How to turn a PDF or lecture into flashcards

Most of your study material is already sitting in front of you: the lecture slides as a PDF, the chapter you scanned, the article in a browser tab, the recording of a lecture you half-listened to. The problem was never a lack of material. It was the hour it takes to turn material into cards — the hour that makes most people give up on flashcards before they ever benefit from them. Here is how to close that gap, whatever form your material takes, and how to tell a card worth keeping from one worth deleting.

Your material already contains the cards

Before any tool, a mindset: you are not transcribing the document, you are extracting the questions it answers. Skim a page and ask what you would need to produce later — a definition, a mechanism, a cause-and-effect, a contrast between two things easily confused. Those become cards. The connective prose, the examples that merely illustrate, the sentences you would nod at but never be asked about: those do not. A dense page might yield five good cards. Extracting five is the skill; extracting fifty is the mistake.

From a PDF or a scan

Slides and textbook chapters are the richest source, and the most tedious to process by hand. Import the PDF — or a scan of a printed page or handwritten notes — and let the app draft cards from it, then do the part only you can do: read each draft and ask whether it captures a concept or just a keyword. Rewrite the ones that are too shallow into the question you would actually face. A draft that says “Enzyme X — location” is a keyword; “Where does enzyme X act, and what happens downstream when it fails?” is a concept. The draft saves you the typing; the editing is what makes it yours.

From a link or pasted text

An article, a documentation page, a paragraph from a forum that finally explained something: paste the text or drop in the link and turn it into cards while the understanding is fresh. This is the fastest way to capture the learning that happens outside class, and it is where most people lose knowledge they genuinely had for an afternoon.

From a recorded or dictated lecture

Some material never exists as text — a lecturer’s explanation, a mentor’s answer, your own thoughts on the walk home. Dictate it. Speak the key points into the app and let them become cards. The point is to capture the concept in the moment you understood it, in your own words, before it evaporates.

What makes a card worth keeping

Whatever produced the draft, apply the same test to every card. One idea per card. Phrased as a question you must answer, not a heading you recognise. Answer side short enough to check at a glance. If you would not be able to say what the “correct” answer is, the card is ambiguous — fix it or delete it. Add tags so you can find related cards later, and reverse a card when both directions matter. Deleting freely is part of the craft: a lean deck of cards you actually answer beats a bloated one you dread. This is active recall in practice, and the testing effect is why the question form matters so much — we go deeper in active recall vs. re-reading.

Then let the scheduler take over

Once cards exist, stop managing them. A scheduler such as FSRS decides when each card is due and shows it right before you would forget it; your job is to review what is due and rate honestly. Watch the forecast as you import: a big batch of new cards today means a heavier week ahead, so pace new material to the time you actually have. Our FSRS guide explains how the timing works.

A note on credits

Turning material into cards with AI costs something to run, so it is fair that an app charges for it — and fair that it does so transparently. In Repeto, AI generation uses credits measured in cards: around 100 are free to start, credit packs are a one-time purchase, and credits never expire. Everything else — learning, the FSRS scheduler, writing cards by hand and sync — is free without limits. You can see the exact model in the pricing section.

How Repeto fits

Repeto accepts all of the sources above: import from PDF, scan, web link, pasted text or dictation, or draft cards from a topic with AI, then refine every card in the editor with images, tags and reverse cards. The FSRS scheduler takes over from there, with a daily goal, streak, forecast and activity calendar to keep the pace visible, and loss-proof sync between iPhone and iPad so the cards you made on one device are waiting on the other. For the full checklist of what an app should offer, see what to look for in a spaced-repetition app.

Open the PDF you have been avoiding. Extract five good cards. That is the whole method.