tuanh123789/Train_Hifigan_XTTS

Issues

• Speaker Consistency Loss is Missing

  #16 opened a month ago by talipturkmen
  1

• Output model generates noisy sounds, robotic voices

  #15 opened a month ago by thucth-qt
  1

• Batch size and expected VRAM usage / training time

  #14 opened 2 months ago by lukaszliniewicz
  0

• generate_latents.py outputs literally nothing but the 4 empty subfolders

  #13 opened 2 months ago by Scarlet-Raine
  1

• Training HiFi-GAN to Upsample from 22,050 Hz to 48,000 Hz

  #10 opened 3 months ago by kikozi2000
  1

• can finetuning hifigan reduce "hoarseness"?

  #12 opened 3 months ago by kunibald413
  1

• Custom dataset resulting in mel.shape[-1] * self.hop_len == audio.shape[-1]

  #11 opened 3 months ago by C00reNUT
  1

• How many hours of audio are needed to train a model?

  #2 opened 7 months ago by bensonbs
  29

• Roadmap for training XTTS for custom language

  #5 opened 6 months ago by VafaKnm
  6

• Missing model state_dict when run test.py

  #9 opened 4 months ago by mpquochung
  1

• XTTSv2 training codes

  #4 opened 6 months ago by NikitaKononov
  0

• Train model does not generate any speech at all

  #3 opened 7 months ago by Many0therFunctions
  23