• python3, < 3.9 (numpy currently broken on 3.9)
• node
• ffmpeg
• google api set up for speech recognition
  • ex: ENV variable GOOGLE_APPLICATION_CREDENTIALS=/path/to/credentials.json
• access to an Elasticsearch cluster
  • localhost or AWS

git clone this repo then inside it:

pip install -r requirements.txt
cd frontend
npm install
npm run build
cd ..
streamlit run app.py