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