Rosalind
English · Narration
Markdown to audiobook for writers in Obsidian, iA Writer or any plain-text editor. Upload the .md file: # and ## headings become chapters, asterisks, links and front matter are cleaned out, and every character gets a voice.
manuscript.md
Markdown
Reading your file…
Lewis Carroll
Alice's Adventures in Wonderland
Alice asks the grinning Cheshire Cat for directions — and gets none.
The Cat only grinned when it saw Alice. It looked good-natured, she thought; still it had very long claws and a great many teeth, so she felt it ought to be treated with respect.
Step-by-step guide
Free to start: no card, nothing to install.
Markdown to Audiobook
What you get, from manuscript to finished audiobook.
Use cases
Where authors and storytellers put it to work.



Demo performances. In the studio you cast from standard and premium voice lists.
English · Narration
English · Narration
English · Narration
English · Narration
English · Narration
In depth
Plenty of fiction now starts life as Markdown: chapters as notes in Obsidian, a manuscript in iA Writer, or a folder of .md files under version control. Markdown to audiobook is the step that keeps that workflow intact. Upload the .md file and the headings you already wrote become chapters, while the symbols that format the text on screen stay out of the narration.
From there FullCastBook does what it does for any manuscript: it splits narration from dialogue, works out who speaks each line, and gives the narrator and every character a voice. The demo above plays a full-cast scene.
Chapters come from your headings. A line that starts with # or ## opens a new chapter; its text, without the hashes, becomes the chapter title, which the narrator reads at the start of the chapter and the M4B shows as a marker. Everything up to the next heading belongs to that chapter.
So use those headings for chapters and parts only. If your draft marks scenes inside a chapter with ## headings, each scene would become a chapter of its own; turn them into plain lines or remove them before you upload. A file with no headings at all is cut into even parts, which is the cue to add them.
Take a note that opens with front matter (a title and two tags between --- lines), then the heading "# Chapter 1: The Letter". Below it: The envelope had *no* stamp. "Who brought this?" Mara asked. "Nobody," said Tomas. "It was under the door." A link to a map note follows, then an embedded sketch of the lighthouse.
Here is what you hear. The narrator announces "Chapter 1: The Letter" and reads "The envelope had no stamp" with no asterisks. Mara's question comes in her voice and Tomas's answer in his. The link is read as its words alone, the sketch is skipped, and the front matter isn't read at all. If Mara's line should be hushed, add [whisper, like sharing a secret] to it in the studio and regenerate only that line.
Markdown is formatting for the eye, which is why Markdown text to speech needs a clean-up step. Fed straight to a text-to-speech engine, a heading can come out as "hash Chapter 1" and bold text as a string of asterisks. The importer strips those marks first: heading hashes go, and so do the asterisks and underscores around emphasis, while the words stay where they were.
Emphasis isn't performed from the asterisks, though. If a word really needs weight, say so with a direction on that line in the studio. That's also the place for sound tags such as [sigh], [laugh] or [breathe].
YAML front matter, the block between two --- lines at the top of a note where Obsidian, Jekyll or Hugo keep the title, tags and dates, is removed before narration. A link keeps its words and loses its address, so a sentence that links to another note reads as if the link weren't there.
Images are dropped, alt text included. If a picture carries something the listener needs, a map or a letter, describe it in the prose.
Markdown to audiobook works either way. A single manuscript.md with a heading per chapter imports in one go. If each chapter is its own note, upload the first one to create the project, then add the next files to the same project with Add chapters; they join in the order you add them. For a long book, joining the notes into one file in reading order first is quicker.
Short pieces can go through the paste box as well, but for Markdown with headings, links and front matter, uploading the .md file is the sure route. For a TXT manuscript, see text to audiobook; for Word files, Word to audiobook uses your heading styles the same way.
The last step of Markdown to audiobook is the export. When the chapters sound right, download one MP3 of the whole book, a ZIP with an MP3 per chapter, or an M4B where every heading is a chapter marker. That makes this an MD to MP3 route that keeps your structure, not one long file with no breaks.
Keep the .md files as your source: if you fix a typo, fix it there and in the studio, and regenerate just that line. For a line or two of plain speech, the text-to-speech playground on the home page is quicker. New accounts get free credits for a short sample chapter, and plans are on the pricing page.
Yes. The studio accepts .md files next to EPUB, DOCX, PDF and TXT. The Markdown is cleaned up before narration and the headings become chapters.
Yes. A line starting with # or ## starts a new chapter, and the heading text becomes the chapter title. Use those headings only for chapters and parts.
No. Heading hashes and emphasis marks are stripped, links are read as their words without the address, images are dropped and front matter is removed.
Yes, one at a time: create the project with the first file, then add the others to the same project with Add chapters. For a long book, joining them into one file in reading order is faster.
Yes. Obsidian keeps every note as a .md file and iA Writer saves Markdown, so you can upload those files as they are.
You can try it free: new accounts get free credits, with no card, enough for a short sample chapter. A whole novel needs a paid plan; see pricing.
Markdown to audiobook in one upload: check the chapters and voices, then export MP3 or M4B. Free to start.