s2-window-10.wav",
      "baseline": {
        "log_mel": 0.3086870610713959,
        "log1p_mel": 0.00981848780065775,
        "mr_stft": 0.9192757606506348
      },
      "treatment": {
        "log_mel": 0.5399096608161926,
        "log1p_mel": 0.024296103045344353,
        "mr_stft": 1.3912194967269897
      },
      "decoder_drift": {
        "log_mel": 0.43302249908447266,
        "log1p_mel": 0.021290261298418045,
        "mr_stft": 1.1127204895019531
      },
      "relative_percent": {
        "mr_stft": 51.33864682152915,
        "log_mel": 74.90518032802078,
        "log1p_mel": 147.45259696423653
      },
      "recorded_metric_max_abs_difference": 4.76837158203125e-07,
      "window_frames": 10,
      "window_ms": 464.39909297052156,
      "scalar_tokens": 100,
      "codebooks": 10,
      "seam_count": 22,
      "id": "fish-dac-s2--podcast-800--log_mel--over-50",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 51.33859495056956,
        "log_mel": 74.90518032802078,
        "log1p_mel": 147.452577993442
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "0-5",
      "audio": "listening-audio/podcast-800-fish-dac-s2-window-224.wav",
      "baseline": {
        "log_mel": 0.3086870610713959,
        "log1p_mel": 0.00981848780065775,
        "mr_stft": 0.9192757606506348
      },
      "treatment": {
        "log_mel": 0.3158700466156006,
        "log1p_mel": 0.010017558000981808,
        "mr_stft": 0.9255833625793457
      },
      "decoder_drift": {
        "log_mel": 0.011032262817025185,
        "log1p_mel": 0.0002648089430294931,
        "mr_stft": 0.04330059885978699
      },
      "relative_percent": {
        "mr_stft": 0.6861490532771741,
        "log_mel": 2.3269474007993285,
        "log1p_mel": 2.0275036682402714
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 224,
      "window_ms": 10402.539682539682,
      "scalar_tokens": 2240,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--podcast-800--log1p_mel--0-5",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 0.6861490532771741,
        "log_mel": 2.3269474007993285,
        "log1p_mel": 2.0275036682402714
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "5-10",
      "audio": "listening-audio/podcast-800-fish-dac-s2-window-182.wav",
      "baseline": {
        "log_mel": 0.3086870610713959,
        "log1p_mel": 0.00981848780065775,
        "mr_stft": 0.9192757606506348
      },
      "treatment": {
        "log_mel": 0.32676631212234497,
        "log1p_mel": 0.010554416105151176,
        "mr_stft": 0.9469383955001831
      },
      "decoder_drift": {
        "log_mel": 0.04468302056193352,
        "log1p_mel": 0.0016005882062017918,
        "mr_stft": 0.1537410020828247
      },
      "relative_percent": {
        "mr_stft": 3.0091770101682735,
        "log_mel": 5.85682178844793,
        "log1p_mel": 7.495332473134275
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 182,
      "window_ms": 8452.063492063493,
      "scalar_tokens": 1820,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--podcast-800--log1p_mel--5-10",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 3.0091770101682735,
        "log_mel": 5.85682178844793,
        "log1p_mel": 7.495332473134275
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "10-20",
      "audio": "listening-audio/podcast-800-fish-dac-s2-window-215.wav",
      "baseline": {
        "log_mel": 0.3086870610713959,
        "log1p_mel": 0.00981848780065775,
        "mr_stft": 0.9192757606506348
      },
      "treatment": {
        "log_mel": 0.32998400926589966,
        "log1p_mel": 0.011294503696262836,
        "mr_stft": 0.9771016836166382
      },
      "decoder_drift": {
        "log_mel": 0.03493419662117958,
        "log1p_mel": 0.001901065232232213,
        "mr_stft": 0.1905924379825592
      },
      "relative_percent": {
        "mr_stft": 6.2903783000952895,
        "log_mel": 6.899203394073594,
        "log1p_mel": 15.033026730513498
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 215,
      "window_ms": 9984.580498866213,
      "scalar_tokens": 2150,
      "codebooks": 10,
      "seam_count": 1,
      "id": "fish-dac-s2--podcast-800--log1p_mel--10-20",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 6.2903783000952895,
        "log_mel": 6.899203394073594,
        "log1p_mel": 15.033026730513498
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "20-30",
      "audio": "listening-audio/podcast-800-fish-dac-s2-window-79.wav",
      "baseline": {
        "log_mel": 0.3086870610713959,
        "log1p_mel": 0.00981848780065775,
        "mr_stft": 0.9192757606506348
      },
      "treatment": {
        "log_mel": 0.3475630283355713,
        "log1p_mel": 0.012280692346394062,
        "mr_stft": 1.0019787549972534
      },
      "decoder_drift": {
        "log_mel": 0.14177919924259186,
        "log1p_mel": 0.0059410203248262405,
        "mr_stft": 0.41757839918136597
      },
      "relative_percent": {
        "mr_stft": 8.99653813215786,
        "log_mel": 12.593973692724303,
        "log1p_mel": 25.07722773329073
      },
      "recorded_metric_max_abs_difference": 8.940696716308594e-08,
      "window_frames": 79,
      "window_ms": 3668.7528344671205,
      "scalar_tokens": 790,
      "codebooks": 10,
      "seam_count": 2,
      "id": "fish-dac-s2--podcast-800--log1p_mel--20-30",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 8.99653813215786,
        "log_mel": 12.594002656350733,
        "log1p_mel": 25.07723721868802
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "30-40",
      "audio": "listening-audio/podcast-800-fish-dac-s2-window-109.wav",
      "baseline": {
        "log_mel": 0.3086870610713959,
        "log1p_mel": 0.00981848780065775,
        "mr_stft": 0.9192757606506348
      },
      "treatment": {
        "log_mel": 0.34482303261756897,
        "log1p_mel": 0.013258828781545162,
        "mr_stft": 1.0190695524215698
      },
      "decoder_drift": {
        "log_mel": 0.13789790868759155,
        "log1p_mel": 0.006809409707784653,
        "mr_stft": 0.4286259114742279
      },
      "relative_percent": {
        "mr_stft": 10.85569706529672,
        "log_mel": 11.706344743039043,
        "log1p_mel": 35.03941799119963
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-07,
      "window_frames": 109,
      "window_ms": 5061.950113378684,
      "scalar_tokens": 1090,
      "codebooks": 10,
      "seam_count": 2,
      "id": "fish-dac-s2--podcast-800--log1p_mel--30-40",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 10.855684097556818,
        "log_mel": 11.706325433954778,
        "log1p_mel": 35.03941799119963
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "40-50",
      "audio": "listening-audio/podcast-800-fish-dac-s2-window-107.wav",
      "baseline": {
        "log_mel": 0.3086870610713959,
        "log1p_mel": 0.00981848780065775,
        "mr_stft": 0.9192757606506348
      },
      "treatment": {
        "log_mel": 0.3615918457508087,
        "log1p_mel": 0.014221764169633389,
        "mr_stft": 1.0618312358856201
      },
      "decoder_drift": {
        "log_mel": 0.16572926938533783,
        "log1p_mel": 0.00810198299586773,
        "mr_stft": 0.49589383602142334
      },
      "relative_percent": {
        "mr_stft": 15.507368010453026,
        "log_mel": 17.138646659108446,
        "log1p_mel": 44.84678759473184
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 107,
      "window_ms": 4969.07029478458,
      "scalar_tokens": 1070,
      "codebooks": 10,
      "seam_count": 2,
      "id": "fish-dac-s2--podcast-800--log1p_mel--40-50",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 15.507342074973241,
        "log_mel": 17.138646659108446,
        "log1p_mel": 44.8468065655264
      }
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "over-50",
      "audio": "listening-audio/podcast-800-fish-dac-s2-window-39.wav",
      "baseline": {
        "log_mel": 0.3086870610713959,
        "log1p_mel": 0.00981848780065775,
        "mr_stft": 0.9192757606506348
      },
      "treatment": {
        "log_mel": 0.42697200179100037,
        "log1p_mel": 0.017059659585356712,
        "mr_stft": 1.1684441566467285
      },
      "decoder_drift": {
        "log_mel": 0.2620460093021393,
        "log1p_mel": 0.012324333190917969,
        "mr_stft": 0.7248119115829468
      },
      "relative_percent": {
        "mr_stft": 27.104858700912573,
        "log_mel": 38.31872327562395,
        "log1p_mel": 73.75037716310928
      },
      "recorded_metric_max_abs_difference": 3.5762786865234375e-07,
      "window_frames": 39,
      "window_ms": 1811.156462585034,
      "scalar_tokens": 390,
      "codebooks": 10,
      "seam_count": 5,
      "id": "fish-dac-s2--podcast-800--log1p_mel--over-50",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-fish-dac-s2-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 27.10489760413224,
        "log_mel": 38.31870396653967,
        "log1p_mel": 73.75037716310928
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "0-5",
      "audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-107.wav",
      "baseline": {
        "log_mel": 0.3232126235961914,
        "log1p_mel": 0.010577854700386524,
        "mr_stft": 1.0321029424667358
      },
      "treatment": {
        "log_mel": 0.3326055705547333,
        "log1p_mel": 0.010959121398627758,
        "mr_stft": 1.0571062564849854
      },
      "decoder_drift": {
        "log_mel": 0.035166457295417786,
        "log1p_mel": 0.0011251101968809962,
        "mr_stft": 0.14530229568481445
      },
      "relative_percent": {
        "mr_stft": 2.4225600944893566,
        "log_mel": 2.906120080964736,
        "log1p_mel": 3.6043858517673
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 107,
      "window_ms": 8560.0,
      "scalar_tokens": 1712,
      "codebooks": 16,
      "seam_count": 1,
      "id": "qwen3-tts-tokenizer-12hz--podcast-800--mr_stft--0-5",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 2.4225600944893566,
        "log_mel": 2.906120080964736,
        "log1p_mel": 3.6043858517673
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "5-10",
      "audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-30.wav",
      "baseline": {
        "log_mel": 0.3232126235961914,
        "log1p_mel": 0.010577854700386524,
        "mr_stft": 1.0321029424667358
      },
      "treatment": {
        "log_mel": 0.37119174003601074,
        "log1p_mel": 0.012076837942004204,
        "mr_stft": 1.1103347539901733
      },
      "decoder_drift": {
        "log_mel": 0.15236765146255493,
        "log1p_mel": 0.004489606246352196,
        "mr_stft": 0.4535512924194336
      },
      "relative_percent": {
        "mr_stft": 7.579845798759455,
        "log_mel": 14.844443854322487,
        "log1p_mel": 14.170957004759256
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-07,
      "window_frames": 30,
      "window_ms": 2400.0,
      "scalar_tokens": 480,
      "codebooks": 16,
      "seam_count": 4,
      "id": "qwen3-tts-tokenizer-12hz--podcast-800--mr_stft--5-10",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 7.579857348895058,
        "log_mel": 14.844453074977949,
        "log1p_mel": 14.170957004759256
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "10-20",
      "audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-22.wav",
      "baseline": {
        "log_mel": 0.3232126235961914,
        "log1p_mel": 0.010577854700386524,
        "mr_stft": 1.0321029424667358
      },
      "treatment": {
        "log_mel": 0.3886074423789978,
        "log1p_mel": 0.013754366897046566,
        "mr_stft": 1.185935378074646
      },
      "decoder_drift": {
