Update Gpt4oDemo.py

Gpt4oDemo.py

@@ -102,32 +102,44 @@ base64Frames = []
 # '''
 transcript=""
 
-def process_video(video_path, seconds_per_frame=2):
-    global base64Frames, audio_path
+def process_video(video_path, seconds_per_frame=2, target_width=320, target_height=180):
+    base64Frames = []
     base_video_path, _ = os.path.splitext(video_path)
 
     video = cv2.VideoCapture(video_path)
     total_frames = int(video.get(cv2.CAP_PROP_FRAME_COUNT))
     fps = video.get(cv2.CAP_PROP_FPS)
     frames_to_skip = int(fps * seconds_per_frame)
-    curr_frame=0
+    curr_frame = 0
 
+    # Retrieve and print the original width and height
+    original_width = int(video.get(cv2.CAP_PROP_FRAME_WIDTH))
+    original_height = int(video.get(cv2.CAP_PROP_FRAME_HEIGHT))
+    print(f"Original width: {original_width}, Original height: {original_height}")
+
+    # Loop through the video and extract frames at specified sampling rate
     while curr_frame < total_frames - 1:
         video.set(cv2.CAP_PROP_POS_FRAMES, curr_frame)
         success, frame = video.read()
         if not success:
             break
-        _, buffer = cv2.imencode(".jpg", frame)
+
+        # Resize the frame
+        resized_frame = cv2.resize(frame, (target_width, target_height))
+
+        _, buffer = cv2.imencode(".jpg", resized_frame)
         base64Frames.append(base64.b64encode(buffer).decode("utf-8"))
         curr_frame += frames_to_skip
+
     video.release()
 
-    audio_path = "./TEST.mp3"
+    # Extract audio from video
+    audio_path = f"./Task 1.mp3"
     clip = VideoFileClip(video_path)
     clip.audio.write_audiofile(audio_path, bitrate="32k")
     clip.audio.close()
     clip.close()
-    # transcribe_video(audio_path)
+    #transcribe_video(audio_path)
     print(f"Extracted {len(base64Frames)} frames")
     print(f"Extracted audio to {audio_path}")
     return base64Frames, audio_path
@@ -189,7 +201,7 @@ def handle_video(video=None):
         # Load example video
         video = "./TEST.mp4"
    
-    base64Frames, audio_path = process_video(video_path=video, seconds_per_frame=100)
+    base64Frames, audio_path = process_video(VIDEO_PATH, seconds_per_frame=1, target_width=320, target_height=180)
     chat_history.append({
         "role": "user",
         "content": [