Common questions, plainly answered

Coming soon to iOS and macOS, with Android later. There's no web app, and there isn't one planned. The toolkit runs on your device, not in a browser.

No date yet. It depends on demand. Sign up on the homepage to be notified and add your voice.

macOS 10.15 or later. The iOS version requirement is shown on the App Store listing.

A lot of what makes the app good (the dictionary on your device, transcription that doesn't leave the device, your saved translations being yours) only really works as a native app. It also means there's no account to create, no password to remember, and nothing to log into. A web version would be a different, weaker product.

On your device. The dictionary, your saved translations, your word lists, your transcripts, your recordings. The app reaches the network only when you ask it to translate or speak something out loud, and only to the provider you chose.

No. Both kinds of audio stay local: recordings you make with the microphone, and audio files you load into the audio clipper. Transcription happens on your device. Only the resulting text is sent to your translation provider, and only when you ask.

Only the text you're translating, or the text you want spoken. Nothing about you, nothing about your other translations, nothing else from the app.

No. The app doesn't track your usage and doesn't send analytics anywhere.

No. There's no signup, no login, no account at all. The app is yours from the moment you install it.

Yes, on iOS. Open Translate, tap the camera, point it at the text. The camera is iOS-only because that's where the unprepared reading actually happens. On macOS the same text is almost always already digital, and pasting it in is the simpler path.

Yes. Tap the microphone in Translate, record what you heard, tap to stop. The transcription happens locally on your device. If it isn't quite right, you can edit it before sending it off to translate.

No. The dictionary is English to Japanese only. If support for another language would help you, get in touch.

MP3 files from your device. Pick anything you've already got: a podcast you've downloaded, a clip from a recording app, audio you've extracted from a video.

Custom collections you build from dictionary entries. Looking up something you want to remember? Add it to a list. Lists are accessible from the home screen.

Open Audio Clipper from the home screen. Load a podcast, a clip, or a recording. Select the range you want to extract. The app splits the audio into segments, and you assign each one to P1 (Person 1), P2 (Person 2), or none. Each turn is transcribed locally, and you get a conversation you can study line by line. You can also build a conversation from scratch by typing or speaking each line.

JLPT (Japanese Language Proficiency Test) is the standard grading scale. N5 is beginner, N1 is the most advanced. The dictionary filter limits results to entries within the levels you choose, so you can study at your range without older or harder vocabulary getting in the way. The filter needs the full dictionary download.

Yes. Tap any kanji in a translation or dictionary entry to see its components, similar kanji, readings, meanings, and JLPT or grade level.

Not at the moment. The study games are designed to be picked up and put down, not turned into another streak to maintain. If progress tracking would help your practice, get in touch.

Translation runs against a cloud model, because local models aren't quite small enough and accurate enough for learning yet. Bringing your own key means you pay only for what you use, your costs stay yours, and there's no markup or middleman.

OpenAI and Google are built in. Beyond that, any OpenAI-compatible endpoint will work, so most providers and many self-hosted setups are supported. You can route text, speech, and image capabilities to different providers if you want to.

On your device, only. They're used to send your requests to the provider you chose, and nothing else.

It depends on your provider's pricing and how often you translate. For most learners, individual translations cost fractions of a penny. The dictionary, transcription, conjugations, kanji breakdowns, search, and study tools all run locally and have no per-use cost.

The app comes with the OS text-to-speech voice built in, which works offline and handles every word, conjugation form, and translation. For higher-quality pronunciation, connect a third-party voice provider in settings. The OS voice can also be improved a little by downloading higher-quality voices in your iOS settings.

The dictionary, conjugations, kanji breakdowns, search, word lists, the random word and word search games, scripted conversations, transcription, and the offline pronunciation voice. Translation and higher-quality voices need a provider.

Mostly, yes. The dictionary, your saved translations, transcription, furigana, conjugations, and study games all work offline. Audio that's already been generated for your translations stays available, along with the OS-provided text-to-speech voice. Only fresh translations and higher-quality pronunciation need the network, since they go to your provider.

The lite version is around 20 MB and covers the basics. The full version is around 400 MB and unlocks JLPT and frequency filters, similar kanji, kanji components, and the complete entry set. Both work offline once installed.

A free tier will always exist, and it covers most of what the app does today. Beyond that, pricing isn't finalised yet. The full dictionary download (around 400 MB) and future sync between devices are the parts with real costs behind them, so they're the most likely candidates for a paid tier when the time comes.

Email is the best route. Get in touch.