YODAS v3 lands on Hugging Face, with a post calling it a 1.1 million-hour audio dataset
Hugging Face posted that YODAS v3 is now available, describing it as a 1.1 million-hour audio dataset spanning 100-plus languages under a CC-BY-3.0 license.
TLDR
Hugging Face posted that YODAS v3 is now available on Hugging Face, describing it as a 1.1 million-hour audio dataset and calling it the biggest audio dataset ever. The post also said it is the first dataset at this scale to include stereo audio at 48 kHz, along with timestamped transcripts and translations, and that it covers more than 100 languages under a CC-BY-3.0 license.
Combined views
1.3K
1 Source, first seen 2h ago
