egs/aishell2/conformer/conf/train_asr_conformer.yaml @@ -84,6 +84,7 @@ num_time_mask: 2 dataset_conf: data_names: speech,text data_types: sound,text shuffle: True shuffle_conf:
