Skip to main content
← Back to news

Parakeet TDT 0.6B v2: local ASR transcription, 24 minutes of audio in one pass

Model releases
Source:nvidia
NVIDIA updates Parakeet TDT 0.6B to v2: the NeMo transcription model (0.6B parameters, English) handles audio up to 24 minutes in a single pass and scores RTFx 3380 on the HF-Open-ASR leaderboard with batch 128. A compact ASR family built for local use.

Get the next local-AI briefing

Once a week, the top selection from iaLocale: articles, news and resources on local AI, and more.

After signing up you receive a confirmation email (double opt-in). Without confirmation we send nothing. No spam, we promise.