p0p

[configs/vits2_vctk_standard.json](https://github.com/p0p4k/vits2_pytorch/blob/1f4f3790568180f8dec4419d5cad5d0877b034bb/configs/vits2_vctk_standard.json#L43) follow that file and change your json to add 'duration_discriminator_type'

if you are talking about this error - `torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 20.00 MiB. GPU 0 has a total capacty of 14.75 GiB of which 5.06...

The KL divergence that author uses is correct. If you want to know more about it, please tag me or ping me on discord (p0p4k). I can explain it for...

Regarding encodec, can it process batched input of wav files? Where do we mask the wav files?

Is there any stylespeechloss in sc-cnn?