| dataset_info: | |
| features: | |
| - name: merged_phonemes | |
| dtype: string | |
| - name: input_ids | |
| sequence: int32 | |
| - name: phonemes | |
| sequence: string | |
| splits: | |
| - name: train | |
| num_bytes: 24927984204 | |
| num_examples: 1949339 | |
| download_size: 11978919561 | |
| dataset_size: 24927984204 | |
| configs: | |
| - config_name: default | |
| data_files: | |
| - split: train | |
| path: data/train-* | |