--- dataset_info: features: - name: audio dtype: audio - name: sentence dtype: string splits: - name: train num_bytes: 2458890204.774 num_examples: 12421 download_size: 2390558125 dataset_size: 2458890204.774 configs: - config_name: default data_files: - split: train path: data/train-* ---