tts-dataset/japanese-singing-voice-vocal-only-audit

Japanese singing voice vocal-only — aggregate audit This one-row audit describes tts-dataset/japanese-singing-voice-vocal-only at revision c3ea48aa3909c23fae8e04ca28bed7ab84054066. It excludes titles, source names, URLs, item IDs, paths, JSON metadata, hashes, and audio. Sparse tar indexing found 4,871 complete JSON/WAV pairs across 40 unique archives, with zero irregular pairs, unsafe paths, walk errors, or unterminated tars. Forty bounded WAV samples were 44.1 kHz stereo… See the full description on the dataset page: https://huggingface.co/datasets/tts-dataset/japanese-singing-voice-vocal-only-audit.