Nvidia Unveils Nemotron 3.5 Transcribe
Nvidia on Wednesday released Nemotron 3.5 Transcribe, a free, open-weight model capable of transcribing audio in up to 40 languages. Alongside it, the company published version 2 of Granary, its open-source speech dataset.
According to Nvidia, Nemotron 3.5 Transcribe can translate 31 languages to and from English, process recordings up to an hour long, and distinguish up to eight separate speakers. The company claims it outperforms competitors on benchmarks, including Hugging Face's Open ASR Leaderboard.
Built for Business and Medical Transcription
These tools target developers building applications for business meetings and medical transcription—two of AI's most established footholds in the workplace. Nvidia says developers can combine the model, the updated Granary dataset, and tools such as its NeMo Speech model to tailor apps that recognize industry-specific jargon in each language. One limitation: Transcribe works with recorded audio, not live streams.
Runs on a Single Chip
Despite its 30 billion parameters, Nvidia promises Transcribe can run on a single chip—from the Jetson AGX Thor robotics controller and the DGX Spark chip powering new laptops to Blackwell data center chips. It joins Nvidia's lineup of open-weight voice models, including the compact yet powerful Parakeet (1.1 billion parameters).
A Crowded Voice AI Market
Nvidia's latest voice entry arrives 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 own voice model.
via TechCrunch AI