{
  "format": "p4mbfa",
  "format_version": 1,
  "model_config": {
    "input_wav_length": 1920,
    "cnn_channels": 192,
    "embed_dim": 512,
    "context_kernels": [
      5,
      3
    ],
    "cnn_dropout": 0.2,
    "head_dropout": 0.3,
    "noise_level": 0.02,
    "frame_blocks": 4,
    "ffn_mult": 4,
    "attn_radius": 0,
    "attn_heads": 8,
    "block_dropout": 0.3,
    "n_phg": 15,
    "lang_group": "cjk",
    "n_tokens": 48,
    "n_tones": 22
  },
  "geometry": {
    "sample_rate": 16000,
    "window_size_ms": 120,
    "window_stride_ms": 80
  },
  "decoder": {
    "min_phone_frames": 6,
    "soft_phone_frames": 10,
    "soft_frame_cost": 1.0,
    "sil_skippable": true,
    "optional_sil": true,
    "class_prior_norm": true,
    "boundary_bonus_weight": 30.0,
    "boundary_contrast_frames": 17,
    "boundary_flux_source": "mel"
  },
  "labels": {
    "lang_group": "cjk",
    "tokens": [
      "<blank>",
      "SIL",
      "noise",
      "<unk>",
      "a",
      "s",
      "i",
      "e",
      "n",
      "t",
      "l",
      "k",
      "o",
      "m",
      "u",
      "ɪ",
      "f",
      "j",
      "ə",
      "p",
      "ɛ",
      "w",
      "ɑ",
      "ɔ",
      "ŋ",
      "ʊ",
      "ɯ",
      "h",
      "ɐ",
      "x",
      "ʂ",
      "y",
      "ɕ",
      "ɵ",
      "œ",
      "ʈ",
      "tʰ",
      "ʐ",
      "kʰ",
      "ɤ",
      "sʰ",
      "pʰ",
      "ɥ",
      "ɕʰ",
      "ʂʰ",
      "ɚ",
      "ɻ",
      "m̩"
    ],
    "special_tokens": {
      "<blank>": 0,
      "SIL": 1,
      "noise": 2,
      "<unk>": 3
    },
    "phg_names": [
      "special",
      "vowel_close",
      "vowel_mid",
      "vowel_open",
      "stop_voiceless",
      "stop_voiced",
      "affricate",
      "fricative_vl",
      "fricative_vd",
      "nasal",
      "lateral",
      "rhotic",
      "glide",
      "glottal",
      "other"
    ],
    "tone_vocab": [
      "",
      "35",
      "21",
      "33",
      "45",
      "51",
      "32",
      "214",
      "55",
      "24",
      "3",
      "312",
      "5",
      "13",
      "2",
      "212",
      "42",
      "44",
      "53",
      "22",
      "4",
      "114"
    ],
    "member_langs": [
      "nan",
      "yue",
      "zh",
      "zh-Hant"
    ],
    "gold_fingerprint": "9438371ed6dd"
  },
  "training": {
    "experiment": "mc01a",
    "note": "first cjk run: trunk + phg_head from latin ma02a, fresh ph_head (48) and tone_head",
    "data_src": "fleurs",
    "train_limit": 100000,
    "val_limit": null,
    "exclude_langs": null,
    "duration_max": 20,
    "ckpt_path": "tmp/training/p4mbfa/ma02a/m_ma02a_final_e30.pt",
    "trunk_init_path": null,
    "reset_fine_heads": true,
    "ramp_frames": 1.0,
    "soft_onset_mix": 1.0,
    "noise_level": 0.02,
    "lr": "1e-4",
    "epochs": 30,
    "epoch": 13,
    "global_step": 476,
    "source_checkpoint": "m_mc01a_epoch=13_step=476_val_loss=2.5963.ckpt",
    "selection": {
      "trained_langs": [
        "cmn",
        "yue"
      ],
      "train_data": "FLEURS cjk, a 62% sample (train_limit 100000): ~9 of 14.9 h, training noise_level 0.02",
      "selected_on": "FLEURS val_loss (best of 30 epochs)",
      "metrics": {
        "val_loss": 2.5963,
        "val_frame_acc": 0.4795,
        "val_frame_acc_groups": 0.612
      },
      "note": "Trunk and phg_head from latin ma02a (final epoch), ph_head and tone_head trained fresh."
    }
  }
}