CV_bn_train

--- dataset_info: features: - name: audio dtype: audio: sampling_rate: 32000 - name: sentence dtype: string splits: - name: train num_bytes: 26456406807.408 num