Wenjun Wang
Results
1
comments of
Wenjun Wang
As we have found, the output spectrogram should be raised by a power of 1.2. Then the voice sounds more clear