adjust output audio and narrator output

app.py

@@ -13,15 +13,19 @@ def describe_image(file_path):
     img_text_pip_output = img_text_pipe(file_path)
 
     description_text = img_text_pip_output[0]['generated_text']
+    print(description_text)
 
     narrated_text = narrator(description_text)
+
+    (narrated_text["sampling_rate"], narrated_text["audio"][0] )
         
-    return narrated_text["audio"][0]
+    return (narrated_text["sampling_rate"], narrated_text["audio"][0])
+
 
 
 iface = gr.Interface(fn=describe_image, 
                      inputs=gr.Image(label="Input image", 
                   type="pil"),
-                   outputs="audio"
+                   outputs=gr.Audio(label="Narration", type="numpy",  autoplay=True)
                    )
 iface.launch()