Below, the part that needs a model: the WeSolve+ AI flashcard generator reads a PDF, a photograph or a camera shot and writes the cards from it, which is work no rule can do because it requires reading the page.
WeSolve+ reads the whole document and writes the questions for you
Upload your PDF, photograph your notebook, or point the camera. WeSolve+ writes questions from that material, explains why each answer is right, reads the chapter back to you as a podcast, and remembers every item you missed until you own it.
The tool below is a small browser-only tool and it is not WeSolve+: paste a few lines and text rules turn them into cards on the spot. The real app, the one that uses AI, is behind the link above.
Flashcard generator
This is a browser-only tool, and that is all it isIt splits the text you paste by rule, and nothing else. WeSolve+ is a different thing entirely: it reads your whole PDF with AI, writes the reasoning behind every question, speaks the chapter back to you, and remembers what you missed so it can return it. Try the real app now, free!
This tool uses text rules, not AI. It looks for lines written as "term: definition" or "term = definition" and makes a two sided item; for any other line it removes the longest word and asks you to supply it. Your text stays in the browser and is never sent anywhere. The copy button produces tab separated rows, the format Anki and Quizlet both accept.
The difference between converting text and generating cards
Converting is what happens above: your line already contains the term and the meaning, and the tool splits it. It is fast, private and free, and it is limited to what your text already says. Generating is a different job. Given a page of prose, a model has to decide which claims are worth a card, phrase a prompt that has one answer, and leave out the sentence that was only there for flow. That decision is the whole product. If a tool claims to generate cards from a PDF without reading it, it is doing the first job and calling it the second.
What the app generates from one file
Upload once and three things come out, all tied to the same source. Flashcards for the terms and relationships worth recalling. An explained quiz, multiple choice and true or false, with a short reason under each answer. A roughly one minute audio recap you can listen to while walking. None of them add outside information and none refer to a figure you cannot see. Your wrong answers collect in a separate list so the second pass starts where it matters. The AI flashcards page shows the card side, audio recap the listening side.
Sources it can read
PDF, image and camera. A lecture slide deck exported as PDF, a photograph of a textbook page, a shot taken straight from your camera. That is the honest list; there is no video import, no PowerPoint import and no audio recording import, and we would rather say so than let you find out after signing up. If your material is in one of the three, the generator has something to read. If it is a video lecture, you would need a transcript first, and that is a different tool.
Why generated cards still need spacing
Generating a deck removes the typing, not the reviewing. The two findings that matter here are old and well replicated. Roediger and Karpicke, 2006: testing yourself on material beat rereading it a week later. Cepeda and colleagues, 2006: across 254 studies, the same minutes spread across days beat the same minutes in one sitting. A generator that hands you 200 cards and no schedule has done half the job, which is why the app reshows cards at increasing intervals rather than dumping the deck on you.
Keeping the deck where you already work
If your review already lives in Anki or Quizlet, the copy button above gives tab separated rows that both accept, and the app can be the generator while your old tool stays the scheduler. If you would rather keep it all in one place, the app stores every card against the file it came from, so a card you get wrong can be traced back to the page that produced it. A card cut loose from the page it came from is a card you cannot settle an argument with. The trade offs are laid out on WeSolve+ and Anki and WeSolve+ and Quizlet.
Free plan and limits
Converting costs nothing and is not rationed, since no request ever leaves the tab. Generating is the part with a real bill behind it, and the free tier still covers a fair amount of it: two files a day, two quizzes a day, two recaps a week, two duels a day. Card count is left uncapped on purpose, because once a file has been read, producing thirty cards from it costs no more than producing three. Higher tiers raise those daily counts through Apple. Numbers on pricing.
Sources used on this page
- Roediger and Karpicke, Test-Enhanced Learning (2006)
- Cepeda et al., distributed practice meta analysis (2006)
- Dunlosky et al., improving students' learning (2013)
- Hermann Ebbinghaus, the forgetting curve (1885)
- Anki manual, importing text files
- Karpicke and Blunt, retrieval practice and concept mapping (2011)
- Bjork and Bjork, desirable difficulties
- WeSolve+ on the App Store
- MDN, the Clipboard API
- MDN, the details element
- Karpicke and Roediger, the critical importance of retrieval for learning (2008)
- Cepeda et al., spacing effects and the optimal retention interval (2008)
- Kornell and Bjork, is spacing the enemy of induction (2008)
- What Works Clearinghouse, organizing instruction and study
- The Learning Scientists, retrieval practice
- Retrieval Practice, the research library
- Cloze test
- Active recall
- Flashcard
- Spaced repetition
- Testing effect
- Leitner system
- SuperMemo, the SM-2 algorithm
- Anki manual, studying and scheduling
- Anki manual, deck options
- MDN, the textarea element
- MDN, the output element
- W3C, the disclosure pattern
- W3C, Web Content Accessibility Guidelines 2.2
| Criterion | This page | WeSolve+ app |
|---|---|---|
| Input | Text you paste | PDF, image or camera shot |
| Decides what is worth a card | No, it splits what you wrote | Yes, that is the generation step |
| Reads a scanned or photographed page | No | Yes |
| Also produces | Quiz prompts from the same text | Explained quiz and an audio recap |
| Review scheduling | None, you export the deck | Spaced repetition inside the app |
| Cost | Free, no limit, no account | Free to start, limits on the free plan |
Is the tool on this page an AI flashcard generator?
No, and it says so. It converts text you already wrote using text rules. The AI generator is in the app, where a model reads a PDF, image or camera shot and decides what deserves a card.
Which file types can the app read?
PDF, image and camera. There is no video, PowerPoint or audio recording import. For a video lecture you would need a transcript first.
How many cards can one upload produce?
As many as the material carries; card count is not rationed. The meter sits on files instead, two a day on the free tier, because reading the file is the expensive step and writing one more card from it is not.
Can I move generated cards into Anki?
The converter on this page exports tab separated rows that Anki imports directly. Cards generated inside the app are reviewed there, where the scheduling lives.
Does anything I paste here get uploaded?
It does not. The converter ships with the page and has nowhere to send anything. Watch the network panel in your developer tools as you type and you will see it stay empty.
Are generated cards as good as ones I write myself?
Writing a card is itself a form of study, so writing your own is not wasted. What the research is clear about is that the retrieval matters more than the authorship: what you do with the card outweighs who typed it.
Last updated: 2026-08-15