{
  "version": 1,
  "dataset": "Seed-TTS test-en",
  "selection": {
    "method": "Two utterances from each natural-target-duration tercile, ranked by SHA256(salt + \"|\" + dataset ID), with six distinct prompt recordings. Selected before reading model metrics; no output-quality filtering.",
    "salt": "loop-f5-demo-20260922",
    "count": 6,
    "population": 1088,
    "bins": [
      {
        "name": "short",
        "population": 362,
        "min_seconds": 1.312333,
        "max_seconds": 3.643667,
        "selected": 2
      },
      {
        "name": "medium",
        "population": 363,
        "min_seconds": 3.645208,
        "max_seconds": 4.80025,
        "selected": 2
      },
      {
        "name": "long",
        "population": 363,
        "min_seconds": 4.8035,
        "max_seconds": 9.169,
        "selected": 2
      }
    ]
  },
  "protocol": {
    "checkpoint_updates": 500000,
    "weights": "EMA",
    "seed": 0,
    "solver": "Euler",
    "cfg": 2,
    "sway": -1,
    "steps": [
      4,
      32
    ],
    "sample_rate": 24000,
    "audio_processing": "Unmodified evaluation WAVs."
  },
  "metric_units": {
    "wer": "percent; per-utterance automatic ASR WER",
    "sim": "WavLM speaker-embedding cosine similarity",
    "utmos": "Automatic quality prediction; not human MOS",
    "missing": "Null means the objective score is not yet available."
  },
  "models": [
    {
      "id": "baseline",
      "label": "Baseline",
      "params_m": 157.965668,
      "layout": "18 × 1"
    },
    {
      "id": "full2",
      "label": "CYCLE 9 × 2",
      "params_m": 83.557988,
      "layout": "9 × 2"
    },
    {
      "id": "sequence",
      "label": "SEQUENCE 9 × 2",
      "params_m": 83.557988,
      "layout": "1,1,2,2,...,9,9"
    },
    {
      "id": "full3",
      "label": "Full ×3",
      "params_m": 58.755428,
      "layout": "6 × 3"
    },
    {
      "id": "prefix",
      "label": "Prefix",
      "params_m": 108.360548,
      "layout": "6 × 2 + 6"
    },
    {
      "id": "middle",
      "label": "Middle",
      "params_m": 108.360548,
      "layout": "3 + 6 × 2 + 3"
    },
    {
      "id": "suffix",
      "label": "Suffix",
      "params_m": 108.360548,
      "layout": "6 + 6 × 2"
    }
  ],
  "samples": [
    {
      "id": "sample-01",
      "source_id": "common_voice_en_30232728-common_voice_en_30232725",
      "text": "Failure would mean automatic eviction.",
      "length_group": "short",
      "reference": {
        "src": "assets/audio/sample-01/reference.wav",
        "duration_seconds": 5.637333,
        "text": "The economic activity of the Pampas district is not oriented to tourism."
      },
      "outputs": {
        "4": {
          "baseline": {
            "src": "assets/audio/sample-01/baseline-4.wav",
            "duration_seconds": 2.922667,
            "wer": 0.0,
            "sim": 0.5327845215797424,
            "utmos": 2.1995866298675537
          },
          "full2": {
            "src": "assets/audio/sample-01/full2-4.wav",
            "duration_seconds": 2.922667,
            "wer": 0.0,
            "sim": 0.3979495167732239,
            "utmos": 2.4436495304107666
          },
          "full3": {
            "src": "assets/audio/sample-01/full3-4.wav",
            "duration_seconds": 2.922667,
            "wer": 0.0,
            "sim": 0.43136364221572876,
            "utmos": 2.099505662918091
          },
          "prefix": {
            "src": "assets/audio/sample-01/prefix-4.wav",
            "duration_seconds": 2.922667,
            "wer": 0.0,
            "sim": 0.4554045796394348,
            "utmos": 2.48953914642334
          },
          "middle": {
            "src": "assets/audio/sample-01/middle-4.wav",
            "duration_seconds": 2.922667,
            "wer": 0.0,
            "sim": 0.4322205185890198,
            "utmos": 2.543182373046875
          },
          "suffix": {
            "src": "assets/audio/sample-01/suffix-4.wav",
            "duration_seconds": 2.922667,
            "wer": 0.0,
            "sim": 0.3939160704612732,
            "utmos": 1.8933424949645996
          },
          "sequence": {
            "src": "assets/audio/sample-01/sequence-4.wav",
            "duration_seconds": 2.922667,
            "wer": 0.0,
            "sim": 0.41230320930480957,
            "utmos": 1.8261113166809082
          }
        },
        "32": {
          "baseline": {
            "src": "assets/audio/sample-01/baseline-32.wav",
            "duration_seconds": 2.922667,
            "wer": 0.0,
