Post by Mistral AI

🚀 Voxtral Mini Transcribe 2 & Voxtral Realtime: The future of speech-to-text is here. Today, we’re launching two models that push the boundaries of transcription: ⚡ Voxtral Realtime - A streaming-first model built for ultra-low latency (configurable down to sub-200ms), enabling real-time voice agents and live applications. It offers edge deployment, industry-leading accuracy at the lowest cost and we’re releasing the model as open weights under Apache 2.0. 🔹 Voxtral Mini Transcribe 2 - Our next-gen batch transcription model, delivering state-of-the-art accuracy with speaker diarization, word-level timestamps, and context biasing in 13 languages. 🎛️ Test them instantly in our new Audio Playground in Mistral AI Studio - experience diarization, timestamps, and real-time transcription in action. The fastest, most precise transcription is now in your hands. Learn more here: 🎙️➡️https://lnkd.in/eKwXxzRH

Video Content