Upload to debug

app.py

@@ -163,6 +163,9 @@ def extract_resampled_audio(video_path, target_sr=16000):
     return resampled_audio_path
 
 @spaces.GPU
+def yolo_detection(frame, verbose=False):
+    return model(frame, verbose=verbose)[0]
+
 def extract_faces(video_file):
     cap = cv2.VideoCapture(video_file)
     fps = cap.get(cv2.CAP_PROP_FPS)
@@ -174,7 +177,8 @@ def extract_faces(video_file):
             break
 
         # Inference
-        results = model(frame, verbose=False)[0]
+        #results = model(frame, verbose=False)[0]
+        results = yolo_detection(frame, verbose=False)
         for box in results.boxes:
             # version 1
             # x1, y1, x2, y2 = map(int, box.xyxy[0])