Update app.py

app.py

@@ -143,17 +143,18 @@ def infer(video_in):
     print(extracted_audio)
 
     # Get description of audio content
-    audio_content_described = get_salmonn(extracted_audio)
+    #audio_content_described = get_salmonn(extracted_audio)
 
     # Assemble captions
-    formatted_captions = f"""
+    '''formatted_captions = f"""
 ### Visual events:\n{string_list}\n ### Audio events:\n{audio_content_described}
 """
     print(formatted_captions)
 
     # Send formatted captions to LLM
     #video_description_from_llm = llm_process(formatted_captions)
-    return formatted_captions
+    '''
+    return extracted_audio
 
 with gr.Blocks() as demo :
     with gr.Column(elem_id="col-container"):
@@ -162,7 +163,7 @@ with gr.Blocks() as demo :
         """)
         video_in = gr.Video(label="Video input")
         submit_btn = gr.Button("Submit")
-        video_description = gr.Textbox(label="Video description")
+        video_description = gr.Audio(label="Video description")
     submit_btn.click(
         fn = infer,
         inputs = [video_in],