            "sim": 0.48271438479423523,
            "utmos": 2.8935658931732178
          },
          "full2": {
            "src": "assets/audio/sample-01/full2-32.wav",
            "duration_seconds": 2.922667,
            "wer": 0.0,
            "sim": 0.46032947301864624,
            "utmos": 3.3310275077819824
          },
          "full3": {
            "src": "assets/audio/sample-01/full3-32.wav",
            "duration_seconds": 2.922667,
            "wer": 0.0,
            "sim": 0.5330569744110107,
            "utmos": 3.177840232849121
          },
          "prefix": {
            "src": "assets/audio/sample-01/prefix-32.wav",
            "duration_seconds": 2.922667,
            "wer": 0.0,
            "sim": 0.5393602252006531,
            "utmos": 3.077620506286621
          },
          "middle": {
            "src": "assets/audio/sample-01/middle-32.wav",
            "duration_seconds": 2.922667,
            "wer": 0.0,
            "sim": 0.45344075560569763,
            "utmos": 3.1795201301574707
          },
          "suffix": {
            "src": "assets/audio/sample-01/suffix-32.wav",
            "duration_seconds": 2.922667,
            "wer": 0.0,
            "sim": 0.41353726387023926,
            "utmos": 3.621750593185425
          },
          "sequence": {
            "src": "assets/audio/sample-01/sequence-32.wav",
            "duration_seconds": 2.922667,
            "wer": 0.0,
            "sim": 0.440796822309494,
            "utmos": 1.7081133127212524
          }
        }
      }
    },
    {
      "id": "sample-02",
      "source_id": "common_voice_en_19540120-common_voice_en_19540122",
      "text": "He originally studied architecture at the University of Cape Town.",
      "length_group": "short",
      "reference": {
        "src": "assets/audio/sample-02/reference.wav",
        "duration_seconds": 3.445833,
        "text": "Activation typically spreads to all the other units connected to it."
      },
      "outputs": {
        "4": {
          "baseline": {
            "src": "assets/audio/sample-02/baseline-4.wav",
            "duration_seconds": 3.285333,
            "wer": 0.0,
            "sim": 0.5396069288253784,
            "utmos": 1.6865925788879395
          },
          "full2": {
            "src": "assets/audio/sample-02/full2-4.wav",
            "duration_seconds": 3.285333,
            "wer": 0.0,
            "sim": 0.5472343564033508,
            "utmos": 3.315905809402466
          },
          "full3": {
            "src": "assets/audio/sample-02/full3-4.wav",
            "duration_seconds": 3.285333,
            "wer": 0.0,
            "sim": 0.5966143608093262,
            "utmos": 2.9763171672821045
          },
          "prefix": {
            "src": "assets/audio/sample-02/prefix-4.wav",
            "duration_seconds": 3.285333,
            "wer": 0.0,
            "sim": 0.5259464979171753,
            "utmos": 2.876567840576172
          },
          "middle": {
            "src": "assets/audio/sample-02/middle-4.wav",
            "duration_seconds": 3.285333,
            "wer": 0.0,
            "sim": 0.5388767719268799,
            "utmos": 2.96877121925354
          },
          "suffix": {
            "src": "assets/audio/sample-02/suffix-4.wav",
            "duration_seconds": 3.285333,
            "wer": 0.0,
            "sim": 0.5260822772979736,
            "utmos": 4.050539016723633
          },
          "sequence": {
            "src": "assets/audio/sample-02/sequence-4.wav",
            "duration_seconds": 3.285333,
            "wer": 0.0,
            "sim": 0.4985881447792053,
            "utmos": 3.5396246910095215
          }
        },
        "32": {
          "baseline": {
            "src": "assets/audio/sample-02/baseline-32.wav",
            "duration_seconds": 3.285333,
            "wer": 0.0,
            "sim": 0.6241576075553894,
            "utmos": 4.109043598175049
          },
          "full2": {
            "src": "assets/audio/sample-02/full2-32.wav",
            "duration_seconds": 3.285333,
            "wer": 0.0,
            "sim": 0.6104271411895752,
            "utmos": 3.852665901184082
          },
          "full3": {
            "src": "assets/audio/sample-02/full3-32.wav",
            "duration_seconds": 3.285333,
            "wer": 0.0,
            "sim": 0.5925915241241455,
            "utmos": 4.137031555175781
          },
          "prefix": {
            "src": "assets/audio/sample-02/prefix-32.wav",
            "duration_seconds": 3.285333,
            "wer": 0.0,
