Batched resampling for the new implementation about alias-free-torch HOT 2 CLOSED

Thank you very much for your comment!

Since I tested for channel size 1, my current implementation did not cause error...

I'm so sorry for too many issuses... I will fix it in a day.

p.s. I'm looking forward to getting official BigVGAN implementation!

from alias-free-torch.

With your comment, all of resample.UpSample1/2d resample.DownSample1/2d and filter.LowPassFilter1/2d are fixed!

   #input [B,C,T]
    def forward(self, x):
        _, C, _ = x.shape
        if self.padding:
            x = F.pad(x, (self.pad_left, self.pad_right),
                      mode=self.padding_mode)
        out = F.conv1d(x, self.filter.expand(C, -1, -1),
                       stride=self.stride, groupds=C)
        return out

from alias-free-torch.