Take any book or article in PDF form and convert it to a deck of flashcards!

• 👀 Recognizes text, equations, images, and (some) handwriting.
• 🧠 Understands the content and context of the PDF.
• ✍🏻 Summarizes the PDF as questions and answers systematically.
• 📕 Reformats everything into .txt, ready to be used in quiz programs like Anki!

ChatGPT is used for understanding the content and MathPix to extract equations and images.

❓ How to use...

1. After installing these scripts as well as the depencies listed below and getting the required API keys, add these in the api_keys.py file.
2. Put the PDF you want to extract flashcards from into this directory, open the main.py file and change the PDF_TITLE = "this_is_your_pdf.pdf" variable to the name of the PDF.
3. Run the script!

📂 Dependencies

You will need to install the following two libraries openai, pypdf, and get API keys for ChatGPT and MathPix. Beware that using both will cost money. Even large textbooks will only cost a few cent though and the decks will be done in minutes not weeks!

---

This is the hacky bare bone version of the software I had written last year trying to add a bunch more features and creating a web app. But I am not a web developer and so eventually lost interest when OpenAI kept releasing more and more cool stuff that made this seem not all that special. I learned a lot about web dev though and if this little thing helped and you like it, come to tell me on X (formerly Twitter) @quentinwach 👋🏻.