leimao/Voice-Converter-CycleGAN

Issues

• Request Help

  #41 opened 10 months ago by CoinKing-Dream
  0

• from preprocess import *

  #39 opened 2 years ago by zahrakhanjani128
  0

• Cant usage pretrained model

  #38 opened 3 years ago by Lip4ik
  1

• Is pertained model available?

  #30 opened 5 years ago by mrgloom
  1

• update to tensorflow 2?

  #37 opened 3 years ago by carcinocron
  1

• No weight files for the pretrained model on Google Drive

  #35 opened 4 years ago by Kokkini
  0

• GPU

  #24 opened 5 years ago by Wajih88
  1

• Using the voice converter for French

  #22 opened 5 years ago by Wajih88
  3

• Why not ues the cross_entropy_loss

  #32 opened 5 years ago by sqsqsqthu
  2

• preprocess.py:170: RuntimeWarning: divide by zero encountered in log

  #29 opened 5 years ago by salisbury-espinosa
  1

• Is there a way to resume training?

  #31 opened 5 years ago by graham-eisele
  0

• ImportError: cannot import name 'Type'

  #28 opened 5 years ago by wueching
  11

• Time in training becomes gradually slow.

  #27 opened 5 years ago by n5-suzuki
  0

• how much speech data for training ?

  #26 opened 5 years ago by peter05010402
  2

• Checkpoints

  #12 opened 6 years ago
  4

• ValueError: Variable generator_A2B/h1_conv/kernel already exists, disallowed. Did you mean to set reuse=True or reuse=tf.AUTO_REUSE in VarScope? Originally defined at:

  #25 opened 5 years ago by yuuSiVo
  9

• Many to one conversion

  #23 opened 5 years ago by Husnain08
  1

• mcep

  #16 opened 5 years ago by assyoucan
  3

• The Number of Training Epochs

  #21 opened 6 years ago by Husnain08
  1

• The Number of Training Epochs

  #20 opened 5 years ago by Husnain08
  0

• Mismatched number of frames between F0 (627), spectrogram (628) and aperiodicty (627)

  #17 opened 6 years ago by Likkkez
  4

• No non-parallel setting

  #18 opened 6 years ago by njellinas
  6

• ImportError: No module named 'librosa'

  #15 opened 6 years ago by c1a1o1
  0

• error when run train.py

  #14 opened 6 years ago by TTTJJJWWW
  4

• raise NoBackendError() audioread.NoBackendError

  #13 opened 6 years ago
  12

• preprocess

  #11 opened 6 years ago by assyoucan
  3

• Continue training with the saved model after interruption

  #7 opened 6 years ago by dawningo
  2

• Store intermediate preprocessing data

  #10 opened 6 years ago by naveenkumarmarri
  1

• Interpretation GLU

  #9 opened 6 years ago by pritishyuvraj
  2

• About Loss Curve and timbre of converted voices

  #8 opened 6 years ago by MorganCZY
  4

• It's not an issue, but if i want to map the accent as well as the voice. so let's say i have a phonetic transcription for the audio data and i want to be able to convert accent from domain A to domain B and vice versa. How to achieve this? I thought of the conditional gan approach where you include the additional data to train the model on it as well as the audio data

  #5 opened 6 years ago by nawawy
  3

• AttributeError about tf.contrib.layers.instance_norm

  #6 opened 6 years ago by MorganCZY
  5

• when training and converting, I econtered the problem of "RuntimeWarning: divide by zero "

  #3 opened 6 years ago by hujinsen
  3

• 'ReplayBuffer' is not defined

  #4 opened 6 years ago by hujinsen
  0

• Discriminator tensor shape

  #2 opened 7 years ago by edchengg
  13

• The output of the conversion basically sounds like the input

  #1 opened 7 years ago by fazlekarim
  3