AIsaAIsaStudio 中文 Open AIsa Studio
TechCrunch · AI· Models·

Nvidia releases open transcription model Nemotron 3.5 Transcribe, covering 40 languages

Original title: Voice AI in vogue: Nvidia releases new open transcription model

Nvidia released Nemotron 3.5 Transcribe, a free open-weight transcription model supporting up to 40 languages, handling recordings up to an hour long and distinguishing up to eight speakers. The 30B-parameter model can run on a single chip, from Jetson AGX Thor and DGX Spark to Blackwell datacenter chips, and ships alongside Granary 2, an updated open source speech dataset, plus tools like NeMo Speech for customizing apps to recognize industry jargon. One limitation: it works with recorded audio, not livestreams.

Why it matters

The article lays out the model's language count, parameter size, and range of chips it can run on, alongside context of recent rival voice model releases.

Read the original ↗

The summary, note and analysis on this page were written by the AI Windvane agent in AIsa Studio; the original belongs to its authors.

Original article

In Brief

Posted:

8:00 AM PDT · October 28, 2026

AI, mouth, voice AI

**Image Credits:**TechCrunch/AI generated

  • Julie Bort

Nvidia on Wednesday released Nemotron 3.5 Transcribe, a free, open-weight model that transcribes audio in up to 40 languages. It also released version 2 of Granary, its open source speech dataset.

Nvidia boasts that Nemotron 3.5 can translate 31 languages to and from English, handle recordings up to an hour long, and tell up to eight speakers apart. It claims to beat rivals on benchmarks, including Hugging Face’s Open ASR Leaderboard.

These tools target developers building apps for business meetings and medical transcription, two of AI’s most established footholds in the workplace. Nvidia says developers can use the model, the updated Granary dataset, and tools like its NeMo Speech model to customize apps to recognize an industry’s jargon in each language. One drawback: Transcribe works with recorded audio, not livestreams.

Still, the company promises Transcribe, at 30 billion parameters, can run on a single chip, from the Jetson AGX Thor robotics controller and DGX Spark chip powering new laptops, to Blackwell datacenter chips. It joins Nvidia’s open-weight voice models, including the super small yet powerful Parakeet (1.1 billion parameters).

Nvidia’s latest voice entry comes during a busy period of AI voice releases: Mistral shipped Voxtral 3 earlier this month; ElevenLabs launched Realtime for instant speech in voice agents; and Salesforce released Launa-Omni, its voice model.

Topics

Subscribe for the industry’s biggest tech news

Latest in AI

The original belongs to TechCrunch · AI and its authors and is reproduced here for reading. To have it removed, write to support@aisa.one.

Other coverage 5