jxzhanggg/nonparaSeq2seqVC_code

Issues

• 代码全是坑

  #51 opened 2 years ago by vcbeaut
  3

• ValueError: too many values to unpack (expected 3)

  #52 opened 2 years ago by JaistZhanhangZhang
  1

• Feature Request

  #44 opened 2 years ago by ChrisDelClea
  1

• Include pre-trained models

  #1 opened 5 years ago by JRMeyer
  4

• Transition agent in forward attention

  #47 opened 3 years ago by narendranp
  1

• Keeping prosodic features of reference Speaker

  #41 opened 4 years ago by jucasansao
  4

• > They are my prepared training list.

  #43 opened 4 years ago by JeffC0628
  2

• Fine-tuning help

  #38 opened 4 years ago by leonardw86
  1

• Documentation: Audio + Text Feature Extraction

  #2 opened 4 years ago by JRMeyer
  26

• assignment statement of y_tts_pred & y_vc_pred

  #34 opened 4 years ago by inconnu11
  2

• Multi-GPU training

  #39 opened 4 years ago by ivancarapinha
  3

• list file format

  #33 opened 4 years ago by leonardw86
  2

• The mechanism of alignment between text encoder output and audio_seq2seq output

  #37 opened 4 years ago by inconnu11
  1

• Speaker and linguistic embedding visualizations do not look good as in the paper

  #29 opened 4 years ago by huukim136
  10

• Training the model for a different language

  #36 opened 4 years ago by ivancarapinha
  1

• Pre-train model results

  #35 opened 4 years ago by ivancarapinha
  3

• GPU memory requirements

  #32 opened 4 years ago by ivancarapinha
  2

• Interpreting results

  #26 opened 5 years ago by JRMeyer
  15

• Did you use a silence symbol?

  #31 opened 4 years ago by youngsuenXMLY
  1

• Alignment problem

  #23 opened 4 years ago by Approximetal
  2

• How much time is the training dataset?

  #30 opened 4 years ago by youngsuenXMLY
  4

• deduce num_speakers from *list files

  #18 opened 4 years ago by JRMeyer
  1

• Redundant code that is the same in "fine-tune" and "pre-train"

  #11 opened 4 years ago by jxzhanggg
  1

• Normalizing features

  #28 opened 5 years ago by huukim136
  5

• Remove all Legacy Code

  #24 opened 5 years ago by JRMeyer
  3

• Why consistent_loss_w=0.0 ?

  #22 opened 5 years ago by Approximetal
  6

• Change variable names "input_text" and "text_input"

  #25 opened 5 years ago by JRMeyer
  0

• Documentation: Training Time

  #20 opened 5 years ago by JRMeyer
  3

• gate outputs?

  #17 opened 5 years ago by JRMeyer
  5

• Upgrade to Python3

  #4 opened 5 years ago by JRMeyer
  2

• Is there any method to predict speaker code embedding?

  #12 opened 5 years ago by youngsuenXMLY
  1

• text_inputs vs text_targets?

  #16 opened 5 years ago by JRMeyer
  1

• Why skip long utterances?

  #3 opened 5 years ago by JRMeyer
  1

• Upgrade to TensorFlow 2.0 (or not?)

  #6 opened 5 years ago by JRMeyer
  1