Đỗ Trí Nhân

Just for ref, I was successful in converting when using pytorch in CPU, I wonder if should we check we are running with cpu or gpu

> Is there any confirmation on the quality of the Transformer encoder or Conformer, I found that the conformer in DelightfulTTS is quite different from ASR a little bit.

> Thanks for your help. I'll try it! I reach the same. Is there any problem when I keep training with that problem: "No module named 'lightconv_cuda'" ? If you...

Should we evaluate the audios with MCD score?

> @tekinek Have you solved this, is the preprocessing step affect the training process which use difference sample rate:  And the weight which is generated from TasNet work in...

> I install all the dependencies with ' pip install -r requirements.txt', written in README.md. > But when I try to run the training script 'python train.py --gin iam/iam.gin', >...

As I remember I have solved this problem by install apex from the source and try to use the correct version set of cuda, cudnn, pytorch. Check your env, the...

For FID, I find an implement from torchmetric: https://torchmetrics.readthedocs.io/en/stable/image/frechet_inception_distance.html But it required input much be in dtype=torch.unit8, while spectrogram I generate is in float, if I try to convert to...

> I want to confirm that the issues I'm experiencing are a fundamental issue with the loss and not my implementation (which is a slight modification of this). > >...

> > Alternatively, since only the minor version differs, you could also try to disable the minor version check (you should get an error with a link to more information...