A demo of the project can be visited at:
https://podcast-book-recommendation-website-new.vercel.app/
This repo is a mono repo, organized by "projects":
./nlp
: Jupyter Notebook scripts to train the model + spaCy training and test set./book-recommendation-extractor-api
: FastAPI Deployment of SpaCy Model. You can call it to get Book Recommendations (with all the Amazon information)./podcast-transcript-scraper
: Selenium scraper that connects to Airtable to receive new added Podcasts and goes to Happyscribe to scrape Podcast transcript- Uploads them to S3
- Runs dockerized on DigitalOcean
./website
: Website to show the Book Recommendations- Based on Next.js + React + TailwindCSS + CosmicJS boilerplate
./cosmic-js-uploader
: Scripts to upload Book Recommendations to Content Management System- Also includes all the transcript files + metadata
All projects have their own README file that explains how to install and run the different projects.