xcmyz/FastSpeech

Issues

• Objective function for fast speech model

  #102 opened a month ago by MordehayM
  0

• Is this still maintained?

  #100 opened 2 years ago by Akul2010
  0

• How to extract alignment from tacotron2?

  #88 opened 4 years ago by CanKorkut
  6

• long int 转换成float erro

  #89 opened 3 years ago by li-xx-5
  0

• denoiser

  #90 opened 3 years ago by li-xx-5
  1

• Expected object of backend CUDA but got backend CPU for argument #3 'index'

  #93 opened 3 years ago by li-xx-5
  1

• wav in chinese

  #95 opened 3 years ago by li-xx-5
  1

• BUG:OSError: sndfile library not found

  #97 opened 3 years ago by ywh-my
  1

• RuntimeError: stack expects each tensor to be equal size, but got [40, 240] at entry 0 and [78, 202] at entry 1

  #98 opened 3 years ago by khainh3101
  0

• how to solve this bug?

  #54 opened 5 years ago by lzkzls
  4

• Error when training new model for another language

  #96 opened 4 years ago by EuphoriaCelestial
  0

• duration loss calulated in log domain or linear domain

  #94 opened 4 years ago by MorganCZY
  0

• onnxruntime.capi.onnxruntime_pybind11_state.Fail: [ONNXRuntimeError] : 1 : FAIL : Load model from fastwave.onnx failed:Type Error: Type parameter (T) bound to different types (tensor(double) and tensor(float) in node ().

  #92 opened 4 years ago by li-xx-5
  0

• fastspeech model to torch script convert for c++ inference

  #61 opened 5 years ago by alokprasad
  4

• some question about squeezewave denoiser

  #91 opened 4 years ago by li-xx-5
  0

• 请问训练多久得到的pretrain model呢？然后，请问使用了多少GPU呢？

  #86 opened 4 years ago by PeiyanFlying
  1

• Resume training from checkpoint result in NaN?

  #87 opened 4 years ago by Approximetal
  1

• What is the difference between postnet and CBHG?

  #85 opened 4 years ago by Approximetal
  3

• How to get alignment?

  #82 opened 4 years ago by LLianJJun
  8

• Preprocess.py got stuck: Tried to debug

  #83 opened 4 years ago by deepkshikha
  0

• Have anyone tried using LSTM to replace FFT block?

  #84 opened 4 years ago by BuaaAlban
  0

• error in new commit

  #81 opened 4 years ago by Ahmad-noborders
  9

• ModuleNotFoundError: No module named 'glow'

  #77 opened 4 years ago by ccuiyuhan
  3

• how to generate new alignment files?

  #67 opened 4 years ago by alokprasad
  2

• RuntimeError: shape '[1, 1, 155520]' is invalid for input of size 311040 [custom data training]

  #79 opened 4 years ago by Ahmad-noborders
  3

• training stops in few seconds and no checkpoint file created

  #80 opened 4 years ago by Ahmad-noborders
  3

• How to Combine Mel genereted from Fastspeech and using lpcnet as vocoder.

  #57 opened 4 years ago by alokprasad
  5

• FastSpeech2 support?

  #76 opened 4 years ago by LLianJJun
  2

• RuntimeError: shape '[1, 1, 47937]' is invalid for input of size 95874

  #58 opened 5 years ago by aguazul
  5

• Training with custom dataset

  #74 opened 4 years ago by sujeendran
  1

• synthesized sentence is not complete

  #75 opened 4 years ago by zhoulinmin
  1

• The training ended early

  #78 opened 4 years ago by Text2-m
  1

• synthesis error

  #69 opened 4 years ago by BuaaAlban
  2

• why duration_loss use L1 loss?

  #72 opened 4 years ago by superhg2012
  1

• Synthesis Result of Chinese Corpus

  #55 opened 5 years ago by Liujingxiu23
  6

• Specific batch loading in train.py

  #71 opened 5 years ago by alxmamaev
  1

• alignment mel frames does not math mel_target frame number

  #70 opened 5 years ago by superhg2012
  12

• FastSpeech quality?

  #66 opened 5 years ago by qo4on
  1

• Early stopping? Loss

  #68 opened 5 years ago by BuaaAlban
  2

• Can we add and control break when inference new text?

  #65 opened 5 years ago by xiaoyangnihao
  0

• why does vibrato exist when training and inferencing chinese sentences?

  #64 opened 5 years ago by xiaoyangnihao
  0

• preprocess.py 好像运行不了

  #62 opened 5 years ago by cythc
  1

• 是否有基于中文音素和mel的alignment model

  #63 opened 5 years ago by xingshi19977
  1

• integration of fastspeech with Squeezewave vocoder

  #59 opened 5 years ago by alokprasad
  3

• RuntimeError: The size of tensor a (1248) must match the size of tensor b (156) at non-singleton dimension 1

  #60 opened 5 years ago by cutekih
  2

• 我想使用新的alignment，但是并未找到alignment.py文件

  #51 opened 5 years ago by victor-wzq
  6

• How to combine FastSpeech with voice synthesizer like melGAN?

  #56 opened 5 years ago by nikawool
  1

• Which License?

  #52 opened 5 years ago by Ina299
  1

• The pretrained model on 百度网盘 is broken?

  #53 opened 5 years ago by sniperwrb
  1

• running preprocess.py occurs NAN

  #50 opened 5 years ago by runningJ
  4