MistralAITTS
Text-to-speech with Mistral’s Voxtral voices, using your own Mistral key. Expressive multilingual speech with zero-shot voice cloning from a short clip. Authenticates with MISTRALAI_API_KEY from your environment, or an explicitapi_key.
Fields
str
default:"UNSET"
Voxtral model, e.g. “voxtral-mini-tts-3-0” or “voxtral-mini-tts-2603”. Use “mistralai/Voxtral-4B-TTS-2603” with a self-hosted base_url. Defaults to ‘voxtral-mini-tts-3-0’.
Optional[str]
default:"UNSET"
Voice ID: a Voxtral 3 preset such as “voice_478”, a Voxtral 2 slug such as “gb_jane_neutral”, or a voice saved in your Mistral workspace. Defaults to None (not set).
Optional[str]
default:"UNSET"
Base64 audio clip of a few seconds to clone a voice from, used when no voice is set. Defaults to None (not set).
Optional[float]
default:"UNSET"
Strength of expressive delivery, 1.0 to 5.0. Higher is more emotive but less like the chosen voice. Leave unset to let Mistral decide. Defaults to None (not set).
Optional[str]
default:"UNSET"
Address of a self-hosted vLLM-Omni server running Voxtral, e.g. “http://host:8000/v1”. Defaults to None (not set).
bool
default:"UNSET"
Synthesize each sentence as soon as it is ready instead of the whole reply at once. Faster first audio for long replies, at some cost to natural flow. Defaults to False.