Mistral AI releases Voxtral TTS speech generation model with open weights
Original title:Research Speaking of Voxtral Voxtral TTS: A frontier, open-weights text-to-speech model that’s fast, instantly adaptable, and produces lifelike speech for voice agents. March 23, 2026 By Mistral
AI summary
Mistral AI released Voxtral TTS, a 4B-parameter text-to-speech model supporting 9 languages and adapting to a new voice with just 3 seconds of reference audio. Human evaluations show its naturalness surpasses ElevenLabs Flash v2.5 and matches ElevenLabs v3 quality, with 70ms model latency for a 10-second sample plus 500-character input.
Key Takeaways
- Mistral AI released Voxtral TTS,a 4B-parameter text-to-speech model supporting 9 languages and…
- Human evaluations show its naturalness surpasses ElevenLabs Flash v2.
- 5 and matches ElevenLabs v3 quality,with 70ms model latency for a 10-second sample plus…
Quick Facts
- Published
- Updated
- Source
- Mistral AI:News(网页)
- Category
- Model Releases
- AI score
- 64
官方发布首个 TTS 模型,给出了人类评测对比、延迟数据和开源权重入口,可了解其与 ElevenLabs 的实际差距。
Related entities
Source & copyright
This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.
How to Cite
When citing this article, keep the title, date, and permalink, and credit Orange News as the source.
Orange News: Mistral AI releases Voxtral TTS speech generation model with open weights. March 23, 2026. https://new.nat6.net/en/news/cmnwsvbvi0032slteuam5waiq/