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) — SkimNews
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- Nemotron 3.5 Transcribe was released as an open-weight model for multilingual audio transcription and translation, with open speech recognition spanning 40 languages.
- Nvidia introduced Granary v2, an open-source speech dataset, in the same announcement as the leaderboard-winning Nemotron 3.5 Transcribe model.
Why it matters: Developers seeking speech recognition across 40 languages gain access to both Nemotron 3.5 Transcribe's weights and the Granary v2 speech dataset through Nvidia's open release. The launch supplies a leaderboard-winning transcription and translation model alongside a corresponding dataset, giving developers two open components in the same release.
Ask SkimNews
