Mistral launches Voxtral 3 real-time voice agent stack
Original title: Mistral AI launches Voxtral 3, a full voice agent stack that listens, thinks, and talks in real time
The stack pairs transcription, speech synthesis and a LiveKit SDK, priced from $0.0035 per minute.
Mistral AI released Voxtral 3, a family of models for real-time voice agents. It combines Voxtral Small Transcribe 3 for speech-to-text, Voxtral Mini TTS 3 for text-to-speech, and a voice agent SDK based on LiveKit in one streaming pipeline that can call external tools during a call. Transcription covers more than 50 languages; speech output supports over 35 languages and 50 dialects. Mistral says it can run on customer infrastructure or through EU-hosted APIs.
- Agents can use external systems during live calls, such as booking appointments.
- Text-to-speech includes emotional adaptation.
- Transcription pricing starts at $0.0035 per minute.
- Vendors
- Mistral AI
- Products
- Voxtral 3
- Organizations
- LiveKit
From the original article
1 min read at the-decoder.com
Mistral AI launched Voxtral 3, a new family of language models built for real-time voice agents. The system combines speech-to-text, text-to-speech, and a voice agent SDK, supports more than 50 languages, and lets agents tap into external tools during live conversations. It can run on your own infrastructure or through EU-hosted APIs to keep data under your control.
By Matthias Bastian · the-decoder.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.