Create EPUB turns a manuscript into a valid EPUB 3 ebook in a few clicks. Paste or type your text, or upload one or more Word (DOCX), Markdown, plain text or HTML files, then fill in the title, author and language. The tool splits the content into chapters at the headings you choose, builds the table of contents automatically, and packages everything with a cover: either an image you upload or a simple generated cover with your title and author. In plain text, lines such as Chapter 1, Prologue or Part Two are recognised as chapter headings, and the same works for Serbian, Croatian, German, French, Spanish and other languages. Word documents keep their headings, bold and italic text, lists, tables and embedded images. A live chapter list shows exactly how the book will be divided before you download it. Optional details such as publisher, description, tags and series are written into the metadata so the book sorts correctly in Calibre, Apple Books and other library apps. The finished file includes both an EPUB 3 navigation document and an EPUB 2 table of contents, so older readers open it too. Check the result in EPUB Reader, fix details later with EPUB Metadata Editor, or turn the same book into a printable file with EPUB to PDF Converter.
A valid EPUB has a few strict rules that many hand-made files break. The archive must start with an uncompressed file named mimetype containing exactly application/epub+zip, so that readers can identify the format without unpacking it. A META-INF/container.xml file must point to the package document, and that package must declare a unique identifier, a title, a language and a modified date, list every file in its manifest with the correct media type, and define the reading order in its spine. EPUB 3 also requires a navigation document marked as the table of contents. This tool generates all of that for you and writes each chapter as well formed XHTML, the stricter XML flavour of HTML that ebook readers expect. Because XHTML is unforgiving, anything that would break a reader is cleaned out: scripts, forms, embedded frames and unknown tags are removed, while headings, paragraphs, emphasis, lists, quotes, code, tables and images stay. Markdown is converted with the standard rules, so a single # makes a top level heading, ** makes bold text, and a line starting with > becomes a quote. Plain text is treated as paragraphs separated by blank lines, with wrapped lines inside a paragraph joined back together, which suits text exported from word processors and Project Gutenberg files. The language setting matters more than it looks: readers use it to pick hyphenation rules and the correct voice for text to speech, so set it to the language the book is written in.