narratorRL

Image to text to speech app with LLM and AI enhanced features.

---

Table of Contents

• Table of Contents
• About
• Installation/Usage
  • Chrome Extension
  • Backend Server

About

We are a group of 4 University of Waterloo Math and CS students. This project was made for the MetHacks 2023 hackathon. narratorRL targets the challenges issued by the following companies: Cohere, Domain.com, and Microsoft Cloud.

The backend of the server projects uses Python's Django framework to serve a RESTful API. The frontend is a React Native application with JavaScript.

Installation/Usage

Note: to install and use narratorRL, you must set up BOTH the React Native app and the Django server.

Django Server

1. Clone this repository and optionally create a virtual environment
2. Install the dependencies with pip install -r requirements.txt
3. Add a file named env.py in the server directory
   1. Generate a Django secret key (see command below) and set it to DJANGO_SECRET_KEY
   2. Add a Cohere secret key to COHERE_SECRET_KEY
4. Install Tesseract OCR (for all users by default, or you'll need to change the command path variable)
5. Add your device's IPv4 address to ALLOWED_HOSTS in server/settings.py
6. Start the server with python manage.py runserver 0.0.0.0:8000

from django.core.management.utils import get_random_secret_key
print(get_random_secret_key())

Backend Server

1. Enter the mobile directory
2. Install the dependices with npm install
3. Add your device's IPv4 address to the fetch command in mobile/Home.js
4. Start Expo with npx expo start
5. Connect however you wish