Built for immersion learners

Automate your mining
with ease.

Upload any PDF. Get a study-ready vocabulary deck in minutes, not hours. FSRS-powered spaced repetition included.

採掘 — Novel: 1Q84 by Haruki Murakami
Card 47 of 312

青豆は、その________を黙って見つめていた。

Aomame silently gazed at that ________.

Mining shouldn't take longer than reading

You know the drill. Look up a word, copy the sentence, format the card, add it to Anki... repeat 50 times. By then, you've forgotten what you were reading.

30-45 minutes per session

Creating cards manually eats into your actual study time. Most people give up before finishing.

Kills your reading flow

Context switching between reading and card creation destroys immersion. The story becomes a chore.

5 tools, one task

PDF reader, dictionary, spreadsheet, Anki, maybe a tokenizer. The setup alone is exhausting.

Everything you need to mine smarter

Upload a PDF, and we handle the rest. Tokenization, definitions, example sentences, spaced repetition scheduling — all automatic.

PDF-to-Deck Pipeline

Drop any PDF — novels, articles, textbooks. We extract text, tokenize properly for your target language, and build your deck.

FSRS v5 Scheduling

State-of-the-art spaced repetition built in. No Anki setup required — though you can export if you want.

Cloze Deletion Cards

Every card uses context sentences from YOUR content. Not generic dictionary examples — real sentences you care about.

Frequency Ranking

Words ranked by how often they appear — both in your document and in the language overall. Learn high-value words first.

Key

Deep Anki Integration

Import your existing Anki decks — we skip words you already know. Export new decks as .apkg. No duplicate cards, ever.

Pre-built Decks

Not ready to mine yet? Start with our Core 500/1000/2000 frequency decks to build a foundation.

Three steps. That's it.

No setup guides, no plugin installations, no configuration files.

1

Upload your PDF

Drag and drop any PDF — a novel you're reading, an article, study materials. We accept Japanese, Spanish, French, German, and Portuguese.

1Q84_Vol1.pdf
2.4 MB
2

We do the mining

Text extraction, tokenization, lemmatization, dictionary lookups, frequency analysis, card generation. Usually done in under 5 minutes.

Text extracted
Tokenized (312 words)
Adding definitions...
3

Start studying

Your deck is ready. Review in-app with FSRS scheduling, or export to Anki. Either way, you're learning in minutes instead of hours.

312
cards ready to review

Simple pricing

Free forever for casual learners. Upgrade when you're ready to go unlimited.

Free

$0 /month
  • 30 reviews per day
  • 3 PDF uploads per month
  • 5 decks maximum
  • 1 Anki export per month
  • Pre-built frequency decks
Get Started Free
Most Popular

Pro

$10 /month

or $60/year (2 months free)

  • Unlimited reviews
  • Unlimited PDF uploads
  • Unlimited decks
  • Unlimited Anki exports
  • Priority processing
Start 7-Day Free Trial

Common questions

Which languages are supported?

Japanese is our primary focus at launch, with full tokenization and dictionary support. Spanish, French, German, and Portuguese are supported with stemming-based tokenization. More languages coming soon.

I already use Anki. Why would I switch?

You don't have to — Saikutsu integrates deeply with Anki. Import your existing decks and we'll skip words you already know when mining new content. Export new decks as .apkg. No duplicates, no wasted time relearning. Use both tools together.

What's FSRS?

Free Spaced Repetition Scheduler — a modern, open-source algorithm that's more efficient than SM-2 (what Anki uses by default). It learns your memory patterns and optimizes review timing. Research shows 20-40% fewer reviews for the same retention.

Can I use scanned PDFs?

Not yet. We need digital text to extract vocabulary. Scanned images and handwritten documents aren't supported. If your PDF has selectable text, you're good.

How accurate is the tokenization?

For Japanese, we use MeCab-compatible tokenization with proper dictionary form extraction. You'll get clean vocabulary words, not conjugated forms. European languages use stemming which works well for most content.

What if I find a bug or want a feature?

Email us or open an issue on GitHub. We're a small team building this because we use it ourselves. Feedback shapes the roadmap.

Ready to automate your mining?

Upload your first PDF and see the difference. Free to start, no credit card required.