Update app.py

app.py

@@ -15,8 +15,8 @@ def infer(image_input, manual_caption, duration_in):
         print("gpt2 caption: " + cap)
     else:
         cap = manual_caption
-        print("manual caption: " + cap)
-    sound = audio_gen(cap, 10, 2.5, 45, 3, fn_index=0)
+        print("manual captiony: " + cap)
+    sound = audio_gen(cap, duration_in, 2.5, 45, 3, fn_index=0)
     
     return cap, sound[1], gr.Group.update(visible=True)