Issues
- 2
ZeroDivisionError when extracting duration
#17 opened by Charlottecuc - 4
Speedyspeech loss
#1 opened by junkeon - 0
error in train
#60 opened by herokarimpoor - 1
if sample_rate is 16k, the reference hyperparamerter hop_size, win_size ..... how to set
#59 opened by jiahong3837 - 0
create_dataset
#58 opened by herokarimpoor - 0
dataset
#57 opened by herokarimpoor - 0
create_dataset
#56 opened by herokarimpoor - 10
Unclear inference result
#3 opened by junkeon - 0
question about the SSIM loss
#53 opened by binbinxue - 1
chain of exponentials
#52 opened by binbinxue - 3
Positional encoding
#46 opened by VigneshBaskar - 2
Residual shape in documentation
#45 opened by VigneshBaskar - 10
- 1
pip install -r requirements.txt ERROR
#37 opened by brentcty-2020 - 2
does speedyspeech support transfer learning?
#40 opened by bespsm - 0
optimizations
#42 opened by rave974 - 1
how much time to train it?
#41 opened by rave974 - 2
Why is there no dropout in the code?
#35 opened by wizardk - 7
- 3
- 6
A mistake in positional encoding
#23 opened by iclementine - 10
Extract durations from the trained model
#16 opened by adnan-mehremic - 5
Time for training teacher model is quite slow.
#33 opened by ductho9799 - 2
wav file quality - melgan vs griffinlim
#32 opened by dsplog - 4
- 5
- 3
Unstable audio generation towards the end of longer sentences using interference.py
#13 opened by eqikkwkp25-cyber - 7
Is it possible to use other vocoders?
#20 opened by nmichaud - 5
Issues for multi_speedyspeech?
#26 opened by TaoTaoFu - 2
- 2
Should the teacher model predict next frame?
#22 opened by iclementine - 0
Add demo inference server
#15 opened by janvainer2 - 2
better than Tacotron2?
#4 opened by joan126 - 6
the choice of positional encoding
#10 opened by MorganCZY - 1
running error
#12 opened by unwritten - 2
About SSIM loss
#14 opened by alexdemartos - 2
I have support for other language?
#9 opened - 4
How to traini MelGAN
#8 opened by Moon-sung-woo - 5
- 4
- 3
- 2