Experiment with Voxtral TTS directly in the Mistral Studio playground. Select one of the Mistral voices or record your own.

State-of-the-art performance. In zero-shot custom voice tests, Voxtral TTS outperformed ElevenLabs v2.5 Flash - judged by native speakers for naturalness, accent accuracy, and similarity to the original voice.

Voxtral TTS is built for global applications supporting 9 languages and powering voice workflows. ✅ Full audio intelligence: Works with Voxtral Transcribe for end-to-end speech-to-speech, or plugs into any STT + LLM stack. ✅ Built for business: From customer support to real-time translation, it’s the output layer that passes the human test. 🎥 See it in action:

🔊Introducing Voxtral TTS: our new frontier open-weight model for natural, expressive, and ultra-fast text-to-speech 🎭Realistic, emotionally expressive speech. 🌍Supports 9 languages and accurately captures diverse dialects. ⚡Very low latency for time-to-first-audio. 🔄Easily adaptable to new voices

Check out our blog post for details:

Thanks for the tutorial! It was needed, because I was not able to find it without the video. Maybe you can add "My Voices" directly under Model (under Oliver - Excited)

how good is that compared to Omnivoice?

Making the model easy to test directly is important. The faster teams can experiment, the faster they can move from curiosity to real implementation.

À noter : un playground vocal embarqué dans Mistral Studio = friction d'évaluation TTS proche de zéro pour les développeurs entreprise. Le facteur décisif n'est plus la qualité brute mais la facilité d'AB-test rapide en pré-prod. ElevenLabs avait pris l'avance côté UX dev.

How do you have “Improve”?

The open-weight play matters more than latency benchmarks here. Audio never leaves your infra — that’s the real deployment blocker for enterprise voice in healthcare, finance, and customer service. ElevenLabs quality is good, but API-only is a non-starter for those verticals.

Would love to connect with the right contact for API partnership / volume pricing discussions. Happy to share more context over email.

It appears that there are only English and French language voices available?