Voxtral Transcribe 2

Mistral has released Voxtral Transcribe 2, an open-source speech-to-text model family licensed under Apache 2.0 that includes speaker diarization and ultra-low-latency processing. The models run on-device, support 13 languages, and are designed to transcribe live audio faster and at lower cost than comparable solutions.

本文综合以下全部报道写成，而非依据任何单一来源。