        "log_mel": 0.18274103105068207,
        "log1p_mel": 0.006915897596627474,
        "mr_stft": 0.5819826126098633
      },
      "relative_percent": {
        "mr_stft": 14.904757004204372,
        "log_mel": 20.2327551613541,
        "log1p_mel": 30.029833899533244
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 22,
      "window_ms": 1760.0,
      "scalar_tokens": 352,
      "codebooks": 16,
      "seam_count": 6,
      "id": "qwen3-tts-tokenizer-12hz--podcast-800--mr_stft--10-20",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 14.904780104475602,
        "log_mel": 20.23272749938776,
        "log1p_mel": 30.029833899533244
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "20-30",
      "audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-7.wav",
      "baseline": {
        "log_mel": 0.3232126235961914,
        "log1p_mel": 0.010577854700386524,
        "mr_stft": 1.0321029424667358
      },
      "treatment": {
        "log_mel": 0.45063069462776184,
        "log1p_mel": 0.016019441187381744,
        "mr_stft": 1.2826719284057617
      },
      "decoder_drift": {
        "log_mel": 0.29900288581848145,
        "log1p_mel": 0.010883515700697899,
        "mr_stft": 0.8221750259399414
      },
      "relative_percent": {
        "mr_stft": 24.277518804487052,
        "log_mel": 39.422368351169766,
        "log1p_mel": 51.44319563017235
      },
      "recorded_metric_max_abs_difference": 2.384185791015625e-07,
      "window_frames": 7,
      "window_ms": 560.0,
      "scalar_tokens": 112,
      "codebooks": 16,
      "seam_count": 18,
      "id": "qwen3-tts-tokenizer-12hz--podcast-800--mr_stft--20-30",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 24.27754190475828,
        "log_mel": 39.422368351169766,
        "log1p_mel": 51.443178021260124
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "30-40",
      "audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-4.wav",
      "baseline": {
        "log_mel": 0.3232126235961914,
        "log1p_mel": 0.010577854700386524,
        "mr_stft": 1.0321029424667358
      },
      "treatment": {
        "log_mel": 0.5103878378868103,
        "log1p_mel": 0.018271472305059433,
        "mr_stft": 1.3765910863876343
      },
      "decoder_drift": {
        "log_mel": 0.3819350302219391,
        "log1p_mel": 0.013622327707707882,
        "mr_stft": 1.0004022121429443
      },
      "relative_percent": {
        "mr_stft": 33.377304699623124,
        "log_mel": 57.91086134199632,
        "log1p_mel": 72.73325095297231
      },
      "recorded_metric_max_abs_difference": 5.960464477539062e-07,
      "window_frames": 4,
      "window_ms": 320.0,
      "scalar_tokens": 64,
      "codebooks": 16,
      "seam_count": 33,
      "id": "qwen3-tts-tokenizer-12hz--podcast-800--mr_stft--30-40",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 33.377362450301206,
        "log_mel": 57.910842900685424,
        "log1p_mel": 72.73325095297231
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "40-50",
      "audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-2.wav",
      "baseline": {
        "log_mel": 0.3232126235961914,
        "log1p_mel": 0.010577854700386524,
        "mr_stft": 1.0321029424667358
      },
      "treatment": {
        "log_mel": 0.5933395028114319,
        "log1p_mel": 0.02173606865108013,
        "mr_stft": 1.5029634237289429
      },
      "decoder_drift": {
        "log_mel": 0.48997175693511963,
        "log1p_mel": 0.0180498119443655,
        "mr_stft": 1.2368857860565186
      },
      "relative_percent": {
        "mr_stft": 45.621464864429704,
        "log_mel": 83.57559683458587,
        "log1p_mel": 105.48654965251015
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 2,
      "window_ms": 160.0,
      "scalar_tokens": 32,
      "codebooks": 16,
      "seam_count": 66,
      "id": "qwen3-tts-tokenizer-12hz--podcast-800--mr_stft--40-50",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 45.621464864429704,
        "log_mel": 83.57559683458587,
        "log1p_mel": 105.48654965251015
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "mr_stft",
      "band_id": "over-50",
      "audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-1.wav",
      "baseline": {
        "log_mel": 0.3232126235961914,
        "log1p_mel": 0.010577854700386524,
        "mr_stft": 1.0321029424667358
      },
      "treatment": {
        "log_mel": 0.6995828151702881,
        "log1p_mel": 0.026284461840987206,
        "mr_stft": 1.6217248439788818
      },
      "decoder_drift": {
        "log_mel": 0.626590371131897,
        "log1p_mel": 0.023373715579509735,
        "mr_stft": 1.4673494100570679
      },
      "relative_percent": {
        "mr_stft": 57.12820662083804,
        "log_mel": 116.44662494504487,
        "log1p_mel": 148.48575240900925
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 1,
      "window_ms": 80.0,
      "scalar_tokens": 16,
      "codebooks": 16,
      "seam_count": 132,
      "id": "qwen3-tts-tokenizer-12hz--podcast-800--mr_stft--over-50",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 57.12820662083804,
        "log_mel": 116.44662494504487,
        "log1p_mel": 148.48575240900925
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "0-5",
      "audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-123.wav",
      "baseline": {
        "log_mel": 0.3232126235961914,
        "log1p_mel": 0.010577854700386524,
        "mr_stft": 1.0321029424667358
      },
      "treatment": {
        "log_mel": 0.3311849534511566,
        "log1p_mel": 0.010840047150850296,
        "mr_stft": 1.0449037551879883
      },
      "decoder_drift": {
        "log_mel": 0.017283739522099495,
        "log1p_mel": 0.000520397734362632,
        "mr_stft": 0.07838082313537598
      },
      "relative_percent": {
        "mr_stft": 1.2402651125728248,
        "log_mel": 2.4665898770480865,
        "log1p_mel": 2.478692115653569
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 123,
      "window_ms": 9840.0,
      "scalar_tokens": 1968,
      "codebooks": 16,
      "seam_count": 1,
      "id": "qwen3-tts-tokenizer-12hz--podcast-800--log_mel--0-5",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 1.2402651125728248,
        "log_mel": 2.4665898770480865,
        "log1p_mel": 2.478692115653569
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "5-10",
      "audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-82.wav",
      "baseline": {
        "log_mel": 0.3232126235961914,
        "log1p_mel": 0.010577854700386524,
        "mr_stft": 1.0321029424667358
      },
      "treatment": {
        "log_mel": 0.34744930267333984,
        "log1p_mel": 0.011191904544830322,
        "mr_stft": 1.1024134159088135
      },
      "decoder_drift": {
        "log_mel": 0.06946462392807007,
        "log1p_mel": 0.002059906953945756,
        "mr_stft": 0.28484171628952026
      },
      "relative_percent": {
        "mr_stft": 6.8123508372173625,
        "log_mel": 7.498679602139791,
        "log1p_mel": 5.805050852337379
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 82,
      "window_ms": 6560.0,
      "scalar_tokens": 1312,
      "codebooks": 16,
      "seam_count": 1,
      "id": "qwen3-tts-tokenizer-12hz--podcast-800--log_mel--5-10",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 6.8123508372173625,
        "log_mel": 7.498679602139791,
        "log1p_mel": 5.805050852337379
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "10-20",
      "audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-30.wav",
      "baseline": {
        "log_mel": 0.3232126235961914,
        "log1p_mel": 0.010577854700386524,
        "mr_stft": 1.0321029424667358
      },
      "treatment": {
        "log_mel": 0.37119174003601074,
        "log1p_mel": 0.012076837942004204,
        "mr_stft": 1.1103347539901733
      },
      "decoder_drift": {
        "log_mel": 0.15236765146255493,
        "log1p_mel": 0.004489606246352196,
        "mr_stft": 0.4535512924194336
      },
      "relative_percent": {
        "mr_stft": 7.579845798759455,
        "log_mel": 14.844443854322487,
        "log1p_mel": 14.170957004759256
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-07,
      "window_frames": 30,
      "window_ms": 2400.0,
      "scalar_tokens": 480,
      "codebooks": 16,
      "seam_count": 4,
      "id": "qwen3-tts-tokenizer-12hz--podcast-800--log_mel--10-20",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 7.579857348895058,
        "log_mel": 14.844453074977949,
        "log1p_mel": 14.170957004759256
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "20-30",
      "audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-13.wav",
      "baseline": {
        "log_mel": 0.3232126235961914,
        "log1p_mel": 0.010577854700386524,
        "mr_stft": 1.0321029424667358
      },
      "treatment": {
        "log_mel": 0.4050700068473816,
        "log1p_mel": 0.013535328209400177,
        "mr_stft": 1.1792773008346558
      },
      "decoder_drift": {
        "log_mel": 0.22300095856189728,
        "log1p_mel": 0.007228419650346041,
        "mr_stft": 0.6074103116989136
      },
      "relative_percent": {
        "mr_stft": 14.25965882978415,
        "log_mel": 25.326171465833426,
        "log1p_mel": 27.959105062254096
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-07,
      "window_frames": 13,
      "window_ms": 1040.0,
      "scalar_tokens": 208,
      "codebooks": 16,
      "seam_count": 10,
      "id": "qwen3-tts-tokenizer-12hz--podcast-800--log_mel--20-30",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 14.259670379919775,
        "log_mel": 25.326171465833426,
        "log1p_mel": 27.959105062254096
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "30-40",
      "audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-10.wav",
      "baseline": {
        "log_mel": 0.3232126235961914,
        "log1p_mel": 0.010577854700386524,
        "mr_stft": 1.0321029424667358
      },
      "treatment": {
        "log_mel": 0.4332602024078369,
        "log1p_mel": 0.014396782964468002,
        "mr_stft": 1.2409396171569824
      },
      "decoder_drift": {
        "log_mel": 0.2557380497455597,
        "log1p_mel": 0.008514409884810448,
        "mr_stft": 0.7206931114196777
      },
      "relative_percent": {
        "mr_stft": 20.234093528609165,
        "log_mel": 34.048044778453466,
        "log1p_mel": 36.10305087610941
      },
      "recorded_metric_max_abs_difference": 7.152557373046875e-07,
      "window_frames": 10,
      "window_ms": 800.0,
      "scalar_tokens": 160,
      "codebooks": 16,
      "seam_count": 13,
      "id": "qwen3-tts-tokenizer-12hz--podcast-800--log_mel--30-40",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 20.234162829422853,
        "log_mel": 34.048044778453466,
        "log1p_mel": 36.103024462741075
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "40-50",
      "audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-6.wav",
      "baseline": {
        "log_mel": 0.3232126235961914,
        "log1p_mel": 0.010577854700386524,
        "mr_stft": 1.0321029424667358
      },
      "treatment": {
        "log_mel": 0.45994555950164795,
        "log1p_mel": 0.016962477937340736,
        "mr_stft": 1.3116974830627441
      },
      "decoder_drift": {
        "log_mel": 0.3140912353992462,
        "log1p_mel": 0.01173525769263506,
        "mr_stft": 0.883681058883667
      },
      "relative_percent": {
        "mr_stft": 27.08979202479307,
        "log_mel": 42.30433031485956,
        "log1p_mel": 60.35839419046769
      },
      "recorded_metric_max_abs_difference": 2.9802322387695312e-08,
      "window_frames": 6,
