Update app.py

app.py

@@ -40,7 +40,7 @@ def generate(prompt,history):
     for response in stream:
         output += response.token.text
         yield [(prompt,output)], ""
-    return [(prompt,output)], output
+    yield [(prompt,output)], output
 
 def load_mod(model):
     yield f"Loading: {model}"
@@ -68,7 +68,7 @@ with gr.Blocks() as iface:
     iface.load(load_mod,names,msg)
     sub_b = submit_b.click(generate, [prompt,chatbot],[chatbot,stt])
     sub_e = prompt.submit(generate, [prompt, chatbot], [chatbot,stt])
-    aud_out = stt.change(pp.stream_tts,[stt,names,length,noise,width,sen_pause],aud)
+    stt.change(pp.stream_tts,[stt,names,length,noise,width,sen_pause],aud)
     names_change=names.change(load_mod,names,msg)
-    stop_b.click(None,None,None, cancels=[sub_b,sub_e,aud_out,names_change])
+    stop_b.click(None,None,None, cancels=[sub_b,sub_e,names_change])
 iface.queue(default_concurrency_limit=10).launch()