Apr 30 02:33 Update annotations with mispronunciations field and regenerate parquet files jimbozhang +1: Claude Opus 4.6 Regenerated parquet data from the latest scores.json on GitHub, which includes
the mispronunciations field for phones with accuracy score lower than 0.5.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> Jun 9, 2024 14:13 Update README.md Jan 17, 2024 03:52 Delete .gitattributes 03:51 Delete train.jsonl 03:51 Delete test.jsonl 01:24 Update README.md 01:16 delete wav files 01:07 Upload dataset Jan 16, 2024 23:40 Use jsonl meta file to replace python loading script Jan 15, 2024 08:56 Add word/phone level scores 04:18 Update README.md 04:04 Update requirements 03:59 Download tar from github release. Jan 11, 2024 18:37 Update README.md 18:25 Remove old style files 18:05 Update README.md 17:56 Fix builder classname 16:58 Add speaker, gender and age in dataviewer 16:27 Add the loading script 14:25 Add dataviewer.tsv 14:21 train|test tsv for dataviewer 14:17 Update README.md 14:13 Update README.md Nov 25, 2023 16:09 Update README.md 15:54 copy files from https://github.com/jimbozhang/speechocean762 15:50 initial commit 
