handle audio with different shapes

app.py

@@ -29,8 +29,8 @@ def denoise(model_name, inputs):
     print(f"Audio shape: {audio.shape}")
     print(f"Sample rate: {sr}")
 
-    if audio.shape[1] > 1:
-        audio = audio.mean(1, keepdim=True)
+    if audio.shape[0] > 1:
+        audio = audio.mean(0, keepdim=True)
 
     print(f"Audio shape: {audio.shape}")