Commit History
-
fix(*): decrease the row group size for audio datasets
By default, the row group size is set to 100 rows, which is too large for the Hugging Face Dataset Viewer. This change reduces that number to 20 rows. See this issue for more details: https://github.com/google-deepmind/librispeech-long/issues/1a5bb3ef
-
feat(data): upload the entire dataset
Now that the sample dataset is working as expected, this commit contains the entire LibriSpeech-Long dataset.9d21390