The text in any image,read in a second.
Drop a manga page, a menu or a screenshot. GlyphOCR boxes every line, translates it, and letters the translation back onto the picture.
10 free scans a day · no account needed
From screenshot to study list in three steps.
Drop an image
Manga page, game screenshot, menu, sign. Paste from the clipboard or drag it in.
Boxes, then words
Every line is boxed in about a second and the language is worked out on its own.
Read, save, study
Translation lettered back onto the image, plus vocab, readings and a quiz.
One scan, seven tools.
Every tool starts from the same fast scan.
Image translator
Upload a screenshot, sign or menu and get the translation written back onto the image.
Manga translator
Vertical Japanese is read, translated and lettered back into the speech bubbles.
Image to text
OCR that detects the language on its own, including vertical Japanese and Chinese.
Image summarizer
A short summary and a full translation of screenshots, documents and signs.
Quiz generator
Turn a page into a quiz: meanings, readings, fill-in-the-blank and translation.
Flashcard maker
The useful words with readings, meanings and levels, ready for Anki.
Pronunciation guide
Furigana for Japanese, pinyin for Chinese, romanization for Korean.
Pay for pages, not seats.
Free
- 10 scans a day
- Translation and readings
- Every tool
Pro
- 3,000 scans a month
- Vocabulary and grammar notes
- Anki and CSV export
- Priority speed
API
- Boxes, text and translation
- Streaming NDJSON
- Study data +$15 / 1,000
Under the hood
Next.js and TypeScript on the front, Python and FastAPI on the back, streaming results to the page as they arrive. OCR runs on the GPU: CRAFT text detection, manga-ocr for Japanese, per-script readers for other languages and optional SAM 2.1 box refinement. I designed the interface and built the whole stack.