Nvidia released a free open speech-to-text transcription model supporting up to 40 languages. The model is designed for recording-based transcription rather than live audio, making it particularly useful for meeting and medical transcription use cases.
Background
Nvidia continues expanding its AI infrastructure offerings with open models. The release comes amid growing demand for voice AI tools in enterprise and healthcare verticals.
- Source
- TechCrunch
- Published
- Oct 28, 2026 at 11:00 PM
- Score
- 6.0 / 10