Juntae, Kim

I seems the validation path is incorrect. Please verify.

1. Did you try to use the test code in test_small.py? 2. norm_noisy.mat is very important in this code, which means the mean & variance of whole training set. norm_noisy...

Our provided recorded dataset was manually annotated by two professional engineers. If you want to construct the training set, do the followings: 1. run the VAD to the clean speech...

1. **VAD and speech enhancement (SE) dataset must be different.** The reason is that SE assumes that incoming signal is always noisy speech not noise only. We also have exploited...

the inputs for vaddnoise.m should be step 1 noise(1:length(speech1)), speech1 step 2 noise(length(speech1)+1 :length(speech1)+1 + length(speech2) ), speech2

It seems you import incorrect utils.py, utils.py definitely has bdnn_transform.

We already remove the dependency : ). From py branch you can enjoy full python version.

No while I already have the spectrogram feature based VAD but not arranged yet for uploading. I will upload that version in this month. thank you!

Unfortunately, the training script in python version haven't been upload yet. We will upload the python version training script soon.

Thank you for the detailed question because there are several questions, let's solve your questions step by step. 1. Your claim may be correct. **The TIMIT labels are not perfectly...