Update app.py

app.py

@@ -4,5 +4,5 @@ text_to_speech = pipeline('image-classification')
 input_img = 'abc.jpeg'
 def text_to_speech(input_img):
   return text_to_speech(input_img).wav
-interface = gr.Interface(text_to_speech, gr.Image(shape=(200, 200)), "audio", theme='dark')
+interface = gr.Interface(text_to_speech, gr.Image(), "audio", theme='dark')
 interface.launch()