I have some confusion about the gain g. In denoise.c, the gain is -1 under some conditions such as silence. And in rnn_train.py, the denoise_output of model is the gain, but the activation function of denoise_output is sigmoid, which ranges from 0 to 1, that means the gain_pred can only be trained to (0,1) while the gain_true is actually -1, would it affect the denoised result?
I have some confusion about the gain g. In denoise.c, the gain is -1 under some conditions such as silence. And in rnn_train.py, the denoise_output of model is the gain, but the activation function of denoise_output is sigmoid, which ranges from 0 to 1, that means the gain_pred can only be trained to (0,1) while the gain_true is actually -1, would it affect the denoised result?
Thanks for your help.