-
Have you tried changing the vocoder from Waveglow to HiFi-GAN? HiFi-GAN is faster and requires less VRAM. Alternatively, you could try adding a different vocoder.
-
As per https://pixl.cs.princeton.edu/pubs/Su_2021_HSS/Su-HiFi-GAN-2-WASPAA-2021.pdf
-
Have you got good audio?
-
Many options have,
LJ_V1
LJ_V2
LJ_V3
And from here: https://drive.google.com/drive/folders/1-eEYTB5Av9jNql0WGBlRoi-WH2J7bp5Y
Which one?
Thanks you
-
Hi, thanks for opensourcing your code! During the training process, I met an error with the command bellow.
**COMMAND**
python train.py -c config.yaml
**ERROR**
Initializing Training Process..…
-
The shape of mel output of Tacotron2 is bigger than mel extracted from audio and the model has issue
```
File "train.py", line 113, in train
for i, batch in enumerate(train_loader):
File "/…
-
如题,现在采样率有点低(16kHz),我看仓库里放了很多效果器,直接应用效果不是很理想
试了下使用[hifi-gan-bwe](https://github.com/brentspell/hifi-gan-bwe)进行超采样
目前使用了三个预训练模型进行超采样:
输入源是opencpop中2018.wav
推理直出
[svc_out.zip](https://github.com…
-
https://github.com/NVIDIA/BigVGAN?tab=readme-ov-file#news
-
I wonder whether tried to test the score for all the checkpoints you trained. Your model juest output the same waveshape you input .Your model can output the input waveshape even you don't train the …
-
We are training fastspeech2 on a single speaker. The output .wav files so far seem robotic. We are thinking that the hifi-gan vocoder, if trained on the same speaker, can give us better results. How c…