Mistral launched Voxtral 3, a family of speech-to-text and text-to-speech models plus a voice agent SDK (built on LiveKit) that combines listening, reasoning, tool calling, and speaking in one real-time streaming pipeline. Transcription supports 50+ languages with diarization and context biasing (Mini Transcribe batch at $0.0035/min, Realtime at $0.008/min, Small at $0.0040/min), while Voxtral Speech 3 TTS covers 35+ languages and 50+ dialects at $13 per 1M characters. Models are available via Mistral's EU/US/global API endpoints or self-deployment, with each pipeline stage swappable so devel…
Read the full story on mistral.ai
What to doTry Voxtral 3 in the Mistral API or Studio playgrounds if you're building voice agents, transcription, or TTS features.
- Provider
- Mistral
- Source
- mistral.ai