meta: title: Translate — べんきょうBox description: Read any Japanese, however it reaches you. Paste, type, speak, or photograph it, and see it broken down word by word. url: /translate format: yaml content: - type: page-header body: A line in a video that catches you out. A sign at the station. A page from a manga. Get it into the app, and see it broken down word by word. heading: Read any Japanese, however it reaches you eyebrow: 01 · TRANSLATE accent: hero - type: card-grid body: Paste, type, microphone, or camera; every input ends up at the same broken-down sentence. heading: However the Japanese reaches you label: Four ways in intro: 'Most of the time you''ve already got the text. Sometimes you''ve only got audio. Sometimes you''ve only got a photograph. The app handles all four, and the result is the same however it got there: a full breakdown you can study.' columns: 4 variant: sub cards: - title: Paste body: Anything from your clipboard, instantly. - title: Type body: If your hands are already on the keyboard. - title: Microphone body: Record it and let the app transcribe. - title: Camera body: Point and shoot, on iOS. - type: strip body: 'Point, shoot, and the text comes through ready to break down. iOS only: on macOS the text is usually digital already, so paste does the job.' heading: The camera lives on iOS, where unprepared reading actually happens label: Camera orientation: text-left paragraphs: - Signs at the station. A page from a book. A t-shirt slogan. A menu. Point, shoot, and the text comes through ready to break down. - On macOS the same text is almost always already digital, so paste does the job. screenshot: label: SCREENSHOT SLOT title: Camera capture view meta: iOS · 4:5 portrait aspect: 4/5 - type: strip body: Record what you heard, transcribe locally, edit if needed, then translate. Your audio doesn't leave the device. heading: From spoken Japanese to broken-down sentence label: Microphone orientation: text-right paragraphs: - Tap the microphone, record what you heard, tap to stop. The app transcribes locally, so your audio doesn't leave the device. - If the transcription isn't quite right, edit it before sending it off to translate. Mishearings happen, yours and the app's. - The full breakdown is saved on your device. screenshot: label: SCREENSHOT SLOT title: Recording / transcript edit meta: Mic flow · 4:5 portrait aspect: 4/5 - type: strip body: Every word colour-matched to its translation, with reading and dictionary form attached. Tap to drill in. Edit if a mapping feels off. heading: A working surface, not a finished answer label: The breakdown orientation: text-left paragraphs: - The breakdown shows you every word in the sentence, colour-matched between Japanese and English, with the reading and dictionary form attached. Tap any word to open its dictionary entry. Tap any kanji to pull it apart into components. - If a word's mapping isn't quite right, or you'd just phrase it differently without losing the meaning, edit it. The breakdown is yours. screenshot: label: SCREENSHOT SLOT title: Translation breakdown meta: Coloured token mapping · 4:3 aspect: 4/3 - type: card-grid body: Furigana and pronunciation, both on your device. heading: Polished where it matters columns: 2 variant: surface cards: - label: Furigana title: Readings on every word, on your device body: Once you've installed the furigana dictionary, every breakdown gets readings without a network round-trip. No waiting for a server, no surprise gaps when you're offline. - label: Pronunciation title: Hear any line spoken body: Every translation has playback. Use the offline voice the app comes with, or connect a higher-quality voice from your provider. - type: card-grid body: Saved translations are kept on your device with their breakdown, audio, and source. columns: 1 variant: surface cards: - label: Saved translations title: Everything you've translated, kept body: Every translation is saved on your device with its breakdown, audio, and source. Search the list, copy a line on long-press, come back to anything later. Nothing gets sent anywhere, they're yours. - type: card-grid body: Bring your own translation provider; switch whenever you want. label: Provider columns: 1 variant: sub cards: - title: Bring your own body: |- Translation runs against a cloud model. Local models aren't quite small enough and accurate enough for learning yet. Use OpenAI, Google, or any OpenAI-compatible endpoint. Keys are stored on-device and used only for API requests. The costs are yours, and the only data sent over the network is the text you're translating. Everything else stays local: your dictionary, your saved translations, your transcripts, your recordings. - type: faq body: Common questions heading: Quick answers items: - question: Does my audio leave my device? answer: No. Microphone recordings are transcribed locally on your device. Only the resulting text is sent to your translation provider. - question: Is camera translation on macOS? answer: No. Camera input is on iOS. On macOS the text is usually already digital, and pasting it in is the simpler path. - question: Can I use it offline? answer: Translation itself needs the network, since it runs against a cloud model. The dictionary, your saved translations, transcription, and furigana all work offline. - question: Which providers can I use? answer: OpenAI, Google, and any OpenAI-compatible endpoint. Your keys are stored on your device and only used to make requests to your provider. - type: card-grid body: 'Coming soon to iOS and macOS. Up next: the dictionary.' columns: 2 variant: sub cards: - label: Get it title: Notify me when it ships → body: Coming soon to iOS and macOS. Android later. href: '#notify' - label: Next title: Dictionary → body: A real Japanese dictionary on your device, behind every word. href: /dictionary tile: tools schema: '@context': https://schema.org '@type': FAQPage name: Translate — べんきょうBox description: Read any Japanese, however it reaches you. Paste, type, speak, or photograph it, and see it broken down word by word. url: https://getbenkyobox.com/translate mainEntity: - '@type': Question name: Does my audio leave my device? acceptedAnswer: '@type': Answer text: No. Microphone recordings are transcribed locally on your device. Only the resulting text is sent to your translation provider. - '@type': Question name: Is camera translation on macOS? acceptedAnswer: '@type': Answer text: No. Camera input is on iOS. On macOS the text is usually already digital, and pasting it in is the simpler path. - '@type': Question name: Can I use it offline? acceptedAnswer: '@type': Answer text: Translation itself needs the network, since it runs against a cloud model. The dictionary, your saved translations, transcription, and furigana all work offline. - '@type': Question name: Which providers can I use? acceptedAnswer: '@type': Answer text: OpenAI, Google, and any OpenAI-compatible endpoint. Your keys are stored on your device and only used to make requests to your provider.