From AI Insider · James Dargan · · 50s
Nvidia Releases Open-Weight Nemotron 3.5 Transcribe Model
Nvidia's free multilingual voice AI model transcribes audio in 40 languages for developers.
In brief
Nvidia's free multilingual voice AI model transcribes audio in 40 languages for developers. Nvidia released an open-weight model that transcribes audio in up to 40 languages for business and medical applications. Originally reported by AI Insider.
Nvidia Releases Nemotron 3.5 Transcribe
Nvidia released a free open-weight model that transcribes audio in up to 40 languages for developers building business meeting and medical transcription applications.
Model Handles 31 Languages and 8 Speakers
The model translates 31 languages to and from English, processes up to one hour of recordings, and distinguishes between eight speakers.
Runs on Single Chip for Business Use
At 30 billion parameters, the model runs on a single chip from Jetson AGX Thor to Blackwell data center processors for business and medical applications.
What matters most
Nvidia released an open-weight model that transcribes audio in up to 40 languages for business and medical applications.
