Mistral announced Voxtral 3, a voice‑agent SDK paired with new speech‑to‑text and text‑to‑speech models. The system supports accurate transcription in more than 50 languages and expressive, emotion‑aware synthesis in 35+ languages. It can run on Mistral’s EU‑hosted API or be deployed on a company’s own servers, giving organizations control over where audio is processed.
Why it matters
Enterprises can add multilingual, low‑latency voice assistants while keeping sensitive conversations on‑premises.
