feat: update

.gitignore

@@ -3,4 +3,4 @@ python39
 heygem
 .vscode
 python39.tar.gz
-examples
+/examples

codes/wenet/examples/aishell/aidata/conf/train_conformer_multi_cn.yaml

@@ -0,0 +1,85 @@
+# network architecture
+# encoder related
+encoder: conformer
+encoder_conf:
+    output_size: 256    # dimension of attention
+    attention_heads: 4
+    linear_units: 2048  # the number of units of position-wise feed forward
+    num_blocks: 12      # the number of encoder blocks
+    dropout_rate: 0.1
+    positional_dropout_rate: 0.1
+    attention_dropout_rate: 0.0
+    input_layer: conv2d # encoder input type, you can chose conv2d, conv2d6 and conv2d8
+    normalize_before: true
+    cnn_module_kernel: 15
+    use_cnn_module: True
+    activation_type: 'swish'
+    pos_enc_layer_type: 'rel_pos'
+    selfattention_layer_type: 'rel_selfattn'
+
+# decoder related
+decoder: transformer
+decoder_conf:
+    attention_heads: 4
+    linear_units: 2048
+    num_blocks: 6
+    dropout_rate: 0.1
+    positional_dropout_rate: 0.1
+    self_attention_dropout_rate: 0.0
+    src_attention_dropout_rate: 0.0
+
+# hybrid CTC/attention
+model_conf:
+    ctc_weight: 0.3
+    lsm_weight: 0.1     # label smoothing option
+    length_normalized_loss: false
+
+# use raw_wav or kaldi feature
+raw_wav: false
+
+# feature extraction
+collate_conf:
+    # waveform level config
+    wav_distortion_conf:
+        wav_dither: 0.1
+        wav_distortion_rate: 0.0
+        distortion_methods: []
+    speed_perturb: true
+    feature_extraction_conf:
+        feature_type: 'fbank'
+        mel_bins: 80
+        frame_shift: 10
+        frame_length: 25
+        using_pitch: false
+    # spec level config
+    # spec_swap: false
+    feature_dither: 0.0 # add dither [-feature_dither,feature_dither] on fbank feature
+    spec_aug: true
+    spec_aug_conf:
+        warp_for_time: False
+        num_t_mask: 2
+        num_f_mask: 2
+        max_t: 50
+        max_f: 10
+        max_w: 80
+
+
+# dataset related
+dataset_conf:
+    max_length: 1300     #40960
+    min_length: 0
+    batch_type: 'static' # static or dynamic
+    batch_size: 40
+    sort: true
+
+grad_clip: 5
+accum_grad: 4
+max_epoch: 240
+log_interval: 100
+
+optim: adam
+optim_conf:
+    lr: 0.0025  #0.0025
+scheduler: warmuplr     # pytorch v1.1.0+ required
+scheduler_conf:
+    warmup_steps: 100000

codes/wenet/examples/aishell/aidata/conf/train_conformer_multi_cn_linear.yaml

@@ -0,0 +1,85 @@
+# network architecture
+# encoder related
+encoder: conformer
+encoder_conf:
+    output_size: 256    # dimension of attention
+    attention_heads: 4
+    linear_units: 1024  # the number of units of position-wise feed forward
+    num_blocks: 6      # the number of encoder blocks
+    dropout_rate: 0.1
+    positional_dropout_rate: 0.1
+    attention_dropout_rate: 0.0
+    input_layer: linear # encoder input type, you can chose linear,conv2d, conv2d6 and conv2d8
+    normalize_before: true
+    cnn_module_kernel: 15
+    use_cnn_module: True
+    activation_type: 'swish'
+    pos_enc_layer_type: 'rel_pos'
+    selfattention_layer_type: 'rel_selfattn'
+
+# decoder related
+decoder: transformer
+decoder_conf:
+    attention_heads: 4
+    linear_units: 1024
+    num_blocks: 3
+    dropout_rate: 0.1
+    positional_dropout_rate: 0.1
+    self_attention_dropout_rate: 0.0
+    src_attention_dropout_rate: 0.0
+
+# hybrid CTC/attention
+model_conf:
+    ctc_weight: 0.3
+    lsm_weight: 0.1     # label smoothing option
+    length_normalized_loss: false
+
+# use raw_wav or kaldi feature
+raw_wav: false
+
+# feature extraction
+collate_conf:
+    # waveform level config
+    wav_distortion_conf:
+        wav_dither: 0.1
+        wav_distortion_rate: 0.0
+        distortion_methods: []
+    speed_perturb: true
+    feature_extraction_conf:
+        feature_type: 'fbank'
+        mel_bins: 80
+        frame_shift: 10
+        frame_length: 25
+        using_pitch: false
+    # spec level config
+    # spec_swap: false
+    feature_dither: 0.0 # add dither [-feature_dither,feature_dither] on fbank feature
+    spec_aug: true
+    spec_aug_conf:
+        warp_for_time: False
+        num_t_mask: 2
+        num_f_mask: 2
+        max_t: 50
+        max_f: 10
+        max_w: 80
+
+
+# dataset related
+dataset_conf:
+    max_length: 1300     #40960
+    min_length: 0
+    batch_type: 'static' # static or dynamic
+    batch_size: 40
+    sort: true
+
+grad_clip: 5
+accum_grad: 4
+max_epoch: 240
+log_interval: 100
+
+optim: adam
+optim_conf:
+    lr: 0.002
+scheduler: warmuplr     # pytorch v1.1.0+ required
+scheduler_conf:
+    warmup_steps: 50000

data/log/dh.log

@@ -1 +1,7 @@
 [2025-05-21 13:53:51,800] [cv_logging.py[line:27]] [INFO] [Use default multi mode: multi-thread, or you can set env 'CV_MULTI_MODE' to multi-process/torch-process]
+[2025-05-21 14:10:15,702] [cv_logging.py[line:27]] [INFO] [Use default multi mode: multi-thread, or you can set env 'CV_MULTI_MODE' to multi-process/torch-process]
+[2025-05-21 14:10:17,488] [__init__.py[line:533]] [WARNING] [mkdir -p failed for path /home/user/.cache/matplotlib: [Errno 13] Permission denied: '/home/user/.cache/matplotlib']
+[2025-05-21 14:10:17,489] [__init__.py[line:550]] [WARNING] [Matplotlib created a temporary cache directory at /tmp/matplotlib-hz8kuiys because there was an issue with the default path (/home/user/.cache/matplotlib); it is highly recommended to set the MPLCONFIGDIR environment variable to a writable directory, in particular to speed up the import of Matplotlib and to better support multiprocessing.]
+[2025-05-21 14:10:17,657] [font_manager.py[line:1584]] [INFO] [generated new fontManager]
+[2025-05-21 14:10:17,823] [app_local.py[line:233]] [WARNING] [ -> 服务不进行注册]
+[2025-05-21 14:10:17,823] [app_local.py[line:234]] [INFO] [TransDhTask init]