Kyt Dotson / SiliconANGLE : Nvidia debuts Nemotron 3.5 Transcribe, an open-weight model for multilingual audio transcription and translation, and Granary v2, its open-source speech dataset — Nvidia Corp. today released a leaderboard-winning open artificial intelligence model for multilingual audio transcription and translation …
Nvidia debuts Nemotron 3.5 Transcribe, an open-weight model for multilingual audio transcription and translation, and Granary v2, its open-source speech dataset (Kyt Dotson/SiliconANGLE)
In short
The chipmaker's latest artificial intelligence release pairs an open, leaderboard-winning transcription model with public speech data. By sharing these components openly, the company is equipping developers with foundational resources to build and refine multilingual voice technologies. This double release addresses processing across diverse languages through transparent assets. Specifically, the rollout includes Nemotron 3.5 Transcribe alongside Granary v2 to support both training data needs and multilingual translation execution.
Key points
- Developers gain access to an open-weight model engineered for complex multilingual audio…: Developers gain access to an open-weight model engineered for complex multilingual audio processing tasks.
- The release pairs a competitive AI model with an open-source speech dataset for community…: The release pairs a competitive AI model with an open-source speech dataset for community use.
- Nvidia is expanding community access to foundational tools for speech data and language t…: Nvidia is expanding community access to foundational tools for speech data and language translation.