music-spectrogram-diffusion-pytorch
music-spectrogram-diffusion-pytorch copied to clipboard
@Ningzhi-Wang pointed out that maybe using different initialization algorithms can affect the autoregressive baseline.
It seems trill embedding on its own will not give norm value greater than 2. Not sure how the original paper was able to get Trill distance comparable to VGGish...
@yoyolicoris Hey bro! First of all, I wanted to thank you for this repo. It was very educational to me and I the pre-trained models are fantastic :) The reason...