Translate scanned PDFs
without breaking the layout
Same layout, any language
Drag the handle to compare. Whatever the language, every column, table, figure, and number stays exactly where it was — only the words change.
Choose your plan
Simple, scan-friendly pricing. Pages are pages — no multiplier for scanned files.
- $0.50 / page
- 10 scanned pages, never expire
- Up to 100 MB per file
- $0.13 / page
- 120 pages every month
- Up to 100 MB per file
- Pages reset each month
- $0.06 / page
- 700 pages every month
- Unused pages roll over (up to 2,100)
- Big files up to 300 MB
Translating scanned PDFs — frequently asked
Why does my scanned PDF come out as gibberish or random characters when I translate it?
A scanned PDF is just an image of text with no underlying text layer, so a translator that expects selectable text either returns blank pages or garbled, random characters. The fix is OCR (optical character recognition) to read the image first. Reglyph runs OCR on the scan before translating, so you get real translated text instead of gibberish.
Why do Google Translate or DeepL say 'no text' or fail on my scanned PDF?
Most mainstream translators only read PDFs that already contain a digital text layer. An image-only scan has none, so they report 'text not recognized', hang, or quietly skip pages. Reglyph is built for exactly this case — it recognizes the text in the image, so you don't have to convert or OCR the file yourself first.
How do I translate a scanned PDF without losing the layout, tables, and columns?
Standard tools dump the translated words into a plain wall of text — tables collapse, columns merge, and headings blend into the body. Reglyph erases the original text from the page and re-typesets the translation in the same position, so tables, columns, figures, stamps, and numbers stay exactly where they were.
Do I have to retype or reformat the document after translating it?
With most workflows, yes — people report spending 30–60 minutes per document copying the original formatting back by hand. Reglyph rebuilds the page layout automatically, so the output is ready to use without manual reformatting.
Why doesn't ChatGPT or my PDF tool's AI assistant work on scanned documents?
Many AI assistants can only read the text layer of a PDF and silently do nothing useful on image-only scans — a limitation that often isn't stated until after you've paid. Reglyph treats scanned, image-only PDFs as the primary use case rather than an unsupported edge case.
Can I translate an image-only PDF or a photo of a page without OCR-ing it myself first?
Yes. You don't need to run a separate OCR step or convert the file to editable text beforehand. Upload the scanned PDF or a photo of the page and Reglyph handles recognition, translation, and re-layout in one pass.
Will stamps, signatures, and numbers stay in the right place?
Yes. Reglyph only replaces the text it translates; stamps, seals, signatures, figures, and numeric values keep their original positions and are not altered — which matters for certificates, contracts, and official records.
Can it handle large or multi-page scanned PDFs?
Yes. Many scanned-PDF translators choke on large or multi-page files. Reglyph processes documents page by page, so multi-page scans are translated with the layout preserved on every page. You can try the first 3 pages free.