Voice AI in vogue: Nvidia releases new open transcription model
- ID
- 40398
- Status
- summarized
- Published
- 28 Oct 2026, 11:00 PM
- Fetched
- 28 Oct 2026, 11:58 PM
- Provider
- TechCrunch
- Category
- technology
- Original URL
- https://techcrunch.com/2026/10/28/voice-ai-in-vogue-nvidia-releases-new-open-transcription-model/
- Source URL
- https://techcrunch.com/feed/
Summary
- Score
- 6.5
- Created
- 28 Oct 2026, 11:58 PM
- Tags
- Audience
- developersai_ml_learnersai_agent_users
What happened
Nvidia released Nemotron 3.5 Transcribe, a free open-weight transcription model covering up to 40 languages, translating 31 languages to and from English, handling recordings up to an hour long, and separating up to eight speakers. It ships alongside Granary v2, an updated open speech dataset, and Nvidia claims benchmark wins including on Hugging Face's Open ASR Leaderboard. At 30 billion parameters it is said to run on a single chip, from Jetson AGX Thor and DGX Spark up to Blackwell datacenter parts, but it only handles recorded audio, not livestreams.
Why it matters
If you are building meeting or medical transcription, a free 30B open-weight model that claims to fit on one chip is a genuine self-host-versus-API decision — and the Granary v2 dataset plus NeMo Speech means you can fine-tune for industry jargon per language. But the recorded-audio-only limit rules it out for realtime voice agents, so streaming pipelines still need something like ElevenLabs Realtime or Mistral's Voxtral 3 instead. The catch: all capability claims are Nvidia's own, including the leaderboard comparison, so benchmark it on your own audio before committing.
Discussion angle
Would you actually self-host a 30B transcription model on a single chip, or does batch-only plus unverified benchmarks keep you on hosted ASR APIs — and what would you test first?