correct audio_output

app.py

@@ -90,6 +90,9 @@ import torch
 ds = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation")
 sample = ds[0]["audio"]
 
+global Audio_output
+Audio_output = []
+
 def speech_to_text_loc(audio):
     device = "cpu"
     pipe = pipeline(
@@ -334,8 +337,7 @@ file_list_loaded = ''
 chunk_size = 500
 chunk_overlap = 300
 
-global Audio_output
-Audio_output = []
+
 
 global Filename_Chatbot
 Filename_Chatbot = ""
@@ -1436,7 +1438,7 @@ with gr.Blocks() as demo:
 
     with gr.Column() as main2:
         title = gr.Markdown("""# <center> STLA BABY - YOUR FRIENDLY GUIDE
-                            <center> v0.6: Powered by MECH Core Team - GPT4 REMOTE MODE"""),
+                            <center> v0.6.1: Powered by MECH Core Team - GPT4 REMOTE MODE"""),
         chatbot = gr.Chatbot()
         with gr.Row():
             inputtext = gr.Textbox(