Questions, bug reports, and requests for Mimikara, the Japanese-learning app from Torven Labs.
Mimikara is made by one person at Torven Labs, and messages go straight to them. The fastest route is X:
Bug reports are genuinely welcome. Including the device model, the iOS version, and what happened just before the problem appeared makes a fix far more likely.
No. Everything runs on the device, including the word data, the audio, and all progress. The app works in airplane mode, on a plane, or underground, and it behaves identically whether or not a network is present.
No. There is no sign-in, no account, and no server. Nothing needs to be registered before starting.
Progress lives in a database inside the app's own storage on the device. Because there is no server, it is not synced automatically. Settings includes a manual backup and restore: a backup writes a single file that can be saved anywhere, and restore reads it back. Moving to a new phone means exporting a backup from the old device and restoring it on the new one.
Only from a backup file made beforehand. Since nothing is uploaded, there is no copy on any server to restore from. Taking a backup before deleting or replacing a device is worth the few seconds.
Words are introduced in real-world frequency order, drawn from frequency data over large amounts of contemporary Japanese, so the most common words come first. The intent is that early study time buys the most comprehension.
Japanese distinguishes words by pitch, not only by sound. The graph traces the pitch contour across a word's morae, so the shape can be seen and not merely guessed. The minimal-pair drill trains the distinction directly, on pairs like 箸 and 橋, which differ in pitch alone.
A hands-free mode. The phone can be set down while high-value words flow past on screen and are read aloud, with nothing to tap and nothing to grade. It is built for walking, training, or any moment when touching the screen is impractical.
Check the device's silent switch and the volume level while the app is in the foreground, since iOS tracks media volume separately. Speech uses the voices installed on the device, so a missing or low-quality Japanese voice can be improved in the iOS Settings app under Accessibility, then Spoken Content, then Voices.
Yes, and it stays that way. No ads, no subscription, no paywalled features, and no data sold or collected. There is nothing to buy inside the app.
iPhone running iOS 17 or later. The interface is dark by design, supports Dynamic Type, and respects the system's Reduce Motion setting.
Both go to the same place: @deleonlangure. Requests are read, and the ones that fit the app's purpose have a real chance of shipping.