app.py

import torch
from transformers import pipeline
import streamlit as st
from st_audiorec import st_audiorec 
@st.cache_resource
def load_model():
    pipe=pipeline("automatic-speech-recognition","distil-whisper/distil-large-v2")
    return pipe

speech_to_text_model=load_model()

def make_text(audio):
    global speech_to_text_model
    text= speech_to_text_model(audio)
    extract_text=text['text']
    return extract_text

st.title('speech recognition')
wave_audio_data=st_audiorec()
if wave_audio_data is not None:

    
    text=make_text(wave_audio_data)

    st.write(text)
else:
    st.error('No Audio data')