Translate PDF
Turn a PDF into another language — French, Arabic, Spanish and 28 more — keeping its paragraphs and page order.
✉️ Privacy: external service — the text extracted from your PDF is sent to MyMemory (api.mymemory.translated.net) for translation. Your file itself is never sent. Consent is required below before anything is sent.
Reading your document…
or drop your PDF here
Text not found
This file contains no readable text to translate. If it is a scanned document, run OCR PDF first to add a text layer, then translate it here.
Segments are sent one at a time. You can stop at any point — finished pages are kept.
Your translation
Translation stopped part-way
How to translate a PDF
Choose your PDF
Click the select button or drop your file into the area provided. Its text is read right here, on your device — the file itself goes nowhere.
Pick the languages
The source language is detected for you, or choose it yourself; then pick the target language and, if you like, a page range. Use the swap button to reverse the pair.
Translate and download
Press Translate and watch the segments progress. Download the result as a PDF in the new language, or as a text file — finished pages are kept even if the daily limit stops the run.
Translating a document is often the difference between filing it away unread and actually using it: a contract from a foreign supplier, a study paper, a user manual, a letter from family abroad. This tool reads the text of your PDF, translates it paragraph by paragraph, and rebuilds it as a clean new document with the same page order — ready to read, print or share.
One thing is different here, and it is stated openly rather than hidden: translation itself is done by the free MyMemory service, because no browser can carry a quality translator for thirty languages inside it. Your PDF file is never uploaded — only the extracted text travels, over an encrypted connection. Every other tool on this site works fully offline on your device; this one is the single, clearly-labelled exception.
Frequently asked questions
How does this tool work?
Your PDF is opened in your browser and its text is extracted page by page, on your device. The text is split into short segments, which are sent one at a time to the free MyMemory translation service and returned in the target language. The translated paragraphs are then rebuilt into a new PDF (for most languages) and a text file, keeping the original page order.
Is my file uploaded?
The file itself — no, never. It is opened and read entirely on your device, exactly like our other tools. But its text is sent to MyMemory to be translated, over an encrypted connection; that is how this one tool works, and it is why we ask you not to use it for highly sensitive documents.
Why did translation stop in the middle?
Almost always, the free service's daily limit: MyMemory allows roughly 1,000 words per day per visitor. When the limit is reached, translation stops cleanly, the pages already finished are kept, and you can download them as a text file. Translate the rest tomorrow, or use the page-range option to do a long document in daily slices.
Which output will I get?
For most languages — Latin, Greek, Cyrillic, Arabic and Hebrew scripts — you get a rebuilt PDF plus a TXT file. For Chinese, Japanese, Korean, Hindi and Thai, the translation is delivered as a text file (.txt) only: producing a PDF would need a font that covers those scripts, and rather than ship a broken PDF we give you the complete text. In Arabic, Persian and Urdu, vowel marks (tashkeel) are not carried into the rebuilt PDF — the embedded font's layout engine cannot position them — but the TXT file keeps the translation exactly as received.
Does it keep the original layout and images?
No. The result is a text-only rebuild: the same paragraphs, in the same order, on matching pages — but not the original fonts, columns or pictures. Keeping the exact layout while replacing the text is a server-side AI feature; it is honestly out of scope for a tool that runs in your browser.
My PDF is a scan — what do I do?
A scanned PDF holds pictures, not text, so there is nothing to translate and this page will tell you so. Run the file through OCR PDF first to recognize the text and add a text layer, then translate the result here.
How good is the translation?
It is machine translation from a free service: generally good for understanding a document, drafting a reply, or studying — and noticeably weaker on idioms, legal phrasing and nuance. For anything important or legally binding, treat the output as a draft and have it reviewed by a human translator.