      "window_ms": 480.0,
      "scalar_tokens": 96,
      "codebooks": 16,
      "seam_count": 22,
      "id": "qwen3-tts-tokenizer-12hz--podcast-800--log_mel--40-50",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 27.08979202479307,
        "log_mel": 42.30432109420412,
        "log1p_mel": 60.35839419046769
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log_mel",
      "band_id": "over-50",
      "audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-2.wav",
      "baseline": {
        "log_mel": 0.3232126235961914,
        "log1p_mel": 0.010577854700386524,
        "mr_stft": 1.0321029424667358
      },
      "treatment": {
        "log_mel": 0.5933395028114319,
        "log1p_mel": 0.02173606865108013,
        "mr_stft": 1.5029634237289429
      },
      "decoder_drift": {
        "log_mel": 0.48997175693511963,
        "log1p_mel": 0.0180498119443655,
        "mr_stft": 1.2368857860565186
      },
      "relative_percent": {
        "mr_stft": 45.621464864429704,
        "log_mel": 83.57559683458587,
        "log1p_mel": 105.48654965251015
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 2,
      "window_ms": 160.0,
      "scalar_tokens": 32,
      "codebooks": 16,
      "seam_count": 66,
      "id": "qwen3-tts-tokenizer-12hz--podcast-800--log_mel--over-50",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 45.621464864429704,
        "log_mel": 83.57559683458587,
        "log1p_mel": 105.48654965251015
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "0-5",
      "audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-89.wav",
      "baseline": {
        "log_mel": 0.3232126235961914,
        "log1p_mel": 0.010577854700386524,
        "mr_stft": 1.0321029424667358
      },
      "treatment": {
        "log_mel": 0.3328109383583069,
        "log1p_mel": 0.010843442752957344,
        "mr_stft": 1.0613081455230713
      },
      "decoder_drift": {
        "log_mel": 0.04610522836446762,
        "log1p_mel": 0.0013594392221421003,
        "mr_stft": 0.18264417350292206
      },
      "relative_percent": {
        "mr_stft": 2.8296792746792088,
        "log_mel": 2.969659617659981,
        "log1p_mel": 2.510793162635472
      },
      "recorded_metric_max_abs_difference": 0.0,
      "window_frames": 89,
      "window_ms": 7120.0,
      "scalar_tokens": 1424,
      "codebooks": 16,
      "seam_count": 1,
      "id": "qwen3-tts-tokenizer-12hz--podcast-800--log1p_mel--0-5",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 2.8296792746792088,
        "log_mel": 2.969659617659981,
        "log1p_mel": 2.510793162635472
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "5-10",
      "audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-57.wav",
      "baseline": {
        "log_mel": 0.3232126235961914,
        "log1p_mel": 0.010577854700386524,
        "mr_stft": 1.0321029424667358
      },
      "treatment": {
        "log_mel": 0.34884703159332275,
        "log1p_mel": 0.01135667972266674,
        "mr_stft": 1.0704491138458252
      },
      "decoder_drift": {
        "log_mel": 0.09176228195428848,
        "log1p_mel": 0.002653856063261628,
        "mr_stft": 0.2693573832511902
      },
      "relative_percent": {
        "mr_stft": 3.7153436737077383,
        "log_mel": 7.9311283426720225,
        "log1p_mel": 7.362788054290048
      },
      "recorded_metric_max_abs_difference": 7.152557373046875e-07,
      "window_frames": 57,
      "window_ms": 4560.0,
      "scalar_tokens": 912,
      "codebooks": 16,
      "seam_count": 2,
      "id": "qwen3-tts-tokenizer-12hz--podcast-800--log1p_mel--5-10",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 3.715412974521448,
        "log_mel": 7.931119122016561,
        "log1p_mel": 7.362788054290048
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "10-20",
      "audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-41.wav",
      "baseline": {
        "log_mel": 0.3232126235961914,
        "log1p_mel": 0.010577854700386524,
        "mr_stft": 1.0321029424667358
      },
      "treatment": {
        "log_mel": 0.3793738782405853,
        "log1p_mel": 0.012165657244622707,
        "mr_stft": 1.152051329612732
      },
      "decoder_drift": {
        "log_mel": 0.15291330218315125,
        "log1p_mel": 0.004394506569951773,
        "mr_stft": 0.4705269932746887
      },
      "relative_percent": {
        "mr_stft": 11.621746456736016,
        "log_mel": 17.375947145727654,
        "log1p_mel": 15.0106291796404
      },
      "recorded_metric_max_abs_difference": 1.1920928955078125e-07,
      "window_frames": 41,
      "window_ms": 3280.0,
      "scalar_tokens": 656,
      "codebooks": 16,
      "seam_count": 3,
      "id": "qwen3-tts-tokenizer-12hz--podcast-800--log1p_mel--10-20",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 11.62175800687164,
        "log_mel": 17.375965587038557,
        "log1p_mel": 15.0106291796404
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "20-30",
      "audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-18.wav",
      "baseline": {
        "log_mel": 0.3232126235961914,
        "log1p_mel": 0.010577854700386524,
        "mr_stft": 1.0321029424667358
      },
      "treatment": {
        "log_mel": 0.391950786113739,
        "log1p_mel": 0.013115247711539268,
        "mr_stft": 1.171886682510376
      },
      "decoder_drift": {
        "log_mel": 0.203641876578331,
        "log1p_mel": 0.006324861664324999,
        "mr_stft": 0.5890618562698364
      },
      "relative_percent": {
        "mr_stft": 13.543585072004127,
        "log_mel": 21.267165172182835,
        "log1p_mel": 23.987784697591152
      },
      "recorded_metric_max_abs_difference": 4.76837158203125e-07,
      "window_frames": 18,
      "window_ms": 1440.0,
      "scalar_tokens": 288,
      "codebooks": 16,
      "seam_count": 7,
      "id": "qwen3-tts-tokenizer-12hz--podcast-800--log1p_mel--20-30",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 13.543631272546586,
        "log_mel": 21.26717439283827,
        "log1p_mel": 23.987784697591152
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "30-40",
      "audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-14.wav",
      "baseline": {
        "log_mel": 0.3232126235961914,
        "log1p_mel": 0.010577854700386524,
        "mr_stft": 1.0321029424667358
      },
      "treatment": {
        "log_mel": 0.417098730802536,
        "log1p_mel": 0.014279940165579319,
        "mr_stft": 1.217085361480713
      },
      "decoder_drift": {
        "log_mel": 0.23983731865882874,
        "log1p_mel": 0.008164267055690289,
        "mr_stft": 0.6745748519897461
      },
      "relative_percent": {
        "mr_stft": 17.922865191321648,
        "log_mel": 29.04778475596983,
        "log1p_mel": 34.99845261683843
      },
      "recorded_metric_max_abs_difference": 4.76837158203125e-07,
      "window_frames": 14,
      "window_ms": 1120.0,
      "scalar_tokens": 224,
      "codebooks": 16,
      "seam_count": 9,
      "id": "qwen3-tts-tokenizer-12hz--podcast-800--log1p_mel--30-40",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 17.922818990779188,
        "log_mel": 29.04781241793617,
        "log1p_mel": 34.99843500792621
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "40-50",
      "audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-11.wav",
      "baseline": {
        "log_mel": 0.3232126235961914,
        "log1p_mel": 0.010577854700386524,
        "mr_stft": 1.0321029424667358
      },
      "treatment": {
        "log_mel": 0.42066410183906555,
        "log1p_mel": 0.01538772415369749,
        "mr_stft": 1.2468743324279785
      },
      "decoder_drift": {
        "log_mel": 0.24510549008846283,
        "log1p_mel": 0.009505770169198513,
        "mr_stft": 0.7269539833068848
      },
      "relative_percent": {
        "mr_stft": 20.809105480111988,
        "log_mel": 30.150888649889506,
        "log1p_mel": 45.4711242454031
      },
      "recorded_metric_max_abs_difference": 3.5762786865234375e-07,
      "window_frames": 11,
      "window_ms": 880.0,
      "scalar_tokens": 176,
      "codebooks": 16,
      "seam_count": 12,
      "id": "qwen3-tts-tokenizer-12hz--podcast-800--log1p_mel--40-50",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 20.809140130518845,
        "log_mel": 30.150888649889506,
        "log1p_mel": 45.47113304985919
      }
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "metric": "log1p_mel",
      "band_id": "over-50",
      "audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-4.wav",
      "baseline": {
        "log_mel": 0.3232126235961914,
        "log1p_mel": 0.010577854700386524,
        "mr_stft": 1.0321029424667358
      },
      "treatment": {
        "log_mel": 0.5103878378868103,
        "log1p_mel": 0.018271472305059433,
        "mr_stft": 1.3765910863876343
      },
      "decoder_drift": {
        "log_mel": 0.3819350302219391,
        "log1p_mel": 0.013622327707707882,
        "mr_stft": 1.0004022121429443
      },
      "relative_percent": {
        "mr_stft": 33.377304699623124,
        "log_mel": 57.91086134199632,
        "log1p_mel": 72.73325095297231
      },
      "recorded_metric_max_abs_difference": 5.960464477539062e-07,
      "window_frames": 4,
      "window_ms": 320.0,
      "scalar_tokens": 64,
      "codebooks": 16,
      "seam_count": 33,
      "id": "qwen3-tts-tokenizer-12hz--podcast-800--log1p_mel--over-50",
      "original_audio": "listening-audio/podcast-800-original.wav",
      "full_audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-full.wav",
      "recorded_relative_percent": {
        "mr_stft": 33.377362450301206,
        "log_mel": 57.910842900685424,
        "log1p_mel": 72.73325095297231
      }
    }
  ],
  "controls": [
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "audio": "listening-audio/cartoon-voices-descript-dac-44khz-full.wav",
      "baseline": {
        "log_mel": 0.3338501751422882,
        "log1p_mel": 0.0026197368279099464,
        "mr_stft": 0.9093503952026367
      },
      "resume_baseline_max_abs": 0.0
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "audio": "listening-audio/librispeech-other-500-descript-dac-44khz-full.wav",
      "baseline": {
        "log_mel": 0.2750609219074249,
        "log1p_mel": 0.001427017035894096,
        "mr_stft": 0.9531030654907227
      },
      "resume_baseline_max_abs": 0.0
    },
    {
      "model_id": "descript-dac-44khz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "audio": "listening-audio/podcast-800-descript-dac-44khz-full.wav",
      "baseline": {
        "log_mel": 0.28345760703086853,
        "log1p_mel": 0.007421256974339485,
        "mr_stft": 0.8638533353805542
      },
      "resume_baseline_max_abs": 0.0
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "audio": "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-full.wav",
      "baseline": {
        "log_mel": 0.45535868406295776,
        "log1p_mel": 0.004118511453270912,
        "mr_stft": 1.0745127201080322
      },
      "resume_baseline_max_abs": 0.0
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "audio": "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-full.wav",
      "baseline": {
        "log_mel": 0.40510469675064087,
        "log1p_mel": 0.002791645936667919,
        "mr_stft": 1.1514151096343994
      },
      "resume_baseline_max_abs": 0.0
    },
    {
      "model_id": "fish-400m-morepitchloss1-step3802000",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "audio": "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-full.wav",
      "baseline": {
        "log_mel": 0.3965880870819092,
        "log1p_mel": 0.012173594906926155,
