The everyday Japanese textbooks skip. Say what you heard; AI makes a card to save and review before it fades.




Catch by voice
Say it as you heard it; AI sorts out reading and meaning

Learn by scene
24 scenes of daily life, from the konbini to disasters

Shadowing
Listen, repeat, get scored

Review
Shows up just before you forget. 30 cards a day, no backlog
At the register, at the counter, at a viewing. Heard daily, missing from dictionaries.

01
Tap and repeat the sound as you heard it. No need to know how it is written.
02
AI drafts the reading, meaning and an example. Check it, tap “Save to my words”, and it is in. No photos, no typing.
03
Up to 30 cards a day, only as much as you can keep up with.

Hear the model, then say the same line. Your audio itself is scored for pronunciation: accuracy, fluency and completeness, plus which words were off. If that scoring is unavailable, the transcript is compared with the model instead.
From kana stroke order to N1 vocabulary, a structured course sits alongside the everyday words.

Menus, instructions and error messages in eight languages. Word meanings appear in your language where a translation exists, otherwise in English.
Yes. Signing up, catching words and reviewing are all free.
Yes. The app comes in eight languages, word meanings show in your language where a translation exists (otherwise in English), and the course starts from kana stroke order.
Yes. 720 everyday English words for work, travel and the clinic, grouped by scene.
Capture recordings are used only to transcribe and build the card; Shadowing recordings only to score your pronunciation. Neither is stored. Processing happens at overseas providers (OpenAI, Anthropic, Microsoft Azure).