            "sim": 0.6014081239700317,
            "utmos": 4.168381690979004
          },
          "middle": {
            "src": "assets/audio/sample-02/middle-32.wav",
            "duration_seconds": 3.285333,
            "wer": 0.0,
            "sim": 0.7012832760810852,
            "utmos": 3.925550937652588
          },
          "suffix": {
            "src": "assets/audio/sample-02/suffix-32.wav",
            "duration_seconds": 3.285333,
            "wer": 0.0,
            "sim": 0.6922149062156677,
            "utmos": 4.312444686889648
          },
          "sequence": {
            "src": "assets/audio/sample-02/sequence-32.wav",
            "duration_seconds": 3.285333,
            "wer": 0.0,
            "sim": 0.6428977251052856,
            "utmos": 4.098278522491455
          }
        }
      }
    },
    {
      "id": "sample-03",
      "source_id": "common_voice_en_19622969-common_voice_en_19622970",
      "text": "He makes the frames himself, those being part of the painting.",
      "length_group": "medium",
      "reference": {
        "src": "assets/audio/sample-03/reference.wav",
        "duration_seconds": 4.005333,
        "text": "The woman kisses his head and says, Happy Halloween dear."
      },
      "outputs": {
        "4": {
          "baseline": {
            "src": "assets/audio/sample-03/baseline-4.wav",
            "duration_seconds": 4.266667,
            "wer": 0.0,
            "sim": 0.5717931985855103,
            "utmos": 3.114689350128174
          },
          "full2": {
            "src": "assets/audio/sample-03/full2-4.wav",
            "duration_seconds": 4.266667,
            "wer": 0.0,
            "sim": 0.5420318841934204,
            "utmos": 3.357391119003296
          },
          "full3": {
            "src": "assets/audio/sample-03/full3-4.wav",
            "duration_seconds": 4.266667,
            "wer": 0.0,
            "sim": 0.5015392303466797,
            "utmos": 3.5243618488311768
          },
          "prefix": {
            "src": "assets/audio/sample-03/prefix-4.wav",
            "duration_seconds": 4.266667,
            "wer": 0.0,
            "sim": 0.5528804659843445,
            "utmos": 3.5791308879852295
          },
          "middle": {
            "src": "assets/audio/sample-03/middle-4.wav",
            "duration_seconds": 4.266667,
            "wer": 0.0,
            "sim": 0.5251166820526123,
            "utmos": 3.266080617904663
          },
          "suffix": {
            "src": "assets/audio/sample-03/suffix-4.wav",
            "duration_seconds": 4.266667,
            "wer": 0.0,
            "sim": 0.5406757593154907,
            "utmos": 3.1600849628448486
          },
          "sequence": {
            "src": "assets/audio/sample-03/sequence-4.wav",
            "duration_seconds": 4.266667,
            "wer": 0.0,
            "sim": 0.5607975721359253,
            "utmos": 3.478485107421875
          }
        },
        "32": {
          "baseline": {
            "src": "assets/audio/sample-03/baseline-32.wav",
            "duration_seconds": 4.266667,
            "wer": 0.0,
            "sim": 0.6244938373565674,
            "utmos": 3.6465377807617188
          },
          "full2": {
            "src": "assets/audio/sample-03/full2-32.wav",
            "duration_seconds": 4.266667,
            "wer": 0.0,
            "sim": 0.636162281036377,
            "utmos": 3.546947717666626
          },
          "full3": {
            "src": "assets/audio/sample-03/full3-32.wav",
            "duration_seconds": 4.266667,
            "wer": 0.0,
            "sim": 0.6161614060401917,
            "utmos": 4.056588649749756
          },
          "prefix": {
            "src": "assets/audio/sample-03/prefix-32.wav",
            "duration_seconds": 4.266667,
            "wer": 0.0,
            "sim": 0.6053755879402161,
            "utmos": 3.879164695739746
          },
          "middle": {
            "src": "assets/audio/sample-03/middle-32.wav",
            "duration_seconds": 4.266667,
            "wer": 0.0,
            "sim": 0.6598341464996338,
            "utmos": 3.9970321655273438
          },
          "suffix": {
            "src": "assets/audio/sample-03/suffix-32.wav",
            "duration_seconds": 4.266667,
            "wer": 0.0,
            "sim": 0.6636666059494019,
            "utmos": 4.058320045471191
          },
          "sequence": {
            "src": "assets/audio/sample-03/sequence-32.wav",
            "duration_seconds": 4.266667,
            "wer": 0.0,
