sh-lee-prml/BigVGAN

Issues

• Why did you use low-pass filter twice in AMPBlock?

  #6 opened 2 years ago by hoyden
  5

• How to reduce the parameters of the bigvgan model

  #12 opened 2 years ago by 980202006
  8

• Results?

  #9 opened 2 years ago by PiotrDabkowski
  3

• Alphas are not trainable?

  #10 opened 2 years ago by PiotrDabkowski
  4

• inference not working

  #11 opened 2 years ago by eschmidbauer
  1

• Training on different sampling rates

  #7 opened 2 years ago by skol101
  3

• Test with MEL spectrogram

  #8 opened 2 years ago by skol101
  1

• Add link in README to dataset used

  #1 opened 2 years ago by nateraw
  3

• Can this be used as a substitute for HifiGan for VC task?

  #2 opened 2 years ago by skol101
  32

• Single node multi gpu training extremely slow vs single node 1 gpu training

  #5 opened 2 years ago by skol101
  0

• correct eps value?

  #4 opened 2 years ago by skol101
  1

• Use for tokens in vocoder only training

  #3 opened 2 years ago by skol101
  1