Mistral AI Releases Voxtral TTS: A 4B Open-Weight Streaming Speech Model for Low-Latency Multilingual Voice Generation
By Asif Razzaq
Continuing our coverage from [yesterday](/?date=2026-03-28&category=news#item-08505072821c), Mistral AI released Voxtral TTS, a 4B-parameter open-weight text-to-speech model supporting low-latency, multilingual streaming voice generation. Released under a CC BY-NC license, it represents Mistral's first major move into audio generation and positions the company as a competitor to proprietary voice APIs.