added audio

app.py

@@ -82,6 +82,7 @@ async def greet(product,description):
     description, # description
     "", # output - leave this blank for generation!
     )
+    #llama guard
     chat_completion = client.chat.completions.create(
         messages=[
             {
@@ -94,7 +95,8 @@ async def greet(product,description):
     warning_message = chat_completion.choices[0].message.content
     if warning_message != 'safe':
         chat = [chat_completion.choices[0].message.content]
-        yield chat[0]
+        processed_audio = combine_audio_files(text_to_speech([chat_completion.choices[0].message.content]))
+        yield processed_audio, chat[0]
     else:
         output = llm.create_chat_completion(
         messages=[
@@ -113,11 +115,11 @@ async def greet(product,description):
         for chunk in output:
             delta = chunk['choices'][0]['delta']
             if 'content' in delta:
-                #audio_list.append([text_to_speech(delta.get('content', ''))])
-                #processed_audio = combine_audio_files(audio_list)
+                audio_list.append([text_to_speech(delta.get('content', ''))])
+                processed_audio = combine_audio_files(audio_list)
                 partial_message = partial_message + delta.get('content', '')
-                yield partial_message
+                yield processed_audio, partial_message
 
 audio = gr.Audio()
-demo = gr.Interface(fn=greet, inputs=["text","text"], concurrency_limit=10, outputs=["text"])
+demo = gr.Interface(fn=greet, inputs=["text","text"], concurrency_limit=10, outputs=[audio, "text"])
 demo.launch()