handle multi-channel audio

app.py

@@ -47,12 +47,12 @@ def denoise(model_name, inputs):
             clean_chunk = model(audio_chunk).logits
         clean.append(clean_chunk.squeeze(0))
 
-    denoised = torch.concat(clean).flatten()[: audio.shape[-1]].clamp(-1.0, 1.0)
+    denoised = torch.concat(clean, 1)[:, : audio.shape[-1]].clamp(-1.0, 1.0)
     denoised = (denoised * 32767.0).numpy().astype(np.int16)
 
     print(f"Denoised shape: {denoised.shape}")
 
-    return model.config.sample_rate, denoised.unsqueeze(0)
+    return model.config.sample_rate, denoised
 
 
 iface = gr.Interface(