Mistral announces Voxtral 3 voice agent SDK and speech models
Mistral AI
- Developer tools
- Models
- #voxtral
- #voice agents
- #speech synthesis
Browse Developer tools topic page →
Mistral has announced Voxtral 3, described as a voice agent SDK plus speech models for real-time, multi-step conversations in more than 50 languages. According to Mistral, it combines transcription, speech synthesis with emotional adaptability and tool integration, and is available via its EU-hosted API or self-deployment for data sovereignty.
What changed
Mistral introduces a voice agent SDK and speech models combining transcription, speech synthesis and tool integration for multi-step voice conversations.
Who it helps
Developers and enterprises building contact-center, virtual-assistant or hands-free copilot applications, particularly those with EU data-sovereignty requirements.
Context and limitations
- Product claims come from Mistral's announcement and were not independently tested or verified.
- Based on a short source excerpt; the full article was not analyzed.