            "sim": 0.6801847219467163,
            "utmos": 3.636772632598877
          }
        }
      }
    },
    {
      "id": "sample-04",
      "source_id": "common_voice_en_19972684-common_voice_en_19972807",
      "text": "Many reviewers compared its style to that of the \"Mario Kart\" series.",
      "length_group": "medium",
      "reference": {
        "src": "assets/audio/sample-04/reference.wav",
        "duration_seconds": 5.28,
        "text": "His daughter Marcia pursued poetry, while his son Hilary Masters became a novelist."
      },
      "outputs": {
        "4": {
          "baseline": {
            "src": "assets/audio/sample-04/baseline-4.wav",
            "duration_seconds": 4.330667,
            "wer": 33.33333333333333,
            "sim": 0.4689728617668152,
            "utmos": 2.1144332885742188
          },
          "full2": {
            "src": "assets/audio/sample-04/full2-4.wav",
            "duration_seconds": 4.330667,
            "wer": 0.0,
            "sim": 0.50505131483078,
            "utmos": 2.9516196250915527
          },
          "full3": {
            "src": "assets/audio/sample-04/full3-4.wav",
            "duration_seconds": 4.330667,
            "wer": 41.66666666666667,
            "sim": 0.4379902482032776,
            "utmos": 2.403961181640625
          },
          "prefix": {
            "src": "assets/audio/sample-04/prefix-4.wav",
            "duration_seconds": 4.330667,
            "wer": 16.666666666666664,
            "sim": 0.3741147220134735,
            "utmos": 2.4110589027404785
          },
          "middle": {
            "src": "assets/audio/sample-04/middle-4.wav",
            "duration_seconds": 4.330667,
            "wer": 8.333333333333332,
            "sim": 0.29872816801071167,
            "utmos": 2.771986961364746
          },
          "suffix": {
            "src": "assets/audio/sample-04/suffix-4.wav",
            "duration_seconds": 4.330667,
            "wer": 25.0,
            "sim": 0.5017696619033813,
            "utmos": 2.8122732639312744
          },
          "sequence": {
            "src": "assets/audio/sample-04/sequence-4.wav",
            "duration_seconds": 4.330667,
            "wer": 16.666666666666664,
            "sim": 0.44154804944992065,
            "utmos": 2.097655773162842
          }
        },
        "32": {
          "baseline": {
            "src": "assets/audio/sample-04/baseline-32.wav",
            "duration_seconds": 4.330667,
            "wer": 0.0,
            "sim": 0.5407326221466064,
            "utmos": 3.2775721549987793
          },
          "full2": {
            "src": "assets/audio/sample-04/full2-32.wav",
            "duration_seconds": 4.330667,
            "wer": 0.0,
            "sim": 0.4664768874645233,
            "utmos": 3.721987247467041
          },
          "full3": {
            "src": "assets/audio/sample-04/full3-32.wav",
            "duration_seconds": 4.330667,
            "wer": 25.0,
            "sim": 0.533944308757782,
            "utmos": 3.669740676879883
          },
          "prefix": {
            "src": "assets/audio/sample-04/prefix-32.wav",
            "duration_seconds": 4.330667,
            "wer": 16.666666666666664,
            "sim": 0.5180967450141907,
            "utmos": 3.084343671798706
          },
          "middle": {
            "src": "assets/audio/sample-04/middle-32.wav",
            "duration_seconds": 4.330667,
            "wer": 0.0,
            "sim": 0.4594966173171997,
            "utmos": 3.2999112606048584
          },
          "suffix": {
            "src": "assets/audio/sample-04/suffix-32.wav",
            "duration_seconds": 4.330667,
            "wer": 0.0,
            "sim": 0.5213176012039185,
            "utmos": 3.8118419647216797
          },
          "sequence": {
            "src": "assets/audio/sample-04/sequence-32.wav",
            "duration_seconds": 4.330667,
            "wer": 0.0,
            "sim": 0.5153689980506897,
            "utmos": 3.240161657333374
          }
        }
      }
    },
    {
      "id": "sample-05",
      "source_id": "common_voice_en_24523505-common_voice_en_24523504",
      "text": "This type of mount is used in Transit telescopes, designed for precision astronomical measurement.",
      "length_group": "long",
      "reference": {
        "src": "assets/audio/sample-05/reference.wav",
        "duration_seconds": 6.085333,
        "text": "Today, transaction cost economics is used to explain a number of different behaviours."