        "mr_stft": 1.0366172790527344
      },
      "resume_baseline_max_abs": 0.0
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "audio": "listening-audio/cartoon-voices-fish-dac-s2-full.wav",
      "baseline": {
        "log_mel": 0.3437022268772125,
        "log1p_mel": 0.0027478106785565615,
        "mr_stft": 0.8997645378112793
      },
      "resume_baseline_max_abs": 0.0
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "audio": "listening-audio/librispeech-other-500-fish-dac-s2-full.wav",
      "baseline": {
        "log_mel": 0.312560111284256,
        "log1p_mel": 0.0020922308322042227,
        "mr_stft": 0.953180193901062
      },
      "resume_baseline_max_abs": 0.0
    },
    {
      "model_id": "fish-dac-s2",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "audio": "listening-audio/podcast-800-fish-dac-s2-full.wav",
      "baseline": {
        "log_mel": 0.3086870610713959,
        "log1p_mel": 0.00981848780065775,
        "mr_stft": 0.9192757606506348
      },
      "resume_baseline_max_abs": 0.0
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "cartoon-voices",
      "clip_id": "Steve-fim_s5e9-866.113039-871.879677--NEUTRAL",
      "audio": "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-full.wav",
      "baseline": {
        "log_mel": 0.35760822892189026,
        "log1p_mel": 0.0033701565116643906,
        "mr_stft": 1.0018222332000732
      },
      "resume_baseline_max_abs": 0.0
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "librispeech-other-500",
      "clip_id": "5764-299665-0035",
      "audio": "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-full.wav",
      "baseline": {
        "log_mel": 0.3318670094013214,
        "log1p_mel": 0.0022429267410188913,
        "mr_stft": 1.1133737564086914
      },
      "resume_baseline_max_abs": 0.0
    },
    {
      "model_id": "qwen3-tts-tokenizer-12hz",
      "dataset_id": "podcast-800",
      "clip_id": "d98de9e2-a658-4fbb-af5d-9e9244ec29ad_seg00085_0000536937_0000547500",
      "audio": "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-full.wav",
      "baseline": {
        "log_mel": 0.3232126235961914,
        "log1p_mel": 0.010577854700386524,
        "mr_stft": 1.0321029424667358
      },
      "resume_baseline_max_abs": 0.0
    }
  ],
  "assets": {
    "listening-audio/cartoon-voices-original.wav": {
      "sha256": "9a7245d02b51f2f684d07807bd7bcbda08f656946d0845c04e7d462f9f70dfcf",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.31736400723457336
    },
    "listening-audio/librispeech-other-500-original.wav": {
      "sha256": "52de5012854452173b8baca196bab6bc32aec8cd1a5867047d918ba5c72abe61",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.33699092268943787
    },
    "listening-audio/podcast-800-original.wav": {
      "sha256": "d94e59e3656828bb3883a81a1e66c9d40136beb9b65c17058ea65fdd24622e46",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 1.153206467628479
    },
    "listening-audio/cartoon-voices-descript-dac-44khz-full.wav": {
      "sha256": "d60be1433841e6b784009ad33bce46e5a186458c28f16f79311b1ea878317a60",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.2961508333683014
    },
    "listening-audio/cartoon-voices-descript-dac-44khz-window-4.wav": {
      "sha256": "99b3951b3254c681175d67289e4033b08943273a2b0c4378df0fc5552caaa131",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.4155099391937256
    },
    "listening-audio/cartoon-voices-descript-dac-44khz-window-6.wav": {
      "sha256": "7e5b3248612ac985b9d88669b428da27ad56dfdafe39b92b48a340658cf70599",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.4530717730522156
    },
    "listening-audio/cartoon-voices-descript-dac-44khz-window-8.wav": {
      "sha256": "3986463ea4cc337a6c994fcf4e0e9201473ea28ebc57a9bfb3253fcbc0a19397",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.3782024383544922
    },
    "listening-audio/cartoon-voices-descript-dac-44khz-window-10.wav": {
      "sha256": "4cbb106b0392d8f9bf431a6aaf80cee9c3cfd147e2bdfc6399e8da804a2d1bac",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.3849093019962311
    },
    "listening-audio/cartoon-voices-descript-dac-44khz-window-12.wav": {
      "sha256": "e4fd9148dedd62c23c834b975708b8a8cad6488e795b28b116825c265574a1ac",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.3778926134109497
    },
    "listening-audio/cartoon-voices-descript-dac-44khz-window-13.wav": {
      "sha256": "062282fccecae660fee42a374888de58a951195dadb80362bbdb8250352ae1d1",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.3376688063144684
    },
    "listening-audio/cartoon-voices-descript-dac-44khz-window-20.wav": {
      "sha256": "c72d7e2394122406429eb6befc87cc2ec476d0d20f780b9075f66cf3729ee4d3",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.31807228922843933
    },
    "listening-audio/cartoon-voices-descript-dac-44khz-window-21.wav": {
      "sha256": "e82f49ea5c5c812c08e8679b2200ddb749015602c674dbf933445086b565f780",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.350008487701416
    },
    "listening-audio/cartoon-voices-descript-dac-44khz-window-26.wav": {
      "sha256": "9b9d2e935911240840572c507cd9fdd6bbf405e0fc96e82147ec6efddbd85bc1",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.2961508333683014
    },
    "listening-audio/cartoon-voices-descript-dac-44khz-window-30.wav": {
      "sha256": "1c321cd0c14292fecd35c542b14d614670315403adee65fd0286012b4e4ccca3",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.3849092423915863
    },
    "listening-audio/cartoon-voices-descript-dac-44khz-window-38.wav": {
      "sha256": "38e0c5a31c8d063f0fdd6de1e2bdc4c19f148c08184e3248fd7577b63b9348da",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.296150803565979
    },
    "listening-audio/cartoon-voices-descript-dac-44khz-window-59.wav": {
      "sha256": "025468a67aeafdd4ac4f109e04d615c053f32705f1def10f0cec5d15b1274c47",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.4529678225517273
    },
    "listening-audio/cartoon-voices-descript-dac-44khz-window-72.wav": {
      "sha256": "fa7feea3f580d5db34dbe9a018af7b03b33f08a0a8538e9b7ddd9589879a518f",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.3778926134109497
    },
    "listening-audio/cartoon-voices-descript-dac-44khz-window-77.wav": {
      "sha256": "9ad503bf5f9b7e9fcbf2eec81784b275bef2725562618b441a58cff27a3d872d",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.350008487701416
    },
    "listening-audio/cartoon-voices-descript-dac-44khz-window-125.wav": {
      "sha256": "bd692c49564cd0988b54a5f6896adccd8bc091d27fb8cef390a748ffdf90dfd5",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.29662641882896423
    },
    "listening-audio/cartoon-voices-descript-dac-44khz-window-134.wav": {
      "sha256": "7305eda8f6dca4a185fadcfe5c9628c2f8168b76cf380b2047d1d3f4021ad451",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.2961508333683014
    },
    "listening-audio/cartoon-voices-descript-dac-44khz-window-142.wav": {
      "sha256": "5315176424c4a83055329d360b014c5e770d795aac351eb1c2e2b20897184285",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.2961508333683014
    },
    "listening-audio/cartoon-voices-descript-dac-44khz-window-160.wav": {
      "sha256": "cc5b2fbc0546b209b625b128100457cef1cb61d941a4752b9ddb92fb2417d3d9",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.2961508333683014
    },
    "listening-audio/librispeech-other-500-descript-dac-44khz-full.wav": {
      "sha256": "5103fe08d2f873c822a1ae8368ff53b95c4e25c465b4f009f2431814d0730860",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3166876435279846
    },
    "listening-audio/librispeech-other-500-descript-dac-44khz-window-7.wav": {
      "sha256": "788042c617c99fa7902bc95e621b46852f155d5ad4054ecb7c4fc04d26c896a4",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3119216859340668
    },
    "listening-audio/librispeech-other-500-descript-dac-44khz-window-10.wav": {
      "sha256": "e1857b2b20290de88f75643629ccb8adf17504a8c286873a51b63de6884b3339",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3159022629261017
    },
    "listening-audio/librispeech-other-500-descript-dac-44khz-window-11.wav": {
      "sha256": "47051a7780c31ca4b55922816895fc713dd2feabf8d944bcbb062579d25409e6",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.29517826437950134
    },
    "listening-audio/librispeech-other-500-descript-dac-44khz-window-14.wav": {
      "sha256": "adfab89079a99daa30ee8dc0d83577e9ca329dcfe934ecec0ad72cc84eb9238a",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.30625057220458984
    },
    "listening-audio/librispeech-other-500-descript-dac-44khz-window-19.wav": {
      "sha256": "5493be96728118a94ba1e03492b0cfa09e49f5ddc24913c5f04438eba728eee4",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.31682631373405457
    },
    "listening-audio/librispeech-other-500-descript-dac-44khz-window-20.wav": {
      "sha256": "8cb608eb9b0c48fe979e881c73ee018181c9b9dd031d27c2c46b438a5fb69613",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.31275010108947754
    },
    "listening-audio/librispeech-other-500-descript-dac-44khz-window-21.wav": {
      "sha256": "bcc4ebb608aee2285fd70f956e694351c54b1d32d1d4756e8ea4f3605ea2f07c",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.30625057220458984
    },
    "listening-audio/librispeech-other-500-descript-dac-44khz-window-24.wav": {
      "sha256": "5ffaa6d7c507a5f487220ad2cc58f0264fd2d8f1a57c7acb1bce1132263cc028",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.33149436116218567
    },
    "listening-audio/librispeech-other-500-descript-dac-44khz-window-29.wav": {
      "sha256": "e5bc54538579359df37f221e1eed9d66e0eb8de7d19fe206caf8014df0ccd46c",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3166876435279846
    },
    "listening-audio/librispeech-other-500-descript-dac-44khz-window-31.wav": {
      "sha256": "f1bbcb55ca4b1f8c63bc064b1da18c66fdf03ccb2bed8d7e696be90336e2e12b",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3166876435279846
    },
    "listening-audio/librispeech-other-500-descript-dac-44khz-window-32.wav": {
      "sha256": "90d163598f01005e0e935dc55395387b705e725b323f645578aa6b126c9d03c2",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.31668758392333984
    },
    "listening-audio/librispeech-other-500-descript-dac-44khz-window-35.wav": {
      "sha256": "58bd2c8f06d79d211c6bc1eecf65bea664e661035a12cf06b266d50d1a813260",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3166753351688385
    },
    "listening-audio/librispeech-other-500-descript-dac-44khz-window-41.wav": {
      "sha256": "270b1b577d2ef030b8263d08d36ca1dab57563d69fb2fb88d49cdf14e7a6934e",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3136342465877533
    },
    "listening-audio/librispeech-other-500-descript-dac-44khz-window-43.wav": {
      "sha256": "d4551480ba054d6135f8164837f6e359d6ee4d43378f02148ef1a875f1058a5f",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.320721834897995
    },
    "listening-audio/librispeech-other-500-descript-dac-44khz-window-60.wav": {
      "sha256": "2e3f7745dc22bf79982aae928af3b1612b7e25b7ad926f705f0c3e8128b2027b",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3166876435279846
    },
    "listening-audio/librispeech-other-500-descript-dac-44khz-window-73.wav": {
      "sha256": "0fbbf75546fe43dbc2287a755c2b6dd0317980c41144aa7ee387551ae9b883cb",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3166876435279846
    },
    "listening-audio/librispeech-other-500-descript-dac-44khz-window-125.wav": {
      "sha256": "9d2c91e1c07ac23b4af54056dc5572397e2cddf0d8f7019a955aafbe2bad7382",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3166876435279846
