Update app.py

app.py

@@ -26,6 +26,16 @@ Please note that the following list of image descriptions (visual details) was o
 Audio events are actually the entire scene description based only on the audio of the video. Your job is to integrate these multimodal inputs intelligently and provide a very short resume about what is happening in the origin video. Provide a succinct overview of what you understood.   
 """
 
+def trim_video(input_path, output_path, max_duration=10):
+    video_clip = VideoFileClip(input_path)
+    
+    if video_clip.duration > max_duration:
+        trimmed_clip = video_clip.subclip(0, max_duration)
+        trimmed_clip.write_videofile(output_path, audio_codec='aac')
+        return output_path
+    else:
+        return input_path
+
 def extract_frames(video_in, interval=24, output_format='.jpg'):
     """Extract frames from a video at a specified interval and store them in a list.
 
@@ -190,11 +200,17 @@ with gr.Blocks(css=css) as demo :
         <h2 style="text-align: center;">Soft video understanding</h2>
         """)
         video_in = gr.Video(label="Video input")
+        video_cut = gr.Video(label="Video cut")
         submit_btn = gr.Button("Submit")
         video_description = gr.Textbox(label="Video description", elem_id="video-text")
+    video_in.upload(
+        fn = trim_video,
+        inputs = [video_in],
+        outputs = [video_cut]
+    )
     submit_btn.click(
         fn = infer,
-        inputs = [video_in],
+        inputs = [video_cut],
         outputs = [video_description]
     )
 demo.queue().launch()