      },
      "outputs": {
        "4": {
          "baseline": {
            "src": "assets/audio/sample-05/baseline-4.wav",
            "duration_seconds": 6.848,
            "wer": 7.142857142857142,
            "sim": 0.5461001396179199,
            "utmos": 3.618607521057129
          },
          "full2": {
            "src": "assets/audio/sample-05/full2-4.wav",
            "duration_seconds": 6.848,
            "wer": 35.714285714285715,
            "sim": 0.4870768189430237,
            "utmos": 3.6476902961730957
          },
          "full3": {
            "src": "assets/audio/sample-05/full3-4.wav",
            "duration_seconds": 6.848,
            "wer": 14.285714285714285,
            "sim": 0.5056657791137695,
            "utmos": 3.529758930206299
          },
          "prefix": {
            "src": "assets/audio/sample-05/prefix-4.wav",
            "duration_seconds": 6.848,
            "wer": 14.285714285714285,
            "sim": 0.40492480993270874,
            "utmos": 3.692328929901123
          },
          "middle": {
            "src": "assets/audio/sample-05/middle-4.wav",
            "duration_seconds": 6.848,
            "wer": 7.142857142857142,
            "sim": 0.47980114817619324,
            "utmos": 3.678795337677002
          },
          "suffix": {
            "src": "assets/audio/sample-05/suffix-4.wav",
            "duration_seconds": 6.848,
            "wer": 0.0,
            "sim": 0.5170681476593018,
            "utmos": 3.670848846435547
          },
          "sequence": {
            "src": "assets/audio/sample-05/sequence-4.wav",
            "duration_seconds": 6.848,
            "wer": 0.0,
            "sim": 0.5508412718772888,
            "utmos": 4.295727729797363
          }
        },
        "32": {
          "baseline": {
            "src": "assets/audio/sample-05/baseline-32.wav",
            "duration_seconds": 6.848,
            "wer": 0.0,
            "sim": 0.6954357624053955,
            "utmos": 4.308203220367432
          },
          "full2": {
            "src": "assets/audio/sample-05/full2-32.wav",
            "duration_seconds": 6.848,
            "wer": 0.0,
            "sim": 0.6939995288848877,
            "utmos": 4.286873817443848
          },
          "full3": {
            "src": "assets/audio/sample-05/full3-32.wav",
            "duration_seconds": 6.848,
            "wer": 0.0,
            "sim": 0.6755543351173401,
            "utmos": 4.225614070892334
          },
          "prefix": {
            "src": "assets/audio/sample-05/prefix-32.wav",
            "duration_seconds": 6.848,
            "wer": 0.0,
            "sim": 0.6955138444900513,
            "utmos": 3.891843318939209
          },
          "middle": {
            "src": "assets/audio/sample-05/middle-32.wav",
            "duration_seconds": 6.848,
            "wer": 0.0,
            "sim": 0.7115540504455566,
            "utmos": 4.058244705200195
          },
          "suffix": {
            "src": "assets/audio/sample-05/suffix-32.wav",
            "duration_seconds": 6.848,
            "wer": 0.0,
            "sim": 0.6915714740753174,
            "utmos": 4.317072868347168
          },
          "sequence": {
            "src": "assets/audio/sample-05/sequence-32.wav",
            "duration_seconds": 6.848,
            "wer": 0.0,
            "sim": 0.7096030712127686,
            "utmos": 4.249574184417725
          }
        }
      }
    },
    {
      "id": "sample-06",
      "source_id": "common_voice_en_22525318-common_voice_en_22525329",
      "text": "He became a nationally known illustrator before he became an author.",
      "length_group": "long",
      "reference": {
        "src": "assets/audio/sample-06/reference.wav",
        "duration_seconds": 3.970542,
        "text": "He has also appeared in a number of short films."
