perceptual similarity implementation

[Raha304]

Hi, dear Li-Wei

I am working on dysarthric voice conversion and I want to use your idea in the " Generative Adversarial Networks for Unpaired Voice Transformation on Impaired Speech " paper. I saw your repo for this paper but I don't understand which part of the code is related to perceptual similarity implementation. I am so thankful if you can help me with this problem. Pleeeeeeeeeeease

[b04901014]

Hi,

Sorry for the late reply. It's been years since I visit this codebase. If you are referring to the Eq. 4 in the paper, I think it's at https://github.com/b04901014/ISGAN/blob/master/src/model.py#L176 which simply calculates the L1 distance of the discriminator layers between speech segments.