Skip to content
Account

Translate a scanned PDF, keeping the layout

5 credits per pageDeleted after 2 h

Free account · 20 free credits. Sign up to run it.

Opening…

Drop a PDF or a picture here

PDF, PNG or JPG · one file

or paste it here (Ctrl+V)

Nothing is uploaded until you press Translate the scan.

Before and after

The words on a scan translated in place; the order number, the date's digits and the figures as written.

Original scan
A scanned work order dated 12 March 2026, with a table of items and totals.(opens the sample PDF in a new tab)

A scanned work order: a heading, the date and terms, and a table of items and totals.

In Hindi
कार्य आदेश 2026-0912
दिनांक: 12 March 2026
शर्तें: नेट 30

What this tool does

Use this when your PDF is a picture of a page — a scan, a photocopy or a phone photo — rather than a document with text in it. The page is read with OCR, and each block of text is translated by an AI model a page at a time. The old words of each block are then painted out, filling in the paper behind them from what surrounds them, and the translation is set in the same box, in a matching font at the size and colour of the ink on the page. Only the blocks that were translated are touched: each is a patch laid over your scan, and every pixel outside the patches is your original. Faces are never painted over, and the paper is filled by our own image code, never by a generative image model. Numbers, amounts, dates, codes, email addresses and phone numbers are copied exactly as written, and a line whose translation would change one of them, or would not fit its box, stays in the original language and is marked. A page that could not be translated is not charged. Languages written right to left, such as Arabic, Urdu and Hebrew, are not offered yet. The file says in its metadata that it was edited with AI, with a signed provenance manifest beside it; there is no visible mark on the pages.

The edited file is an AI-manipulated version of your document. It is marked as AI-edited in its metadata; using it to misrepresent a real document is prohibited by our rules.

What happens to your file

Deleted from our servers after 2 hours. We do not train AI on your files, and the terms of our AI providers do not allow them to; they may keep request logs for up to 55 days, longer for a request they flag for abuse.

Uploads and outputs are deleted after 2 hours, whether or not you download them. Provider request logs are kept for at most 55 days. Details on the privacy page.

Questions

How does it translate a scan without making it look pasted over?

The page is read with OCR, and each block of text — a paragraph, a heading, a table cell — is translated. Its old words are painted out by filling the paper in from what is around them, and the translation is set in the same box, at the size, colour and slope of the ink that was there. A paragraph is re-justified when the original was. The fill is done by our own image code on our servers, never by a generative image model, because a model painting a page of text invents text.

What does it cost?

5 credits a page, because reading a scan and repainting it costs more than rewriting a PDF with text in it, which is 3 credits a page on the digital translator. The price is shown before anything runs, and a page we could not translate is left as it was and not charged. You need a free account to run it — new accounts get 20 credits.

Is the rest of my scan changed?

No. The page image you uploaded stays in the file as it was. Each translated block is a patch laid on top of it, so outside those patches every pixel is your original scan. Faces are never painted over, and a line whose translation would change a number or a code, or would not fit its box, is left as it was and marked in the result.

Which languages can it translate into?

Hindi, Bengali, Marathi, Tamil, Telugu, Gujarati, Kannada, Malayalam, Punjabi and Odia, and English, French, German, Spanish, Italian, Portuguese and Russian. The language of the scan is worked out from the page. Languages written right to left, such as Arabic, Urdu and Hebrew, are not offered yet.

What happens to the file I upload?

It is deleted from our servers after two hours, whether or not you download the result, and it is not used to train anything. The page image goes to the OCR provider that reads it and its text to the AI provider that translates it; they do not train on it and keep request logs for at most 55 days, longer only for a request they flag for abuse. The file you download says in its metadata that it was edited with AI and carries a signed provenance manifest; there is no visible mark on the pages.

Translate a PDF, keeping the layout

Edit a document

Guides and free sample files