avocodo
avocodo copied to clipboard
ncsoft
Official implementation of "Avocodo: Generative Adversarial Network for Artifact-Free Vocoder" (AAAI2023)
As I saw in HiFiGAN, after training using GT mels, they further used teacher forcing mels from TTS inference to fine tune the model, and got better result. Is this...
Nice work! The example results sound promising. It would be better if you could provide some pretrained models.
Added checkpoint argument to load while training for finetuning purposes, or resumption from a training state.
Hello, Thank you for presenting awesome ideas with your work and addressing fundamental issues in previous works. In the **Training Setup** section of your paper the learning rate is mentioned...
Thank you for great research! I just remove some repeated ops in implementation to reduce training time. - Remove repeated PQMF filtering in `Avocodo` - Define `pqmf_list` as a `nn.ModuleList`...
Hi, Thanks for sharing Avocodo, I'd like to use this vocoder on higher sampling rate, 32Khz. Can you give me some suggestions on how to change the PQMF part when...
