Update app.py

app.py

@@ -77,7 +77,6 @@ def transcribe_and_chat(audio):
     response, audio_path = asyncio.run(chat_with_ai(text))
     return response, audio_path
 
-# Define the Gradio interface
 def create_demo():
     with gr.Blocks() as demo:
         gr.Markdown("# AI Voice Assistant")
@@ -90,6 +89,10 @@ def create_demo():
                 chat_output = gr.Textbox(label="AI Response")
                 audio_output = gr.Audio(label="AI Voice Response", autoplay=True)
 
+        def process_audio(audio):
+            response, audio_path = transcribe_and_chat(audio)
+            return response, audio_path, None  # Return None to clear the audio input
+
         demo.load(None, js="""
             function() {
                 document.querySelector("audio").addEventListener("stop", function() {
@@ -98,30 +101,27 @@ def create_demo():
                     }, 500);
                 });
                 
-                // Function to play the assistant's audio response
                 function playAssistantAudio() {
                     var audioElements = document.querySelectorAll('audio');
                     if (audioElements.length > 1) {
-                        var assistantAudio = audioElements[1];  // The second audio element is the assistant's response
+                        var assistantAudio = audioElements[1];
                         if (assistantAudio) {
                             assistantAudio.play();
                         }
                     }
                 }
 
-                // Play assistant's audio when it's loaded
                 document.addEventListener('gradioAudioLoaded', function(event) {
                     playAssistantAudio();
                 });
 
-                // Also try to play the audio when the interface updates
                 document.addEventListener('gradioUpdated', function(event) {
-                    setTimeout(playAssistantAudio, 100);  // Small delay to ensure audio is ready
+                    setTimeout(playAssistantAudio, 100);
                 });
             }
         """)
 
-        audio_input.change(transcribe_and_chat, inputs=audio_input, outputs=[chat_output, audio_output])
+        audio_input.change(process_audio, inputs=[audio_input], outputs=[chat_output, audio_output, audio_input])
 
     return demo