    },
    "listening-audio/librispeech-other-500-descript-dac-44khz-window-126.wav": {
      "sha256": "22c7fa726c9c6010d63ff28dd57aae6b01c7d49921f988e0ebdd22f48340e06a",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3166876435279846
    },
    "listening-audio/librispeech-other-500-descript-dac-44khz-window-151.wav": {
      "sha256": "974335a59172fcf3e2c858a8efe6e5009a05b68cdf3e280f7aa14e369e1ae5b8",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3166876435279846
    },
    "listening-audio/librispeech-other-500-descript-dac-44khz-window-158.wav": {
      "sha256": "403e731bc69e761548159658ff554025399e1a55dfadbd2a03b9108457a7e54e",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3166876435279846
    },
    "listening-audio/librispeech-other-500-descript-dac-44khz-window-320.wav": {
      "sha256": "9c99108488bdc29481a964071312776b99275f42b29ab43f58c371dfa4fefb2a",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3166876435279846
    },
    "listening-audio/podcast-800-descript-dac-44khz-full.wav": {
      "sha256": "e0b9c086b01e11139329269344edf511da4a78eb1f3e9746f22993ca0532d4aa",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9982858896255493
    },
    "listening-audio/podcast-800-descript-dac-44khz-window-3.wav": {
      "sha256": "3e32bff490a1f5528a37a87e68561533f48f78b3541a38c3f34f8a05cd0f2324",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 1.003509521484375
    },
    "listening-audio/podcast-800-descript-dac-44khz-window-5.wav": {
      "sha256": "d18a7619011fceadfc801bce0aefe340ace3d4220e2ca825bd802953633c2fe6",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.998826265335083
    },
    "listening-audio/podcast-800-descript-dac-44khz-window-6.wav": {
      "sha256": "aa02668142f77d7b12d384b6b88a1cb0e4ba7518caf2558a9ddd469aba05fd01",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9999992847442627
    },
    "listening-audio/podcast-800-descript-dac-44khz-window-7.wav": {
      "sha256": "0304aaac3f9fa8f1d7fce86017675352da649993b7c440faf6bdc6a86be65054",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 1.0003595352172852
    },
    "listening-audio/podcast-800-descript-dac-44khz-window-9.wav": {
      "sha256": "f7d48eb25668f3b28fdd6c7c9c770612bf9e7ed905409f75f23ee65f8e8a01b4",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.997780978679657
    },
    "listening-audio/podcast-800-descript-dac-44khz-window-10.wav": {
      "sha256": "a1fead6d0614bbf25f29c5e92c34893f4e9aaef9db8caa598318ae06741daed8",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9982442259788513
    },
    "listening-audio/podcast-800-descript-dac-44khz-window-14.wav": {
      "sha256": "7235ad204eb71ca94b72b7f8c61898e173fa7e8caa52878216ee0fd3bdb5e1f6",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.999575138092041
    },
    "listening-audio/podcast-800-descript-dac-44khz-window-15.wav": {
      "sha256": "5e7eea37d726520a9c77649d0342bd3eed0445a7bdb132b19fcaa6c45c2ac167",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9981449246406555
    },
    "listening-audio/podcast-800-descript-dac-44khz-window-19.wav": {
      "sha256": "8815605a24f4030017a102ba3306fdd59305e2779216b43487f0cf2e8d3f4744",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9996904134750366
    },
    "listening-audio/podcast-800-descript-dac-44khz-window-20.wav": {
      "sha256": "30785cf43a381a39d6bc4ebcecc1f2d432183c298fa36dd6b01d16f6ecb28a1b",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9983146786689758
    },
    "listening-audio/podcast-800-descript-dac-44khz-window-27.wav": {
      "sha256": "301f5e66214f9769dc89fd2ca2016321220b79c5f89944cff81d45aaea92759a",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9982858896255493
    },
    "listening-audio/podcast-800-descript-dac-44khz-window-33.wav": {
      "sha256": "56b69f9733de5dbe263a920593721b1bfec95d184f89180d4de4665693a52de8",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9982858896255493
    },
    "listening-audio/podcast-800-descript-dac-44khz-window-38.wav": {
      "sha256": "44705713164496024ec2e47bfcf9eceb22578070061d934a1a8634ecc46585d2",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9996904134750366
    },
    "listening-audio/podcast-800-descript-dac-44khz-window-44.wav": {
      "sha256": "331e9694534f8c069ff93cd931873b3bcc88987af4239046bb0dade929297db6",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9982858896255493
    },
    "listening-audio/podcast-800-descript-dac-44khz-window-64.wav": {
      "sha256": "1371975d43d6278dd1b7c28197da8ca81a64341c2728935b7321160456d715a4",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9982858896255493
    },
    "listening-audio/podcast-800-descript-dac-44khz-window-86.wav": {
      "sha256": "82580e5cccbfe88336c7801a4e7cfade34ef043f7ec7d6aa94f8c0487b699fe2",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9982858896255493
    },
    "listening-audio/podcast-800-descript-dac-44khz-window-88.wav": {
      "sha256": "098e8c07a5837c97de308166f48a6127411e66cbbb6ca83589940860ea5ea5fa",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9982858896255493
    },
    "listening-audio/podcast-800-descript-dac-44khz-window-135.wav": {
      "sha256": "f1e88d1061de9b6aa5704f1c270b7a4b8c20d0525dcb1648f3c8c38b3308ada5",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9982858896255493
    },
    "listening-audio/podcast-800-descript-dac-44khz-window-142.wav": {
      "sha256": "2d54845b3f7447ad973539b02c34df6176d3d8c9a705f121d86c26bb078f2e36",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9989706873893738
    },
    "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-full.wav": {
      "sha256": "578b9a1aa1459677d67c18962cf0048e7274095eeaa664a96719687b9ca0b30f",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.37181031703948975
    },
    "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-1.wav": {
      "sha256": "adcff8be75a982a38abb08d8ac6d8dd0be97972e0971a77368ccb97c9a9bbfb7",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.5534634590148926
    },
    "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-2.wav": {
      "sha256": "8efa9b9ed905a3b0aa9b476c61b640677150b3ed5f1864590f66bc673f97029b",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.5534634590148926
    },
    "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-3.wav": {
      "sha256": "612f8067462437d5a6ada98505d9e4c86c7002543b91c7086fbbabc6f8a37933",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.5534634590148926
    },
    "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-4.wav": {
      "sha256": "958d4c02da71ff3e9bcb38b6271d6c62bd821432bd9b16f6ef687f508eff49cb",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.3589687943458557
    },
    "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-5.wav": {
      "sha256": "0c3330afaa776d13383cd5d1504470db888a255912db20c506715ed554d4f6de",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.5534635782241821
    },
    "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-7.wav": {
      "sha256": "27f0fb2c16c4d83ac73ccd997a22cb55444dacad28aad994d5604faef76fdcba",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.37906184792518616
    },
    "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-8.wav": {
      "sha256": "a0eddd793dea1385902df4d08f1253fdfa2bf46cfd18beb184efa3d494faa98b",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.37961509823799133
    },
    "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-9.wav": {
      "sha256": "5c002a2719a9846e0744b9bc58d2b66d857878131591388effdd361fcbf889c3",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.37906184792518616
    },
    "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-10.wav": {
      "sha256": "1b7a9cc595f346a8339c54787c0acba5fc69d2669db2733d6ee2ddb8c2a90673",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.5534634590148926
    },
    "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-12.wav": {
      "sha256": "32a6c1772fea7a662b30bc073bf58ace1aa2dd801d37d2ca8ec1aad0da539786",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.3373153805732727
    },
    "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-14.wav": {
      "sha256": "c4e67ba7fdcf0e322962e0522ca1dcb290885733ba26ed095ed8945fbd00194a",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.35896891355514526
    },
    "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-15.wav": {
      "sha256": "2e06d80076bd0a9ee0d715d792e89e075799a87f4307ae06a119ed1c76ef02a9",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.5534635782241821
    },
    "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-16.wav": {
      "sha256": "24f4816c98c80e11a7a1e12706c7213956766055d77c40f333d0a474c64cfe05",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.36001989245414734
    },
    "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-30.wav": {
      "sha256": "fbf8d2f3f2934af45a483cfae2d1e2b185aadf058216cac6cde21b9f84976315",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.5534634590148926
    },
    "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-32.wav": {
      "sha256": "fd3188f294bbc5e292861d2f3c57732fa2bc9e4b61bff5baf55239c828d61b82",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.356831431388855
    },
    "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-36.wav": {
      "sha256": "6764dde2edaa8d03c4bd91c346a33ecb9772bd326d3a660357532b7b0c7e96a2",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.35277479887008667
    },
    "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-38.wav": {
      "sha256": "c66ac8155e82be77a60914d621f0ece3839601416273c42f385248f64e6975fb",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.3555578291416168
    },
    "listening-audio/cartoon-voices-fish-400m-morepitchloss1-step3802000-window-40.wav": {
      "sha256": "08b6d8950257c68a72ada96f60b08a8ab6c9100f22ca127e17e9e93e8de8c139",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.36001986265182495
    },
    "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-full.wav": {
      "sha256": "3a8c3a5063e847c665f0709c542927c4bc4966cf575dc7a117b79c7b3adebb10",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.32620948553085327
    },
    "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-2.wav": {
      "sha256": "fedef0b7f46d5e2f1d4c2181bfb3600134d82004fa876d62cb7e4cc498142bc5",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.39022234082221985
    },
    "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-3.wav": {
      "sha256": "45db51084b77bbaf3ba3784bb66b47b9b4d59022529420e2162aeb2026248c76",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.2924651801586151
    },
    "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-4.wav": {
      "sha256": "f97ff55a0f922409242083902de8dc039d8f2ca78c140ea8d5fa566c88dc3848",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.27036598324775696
    },
    "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-5.wav": {
      "sha256": "024402681ac814bedc79465ba87bb8ddd4e3fd19e3c6e285e8f3d77cd5ffb940",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3005635738372803
    },
    "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-6.wav": {
      "sha256": "947a6ca5914ac5c64defcb82ab53b7d65bb9299d1a3159a2af6763c0160e3774",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.28547751903533936
    },
    "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-8.wav": {
      "sha256": "56c01e1cafafc009d50f42a026103315b81fe47a3af7145d9e18d0c79d642459",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.273062527179718
    },
    "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-9.wav": {
