Skip to content
Go back

Booksum - Scan Book Pages, Extract Text with OCR, and Actually Remember What You Read

Snap a page, extract the text, get a summary, make flashcards. One app.

Get it on Google Play →

I read a lot of non-fiction. The problem isn’t reading, it’s retaining. You highlight a passage, close the book, and two weeks later you couldn’t paraphrase it if someone paid you. Research backs this up: the Ebbinghaus forgetting curve shows we lose roughly 70% of new information within 24 hours. Brutal.

I built Booksum to fix my own workflow. Photograph a page, extract the text automatically, get an AI summary if the passage is dense, and generate flashcards that use spaced repetition to actually burn it into memory. Everything stays in a searchable library organized by book. No more shoeboxes of sticky notes, no more “wait, which book was that quote from?”

Book Library with ISBN Barcode Scanning

The home screen is your library. Every book you’ve captured from shows up with cover art, author, page count, and how many snippets you’ve saved. Point your camera at a book’s ISBN barcode and Booksum auto-fills the metadata from Google Books. No typing, no hunting for the exact edition.

Booksum book library showing cover art, authors, page counts, and snippet counts per book

Mark books as Read, In Progress, or To Read. Search across your entire library by title, author, or keyword. Sort and filter to find what you need.

Snippets: Your Captured Passages

Tap into any book and you see all your captured snippets: photographed pages with extracted text, page numbers, and timestamps. Each snippet is a passage you thought was worth saving.

Book detail view showing captured snippets with page photos, OCR text, and AI summarize buttons

The camera is optimized for book pages. It supports volume-button shutter so you can hold a book open with one hand and snap with the other. (Anyone who’s tried to photograph a page while wrestling a hardcover knows this matters.) Crop and adjust after capture.

On-Device OCR Text Extraction

This is the core feature. Photograph any page and Booksum extracts the text using ML Kit OCR, entirely on-device. No cloud processing, no data leaving your phone. The extracted text becomes searchable, copyable, and shareable.

OCR results showing extracted text from a photographed book page about Dependency Injection

The OCR handles most printed text well. It’s not magic on every font or lighting condition, but for standard book pages it’s reliable enough that I stopped typing quotes manually. Which, if you’ve ever tried to transcribe a paragraph from a novel into your phone, you’ll understand is a small miracle.

Annotation and Markup Tools

Sometimes you want to circle a concept, underline a key phrase, or scribble a note in the margin. Booksum has a full annotation editor with drawing brushes, text overlays, emoji stickers, and a color picker. Undo/redo history so you can experiment freely.

Annotation editor with drawing tools, color picker, and markup on a book page photo

Annotations persist on the snippet image. Useful for visual learners who think spatially about where information lives on a page. Also useful for leaving your future self a “this is important!” arrow that you’ll actually notice.

AI-Powered Summaries

For dense passages like academic textbooks, technical writing, anything that takes three reads to parse, hit the AI Summarize button. Booksum sends the OCR text to an AI model and returns a concise summary with key takeaways.

AI summary of a book passage, condensing the key concepts into a clear overview

This also works for articles. Paste any URL into the Article Digest feature and Booksum extracts the content, summarizes it, and stores it alongside your book snippets. Your “read it later” pile finally becomes your “read it later and remember it” pile.

Spaced Repetition Flashcards

This is where retention actually happens. Booksum can auto-generate flashcards from your highlighted passages with a single tap. No manual card creation, no staring at a blank Anki template wondering how to phrase the question.

The flashcards use the SM-2 spaced repetition algorithm (the same system Anki uses, the same method medical students rely on for board exams). The app schedules review sessions based on how well you remember each card. Cards you struggle with come back sooner. Cards you nail get pushed further out. Over time, the intervals grow and the information sticks.

You can organize flashcards into decks by book, topic, or project. Study sessions track your performance and adjust accordingly.

Reading Insights and Habit Tracking

The insights dashboard shows your reading activity: total books, total snippets, weekly activity, active books, and your most-engaged titles. There are reading streaks and weekly goals to build consistency, because nothing motivates a reader like not wanting to break a streak.

Reading insights dashboard showing books, snippets, weekly activity, and most active book

Booksum also sends daily quote notifications: a random highlight from your saved snippets, surfaced at your preferred time. A small nudge to revisit what you’ve captured.

How the Workflow Fits Together

  1. Capture. Photograph a page or scan an ISBN barcode to add a book.
  2. Extract. On-device OCR pulls the text from the image.
  3. Annotate. Mark up the passage with drawings, notes, or highlights.
  4. Summarize. AI condenses dense text into key takeaways.
  5. Learn. Auto-generated flashcards with spaced repetition reinforce retention.
  6. Track. Reading insights and streaks keep you consistent.

The whole point is eliminating friction between “I just read something important” and “I’ll actually remember this in a month.”

Who This Is For

Privacy and Data

All OCR processing happens on-device via ML Kit. Your library is stored locally. The only network calls are for AI summaries (opt-in, per-snippet), Google Books metadata lookups, and barcode-to-ISBN resolution. Your reading list stays your business.

What’s Free, What’s Premium

The free tier is fully functional for casual use. Premium ($9.99/month, $49.99/year, or $99.99 lifetime) unlocks:

Technical Details


No account required. No cloud sync to configure. Photograph a page, extract the text, make it stick.

Share this article