Docly practical guide

How to Translate a PDF into Another Language

Understanding a paper, a manual or a contract in another language should not mean copying it paragraph by paragraph into a translation box. Translating a PDF takes the document's text as a whole and returns it in a format you can keep working in.

6 minute readUpdated 26 July 2026Docly

Translate a PDF in the browser

Add your file to the AI Translate tool and choose a target language from English, Turkish, German, French, Spanish, Arabic and Russian. Start the job and download the result when it finishes.

The output is an editable Word document whose heading records the source file name and the target language. Long documents are processed in sections, so the time it takes grows with the page count.

Why the output is a Word file

Translation changes the length of text. The same sentence gets shorter in one language and longer in another, and forcing translated text back into the original page layout produces overflowing lines, clipped words and broken tables.

A Word file avoids that entirely: you can read, correct and restyle the translation, or drop it into your own template. Page markers are preserved in the text, so you can trace which page of the source a paragraph came from and check a doubtful sentence against the original.

Translating long papers and books

Because long documents are translated section by section, terminology choices can drift between parts. For academic and technical work, list your key terms before you start and use Find and Replace in Word afterwards to unify them. That single step makes the biggest difference to consistency.

In academic papers, reference lists should normally be left in the original language; translating citation entries makes the sources impossible to look up. Formulas, table headers, figure captions and units all deserve a separate check after translation.

Scanned PDFs and text inside images

Translation reads the document's text layer. A scanned or photographed PDF has no text layer, so there is nothing to extract; run PDF OCR on those files first.

Text baked into images is not translated either. Chart labels, menus in screenshots and stamps printed over a scan stay in their original language. The output contains the translated text; images from the source are not carried into the Word file.

Quick checklist

  • Read the translation next to the original and verify numbers, dates, units and proper names.
  • Machine translation does not replace a certified translation for official use.
  • Translate very long books chapter by chapter to keep control of quality.
  • Apply heading styles in the Word output to rebuild the document structure.

Tools in this guide

Frequently asked questions

Are page references kept?

Yes. Page markers are preserved in the text so you can find where a paragraph appears in the source.

Do images come through into the Word file?

No. The output contains the translated text; images are not transferred.

Can I use this for a certified translation?

No. Processes that require a sworn or notarised translation will not accept this output.

Related guides

View all guides