Update app.py

app.py

@@ -1,6 +1,6 @@
 import gradio as gr
 #import requests
-from moviepy.editor import VideoFileClip
+
 from PIL import Image
 import os
 token = os.environ.get('HF_TOKEN')
@@ -8,21 +8,7 @@ whisper_to_gpt = gr.Blocks.load(name="spaces/fffiloni/whisper-to-chatGPT")
 tts = gr.Interface.load(name="spaces/Flux9665/IMS-Toucan")
 talking_face = gr.Blocks.load(name="spaces/fffiloni/one-shot-talking-face", api_key=token)
 
-def apply_delay(video_path):
-    
-    # Open the video file
-    video = VideoFileClip(video_path)
-    
-    # Delay the audio by 1 ms
-    delayed_audio = video.audio.fx(afx.delay, 0.5)
-    
-    # Overwrite the original audio with the delayed audio
-    video = video.set_audio(delayed_audio)
-    
-    # Save the modified video
-    video.write_videofile("delayed_video.mp4")
 
-    return "delayed_video.mp4"
 
 def infer(audio):
     gpt_response = whisper_to_gpt(audio, "translate", fn_index=0)
@@ -31,10 +17,10 @@ def infer(audio):
     #image = Image.open(r"wise_woman_portrait.png")
     portrait_link = talking_face("wise_woman_portrait.png", audio_response, fn_index=0)
     
-    final_vid = apply_delay(portrait_link)
+    
     #portrait_response = requests.get(portrait_link, headers={'Authorization': 'Bearer ' + token})
     #print(portrait_response.text)
-    return final_vid
+    return portrait_link
 
 title = """
     <div style="text-align: center; max-width: 500px; margin: 0 auto;">
@@ -95,7 +81,7 @@ with gr.Blocks(css=css) as demo:
         gr.HTML(title)
         
         with gr.Row():
-            record_input = gr.Audio(source="microphone",type="filepath", show_label=False,elem_id="record_btn")
+            record_input = gr.Audio(source="microphone",type="filepath", show_label=True,elem_id="record_btn")
             
         with gr.Row():