app.py

import streamlit as st
import os
import soundfile as sf
import torch
from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor, pipeline



device = "cuda:0" if torch.cuda.is_available() else "cpu"
torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32

model_id = "distil-whisper/distil-large-v2"

model = AutoModelForSpeechSeq2Seq.from_pretrained(
    model_id, torch_dtype=torch_dtype, low_cpu_mem_usage=True, use_safetensors=True
)
model.to(device)

processor = AutoProcessor.from_pretrained(model_id)

pipe = pipeline(
    "automatic-speech-recognition",
    model=model,
    tokenizer=processor.tokenizer,
    feature_extractor=processor.feature_extractor,
    max_new_tokens=128,
    chunk_length_s=15,
    batch_size=16,
    torch_dtype=torch_dtype,
    device=device,
)

def transcribe_audio(audio_file):
    # Save the audio file to a temporary file
    with open("temp_audio_file", "wb") as f:
        f.write(audio_file.getbuffer())
    
    # Transcribe the audio file using the Whisper model
    result = pipe("temp_audio_file")
    return result["text"]

# Streamlit app
def main():
    st.title('BETTER TRANSCRIBER')
    
    # Audio file uploader
    uploaded_file = st.file_uploader("Upload an audio file", type=["wav", "mp3", "m4a", "ogg", "flac"])

    if uploaded_file is not None:
        # Show a button to start the transcription process
        if st.button('Transcribe'):
            # Show a message while transcribing
            with st.spinner('Transcribing...'):
                text = transcribe_audio(uploaded_file)
            
            # Show the transcription
            st.subheader('Transcription:')
            st.write(text)
        else:
            st.write('Upload an audio file to get started.')

if __name__ == "__main__":
    main()