yl4579/StyleTTS2

Issues

• Low GPU Utilization during training

  #217 opened 9 months ago
  10

• weird pulse at the end of the model

  #216 opened 9 months ago
  5

• StyleTTS2 Training for vietnamese

  #215 opened 9 months ago
  1

• Wrong kernel sizes in the JDC model

  #214 opened 9 months ago
  1

• Did you add data beyond that of OOD_tests.txt ?

  #212 opened 9 months ago
  0

• self._shortcut(x) + self._residual(x) - The size of tensor a (34) must match the size of tensor b (35) at non-singleton dimension 2

  #208 opened 9 months ago
  4

• error in first train gen loss=0.0

  #206 opened 9 months ago
  6

• Train in one language and synthesize in English

  #205 opened 9 months ago
  3

• Same voice with different emotion

  #203 opened 9 months ago
  5

• Voice Quality issue using Librispeech

  #201 opened 9 months ago
  1

• Finetune duration_proj return NaN

  #200 opened 9 months ago
  2

• Understanding Style tts training process

  #198 opened 10 months ago
  1

• Fine-tuning or training from scratch in a differente language?

  #197 opened 10 months ago
  27

• Can we work on why there's the pop or hiss in some generate voice samples

  #194 opened 10 months ago
  4

• train_second.py model.decoder error (output tensor is nan)

  #193 opened 10 months ago
  5

• Chinese data

  #188 opened 9 months ago
  4

• OOD data for LibriTTS-460 training?

  #187 opened 9 months ago
  1

• Training a Japanese model, pitch accent and IPA

  #186 opened 9 months ago
  1

• training single speaker different accent/language results

  #185 opened 10 months ago
  4

• Preparing text and data

  #182 opened a year ago
  2

• Finetuning kernel size issue

  #180 opened a year ago
  1

• Train a zero-shot voice adaptation model for a different accent/language

  #179 opened 9 months ago
  1

• When start firtst_train give errors. I have 96 Gb Ram and 3 P40/24GB/ 1 T4 /16GB/ ??

  #178 opened 9 months ago
  10

• Fine-tuning worsens the quality of speech-synthesis.

  #177 opened a year ago
  6

• 你好，请问模型支持流式tts吗？

  #176 opened 9 months ago
  1

• stage1 training issue

  #175 opened 9 months ago
  6

• Speech-to-speech possible?

  #174 opened 9 months ago
  4

• Stage 2 training bug (after joint training)

  #171 opened 10 months ago
  7

• An Error From LJspeech Dataset

  #170 opened a year ago
  2

• Using a smaller Hifigan

  #168 opened a year ago
  1

• Some of FineTuning has this error

  #167 opened a year ago
  5

• Noise on long sentences

  #165 opened a year ago
  1

• Testing foundation layer needed!

  #163 opened a year ago
  0

• Current code doesn't work with hifigan

  #161 opened a year ago
  4

• styletts2 inference pip package

  #160 opened a year ago
  1

• All the term of training losses are 0.000 except mel loss

  #159 opened a year ago
  1

• Is It Possible to Get Word-Level Timestamps for Generated Audio?

  #156 opened a year ago
  2

• ValueError: cannot convert float NaN to integer

  #154 opened a year ago
  1

• Solutions to punctuation pause problems

  #153 opened a year ago
  2

• how to generate a more diverse voice by preset a random latent style embedding?

  #150 opened a year ago
  2

• How do I change default voices?

  #149 opened a year ago
  3

• how to generate text sequence ?

  #148 opened a year ago
  4

• the size of tensor a (251) must match the size of tensor b (51) at non-singleton dimension 2

  #147 opened 9 months ago
  4

• slmadv using differentiable duration modeling may not be helpful and even bad

  #146 opened a year ago
  6

• a typo in train_second.py?

  #145 opened a year ago
  0

• Fine Tuning fails on multi GPU

  #143 opened a year ago
  3

• RuntimeError: shape '[540672, 1]' is invalid for input of size 655360

  #141 opened a year ago
  7

• Use XPhoneBERT instead of the provided PL-BERT checkpoints.

  #140 opened a year ago
  3

• mandrain results sharing and training support

  #139 opened 9 months ago
  15

• Weird sound in the beginning and in the end of audio

  #138 opened a year ago
  1