Mistral releases the Voxtral 3 speech models in public preview
Mistral released four Voxtral 3 models in public preview: Voxtral TTS 3 with zero-shot voice cloning, two offline transcription models with diarization, and a real-time transcription model. It also launched a managed Voice Agents pipeline.
Source
"We released the Voxtral 3 audio family, our next generation of speech-to-text and text-to-speech models. All four models are in Public Preview."
Published