Update app.py

app.py

@@ -28,8 +28,10 @@ def decipher(audio):
     )
 
     system_message = response["choices"][0]["message"]["content"]
-    audio_response = gTTS(text=system_message, lang='en', slow=False)
-    audio_response.save("response.mp3")
+    myobj = gTTS(text=system_message, lang='en', slow=False)
+    myobj.save("welcome.mp3")
+    # Playing the converted file
+    os.system("xdg-open welcome.mp3")
 
     messages.append({"role": "assistant", "content": system_message})
 
@@ -38,11 +40,9 @@ def decipher(audio):
         if message['role'] != 'system':
             chat_transcript += message['role'] + ": " + message['content'] + "\n\n"
 
-    return chat_transcript, gr.outputs.Audio("response.mp3", type="audio")
+    return chat_transcript
 
 # Using Gradio's audio Interface 
-interface = gr.Interface(fn=decipher, inputs=gr.Audio(source="microphone", type="file"), 
-                         outputs=["text", gr.outputs.Audio("response.mp3", type="filepath")])
-
-
+interface = gr.Interface(fn=decipher, inputs=gr.Audio(
+    source="microphone", type="filepath"), outputs="text")
 interface.launch()