EPUB to TXT

This tool pulls the readable prose out of an EPUB and hands you a clean .txt file. It fits the moment you need the words without the packaging: feeding a manuscript into a word counter, grammar checker, or readability tool that only accepts plain text; pasting a chapter into a translation or summarization workflow; or quoting passages into notes, a CMS, or an email without dragging along EPUB styling and markup.

For self-publishers and editors it is a fast way to get a proofreading copy or a diffable text snapshot of a draft before another revision pass. Students and researchers can lift a single chapter for citation or annotation. Developers and data folks get a predictable input for indexing, search, or text-analysis scripts, since the output is UTF-8 plain text with no hidden formatting to clean up first.

Drop or select a single .epub file. Everything runs in your browser using JSZip to unpack the archive and the browser's own HTML parser to read each chapter, so nothing is uploaded and nothing remains after you close the tab. The tool follows the EPUB's spine in reading order, skips non-linear and non-HTML items, and strips out navigation, scripts, and styles so only body text survives.

Each chapter appears as a checkbox with its title and word count, so you can include the whole book or pick specific sections. Titles come from the table of contents (nav.xhtml or toc.ncx) when present, falling back to the first heading on the page. Two toggles control the result: include chapter headings (title plus an underline rule) and separate chapters with a --- divider. When you are happy, download the .txt or copy the text straight to your clipboard.

What it does not do: it will not edit the EPUB, keep bold/italic or links, preserve page numbers, run OCR on scanned images, or open DRM-protected files. Files carrying Adobe or similar DRM are detected and refused rather than producing scrambled output.

Because extraction follows the spine and pulls text content directly, anything that lives outside flowing HTML is lost. Text baked into images, complex tables, equations, and footnote layouts may come through flattened, out of order, or not at all. Poetry, code blocks, and verse can lose their line breaks since spacing is normalized into single spaces and collapsed newlines.

Chapter splitting mirrors how the book was built, not how you read it. A publisher who packed many chapters into one HTML file will surface as one big chunk; one who split a chapter across files will show several entries. Front matter, copyright pages, and ad pages count as chapters too, so scan the list before exporting. Untitled sections fall back to the file name or a generic Chapter N label, and very short fragments under about ten characters are dropped entirely.

Two common mistakes: assuming a DRM-protected library book will convert (it will be refused), and expecting MOBI or AZW3 to work here (this tool accepts .epub only).

Open the downloaded file and skim end to end. Confirm the chapters you wanted are present and in the right order, that headings line up with the actual sections, and that no chapter is silently truncated or merged. Spot-check the start and end of a few chapters where split points and front/back matter problems tend to hide, and look for runs of words with collapsed line breaks if the book contained verse or code.

Watch for stray artifacts: page-number remnants, repeated running headers, or footnote markers that landed mid-sentence. If a section you expected is missing, it was likely an image, a non-linear item, or a fragment too short to keep. Compare the on-screen word count against your expectation as a quick sanity signal.

For anything official, financial, or legal, treat the output as a draft and verify it against the original EPUB by eye. Automated text extraction can reorder or omit content without warning, so confirm exact wording and completeness before quoting, filing, or publishing.