Skip to content

IndexTTS Added to External TTS Engines โ€‹

Updated

Set engine to index-tts inside ExtParam and IndexTTS runs on the existing TTS APIs. With this addition, MPS now offers six external TTS engines.

  1. Select the engine: pass {"engine": "index-tts"} in ExtParam (a JSON string). Use synExt.model to pick index-tts-2 (the default) or index-tts-1.5.
  2. Select the voice: put the reference audio URL straight into VoiceId. There are no voice names and no registration step โ€” zero-shot cloning is the default behaviour.
  3. Select the emotion: under engineCustom, choose one of an 8-dimension vector (emo_vector), reference audio (emo_audio_url + emo_alpha), a text description (emo_text), or random sampling (use_random).
  • It is the only engine that can take voice and emotion from two different clips: VoiceId decides who is speaking, emo_audio_url decides with what emotion.

  • The emotion parameters are supported only by index-tts-2; selecting index-tts-1.5 makes them silently no-ops.

  • It is slow, so the async API is recommended. Three APIs are supported: TextToSpeech, TextToSpeechAsync and DescribeTextToSpeechAsyncTask.

  • Audio parameters such as format, sampleRate and speed are not declared in the integration guide. Do not copy another engine's parameter table โ€” test with short text first. On the output side, ExtInfo.duration returns the total length in seconds.

  • TTS-Index API docs

  • MPS TTS Engines Compared

  • AI Dubbing docs

Built with VitePress ยท Deployed on Tencent EdgeOne Pages