updated

app.py

@@ -1,4 +1,5 @@
 import gradio as gr
+import time
 from transformers import pipeline, AutoModelForCTC, Wav2Vec2Processor, Wav2Vec2ProcessorWithLM
 
 
@@ -12,9 +13,9 @@ p=pipeline("automatic-speech-recognition", model="kresnik/wav2vec2-large-xlsr-ko
 def transcribe(audio, state=""):
     time.sleep(2)
     text = p(audio)["text"]
-    #state+= text+ " "
-    #return state, state
-    return text
+    state+= text+ " "
+    return state, state
+    #return text
 
 gr.Interface(
     fn=transcribe,
@@ -37,6 +38,6 @@ gr.Interface(
     allow_screenshot=False,
     allow_flagging="never",
     theme="grass",
-    live=False
+    live=True
 ).launch()