Companies working with Japanese partners routinely need English contracts, product manuals and compliance documents rendered in Japanese, and residents in Japan often have to hand English paperwork to a municipal office in Japanese. Reglyph reads a scanned or digital English PDF with OCR, erases the original text from the page image, and typesets the Japanese translation back in place — so a 40-page manual keeps its numbered diagrams, tables and warning boxes exactly where they were.
Japanese mixes kanji, hiragana and katakana, and a translated line often runs longer than the English original — especially inside narrow table cells, where dense kana strings can overflow a column sized for English. Because Reglyph rebuilds each page rather than overlaying text, the Japanese translation is fitted back into the original cells and text blocks, so tables, stamps and figure callouts keep their positions.


Drag to compare — every table, figure, and number stays in place.
These are the English-language documents people most often translate to Japanese — each keeps its original layout, seals, and tables.
Drop in a scanned English PDF, a digital PDF, or a photo of the page. Image-only files are fine — no text layer needed.
OCR extracts the English text, then it's translated to Japanese while the original text is erased from the page.
Get a clean PDF where tables, figures, stamps, and numbers sit exactly where they were.
Yes. Reglyph erases the English text from the page image and typesets the Japanese in the same positions, so clause numbers, tables and signature blocks stay where they were. It is machine translation, so have a bilingual reviewer check legal wording before anything is signed.
Requirements vary by office; many accept a translation as long as someone takes responsibility for it, while others require specific translator details. Use Reglyph to produce the translated document with the layout intact, then have a qualified translator verify and attest it where the office requires that.
Yes — photographed and scanned documents are supported, and clear printed pages work best. Heavy handwriting is unreliable for OCR, so a handwritten letter may come out incomplete. You can try it on 5 pages free without a credit card.
Simple, scan-friendly pricing. Pages are pages — no multiplier for scanned files.
Actes de naissance with margin annotations, mairie stamps, diplomas from three countries — the practical way to get English versions that authorities will accept.
Read guideGeburtsurkunde, Arbeitszeugnis, a stack of scanned contracts — here's the realistic way to get accurate English versions that still look like the originals.
Read guideThe rule itself is short. The mistakes people make around it — partial translations, missing seals, uncertified machine output — are what generate RFEs.
Read guide