Update app.py

app.py

@@ -72,6 +72,23 @@ def transcribe(audio: tuple[int, np.ndarray], transformers_convo: list[dict], gr
 
 
 with gr.Blocks() as demo:
+    gr.HTML(
+    """
+    <h1 style='text-align: center'>
+    Talk to Qwen2Audio (Powered by WebRTC ⚡️)
+    </h1>
+    <p style='text-align: center'>
+    Once you grant access to your microphone, you can talk naturally to Qwen2Audio.
+    When you stop talking, the audio will be sent for processing.
+    </p>
+    <p style='text-align: center'>
+    There will be some delay in responding due to acquiring the ZeroGPU resources.
+    </p>
+    <p style='text-align: center'>
+    Each conversation is limited to 90 seconds. Once the time limit is up you can rejoin the conversation.
+    </p>
+    """
+    )
     transformers_convo = gr.State(value=[])
     with gr.Row():
         with gr.Column():
@@ -84,8 +101,8 @@ with gr.Blocks() as demo:
         with gr.Column():
             transcript = gr.Chatbot(label="transcript", type="messages")
 
-    audio.stream(ReplyOnPause(transcribe), inputs=[audio, transformers_convo, transcript], outputs=[audio])
-    audio.on_additional_outputs(lambda s,a: (s,a), outputs=[transformers_convo, transcript])
+    audio.stream(ReplyOnPause(yield_audio), inputs=[audio], outputs=[audio])
+    audio.on_additional_outputs(respond, outputs=[transformers_convo, transcript])
 
 if __name__ == "__main__":
     demo.launch()