BSC-LT/distilled-yodas-spanish

Distilled YODAS Spanish is a high-quality subset of the Spanish portion of the YouTube-Oriented Dataset for Audio and Speech (YODAS). While the full YODAS corpus contains over 37,000 hours of Spanish speech across 43 million files, this dataset provides a distilled version of approximately 8,000 validated hours.