Overview

NVIDIA Nemotron 3.5 Transcribe 30B A4B is an open-weight, offline speech-and-audio model that turns recordings and text instructions into text. It supports transcription, translation, speaker-attributed and contextual transcription, timestamps, language identification, and audio question answering across 40 languages. The model is maintained by nvidia.

Best use cases

Transcribe recordings verbatim, or produce readable text by omitting fillers and accidental repetitions.

Improve recognition of names and specialist vocabulary by supplying context terms.

Create speaker-attributed transcripts and request word timestamps for meeting review.

Translate speech into a specified target language.

Identify the language spoken or ask questions about audio.

Process long recordings offline; the model supports recordings up to one hour.

Limitations

This is an offline model, not a real-time streaming ASR service; internal audio chunking does not change that.…