Turn an Ulysses Project Into an Audiobook

Ulysses is a favorite writing app for a lot of Mac and iOS novelists, and for good reason: distraction-free sheets, a clean library, and reliable export. Once the manuscript is done, though, a lot of writers want a second format for it: an audiobook. The good news is that you do not need any special integration to get there. Ulysses already exports the two file types AudioProducer reads, so the whole path is a short one. This guide walks it end to end.

One thing to set straight up front, because it saves confusion later: AudioProducer produces a finished audio file that you download. We render your text into a narrated MP3. We do not distribute or publish it for you. Where the audiobook ends up after that (your own site, a retailer, a podcast feed, a file you hand to beta readers) is entirely your call and happens outside our tool. Keep that mental model and the rest is simple.

Exporting from Ulysses (Markdown or DOCX)

There is no native Ulysses integration, and you do not need one. AudioProducer imports plain Markdown and Word (DOCX) files, and Ulysses exports both cleanly. Pick a sheet or a group in your library, then use File > Export (or the export icon in the toolbar) and choose one of these:

• Markdown / Plain Text. This is the lightest option and the one we recommend. Ulysses writes plain Markdown, so what you get is your prose with simple heading and emphasis marks and nothing else to clean up.
• DOCX (Word). If you would rather hand off a formatted document, export to DOCX. It carries your chapter headings as real Word headings, which import fine.

Both formats work. If you are unsure, export Markdown first; it is the least likely to bring along stray formatting.

Cleaning up sheets and groups for narration

Ulysses lets you keep a book as many small sheets inside a group, which is great for writing and slightly awkward for narration. A narrator reads top to bottom, so the order and boundaries of your text matter more than they did while drafting. Before you export, spend a few minutes here:

• Order the sheets the way the book should be heard. Drag them into reading order inside the group so the export concatenates cleanly.
• Merge or separate chapters deliberately. If each sheet is a chapter, exporting the whole group gives you one continuous file. If you want chapters as distinct audio sections later, keep clear heading lines between them.
• Strip out anything that should not be read aloud. Author notes, keywords, comment sheets, and to-do markers are useful while writing but confusing in narration. Remove them from the exported version or keep them in a separate group.
• Spell out or fix tricky names. A neural voice reads what it sees. If a character name or an invented place gets mispronounced, you can adjust it after import, so make a quick list of the ones you expect to need attention.

A clean, ordered export is the single biggest thing you can do to make the narration sound right on the first pass.

Importing into AudioProducer

Open AudioProducer and start a new project, then upload the file you exported from Ulysses. We read the Markdown or DOCX, keep your paragraph and chapter structure, and lay the text out so you can see it the way it will be narrated. You can sign up and try this on the free tier: 1,200 words, no card required, which is enough to run a chapter or two through the full flow and hear how it sounds before you commit. Paid plans start from $39.99 per month when you are ready for a whole book.

Once the text is in, read through it once inside the project. This is the moment to catch anything the export dragged along and to confirm the chapters landed where you expected. Small fixes here are cheap; fixing them after you have generated audio means regenerating.

Assigning character voices

If your book is mostly narration, you can assign a single voice and move on. Where AudioProducer earns its keep is dialogue-heavy fiction: you can route different characters to different voices so a conversation actually sounds like two people. Pick a voice from the library for the narrator, then assign voices to the speakers you care about most. You do not have to voice every minor character; the leads carry the scene.

A note on voice cloning, since writers ask: cloning a specific voice requires consent. You can clone your own voice, or a voice you have explicit permission to use, and nothing else. For most books a library voice is the faster and simpler choice anyway.

Exporting your MP3

When the text reads the way you want and the voices are assigned, generate the audio and export it as an MP3. That file is yours to download and keep. From there you publish it wherever you already publish: upload it to your store page, hand it to your beta readers, add it to a members area, or send it to a distributor yourself. AudioProducer stops at the finished file on purpose, so you keep full control of where the audiobook goes and how it is listed.

If something sounds off after you listen back, a mispronounced name or a chapter you want repaced, go back into the project, make the edit, and regenerate that part. The loop is quick, and you only touch what needs touching.

FAQ

That is the whole path: export from Ulysses, tidy the sheets, import, assign voices, and download your MP3. Because Ulysses already gives you clean Markdown or DOCX, the handoff is the easy part. For the bigger picture, see our guide on how to make an audiobook with AI, and if you export a different way, we have companion walkthroughs for converting Markdown to an audiobook and converting a Word manuscript to an audiobook. Writing in a different tool? Here is how to turn a Sudowrite draft into an audiobook or turn your novel into an audiobook.