from transformers import pipeline
import gradio as gr
import time
import unicodedata
p = pipeline("automatic-speech-recognition",model="openai/whisper-tiny")

def transcribe(audio, state=""):
    time.sleep(10)
    text = p(audio)["text"]
    state += unicodedata.normalize("NFC",text) + " "
    return state, state

################### Gradio Web APP ################################

title = "Real-Time ASR"



gr.Interface(
    fn=transcribe, 
    inputs=[
        gr.Audio(source="microphone", type="filepath", streaming=True), 
        "state"
    ],
    outputs=[
        "textbox",
        "state"
    ],
    title=title,
    theme='EveryPizza/Cartoony-Gradio-Theme',
    live=True).launch()