Last updated: 2026-10-16
Pay once for your whole book. Text-based PDF in, translated EPUB out. No scanned pages or original PDF layout.
Minimum $3.99 per book. Sign in after selecting your file to see its price.
EPUB and text-based PDF
Read format and pricing details before you upload. No OCR or translated PDF output.
Have a foreign-language novel, memoir or nonfiction book as a text-based PDF? EPUB Translator extracts its text, rebuilds a reading structure and translates the book into your chosen language. You download an EPUB for reflowable reading, with a translation-only edition or a bilingual edition that places each original paragraph above its translation.
This is a paid whole-book translation service. You do not need a subscription, an API key or a separate conversion tool. The PDF-to-EPUB step is part of the existing book translation workflow. The result is an EPUB, not a translated PDF.
Pricing uses the character count of the EPUB extracted from your PDF, with a $3.99 minimum per book. Current rates are $0.004–$0.0125 per 1,000 characters, with lower rates on later portions of longer books. A 500,000-character book costs $4.75 before any applicable coupon. This is a size example, not a quote for your PDF: page count alone cannot tell you the price.
Select your file and sign in. Once the server has converted and parsed it, you see that book's price before choosing a target language or paying. Selecting a file does not buy a translation. See how pricing works for the current tiers and the refund policy before paying.
Try selecting and copying a paragraph from a few pages. If the pasted text is readable and in the right order, your file is a better candidate than an image-only scan. This quick check is not a guarantee that every page will extract correctly.
| Your book | What to expect |
|---|---|
| Selectable prose in a normal single-column layout | The best fit. Text is rebuilt into paragraphs and detected chapters for reading in EPUB. |
| A scanned book or pages that contain only images | Not supported. There is no OCR to turn page images into text. |
| Two-column text, footnotes or a technical textbook | Some structures are handled, but reading order and paragraph boundaries can still be wrong. Check the delivered book against the original. |
| Illustrations, diagrams or dense tables | Image and table retention is limited. Vector diagrams and the PDF cover are not carried over; some raster figures may be retained. Text inside figures is not translated. |
| Formulas or code | Recognized formulas and code are kept as source text rather than translated. Recognition is imperfect. |
| A password-protected or DRM-restricted PDF | Provide an accessible file you have the right to translate. This service does not remove protection. |
PDF page limits are checked against the current server limit when you select a file, where the browser can read its page count. Files that cannot be read in the browser still go through server validation. Large or complex PDFs can exceed processing or memory limits even within the page limit; a small file size does not guarantee acceptance. The upload flow reports failures before translation starts.
An EPUB adapts text to your screen and font size. It does not reproduce the PDF's original page numbers, page breaks, columns or print layout. If you need a translated PDF with the same pages, a print-ready edition, or a guaranteed reconstruction of every diagram, this service does not provide that deliverable.
AI translation can miss meaning, leave passages untranslated or introduce errors. Automated checks cover eligible prose and can miss problems; retries do not guarantee that every issue is repaired. PDF extraction adds another possible source of omissions before translation. Keep your original and review important passages, particularly technical material. This is a reading copy, not a human-reviewed publication or certified translation.
Read how the translation workflow works, file handling and upload safety, and supported target languages before placing your order.