      "sha256": "9156a9e07f27ecbc03b907c62235816014714c264b487050dcac845354d76a98",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.2854774594306946
    },
    "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-11.wav": {
      "sha256": "37aaca4c733f1ea5ce3825056a4cb1039e74758b42dd6bba9a74dccd156fafb6",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.30634376406669617
    },
    "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-14.wav": {
      "sha256": "1fb767993b63e363989bfe175cb3419c6df06a950c1d2fb2b101d606008b0978",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.39022207260131836
    },
    "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-16.wav": {
      "sha256": "9268b3e433f145a054c74e330a15d6a863c4a86a8b4d1cd64b6f5d36941fb4a1",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3262094557285309
    },
    "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-17.wav": {
      "sha256": "17cb0df51d4d4be6aaca58f61b318e97ade3be92dcd8cf5ce7752249cb426850",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3262096345424652
    },
    "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-18.wav": {
      "sha256": "4c0ddec456ee6a3c0e2978fba4907c402f691853b5dab49b56622473d64845cb",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3262094259262085
    },
    "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-19.wav": {
      "sha256": "acafba86764cadb4272c3e48d41d06350a64c4bd58182d68a51141ffbd7dba03",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3423286974430084
    },
    "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-38.wav": {
      "sha256": "319a22c4ed6463148a07b81ab41d3bc1e1af7472b45ffba00c143661b20190c4",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.34232866764068604
    },
    "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-39.wav": {
      "sha256": "eff177a59f440c074d7e203ccb286c2c95248c164a742dfb53b9f31448e0617b",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.32620957493782043
    },
    "listening-audio/librispeech-other-500-fish-400m-morepitchloss1-step3802000-window-96.wav": {
      "sha256": "ba17bb31145ebbc5850a4cf38c674211ef2f9dd701f7858be33a86c1fde2d6a3",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.32620951533317566
    },
    "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-full.wav": {
      "sha256": "60f83268a3116f6c05efc0837ef726ce04b4ffb55e62b17715db0ac0a47e21e4",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 1.0010534524917603
    },
    "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-1.wav": {
      "sha256": "fa35ce9f71a66c591a3bcdd41961c702ca98c7ba2ec90b04364844e13d3f9754",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 1.0037497282028198
    },
    "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-2.wav": {
      "sha256": "cf92c716ae91316b86c2b877d9fd862268450d578ca4baa22bfaff4dc0d8cb44",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 1.0005332231521606
    },
    "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-3.wav": {
      "sha256": "31fc5a3812b936bf0a42ae495a989f2a7c37cf9cbb66777e4add963f80082f32",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 1.00103759765625
    },
    "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-4.wav": {
      "sha256": "5c8cd627a19b114311f35dc33a8654426d210241139a9437cab63fc2787a3828",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 1.0005333423614502
    },
    "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-5.wav": {
      "sha256": "45346bb8b42d0062f47f75627c2d3cc9a4d2fa52fc3073a1bb17336516c956d1",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 1.0022244453430176
    },
    "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-6.wav": {
      "sha256": "5dbba8f5321ef53fd6b99781eaf77d1e0cf230f9664506263692a26ec46cee13",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 1.0005333423614502
    },
    "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-7.wav": {
      "sha256": "1820524643d3c54225f1cdaff80f9a0cc398cdb092ddebd5460741cb99a0da84",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9985457062721252
    },
    "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-8.wav": {
      "sha256": "488a4c58aec6bb49be7151f164178fcd5791a3a631832161d284716147676a48",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9987238645553589
    },
    "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-10.wav": {
      "sha256": "6abe11d95e3d08d00f03ca22766bb714600b798ff3709455cd8c0d1232ae43da",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9983984231948853
    },
    "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-11.wav": {
      "sha256": "4cc9b56cbb036fe4dcb62a0b1fa5997630c352b7b9fc86dd57153d00a3034a13",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 1.0010948181152344
    },
    "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-12.wav": {
      "sha256": "88ffecbdbea946f79fbe9918bf760076ab421e2029881f385d520aad6607f01a",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 1.0005332231521606
    },
    "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-13.wav": {
      "sha256": "382fa46fb39ad65622f726328c31f13ab19c1fb31d0a07b268d672bc64e1f8c3",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 1.0004730224609375
    },
    "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-17.wav": {
      "sha256": "2fef37f75053876b02609e6ee3b92393822a317bb430a2e0fe645dc883c9bff9",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 1.0022244453430176
    },
    "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-29.wav": {
      "sha256": "f7c1445e4f4907bee238d520e005eb3a37c103016ef53dd139ca8603ea988bae",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.997677206993103
    },
    "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-32.wav": {
      "sha256": "6be73da7daa608b657e2b521fc0901985bf28c375c0a4ded118efb189b1d08ed",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9981289505958557
    },
    "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-35.wav": {
      "sha256": "ae4555c87b5ac50450a9fd8620d4bef10414022b5c0f5c42b424fa9cd770e7aa",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 1.001283884048462
    },
    "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-38.wav": {
      "sha256": "bc80e3e8e147d3b13245171fce4ce263905b2612bc6c00e35076c21b70a48b59",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9969750642776489
    },
    "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-49.wav": {
      "sha256": "5b8476063157279276ef29ff664664e31247edb2fdd3dd35f8a9b61906b8d188",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9976968765258789
    },
    "listening-audio/podcast-800-fish-400m-morepitchloss1-step3802000-window-56.wav": {
      "sha256": "0f68842bb7b60fab87354462c77b2285c6f462e42b3e685584eb24fbfd2cf129",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9971923828125
    },
    "listening-audio/cartoon-voices-fish-dac-s2-full.wav": {
      "sha256": "2ef6adcad4679b90a8522838624987164c034cf7db72353f210ceff83760f24c",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.2939414083957672
    },
    "listening-audio/cartoon-voices-fish-dac-s2-window-8.wav": {
      "sha256": "b5dee0012e6fb5c151acf6f3b8460d6ddb5e0ac8cd8851f1d1e3ef5f9e68e920",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.7225018739700317
    },
    "listening-audio/cartoon-voices-fish-dac-s2-window-20.wav": {
      "sha256": "a1eedf1e7db18353a6dabae26fa1a8c3524cbdf5db3149763dea912bb23d9cab",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.53863126039505
    },
    "listening-audio/cartoon-voices-fish-dac-s2-window-31.wav": {
      "sha256": "7c768f59bd18aa654c241286dfa453549e450153460e5d8b0f104e92ba15d91c",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.4813039004802704
    },
    "listening-audio/cartoon-voices-fish-dac-s2-window-34.wav": {
      "sha256": "8c05d2efa5cdcf3c79f61b233d7b51fb411662194ab55531780f467242580d1f",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.5303155183792114
    },
    "listening-audio/cartoon-voices-fish-dac-s2-window-36.wav": {
      "sha256": "d80be2a1caca0006959d4bf90e55c20a8701156eb6d51cdaf7d7b6268e16af99",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.4782605767250061
    },
    "listening-audio/cartoon-voices-fish-dac-s2-window-52.wav": {
      "sha256": "544fbcc853024cb6fca9bb6587bb41e32224c6e5066ff260d86df791fe7a5151",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.603863537311554
    },
    "listening-audio/cartoon-voices-fish-dac-s2-window-60.wav": {
      "sha256": "9b669ef5ab00973a9be0b5e644b1ea016a73c1eaad0de743ee936c512dd95636",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.53863126039505
    },
    "listening-audio/cartoon-voices-fish-dac-s2-window-62.wav": {
      "sha256": "ec64ec02307a6b7ad0588e889b4e6aa48ff4662776c6ecf2730039bda17532d5",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.35199305415153503
    },
    "listening-audio/cartoon-voices-fish-dac-s2-window-70.wav": {
      "sha256": "ca0a4e89fa74ec0872f8bdb386c14f150b694b53dd0e0d9f4a222665083c29d1",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.4108578860759735
    },
    "listening-audio/cartoon-voices-fish-dac-s2-window-71.wav": {
      "sha256": "65315077ee74eba2d1c8eb7959ee5bca7678e548fcdcfcfbe20b682464b10bdc",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.42528894543647766
    },
    "listening-audio/cartoon-voices-fish-dac-s2-window-78.wav": {
      "sha256": "310010965a735bdf6fa1ba9d3d873edd5c3665b0e807dc3ea79022b9b2acee44",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.39043352007865906
    },
    "listening-audio/cartoon-voices-fish-dac-s2-window-88.wav": {
      "sha256": "1afb9848387241a2a4c76759d624f851409f4622d9d5793140ce93904307d9bb",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.5196765065193176
    },
    "listening-audio/cartoon-voices-fish-dac-s2-window-98.wav": {
      "sha256": "44df57ea345bcabec750e1c233b4f6f19a890bd6232628851cde52c78cb6e904",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.4442223012447357
    },
    "listening-audio/cartoon-voices-fish-dac-s2-window-110.wav": {
      "sha256": "6cc1ea6936d83ed140585ba9a0d17f4600309428a751904a503feeb770fb8082",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.572195827960968
    },
    "listening-audio/cartoon-voices-fish-dac-s2-window-117.wav": {
      "sha256": "13598afda3ff02a85c1934865a36aa2fb5330946b2860ea5875538fbf0eae29b",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.3239482045173645
    },
    "listening-audio/cartoon-voices-fish-dac-s2-window-118.wav": {
      "sha256": "d3b16817de97022bacc5c39101ab88c051917c45783718b93803909df5860b4d",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.2939414083957672
    },
    "listening-audio/cartoon-voices-fish-dac-s2-window-120.wav": {
      "sha256": "6400e121aed190824639d3bb1f5e8b0abee0e1d95268f7034b18ab85d7b67469",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.2939414083957672
    },
    "listening-audio/cartoon-voices-fish-dac-s2-window-121.wav": {
      "sha256": "0beb0c10a9400506b8b8ccb2e1a113b1df2b5cbd9f931991878f5e8c0e79c816",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.2939414381980896
    },
    "listening-audio/librispeech-other-500-fish-dac-s2-full.wav": {
      "sha256": "0aafe0fac5eb2c223762ea07b95569b0afe000d6ca9260132e3ea0109c37a5d5",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.2911108732223511
    },
    "listening-audio/librispeech-other-500-fish-dac-s2-window-13.wav": {
      "sha256": "bdf9bfea318cc2f2d36f34af9bff627fd812a6e5f40aba323e82ef94a4bcb4b2",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.4183698892593384
    },
    "listening-audio/librispeech-other-500-fish-dac-s2-window-14.wav": {
