Mistral AI 发布了 Voxtral TTS，这是一个开放权重（open-weights）的文本转语音模型，其性能可与 ElevenLabs 相媲美，同时效率显著更高。这个拥有 40 亿参数的模型支持九种语言，并采用了一种新颖的架构，将自回归语义令牌生成与用于声学令牌的流匹配（flow-matching）相结合。此次发布凸显了 Mistral 在开放研究和拓展多模态人工智能能力前沿方面的承诺。 AI
排序理由 发布了一个开放权重（open-weights）的 TTS 模型，其新颖的架构细节在播客和配套论文中进行了讨论。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
<p>Mistral has been on an absolute tear - with frequent successful model launches it is easy to forget that they raised <a href="https://mistral.ai/news/mistral-ai-raises-1-7-b-to-accelerate-technological-progress-with-ai" target="_blank">the largest European AI round in history<…
Mistral is one of the world's leading frontier model labs, and has just raised $900m to build their European data center hub. Last year marked their first ventures into multimodal models, with Pixtral and then Voxtral in 2025, ending the year with a monster €1.7B fundraise at a €…