Skip to content
  1. Home
  2. For e-readers
  3. PDF to EPUB

PDF to EPUB Converter

Turn a PDF book into an EPUB whose text reflows on any e-reader: page numbers and running heads removed, lines joined into paragraphs, chapters in a table of contents. Convert PDF to EPUB on your own device.

Nothing is uploaded Free, no account Scanned books too

Drop a PDF book here

or choose a file. It is read on this device and never uploaded.

Processed in your browser. Nothing is uploaded.

How to convert a PDF to EPUB

  1. Add the PDF book

    Drop the file or choose it. It is read here; if pages are scans, you pick the language and the text is recognized.

  2. Check the chapters

    Chapters come from the PDF bookmarks or from large headings. Switch the method if the table of contents looks wrong.

  3. Read the preview

    Open any chapter and see the paragraphs, headings, footnotes and pictures as they will be in the book.

  4. Download the EPUB

    Save it and send it to Kindle, copy it to a Kobo or open it in Apple Books.

What you get

THE QUIET HARBOUR re- 17
PDF page: fixed lines, running head, page number
Chapter 2
EPUB: text reflows to the screen

A PDF keeps every line where it was printed, so on a phone or an e-reader you zoom and pan across a page made for paper. An EPUB stores the text itself and lets the reader lay it out for the screen and the font size you choose. This converter rebuilds a PDF book as an EPUB: it finds the paragraphs, chapters and footnotes and drops what only made sense on paper.

Convert a PDF book to EPUB on your device

The conversion runs in your browser. The book is not uploaded anywhere, which matters for manuscripts, course packs and anything you would rather keep to yourself. There is no account, no page limit and no watermark. A typical novel of 300 pages is read in under a minute on a laptop.

What gets fixed: paragraphs, headers, page numbers, chapters

What the converter changes when it turns a PDF into an EPUB
In the PDF In the EPUB
Line breaks Whole paragraphs, found by indents and spacing, that reflow on any screen
Hyphenated words A word broken at the end of a line is joined again
Running heads Removed, with page numbers: lines repeated at the top or bottom of many pages
Bookmarks, headings Chapters and a table of contents
Footnotes Notes at the end of the chapter, linked from the number in the text
Pictures Pictures and drawings in the same place in the text
Title, author Book title and author in the EPUB, cover made from page 1

Each change can be switched off: keep running heads, keep footnotes on their page, drop the pictures for a smaller file, or choose how chapters are found.

Scanned PDFs: text recognition first

A scanned book is a stack of photos with no text inside, so there is nothing to reflow. When the converter finds such pages it asks for the language of the book and recognizes the text first (this is called OCR). Recognition runs on your device too; only the language model is downloaded once, about 2 to 15 MB. It takes time: about 2 to 5 seconds a page on a computer and several times longer on a phone, so a 300 page scan takes 10 to 25 minutes on a computer. Clean, straight scans at 300 dpi give the best text; you can straighten crooked ones with Deskew PDF first.

Preview chapters before you download

Before anything is saved you see the table of contents and the chapters as they will read on an e-reader. If chapters are missing or a running head slipped into the text, change the settings and the preview updates at once. When it looks right, make the EPUB.

Read on Kindle, Kobo, Apple Books

  • Kindle: send the EPUB with Send to Kindle (the app, the website or by email). Amazon converts it for your Kindle and it syncs to the Kindle apps.
  • Kobo, PocketBook, Tolino and other readers: they open EPUB directly. Connect the reader by USB and copy the file, or add it to the library app you use.
  • Apple Books: open the EPUB on an iPhone, iPad or Mac and it is added to your library.
  • Android and Windows: any EPUB reading app opens it, for example Google Play Books after an upload to your library.

How it works

The PDF is read with pdf.js, the PDF engine used by Firefox: every piece of text comes with its position, size and style. Lines that repeat in the same place on many pages are treated as running heads; lines of small text at the bottom of a page that start with a number are footnotes. Lines are joined into paragraphs where the spacing and indents say a new paragraph begins. Scanned pages go through Tesseract, an open source text recognition engine. The EPUB is then written in the browser with a table of contents that every reader understands.

When PDF to EPUB will not work well

It is made for books that are mostly running text: novels, memoirs, essays, reports. Be ready for rough results with:

  • Text in two or more columns, as in magazines and some papers: columns may be read in the wrong order.
  • Tables: they come out as lines of text, without the grid.
  • Math formulas: symbols lose their layout.
  • Textbooks with side boxes, captions and margin notes: these land between paragraphs.
  • Large scanned books: recognition takes minutes, and longer on a phone; mistakes in the scan become mistakes in the text.

For such documents a PDF is often better on a tablet. To print an e-book instead, use EPUB to PDF.

Questions and answers

Will the EPUB keep formatting?

It keeps what matters for reading: paragraphs, chapter and section headings, bold and italic, footnotes and pictures. It does not keep the page layout, fonts or line breaks of the PDF, because the point of EPUB is that your reader lays the text out for its screen.

Can I convert a scanned PDF?

Yes. Pages without a text layer are recognized first: choose the language of the book and the text is read from the images on your device. Allow 2 to 5 seconds a page on a computer and several times longer on a phone.

Can I read it on a Kindle?

Yes. Send the EPUB with Send to Kindle from the app, the website or by email and Amazon delivers it to your Kindle and the Kindle apps.

Is my file uploaded?

No. The PDF is read and converted in your browser and the EPUB is made there too. Only for scanned books is the text recognition model downloaded; your file stays on your device.

Why are some pages still images?

Pictures, charts and drawings are kept as images, as they are in the PDF. If text appears as a picture, it was inside a drawing or a scanned figure. A page that is all scan is recognized as text unless you chose to skip scanned pages.

How long does conversion take?

A text PDF of a few hundred pages takes seconds to a minute. Scanned pages take 2 to 5 seconds each on a computer and several times longer on a phone. The progress bar shows the time left and you can cancel at any moment.

Do I need Calibre?

No. Nothing has to be installed. Calibre is a good library manager, and you can add the EPUB made here to it, but this converter also removes running heads, joins paragraphs and recognizes scans in the browser.

Is it free?

Yes, with no account, no watermark and no page limit.

Updated Works offline once the page has loaded; scanned books need the recognition model once