Rayhane-mamah Tacotron-2 issues

Rayhane-mamah / Tacotron-2

DeepMind's Tacotron-2 Tensorflow implementation

MIT License

2.28k stars 905 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

attention can't be learned when setting magnitude_power=1

#381 NewEricWang opened 5 years ago
0
output_per_step around 2~5,but final-mel-output is only one frame

#380 Liujingxiu23 opened 5 years ago
1
how can i improve the training speed of 'Tacotron GTA Synthesis'

#379 Zero0one1 closed 5 years ago
1
linear_wav = self.session.run(self.linear_wav_outputs, feed_dict={self.linear_spectrograms: linear})这句用了10s

#378 lzj520 opened 5 years ago
11
train mandarin ,eval stop_token_loss increase

#377 ChristopheZhao closed 5 years ago
36
Build issue: PyAudio

#376 xykong1958 opened 5 years ago
1
overtraining

#375 DLZML001 opened 5 years ago
0
bad voice when setting tacotron_dropout_rate=0.0 in synthesis

#374 NewEricWang opened 5 years ago
4
FloWaveNet vocoder

#373 ryhorv opened 5 years ago
0
wavenet train.py loss exloded

#372 11721206 opened 5 years ago
2
Removing dropout in conv1d does not affect synthesized results but reduces loss more

#371 begeekmyfriend opened 5 years ago
0
Bad inference for really short sentences

#370 alexdemartos closed 5 years ago
1
Synthesize error Checkpoint

#369 C5YS opened 5 years ago
4
Training Tacotron and Wavenet

#368 manuel3265 opened 5 years ago
1
What is your CUDA and CUDNN config?

#367 ThongTranHCM opened 5 years ago
0
I am using `griffin-lim` branch which does no contain `split_infos` and have implemented batch synthesis on it and merged into `mandarin-new` branch. I am afraid there is something in code not compitable with that branch. Sorry about it.

#366 ZLZ-Nancy opened 5 years ago
4
Alignment off for particular dataset

#365 ZohaibAhmed closed 5 years ago
2
preprocess.py will fail on TensorFlow 2.0 due to contrib deprecation

#364 RoelVdP opened 5 years ago
4
During the inference, input the same data, the output will be different

#363 iamxiaoyubei closed 5 years ago
4
TTS for Long Sentences

#362 Pawankhatri2k13 opened 5 years ago
4
Why is kernel size 3 in SubPixelConvolution?

#361 hccho2 opened 5 years ago
0
_init_kernel of the SubPixelConvolution

#360 hccho2 opened 5 years ago
0
Run the synthesize.py script and report the error as follows

#359 darkmarbo opened 5 years ago
1
Running into OOM, is my GPU too bad ?

#358 L2ne opened 5 years ago
5
There is a error with the restore trained model

#357 unparalleled-ysj closed 5 years ago
2
Amount of data needed if training new speaker with tacotron_fine_tuning = True?

#356 wanshun123 opened 5 years ago
0
Small batch_size and outputs_perstep can't alignment?

#355 shahuzi opened 5 years ago
10
What dose the scale by 127.5 mean?

#354 zhangyi02 opened 5 years ago
0
Is pretrained model availible?

#353 mrgloom opened 5 years ago
4
Ask For Help!!!

#352 WhiteFu opened 5 years ago
0
How to Train Tacotron2 with CPU?

#351 chazo1994 opened 5 years ago
4
help! I get the following error in training.

#350 whfh3900 opened 5 years ago
0
Does Tacotron-2 support cmudict ?

#349 brucelin closed 5 years ago
0
How to fixed the eval output sentence?

#348 shahuzi closed 5 years ago
0
What makes a good TTS dataset?

#347 wanshun123 opened 5 years ago
0
Guided Attention Loss

#346 begeekmyfriend opened 5 years ago
14
Loss exploded on some datasets

#345 XiaooaiX closed 5 years ago
4
Updated audio samples or pretrained models

#344 wanshun123 opened 5 years ago
1
Failed to restore with saved model instead of checkpoint

#343 begeekmyfriend closed 5 years ago
3
Effective judgement of whether quick alignment or not for this project by the plot

#342 begeekmyfriend closed 5 years ago
9
Tacotron-2 mandrain-new branch demo wrong

#341 unikcc opened 5 years ago
34
Help a Noob train Tacotron 2

#340 rohan6366 closed 5 years ago
1
Bypass the plots when testing WaveNet

#339 LexXia opened 5 years ago
0
Location Sensitive Attention vs BahdanauMonotonic Attention

#338 hccho2 opened 5 years ago
6
ZeroDivisionError while training wavenet

#337 caixxiong opened 5 years ago
0
L1 loss function helps quick alignment

#336 begeekmyfriend closed 5 years ago
3
Multithread feeder to reduce data loading times

#335 alexdemartos closed 5 years ago
3
Training loss never achieves < 0.5 loss

#334 ghost opened 5 years ago
2
Tacotron Generates Unrelated Synthesis for Some Text Inputs

#333 fatihkiralioglu opened 5 years ago
0
Noisy output after training the latest code.

#332 vinayonchip opened 5 years ago
2

Previous Next