Providers
Mistral
Mistral Voxtral open-source text-to-speech through the Speechbase gateway, with voice cloning.
| Prefix | mistral |
| Default model | voxtral-mini-tts-2603 |
| Provider key | Connect under Provider Keys |
Route to Mistral by prefixing the model with mistral/. Voxtral is an
open-source model with streaming and voice cloning.
Models
| Model | Streaming | Voice cloning | Open source | Languages |
|---|---|---|---|---|
voxtral-mini-tts-2603 | Yes | Yes | Yes | 9 |
Usage
curl -X POST https://api.speechbase.ai/v1/audio/speech \
-H "Authorization: Bearer $SPEECHBASE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"mode": "inline",
"model": "mistral/voxtral-mini-tts-2603",
"voice": "default",
"text": "Hello from Speechbase!",
"output": "mp3"
}' --output hello.mp3Replace voice with a Mistral voice ID, or register a cloned voice as a saved
Voice and address it by voiceId.
Voice cloning
Voxtral clones from a reference sample. Clone into a saved
Voice, then use mode: "voice" — inline
requests take a plain voice string only.
Provider options
Anything in providerOptions is forwarded to the Mistral API unchanged.