Update app.py

app.py

@@ -11,11 +11,11 @@ def infer(audio):
     #print(gpt_response)
     audio_response = tts(gpt_response[1], "English Text", "English Accent", "English Speaker's Voice", fn_index=0)
     #image = Image.open(r"wise_woman_portrait.png")
-    #portrait = talking_face("wise_woman_portrait.png", audio_response, fn_index=0)
-    return audio_response
+    portrait = talking_face("wise_woman_portrait.png", audio_response, fn_index=0)
+    return audio_response, portrait
 
 inputs = gr.Audio(source="microphone",type="filepath")
-outputs = [gr.Audio()]
+outputs = [gr.Audio(), gr.Video()]
 
 demo = gr.Interface(fn=infer, inputs=inputs, outputs=outputs)
 demo.launch()