code'

app.py

@@ -65,8 +65,8 @@ frame_rate = model.audio_encoder.config.frame_rate
 
 
 def generate_response(audio):
-    yield None, gr.Textbox(visible=True)
-    question = client.automatic_speech_recognition(audio)
+    gr.Info("Transcribing Audio", duration=5)
+    question = client.automatic_speech_recognition(audio).text
 
     messages = [{"role": "sytem", "content": ("You are a magic 8 ball."
                                               "Someone will present to you a situation or question and your job "
@@ -76,7 +76,7 @@ def generate_response(audio):
     
     response = client.chat_completion(messages, max_tokens=128, seed=random.randint(1, 5000))
     response = response.choices[0].message.content
-    yield response, gr.Textbox(visible=True)
+    return response, gr.Textbox(visible=True)
 
 
 @spaces.GPU
@@ -121,7 +121,7 @@ with gr.Blocks() as block:
     )
     with gr.Group():
         with gr.Row():
-            audio_out = gr.Audio(visible=False, streaming=True)
+            audio_out = gr.Audio(visible=False, streaming=True, autoplay=True)
             answer = gr.Textbox(visible=False, label="Answer")
             state = gr.State()