Jun Chen
You should load the best_model.tar. You'd better choose the pytorch 1.7.1. Hope this answer can help you!
First of all, thank you very much for your correction on this code! I will fix this part later. Secondly, I don't quite understand what you said about clipping, because...
Thank you very much for your response, it has been quite helpful to me！
Sorry, I don't quite understand your question. Generally speaking, the length of speech that needs to be enhanced is generally the same as the input. If you want to adjust...
Oh, I see. Our model is end-to-end, so the degree of noise reduction depends on the data seen during training. If you want to reduce the degree of speech suppression,...
Thank you for your positive review of FullSubNet+! As I described in my paper, this work was indeed inspired by ChannelSElayer. But we wanted to make more use of some...
Thanks for your detailed comments! > Thanks for your reply! I think there might be some confusion here. What confused me is that AvgPool1d appeared after Conv1d seems to make...
I didn't think about this as much as you did when I did this work before! So I haven't tried this point you mentioned either. Your idea is really valuable,...
Have you checked the folder enhanced_0194？ There should be the enhanced audio.
Maybe I know where your bug is. The _dataset.path_ refers to a directory of files, not to a specific file. Can you try changing **testdata/input_data/1.wav** to **testdata/input_data**? The same applies...