app.py

# Transform an audio to text script with language detection.
# Author: Pratiksha Patel

# Description: This script record the audio, transform it to text, detect the language of the file and save it to a txt file.
# import required modules
import os
import streamlit as st
from audio_recorder_streamlit import audio_recorder
import whisper
from langdetect import detect

# Function to open a file
def startfile(fn):
    os.system('open %s' % fn)
    
# Function to create and open a txt file
def create_and_open_txt(text, filename):
    # Create and write the text to a txt file
    with open(filename, "w") as file:
        file.write(text)
    startfile(filename)

# Ask user to record audio
st.title("Audio to Text Transcription..")
audio_bytes = audio_recorder(pause_threshold=3.0, sample_rate=16_000)

# Load the base model and transcribe the audio
model = whisper.load_model("base")
result = model.transcribe(audio_bytes)
transcribed_text = result["text"]
print(transcribed_text)
st.write("Transcription:")
st.write(transcribed_text)

# Detect the language
language = detect(transcribed_text)
st.write(f"Detected language: {language}")

# Create and open a txt file with the text
create_and_open_txt(transcribed_text, f"output_{language}.txt")