      "sha256": "062cdd31c4b0d662165ee669c075654e1d40d1341e8d829d2fe5543c9a2e7cd2",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.34125441312789917
    },
    "listening-audio/librispeech-other-500-fish-dac-s2-window-24.wav": {
      "sha256": "1d85b78d5ef33427166fc31daf5bd41ee6542a7d9f3d7ab40fe8046b266451da",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3135697841644287
    },
    "listening-audio/librispeech-other-500-fish-dac-s2-window-26.wav": {
      "sha256": "f20404d4ddeb31be37384497d25593cf090939040898f9e80c7c60f71feb2804",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3237287998199463
    },
    "listening-audio/librispeech-other-500-fish-dac-s2-window-30.wav": {
      "sha256": "3741429265ff4ed7cdbbc71863df0166c423e24b2fc04068365c1e5023645b99",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3709052503108978
    },
    "listening-audio/librispeech-other-500-fish-dac-s2-window-33.wav": {
      "sha256": "9f77940536c81591b33905c6e500864d0f16a4597f84435ac5a8102a63beef28",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.43204548954963684
    },
    "listening-audio/librispeech-other-500-fish-dac-s2-window-35.wav": {
      "sha256": "1ca593bf3160e3ace7e0dfa9f84f454a04b36de6169ef58fadb4c2478de8effa",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.43078362941741943
    },
    "listening-audio/librispeech-other-500-fish-dac-s2-window-39.wav": {
      "sha256": "f159a92494fb01d7815b8b9a27bb68babb50e2caa924ad7679a70e53a8b164e6",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.41837000846862793
    },
    "listening-audio/librispeech-other-500-fish-dac-s2-window-61.wav": {
      "sha256": "c2df33a00cc24ff38bf2053ea293fa8646f85c167219c7f8ba9731042269df0f",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.2911108136177063
    },
    "listening-audio/librispeech-other-500-fish-dac-s2-window-65.wav": {
      "sha256": "9fc0ea4f91098ee6459e035ff7a32e6b903935e265572f0d245f6c4424349a2b",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.29111066460609436
    },
    "listening-audio/librispeech-other-500-fish-dac-s2-window-69.wav": {
      "sha256": "d48dece389f5bda3788fa60d85f58cac754ac67437ffad537db1c0703ad85986",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.2911108732223511
    },
    "listening-audio/librispeech-other-500-fish-dac-s2-window-79.wav": {
      "sha256": "f1fb3eb41526e75906e49795be402f31ae222b0003830e03dd490c38cbbd24a6",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.2911107838153839
    },
    "listening-audio/librispeech-other-500-fish-dac-s2-window-81.wav": {
      "sha256": "1fb1aa92d09f292332327c12d63c8dcfce95ba42d66334ab0c9d093bb04cead8",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.29111069440841675
    },
    "listening-audio/librispeech-other-500-fish-dac-s2-window-84.wav": {
      "sha256": "33eeb2a556572d9d47bc5aab0ca61cadf31aae1cf971a5f33fb72bc424699036",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.29111066460609436
    },
    "listening-audio/librispeech-other-500-fish-dac-s2-window-89.wav": {
      "sha256": "0e31db29f2ce2d90765c66a7da97edb9dbc85ea67b06bd79ddfe96d90e8882c6",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.29111069440841675
    },
    "listening-audio/librispeech-other-500-fish-dac-s2-window-94.wav": {
      "sha256": "c2b086b2bc150ac3d9c4d74f582f7c8f71470d877ed6d7e2fb52973500e987d3",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.29111066460609436
    },
    "listening-audio/librispeech-other-500-fish-dac-s2-window-97.wav": {
      "sha256": "136688810d8bc543a346f7f48b0c4bc0b906cdae0201dd328c1fc18f0898d59f",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.2911108732223511
    },
    "listening-audio/librispeech-other-500-fish-dac-s2-window-102.wav": {
      "sha256": "0c376aa9eac7e6b4cdf31df7d4f4ef9aa89abd62b2674163250d4c6255b21d10",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.29111066460609436
    },
    "listening-audio/librispeech-other-500-fish-dac-s2-window-104.wav": {
      "sha256": "2682d41298caa0ff9350472eb372d2cfdfb342c7ae354ec6440ea3977e5f3aec",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.29111066460609436
    },
    "listening-audio/librispeech-other-500-fish-dac-s2-window-107.wav": {
      "sha256": "59cac3ee8aafd8eb8d0aeb72dce997c230579bcadbc8a81f9b485c9e0b248bdd",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.2911108732223511
    },
    "listening-audio/podcast-800-fish-dac-s2-full.wav": {
      "sha256": "1f15f932b08b5223969fd0d99fb3306d620e470b5cea46156c57f62ae1dbd0d9",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.8985578417778015
    },
    "listening-audio/podcast-800-fish-dac-s2-window-3.wav": {
      "sha256": "db1471e301a3de08b509ccd6cb1bccee14a8a159912d9d9d4201c9c1e583b7e7",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9176082015037537
    },
    "listening-audio/podcast-800-fish-dac-s2-window-10.wav": {
      "sha256": "0c0b53cbfcdfb65a579095dc21309459f4ed0929b2b0a449d7025cdcc572c42a",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9314402341842651
    },
    "listening-audio/podcast-800-fish-dac-s2-window-24.wav": {
      "sha256": "8e046100c3e70b3ecf58b3425315b008955b63fccac6b0c8c9d2016ac7b1352e",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9438562989234924
    },
    "listening-audio/podcast-800-fish-dac-s2-window-26.wav": {
      "sha256": "15ad689bcbaccf0420a99f31b079e7b3d083460e78051dfea6d9d28967fd8533",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9459750056266785
    },
    "listening-audio/podcast-800-fish-dac-s2-window-36.wav": {
      "sha256": "be4c6134da3d38a5dda8eb690d1be6074cc68fee83f0a20b4ed5e57ce9079dd0",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9176083207130432
    },
    "listening-audio/podcast-800-fish-dac-s2-window-38.wav": {
      "sha256": "a4590867fcc916be27721c4fd6ce54c0aa15898218eabb78f02f6d1d4aebb8a1",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.8985579609870911
    },
    "listening-audio/podcast-800-fish-dac-s2-window-39.wav": {
      "sha256": "5af1d06487d4c8d3bccc1f2a3de4249ac8a05138c7832b6b2ec700e51dc50c46",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9019877910614014
    },
    "listening-audio/podcast-800-fish-dac-s2-window-70.wav": {
      "sha256": "2ee94ce62c4c820ae02d74102511a8c75821c445925f34c964d23e3e0d842547",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.8985578417778015
    },
    "listening-audio/podcast-800-fish-dac-s2-window-76.wav": {
      "sha256": "48deafd8167368cc9c7bcc37f9b0cefd3282c1583d1f262fc43fb79838c1e51d",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.8985579013824463
    },
    "listening-audio/podcast-800-fish-dac-s2-window-79.wav": {
      "sha256": "250cd4d2b3182c28d27d9616b7a2add1387bb3a0681c9f9770956f0d30d75b5b",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9111092686653137
    },
    "listening-audio/podcast-800-fish-dac-s2-window-93.wav": {
      "sha256": "07e99e9b74f6e7060c76a12ff7db9ca6fdccae430b36c71cc1ec5c3d233efd4b",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.8985579013824463
    },
    "listening-audio/podcast-800-fish-dac-s2-window-95.wav": {
      "sha256": "f33d15d0b4968e08c40eda93e138170878326b4cfa42844b09ee2a0817e8e5eb",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.8985579013824463
    },
    "listening-audio/podcast-800-fish-dac-s2-window-107.wav": {
      "sha256": "92b09300b1ae23e4ecae429dfef0403098ae837ed4b00216e67c9b15b1824424",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9320656657218933
    },
    "listening-audio/podcast-800-fish-dac-s2-window-109.wav": {
      "sha256": "3384b747566ac8c9753b86bd4c34cff81fbe2dfeda3b10ba9b559e127793cf07",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.924902081489563
    },
    "listening-audio/podcast-800-fish-dac-s2-window-179.wav": {
      "sha256": "1b9c80b6c5e8f29cfd2682592af15abd29461a66b6f642d64cc23e8d3f59479b",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.8985580205917358
    },
    "listening-audio/podcast-800-fish-dac-s2-window-182.wav": {
      "sha256": "4fc8a127e78934b818b2d919000d48e746ba75bd67848762bf34ee542e088800",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.898557722568512
    },
    "listening-audio/podcast-800-fish-dac-s2-window-193.wav": {
      "sha256": "8149c3735205af42785b75602c1a77afa98a0f0e9c5f53cc7efcd34c52c4b2b1",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.8985581398010254
    },
    "listening-audio/podcast-800-fish-dac-s2-window-196.wav": {
      "sha256": "7cab400246a6dc96e176f056124e9a438800cd412662dd9aee9da802c781d8d3",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.8985580801963806
    },
    "listening-audio/podcast-800-fish-dac-s2-window-207.wav": {
      "sha256": "0f2607fcd435656f2520eed55a4ac2a98f2feddc8c7597f3377644944714232e",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9084176421165466
    },
    "listening-audio/podcast-800-fish-dac-s2-window-215.wav": {
      "sha256": "4b981aa7747fc1a7d95ae2172b0f9e30ad89dfb985d6b3e09e6f35397d71a9d3",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.8985579013824463
    },
    "listening-audio/podcast-800-fish-dac-s2-window-224.wav": {
      "sha256": "24c16145ac294fe611bd9ca9347e9ba16cc877459830338e73976aab5ac436fd",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.8985580205917358
    },
    "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-full.wav": {
      "sha256": "53fb73580731c53f98afa3cc7270c442a8e19a6694d75e5ea7882a77f7eff811",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.2765774130821228
    },
    "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-1.wav": {
      "sha256": "5e2cff75a36d7ac8e603662e1731f2c48239ed8d7bee294ea8c5544ddededa3f",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.41996315121650696
    },
    "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-3.wav": {
      "sha256": "06915b457561bf5c93e2ee9cd68646d7c98a6795c12398fb54525871c42b8ea7",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.3473474681377411
    },
    "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-8.wav": {
      "sha256": "302dcf654e09d56286e967e1bfefc7edaf12f6c6cc0398e8d3f6795c2330abd0",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.28327158093452454
    },
    "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-9.wav": {
      "sha256": "c961261e5daef7e094ecbe6bbc6413ed5ac7b230188bf69bc748124c24f63817",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.31614169478416443
    },
    "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-10.wav": {
      "sha256": "31a848bce9665d82b8682f3df3cf2370239d8b226970c5fe39809bc6ec280b80",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.2819381654262543
    },
    "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-14.wav": {
      "sha256": "5848e6a40b17d272da6d9805d1deb2aad79bbc17cad496fe8fef96c5965ba393",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.2961994409561157
    },
    "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-15.wav": {
      "sha256": "66ad4fcd78c5e29d7b00f84ca56ab96bf0d7dc108082cb9c38c2e17af3489e86",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.28193819522857666
    },
    "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-16.wav": {
      "sha256": "0d1c184d2d79ba229643beeb56ef9d5ef6e0d1b3c794c0fc48a1dbd87f1d5eb9",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.2832719683647156
    },
    "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-31.wav": {
      "sha256": "7d186ad07b37f9a8a4ea0018eb0003433e3e11dcda5e93b5bc3f30a0a2fd46a0",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.27373480796813965
    },
    "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-32.wav": {