      },
      "outputs": {
        "4": {
          "baseline": {
            "src": "assets/audio/sample-06/baseline-4.wav",
            "duration_seconds": 5.504,
            "wer": 45.45454545454545,
            "sim": 0.21562828123569489,
            "utmos": 1.3580896854400635
          },
          "full2": {
            "src": "assets/audio/sample-06/full2-4.wav",
            "duration_seconds": 5.504,
            "wer": 0.0,
            "sim": 0.32404375076293945,
            "utmos": 1.5633236169815063
          },
          "full3": {
            "src": "assets/audio/sample-06/full3-4.wav",
            "duration_seconds": 5.504,
            "wer": 9.090909090909092,
            "sim": 0.27467024326324463,
            "utmos": 1.6592671871185303
          },
          "prefix": {
            "src": "assets/audio/sample-06/prefix-4.wav",
            "duration_seconds": 5.504,
            "wer": 0.0,
            "sim": 0.49184471368789673,
            "utmos": 1.49130380153656
          },
          "middle": {
            "src": "assets/audio/sample-06/middle-4.wav",
            "duration_seconds": 5.504,
            "wer": 0.0,
            "sim": 0.3666783571243286,
            "utmos": 1.7649335861206055
          },
          "suffix": {
            "src": "assets/audio/sample-06/suffix-4.wav",
            "duration_seconds": 5.504,
            "wer": 0.0,
            "sim": 0.43240463733673096,
            "utmos": 1.7323060035705566
          },
          "sequence": {
            "src": "assets/audio/sample-06/sequence-4.wav",
            "duration_seconds": 5.504,
            "wer": 18.181818181818183,
            "sim": 0.2602136731147766,
            "utmos": 1.8886125087738037
          }
        },
        "32": {
          "baseline": {
            "src": "assets/audio/sample-06/baseline-32.wav",
            "duration_seconds": 5.504,
            "wer": 0.0,
            "sim": 0.5334051251411438,
            "utmos": 3.61325740814209
          },
          "full2": {
            "src": "assets/audio/sample-06/full2-32.wav",
            "duration_seconds": 5.504,
            "wer": 0.0,
            "sim": 0.5640376806259155,
            "utmos": 3.9696226119995117
          },
          "full3": {
            "src": "assets/audio/sample-06/full3-32.wav",
            "duration_seconds": 5.504,
            "wer": 0.0,
            "sim": 0.5807673931121826,
            "utmos": 4.017979621887207
          },
          "prefix": {
            "src": "assets/audio/sample-06/prefix-32.wav",
            "duration_seconds": 5.504,
            "wer": 0.0,
            "sim": 0.5570367574691772,
            "utmos": 3.5180065631866455
          },
          "middle": {
            "src": "assets/audio/sample-06/middle-32.wav",
            "duration_seconds": 5.504,
            "wer": 0.0,
            "sim": 0.5668193101882935,
            "utmos": 3.4478893280029297
          },
          "suffix": {
            "src": "assets/audio/sample-06/suffix-32.wav",
            "duration_seconds": 5.504,
            "wer": 0.0,
            "sim": 0.5726538896560669,
            "utmos": 4.127363204956055
          },
          "sequence": {
            "src": "assets/audio/sample-06/sequence-32.wav",
            "duration_seconds": 5.504,
            "wer": 0.0,
            "sim": 0.5761797428131104,
            "utmos": 3.465048313140869
          }
        }
      }
    }
  ]
}
