Mistral AI releases free, open-weight text-to-speech model

Mistral AI has launched Voxtral TTS, an open-weight text-to-speech model designed for enterprise use, challenging proprietary models like ElevenLabs. Unlike competitors who offer API access, Mistral is releasing the full model weights, allowing companies to run the AI on their own infrastructure for greater control and data privacy. This move aligns with Mistral's strategy to build a comprehensive, enterprise-owned AI stack, including speech-to-text and language models. The 3-billion-parameter model is efficient, capable of running on standard hardware, and boasts superior performance in human evaluations for voice customization and naturalness.