| import torch | |
| import streamlit as st | |
| import numpy as np | |
| from PIL import Image, ImageDraw | |
| from transformers import pipeline | |
| from tempfile import NamedTemporaryFile | |
| audiopipe = pipeline("automatic-speech-recognition", model="openai/whisper-large-v3") | |
| st.title('Upload an audio file for speech recognition') | |
| uploaded_audio_file = st.file_uploader("Choose an audio file (wav)") | |
| if uploaded_audio_file is not None: | |
| with NamedTemporaryFile(suffix="wav") as temp: | |
| temp.write(uploaded_audio_file.getvalue()) | |
| temp.seek(0) | |
| result = audiopipe(temp.name) | |
| st.write(result) | |