diff --git a/examples/aishell/s0/conf/augmentation.json b/examples/aishell/s0/conf/augmentation.json index 5635d9c8..1987ad42 100644 --- a/examples/aishell/s0/conf/augmentation.json +++ b/examples/aishell/s0/conf/augmentation.json @@ -15,5 +15,20 @@ "max_shift_ms": 5 }, "prob": 1.0 + }, + { + "type": "specaug", + "params": { + "F": 10, + "T": 50, + "n_freq_masks": 2, + "n_time_masks": 2, + "p": 1.0, + "W": 80, + "adaptive_number_ratio": 0, + "adaptive_size_ratio": 0, + "max_n_time_masks": 20 + }, + "prob": 1.0 } ] diff --git a/examples/aishell/s1/conf/chunk_conformer.yaml b/examples/aishell/s1/conf/chunk_conformer.yaml index 8f67ef41..e8dabea5 100644 --- a/examples/aishell/s1/conf/chunk_conformer.yaml +++ b/examples/aishell/s1/conf/chunk_conformer.yaml @@ -93,7 +93,7 @@ training: decoding: - batch_size: 1 + batch_size: 128 error_rate_type: cer decoding_method: attention # 'attention', 'ctc_greedy_search', 'ctc_prefix_beam_search', 'attention_rescoring' lang_model_path: data/lm/common_crawl_00.prune01111.trie.klm @@ -104,11 +104,11 @@ decoding: cutoff_top_n: 0 num_proc_bsearch: 8 ctc_weight: 0.5 # ctc weight for attention rescoring decode mode. - decoding_chunk_size: 16 # decoding chunk size. Defaults to -1. + decoding_chunk_size: -1 # decoding chunk size. Defaults to -1. # <0: for decoding, use full chunk. # >0: for decoding, use fixed chunk size as set. # 0: used for training, it's prohibited here. num_decoding_left_chunks: -1 # number of left chunks for decoding. Defaults to -1. - simulate_streaming: True # simulate streaming inference. Defaults to False. + simulate_streaming: true # simulate streaming inference. Defaults to False. diff --git a/examples/librispeech/s1/conf/chunk_confermer.yaml b/examples/librispeech/s1/conf/chunk_confermer.yaml index 0d07d0a7..ec945a18 100644 --- a/examples/librispeech/s1/conf/chunk_confermer.yaml +++ b/examples/librispeech/s1/conf/chunk_confermer.yaml @@ -56,7 +56,7 @@ model: pos_enc_layer_type: 'rel_pos' selfattention_layer_type: 'rel_selfattn' causal: True - use_dynamic_chunk: True + use_dynamic_chunk: true cnn_module_norm: 'layer_norm' # using nn.LayerNorm makes model converge faster use_dynamic_left_chunk: false @@ -110,6 +110,6 @@ decoding: # >0: for decoding, use fixed chunk size as set. # 0: used for training, it's prohibited here. num_decoding_left_chunks: -1 # number of left chunks for decoding. Defaults to -1. - simulate_streaming: False # simulate streaming inference. Defaults to False. + simulate_streaming: true # simulate streaming inference. Defaults to False.