Nvidia releases new transcription model supporting 40 languages

Nvidia has released a new transcription model that converts speech to text across 40 languages. The model works on recorded audio only, not live streams, making it a fit for meeting and medical transcription tools rather than real-time conversations. That leaves live applications to rivals like ElevenLabs.

Would you use an open transcription model over a paid one for your projects?

#nvidia #speechtotext #opensource #ai