Event ·Official first-hand

Mistral AI 发布 Voxtral 3：语音转写、TTS 与语音智能体 SDK

Mistral AI 发布 Voxtral 3 系列，包含 3 个语音转写模型、1 个文本转语音模型和一个基于 LiveKit 的语音智能体 SDK。其中 Voxtral Small Transcribe 3 的 AA-WER 为 2.1%，在 Artificial Analysis 排行榜上位列第 3。开发者可以直接用该 SDK 搭建语音对话类应用。

Current read

Mistral 把转写、TTS 和 LiveKit 语音 SDK 打包成一整套，转写错误率只有 2.1%，做语音应用的可以直接上手试试。

2 sources55° AI heatLast updated 2026/10/21 06:03:12

Evidence

2 sources

Primary sourcemarktechpost

2026/10/21 06:03:12View source

Related source 1Fish Audio

2026/10/19 16:48:01View source

Conflict check

Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.

Open the story