Discussion is a tool to help users transcribe and determine key topics being discussed in conversations. It uses google's speech to text API to convert a audio stream or file into text. We can use that text to find structure and determine structure using LDA (one of the types of topic modeling).

For this app, you will need both node, python and the google-cloud-speech authentication credentials. Use the requirements.txt (in the server folder) and package.json (in the static folder) to determine dependencies for the server and the front-end.

To Start the Frontend Service,

  cd static
  npm install #only needed for the first run
  npm start

To Start the Backend Service,

  cd server
  pip install -r requirements.txt
  python app.py

Follow this guide by google: to create your google api credentials.