How to Translate a PDF (Privately)
Translating a foreign-language contract, understanding a manual in a language you don't speak, or preparing a document for a human translator all start with extracting the text. Most online PDF translators upload your entire document to their servers — risky for confidential content. Extract the text first, then translate only what's needed.
Why extract first
When you paste raw text into Google Translate or DeepL, the service receives only that text — not the original PDF, not its metadata, not the surrounding pages you didn't extract. For confidential documents, this minimizes what the translation service sees.
How PDFPuddle extracts
Open Translate PDF, upload the document, click Extract Text. PDFPuddle walks the PDF's text layer with PDF.js and returns a clean .txt file with page separators.
Best translation services
DeepL produces the most natural translations for European languages — German, French, Spanish, Italian, Dutch. Google Translate has the broadest language support including East Asian languages. For technical or legal documents, services like SDL or specialized vendors offer human-reviewed machine translation.
Reassembling translated content
Once translated, paste the result into Word, format to taste, then export to PDF using Word to PDF. The result is a translated PDF that mirrors the structure of the original. For pixel-perfect layout preservation, professional translation services use specialized tools like SDL Trados that overlay translations onto the original layout.