Nvidia releases a free speech-to-text model that can transcribe up to 40 languages, but only works on recordings, not live audio.
Read the original at techcrunch.com→Nvidia just released a free speech-to-text model that can transcribe up to 40 languages, but it only works on recordings, not live audio, so it’s a big step for meeting and medical transcription tools while still...
Original headline: "Voice AI in vogue: Nvidia releases new open transcription model"
Coverage timeline
- Oct 28, 15:00 UTC TechCrunch AI lead source Voice AI in vogue: Nvidia releases new open transcription model