A desktop app for language learners — read PDFs, look up words, build your vocabulary with spaced repetition.
These control how the built-in SRS schedules card reviews.
| Parameter | Default | Meaning |
|---|---|---|
request_retention |
85% |
Target probability you'll recall a card when it's due. Higher → shorter intervals, more reviews, stronger memory. Lower → longer intervals, fewer reviews, more forgetting. |
maximum_interval |
365 days |
Hard cap on days between reviews. No card exceeds this even if the algorithm computes a longer interval. |
enable_fuzz |
true |
Adds small random jitter to intervals so cards from the same batch don't all fall due on the same day. Keep it on. |
enable_short_term |
true |
When on, new & forgotten cards go through short (minute-level) steps before entering the long-term cycle. Turn off to skip straight to long-term scheduling. |
learning_steps |
1m, 10m |
Progressive intervals for new cards: rate Good → wait 1 min → see it again → rate Good → wait 10 min → see it again → graduate to long-term. Rate Again to restart the current step. Supports m (minutes), h (hours), d (days). |
relearning_steps |
10m |
Single-step catch-up for forgotten cards (rated Again after graduation). One short interval is usually enough to pull it back into the long-term schedule. |
Available shortcuts can be customized in Settings → Shortcuts.
| Action | Default | Context |
|---|---|---|
| Docs tab | Alt+1 |
Global |
| Review tab | Alt+2 |
Global |
| Outlines tab | Alt+3 |
Global |
| Annotations tab | Alt+A |
Global |
| Summary tab | Alt+S |
Global |
| Toggle left panel | Alt+[ |
Global |
| Toggle right panel | Alt+] |
Global |
| Open settings | Ctrl+, |
Global |
| Toggle quick add | Ctrl+I |
Global |
| Toggle fit width | Ctrl+E |
PDF open |
| Start Learning | Ctrl+S |
Annotations tab |
| Reveal / flip card | Space |
Learning mode |
| Again (grade 1) | Num1 |
Learning mode |
| Hard (grade 2) | Num2 |
Learning mode |
| Good (grade 3) | Num3 |
Learning mode |
| Easy (grade 4) | Num4 |
Learning mode |
| Listen / stop audio | Alt+L |
Learning mode |
| Back from learning | Escape |
Learning mode |
Supports two providers:
- Edge-TTS — local, via
pip install edge-tts. Configurable voice per language, rate, volume, pitch, and binary path. - MiMo — cloud API from XiaoMi. Requires an API key and supports preset voice models as well as voice design/cloning.