config.json

{
  "model_type": "packedtts",
  "library_name": "chichat",
  "bundle_file": "tts.pt",
  "runtime_script": "PackedTTS.py",
  "version": 1,
  "architecture": {
    "text_model": "T3",
    "vocoder": "S3Gen",
    "speaker_encoder": "VoiceEncoder",
    "tokenizer": "EnTokenizer"
  },
  "bundle_layout": {
    "models": {
      "t3_state": "models.t3_state",
      "s3gen_state": "models.s3gen_state",
      "ve_state": "models.ve_state",
      "tokenizer_json": "models.tokenizer_json"
    },
    "collections": {
      "voices": "voices",
      "emotions": "emotions",
      "defaults": "defaults",
      "indexes": "indexes"
    }
  },
  "audio": {
    "sample_rate": 24000,
    "reference_sample_rate": 16000,
    "max_reference_seconds": 10.0
  },
  "defaults": {
    "default_voice": "",
    "default_emotion": "",
    "missing_voice_policy": "random",
    "missing_emotion_policy": "voice_default_else_random",
    "fuzzy_cutoff": 0.72
  },
  "supported_commands": [
    "create",
    "add-voice-dir",
    "add-voice-ref",
    "add-emotion-dir",
    "add-emotion-ref",
    "remove-voice",
    "remove-emotion",
    "list",
    "validate"
  ],
  "input_files": [
    "tts.pt",
    "PackedTTS.py",
    "requirements.txt",
    "README.md"
  ],
  "query_file": true
}