{
  "schemaVersion": "1.0",
  "updates": [],
  "companyNews": [],
  "id": "b897",
  "language": "en",
  "name": "IndexTTS",
  "originalName": "IndexTTS",
  "library": "tools",
  "category": "Audio & music",
  "categoryKey": "音频音乐",
  "categoryKeys": [
    "音频音乐"
  ],
  "categories": [
    "Audio & music"
  ],
  "entryPoints": [],
  "summary": "Synthesize a reference voice with IndexTTS 2.5 across Chinese, English, Japanese, Spanish and Arabic, with emotion, speed and pronunciation controls.",
  "introduction": "IndexTTS extracts voice characteristics from reference audio and synthesizes new text. Current IndexTTS-2.5 separates voice and emotion conditions, with speed scaling and pronunciation correction through Chinese pinyin, English CMU phonemes and Japanese kana. It supports Chinese, English, Japanese, Spanish and Arabic through WebUI and Python workflows. Speed scaling is not arbitrary exact-duration generation. Model and code follow the bilibili Model Use License, including additional conditions for large-scale commercial use.",
  "capabilities": [],
  "company": {
    "name": "bilibili",
    "originalName": "bilibili"
  },
  "officialUrl": "https://github.com/index-tts/index-tts",
  "profileUpdatedOn": "2026-10-22",
  "sources": [
    {
      "kind": "product",
      "url": "https://github.com/index-tts/index-tts"
    },
    {
      "kind": "documentation",
      "url": "https://huggingface.co/IndexTeam/IndexTTS-2.5"
    }
  ],
  "news": [],
  "links": {
    "interactive": "products/indextts--b897.html",
    "reading": "read/en/products/indextts--b897.html",
    "markdown": "read/en/products/indextts--b897.md",
    "json": "read/en/products/indextts--b897.json"
  }
}