      "sha256": "e6371e14c2e06e5be23e22b13379e4b9537529aada2dc957c147901e51dfa84e",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.2832719683647156
    },
    "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-40.wav": {
      "sha256": "4290469dc0841a50eaae098a13b27f4bdd53285d0d4f90fc52106477617603ef",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.2765774130821228
    },
    "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-42.wav": {
      "sha256": "9752c8bb244a3c6c84e73942e89bb0ed333f3fe7d79d2f332d16f209d5839e28",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.2765774130821228
    },
    "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-46.wav": {
      "sha256": "955d973287d6fa33e3fdd2f0d4910587934a49f84a30f144fe56a22668d4ad37",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.2765774130821228
    },
    "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-47.wav": {
      "sha256": "02f18330601171140d60e0102f2862ef3b7696b278dea29430da79ae38f47cf5",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.2765774130821228
    },
    "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-56.wav": {
      "sha256": "6fa4436671811a35644246fa3f815b6632ba154590c0b131aa3d699a59bea0f6",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.276577353477478
    },
    "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-58.wav": {
      "sha256": "19cdaee53b9ea91b4397693121da66702783e3d4fee05e608a0a645a2aa76aa0",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.37462297081947327
    },
    "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-60.wav": {
      "sha256": "d337cc2eb9870b5733c1407e0cc385e548dd4b26eb91ef3fe69c28e2a30d1fd7",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.276577353477478
    },
    "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-64.wav": {
      "sha256": "263ee23b037ea951ba1140515617ed6d2a34e253fdf2c1bbca6b3a76234f03de",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.2832719683647156
    },
    "listening-audio/cartoon-voices-qwen3-tts-tokenizer-12hz-window-66.wav": {
      "sha256": "027a714aba6f6d4c202622c72eb92293ebbd45a681582a8fdaf0a428da9d5d2f",
      "bytes": 549052,
      "sample_rate": 24000,
      "samples": 137243,
      "duration_seconds": 5.718458333333333,
      "subtype": "FLOAT",
      "peak_absolute": 0.2765774130821228
    },
    "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-full.wav": {
      "sha256": "7728a4414372e1e90c8b50fc35b1cf5c57d652417a652e1e1b8af1abd4d81178",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3144649267196655
    },
    "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-1.wav": {
      "sha256": "8d526868f3af12d06218806c569831478d592077aa8f78717b6079892aaeb857",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.33788952231407166
    },
    "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-2.wav": {
      "sha256": "def5ea21ef88d4935980d682668487ab88b107e4a778195db73249e8a2730b26",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3716079890727997
    },
    "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-3.wav": {
      "sha256": "1e818e6e664b4e7a27b6582e094750680d728e5b9170d5c143faa5790ab7bace",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.34125465154647827
    },
    "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-4.wav": {
      "sha256": "733f556b9f1bc28de23bc0d00f84781aedae6ebf4da386cc606159783c64d0dd",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.34753432869911194
    },
    "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-5.wav": {
      "sha256": "f59410e3279d1460958731d448a901b791b2841c93c6f4d1c360cb97537e56ae",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.34753429889678955
    },
    "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-6.wav": {
      "sha256": "22d5e2fbcc615c2a35ef871ee004e86d76cfa26a061332d3f7762d7a509112ba",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3132530450820923
    },
    "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-7.wav": {
      "sha256": "89eb456507bf41aef819f00413676fa659e0164a6c0f006a83b8d0c301d3cafa",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3412546217441559
    },
    "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-10.wav": {
      "sha256": "fa97dc9673d507bfeae5310cc590046afe928f1ee3a4e439ac352f88c5b72c90",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3475343883037567
    },
    "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-12.wav": {
      "sha256": "c56d3cce7ce447bc8e91db0b69ef28f4bc7a5aacac9f93366cb7544fffa027fe",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.31446489691734314
    },
    "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-15.wav": {
      "sha256": "e177ac3bbb35f1a30befdc485d371655eb6498bd7425bfce80a0a2648774bc39",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3144649565219879
    },
    "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-23.wav": {
      "sha256": "fd7ba0633670302ec7be872a1e0e6663fb8cd4ce68ff8f1f52f9f1387c9f8d54",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3144649863243103
    },
    "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-42.wav": {
      "sha256": "a6d10bd28e26e65a6da3e3c4867d8ca589fba0a562b3470e4d7dc3d6fbf747b0",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3144649863243103
    },
    "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-47.wav": {
      "sha256": "7bd96a5916565aa147150de24e3a4490d5d53ab96dd84aa5143d2130cbe539d5",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3144649863243103
    },
    "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-48.wav": {
      "sha256": "89feebf02f4f1e9aa900ccf7cbde2dc4e2505f0be89773a347fae3c10d87fd14",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3144649267196655
    },
    "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-56.wav": {
      "sha256": "0475fd80536654f6dfa9f96273d33be1de960e2a5147ac93eca21a1a6ac90a37",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3144649863243103
    },
    "listening-audio/librispeech-other-500-qwen3-tts-tokenizer-12hz-window-58.wav": {
      "sha256": "425ba3abf838d1174d8ccf96f7e7a2411f2bdaa47045c31c99a822d75fd1a505",
      "bytes": 492560,
      "sample_rate": 24000,
      "samples": 123120,
      "duration_seconds": 5.13,
      "subtype": "FLOAT",
      "peak_absolute": 0.3144649863243103
    },
    "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-full.wav": {
      "sha256": "2aa64917cbd1750311172d73931365a2c4a215332121e3676a10f85184708260",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9834774732589722
    },
    "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-1.wav": {
      "sha256": "947923a4e4ce60230849f8857d6fd4990ee57685d4943bd037323bb43863d2bf",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.894189178943634
    },
    "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-2.wav": {
      "sha256": "21a43221961e9e3e14cafc5e50d53fc2251c5666e3933d6c4e3bf196068b1ee4",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9271305799484253
    },
    "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-4.wav": {
      "sha256": "4988913357f8aa7348c358b12428784f27641f5a6b24f72f66bd7a55cfe31303",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.927130937576294
    },
    "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-6.wav": {
      "sha256": "9f16b10619ab77ff9885586bcf02d64374c699ac2436cd5c329876a404e09d0d",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9344982504844666
    },
    "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-7.wav": {
      "sha256": "f8a2b608a76281409585496215cd399249c1c9c28df95f96fa26d6057050a86e",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9271303415298462
    },
    "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-10.wav": {
      "sha256": "35ab7aa28ce2f36e7f25397d14264cf99624fbaff8eb8ceb4722f12a97b06d41",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9271302819252014
    },
    "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-11.wav": {
      "sha256": "6d7947c4b0173018d65ef839848ecfaa501e41bc992ea891cfdb62aee3ec1090",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9476893544197083
    },
    "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-13.wav": {
      "sha256": "89425424c9e4f1353e477af5d557e54aedccd82947eeb670642e9ac351496da8",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9754003286361694
    },
    "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-14.wav": {
      "sha256": "494db77992daa8cd285465904e694031c25a4d4145469c537bf9245666fab719",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.970105767250061
    },
    "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-18.wav": {
      "sha256": "6318f2c1a58422dd837f6146b864361c49792d6f4c1eaa481cd1952b4e239b0d",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9888814091682434
    },
    "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-22.wav": {
      "sha256": "9564fba4ab1706f5fdd4b51b6baad5d314b2c762629a535da971510cf0ca6a3f",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9271302819252014
    },
    "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-30.wav": {
      "sha256": "88830a085a555d27feb2581a8a4dd592ff9f3abe6d9f501c2d7b4520c6785178",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 1.0
    },
    "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-41.wav": {
      "sha256": "27c554d3e28f85c8cb0a60250f5ab9aad55a3bd8aa1df84ae65160c3f86a972f",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9271308183670044
    },
    "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-57.wav": {
      "sha256": "49778446b8e3d321e369c881fd1f5ec05b40e4a53b22778e4d2f6b03ab944f1b",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9834778904914856
    },
    "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-82.wav": {
      "sha256": "a29c50efeeb1edd4f40036bcc31aeadc1dccf5d79bda1fabddc2b30bf92b0540",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9834773540496826
    },
    "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-89.wav": {
      "sha256": "fea189b856fd42962b335e85660c327eccde94156e56080b3a7e7772a1a4999c",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9834773540496826
    },
    "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-107.wav": {
      "sha256": "2deb44eb0c7852c5941b0ac659b73997e92cf2d7366400f60b5c3db8d361d446",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9834772348403931
    },
    "listening-audio/podcast-800-qwen3-tts-tokenizer-12hz-window-123.wav": {
      "sha256": "ac232a47063dba877dd9eacdb35cbbf1dfd10298fe2a40d823c3cbe0f2b39a70",
      "bytes": 1014128,
      "sample_rate": 24000,
      "samples": 253512,
      "duration_seconds": 10.563,
      "subtype": "FLOAT",
      "peak_absolute": 0.9834772348403931
    }
  },
  "unique_window_conditions": 223,
  "example_count": 252,
  "playback_protocol": "24 kHz mono IEEE float32 WAV, exactly preserving the metric-domain arrays; no lossy compression or loudness normalization",
  "selection": "One clip per dataset, fixed across every model, metric and band. Among clips covering all 84 combinations, choose nearest dataset median duration; each width is closest to the band midpoint (75% for >50%). Illustrative, coverage-selected clips, not population representatives.",
  "band_definition": "Individual clip error increase relative to its full-decode source error, not the pooled confidence-bound thresholds, MOS, or percentage of audio lost.",
  "runtime": {
    "gpu": "NVIDIA L40S",
    "torch": "2.12.0+cu130",
    "cuda": "13.0",
    "tf32": false
  }
}
