config/tts.json

{
  "base_config": "config/base.json",  
  "supported_model_type": [
    "Fastspeech2",
    "VITS",
    "VALLE",
  ],
  "task_type": "tts",
  "preprocess": {
    "language": "en-us",
    // linguistic features
    "extract_phone": true,
    "phone_extractor": "espeak", // "espeak, pypinyin, pypinyin_initials_finals, lexicon (only for language=en-us right now)"
    "lexicon_path": "./text/lexicon/librispeech-lexicon.txt",
    // Directory names of processed data or extracted features
    "phone_dir": "phones",
    "use_phone": true,
    // "spk2id": "spk2id.json", // used for multi-speaker dataset
    // "utt2spk": "utt2spk", // used for multi-speaker dataset
    "add_blank": true
  },
  "model": {
      "text_token_num": 512,
  }

}