update

examples/spectrum_dfnet_aishell/run.sh

@@ -8,7 +8,7 @@ sh run.sh --stage 2 --stop_stage 2 --system_version windows --file_folder_name f
 --speech_dir "E:/programmer/asr_datasets/aishell/data_aishell/wav/train"
 
 
-sh run.sh --stage 1 --stop_stage 3 --system_version centos --file_folder_name file_dir \
+sh run.sh --stage 1 --stop_stage 2 --system_version centos --file_folder_name file_dir \
 --noise_dir "/data/tianxing/HuggingDatasets/nx_noise/data/noise" \
 --speech_dir "/data/tianxing/HuggingDatasets/aishell/data_aishell/wav/train"
 

examples/spectrum_dfnet_aishell/step_2_train_model.py

@@ -320,7 +320,7 @@ def main():
             if torch.any(torch.isnan(lsnr_prediction)) or torch.any(torch.isinf(lsnr_prediction)):
                 raise AssertionError("nan or inf in lsnr_prediction")
 
-            speech_loss = speech_mse_loss.forward(speech_spec_prediction, speech_complex_spec)
+            speech_loss = speech_mse_loss.forward(speech_spec_prediction, torch.view_as_real(speech_complex_spec))
             irm_loss = irm_mse_loss.forward(speech_irm_prediction, speech_irm_target)
 
             lsnr_prediction = (lsnr_prediction - config.lsnr_min) / (config.lsnr_max - config.lsnr_min)
@@ -375,7 +375,7 @@ def main():
                 if torch.any(torch.isnan(lsnr_prediction)) or torch.any(torch.isinf(lsnr_prediction)):
                     raise AssertionError("nan or inf in lsnr_prediction")
 
-                speech_loss = speech_mse_loss.forward(speech_spec_prediction, speech_complex_spec)
+                speech_loss = speech_mse_loss.forward(speech_spec_prediction, torch.view_as_real(speech_complex_spec))
                 irm_loss = irm_mse_loss.forward(speech_irm_prediction, speech_irm_target)
 
                 lsnr_prediction = (lsnr_prediction - config.lsnr_min) / (config.lsnr_max - config.lsnr_min)