Transcriptor / app.py
taham655's picture
Update app.py
0ae7886
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()