DOCX · MD → EPUB20 credits per book — flat, whatever the lengthBeta
Your DOCX or Markdown manuscript → a store-ready EPUB with chapters, table of contents, images, and footnotes intact.
What do you need?
Files are deleted after processing
EXAMPLE
MANUSCRIPT.DOCX — 14 HEADINGS
arrow_forward
BOOK.EPUB + REPORT.TXT
Heading 1: "Prologue" chapter 1
Heading 1: "The Long Road" chapter 2
12 images, 38 footnotes embedded
Structural validation PASSED
What you get
check_circleEPUB 3 with a compatibility toc.ncx — accepted by KDP, Apple Books, and Kobo
check_circleChapters split on your Heading 1/2 styles (or # / ## in Markdown)
check_circleEmbedded images carried into the book, footnotes converted to tappable EPUB footnotes
check_circleStructural validation before delivery — a broken book is never handed to you
check_circlePlain-text conversion report: every chapter, image, footnote, and anything approximated
How it works
1Upload your manuscriptOne DOCX, one Markdown file, or a zip of Markdown plus the images it references.
2Structure is resolvedHeading styles become chapters and sections. If there are none, chapter breaks are detected from candidate lines in the text — and the report says so.
3Download a validated EPUBA store-ready EPUB 3 with TOC, images, and footnotes, plus a conversion report listing everything converted, approximated, or dropped.
Who uses this tool
The same file, different jobs. Tap a prompt to try it.
SELF-PUBLISHING
From Word draft to KDP upload
The DOCX you wrote in becomes a validated EPUB with chapters and TOC — upload it to KDP as-is.
FICTION AUTHORS
Scene breaks that survive
*** separators stay scene breaks inside chapters instead of shattering your book into fragments.
TECHNICAL WRITERS
Markdown book, zipped with images
A zip of Markdown chapters and their screenshots becomes one EPUB with every image embedded.
ACADEMICS & NONFICTION
Footnotes readers can tap
Word footnotes become proper EPUB 3 footnotes with backlinks — not text dumped at the end.
LEGACY MANUSCRIPTS
No heading styles? Still a book
Old typewriter-style drafts with ALL-CAPS chapter lines get their breaks detected from the text.
SMALL PRESSES
A conversion report you can trust
Every chapter, image, and approximation is listed in a report, so nothing is silently dropped.
SECURITY
Built for sensitive documents
Bank statements, medical files, case records. Security isn't a feature we added — it's the foundation.
lock
Encrypted in transit
Encrypted in transit over TLS 1.2+, and stored in access-controlled, encrypted object storage. Files are protected the moment they leave your browser.
admin_panel_settings
Your files stay yours
Workspaces are isolated per account. Role-based access shows teammates only what they need.
auto_delete
Deleted, not stored
Files are deleted after processing. Everything runs on our own hardware and is never sent to an outside AI service, so your data is never used to train models.
dnsProcessed on our own hardwareauto_deleteDeleted after processinglockEncrypted in transit
Questions
The output is a standard EPUB 3 package with a compatibility toc.ncx, and every book is structurally validated before delivery — the container, package file, navigation, manifest, and spine are all checked. If validation fails, the job fails and your credits are refunded instead of handing you a broken file.
From your heading styles first: Heading 1 and Heading 2 in Word, # and ## in Markdown, become chapters and sections deterministically — no AI involved. Only when a manuscript has no usable headings at all do we run a single AI pass over candidate break lines ("Chapter 7", ALL-CAPS lines, *** separators) to place the breaks, and the conversion report tells you whether that pass was used.
Embedded images are carried into the book, and Word or Markdown footnotes become proper EPUB 3 footnotes with backlinks. Tables are flattened to text — EPUB reflow handles complex tables poorly — and anything approximated or dropped is listed in the conversion report so nothing disappears silently.
Yes. Conversion runs on our own servers, your manuscript is never used to train any model, and files are deleted after the job completes. Most conversions use no AI at all — with clear headings the entire pipeline is deterministic code.