--- dataset_info: config_name: hi features: - name: path dtype: string - name: audio dtype: audio: sampling_rate: 16000 - name: sentence dtype: string - name: condition_on_prev sequence: int64 - name: whisper_transcript dtype: string splits: - name: train num_bytes: 645313667.0 num_examples: 738 - name: validation num_bytes: 364122123.0 num_examples: 415 - name: test num_bytes: 512035626.0 num_examples: 590 download_size: 1423863035 dataset_size: 1521471416.0 configs: - config_name: hi data_files: - split: train path: hi/train-* - split: validation path: hi/validation-* - split: test path: hi/test-* ---