egs/aishell2/transformer/conf/train_asr_transformer.yaml @@ -78,6 +78,7 @@ num_time_mask: 2 dataset_conf: data_names: speech,text data_types: sound,text shuffle: True shuffle_conf:
