| import streamlit as st | |
| from transformers import pipeline | |
| st.markdown("# Podcast Q&A") | |
| st.markdown( | |
| """ | |
| This helps understand information-dense podcast episodes by doing the following: | |
| - Speech to Text transcription - using OpenSource Whisper Model | |
| - Summarizes the episode | |
| - Allows you to ask questions and returns direct quotes from the episode. | |
| """ | |
| ) | |
| audio_file = st.file_uploader("Upload audio copy of file", key="upload", type=['.mp3']) | |
| if audio_file: | |
| transcribe_audio(audio_file) | |