fix: manually move hann window to device (#87)

Author: JacobLinCool

audio_separator/separator/uvr_lib_v5/bs_roformer.py

@@ -475,7 +475,7 @@ def forward(
 
         raw_audio, batch_audio_channel_packed_shape = pack_one(raw_audio, '* t')
 
-        stft_window = self.stft_window_fn(device=device)
+        stft_window = self.stft_window_fn().to(device)
 
         stft_repr = torch.stft(raw_audio, **self.stft_kwargs, window=stft_window, return_complex=True)
         stft_repr = torch.view_as_real(stft_repr)

audio_separator/separator/uvr_lib_v5/mel_band_roformer.py

@@ -407,7 +407,7 @@ def forward(
 
         raw_audio, batch_audio_channel_packed_shape = pack_one(raw_audio, '* t')
 
-        stft_window = self.stft_window_fn(device=device)
+        stft_window = self.stft_window_fn().to(device)
 
         stft_repr = torch.stft(raw_audio, **self.stft_kwargs, window=stft_window, return_complex=True)
         stft_repr = torch.view_as_real(stft_repr)