Closing the ‘Expressivity Gap’: How Mistral’s Voxtral TTS is Redefining Multilingual Voice Cloning with a Hybrid Autoregressive and Flow-Matching Architecture

Mistral introduces Voxtral, a text-to-speech system using hybrid autoregressive and flow-matching architecture to address the 'expressivity gap' in voice cloning. The technology aims to improve emotional expressiveness and naturalness in multilingual voice synthesis beyond current capabilities.