--- dataset_info: features: - name: audio dtype: audio: sampling_rate: 16000 - name: transcripts dtype: string splits: - name: train num_bytes: 18234945060.0 num_examples: 101236 - name: test num_bytes: 2410514563.634364 num_examples: 12655 - name: valid num_bytes: 2300143023.7636356 num_examples: 12654 download_size: 21259055664 dataset_size: 22945602647.398003 configs: - config_name: default data_files: - split: train path: data/train-* - split: test path: data/test-* - split: valid path: data/valid-* ---