You can turn a finished book into an audiobook without booking a studio or hiring a narrator. A synthesized voice can read your manuscript, and for many authors that is enough for a first audio edition, a sample for readers, or an accessibility copy. It is not a perfect substitute for a human performance, and not every store accepts synthesized narration, so this guide covers the whole job: your options, how to prepare the text, how to choose voices, what file formats to export, and how to check your results. The final section walks through it step by step in Quillen.
The short answer
Prepare your text, choose a narrator voice (or a cast, with a different voice per character), listen to a sample, export MP3 files, listen to the start, middle and end of every file, and only then check each store's current rules on synthesized voices. In Quillen the whole process lives under Export, then Audiobook, and the voices run on your own computer, so your manuscript is not uploaded anywhere.
What you need
- A finished, edited manuscript. Fix typos before you narrate, because every change after export means re-exporting.
- Time. Making the audio takes about as long as the audiobook lasts, and longer on a slower computer. A 9-hour book is a long job, so plan to leave your computer running.
- Disk space. MP3 files are fairly small, but WAV files are much larger.
- Headphones for the listening check.
Your options for making an audiobook
There are three common routes, each with trade-offs. None is right for every book.
| Route | Strengths | Trade-offs |
|---|---|---|
| Human narrator (hired or yourself) | Most natural performance; accepted everywhere; can convey humor, accents and emotion | Cost or many hours of recording and editing; scheduling |
| Studio production through a publisher or service | Professional result with quality control | Cost, contracts and rights to read carefully |
| Synthesized voice on your own computer | Fast, private, costs nothing extra, easy to redo after edits | Less expressive; some stores restrict or label it; names and unusual words may need checking |
Many authors use synthesized narration for samples, for an accessibility edition, for non-fiction where the text is the point, or as a way to hear their own book before paying a narrator.
Step 1: Prepare the text
A reader's eye forgives things a listener's ear does not. Before you export, read your manuscript with audio in mind:
- Numbers and years. Synthesized voices usually spell numbers out. Years are often read as years (for example, 1984 as nineteen eighty-four), but check any figure that could be read two ways.
- Abbreviations and symbols. "St." can be Street or Saint, and symbols such as % or & may be read oddly. Spell out anything ambiguous.
- Names and invented words. A voice sounds out words it does not know. A fantasy name like Orrin may not come out as you intend, so listen to it in a sample.
- Visual-only features. Tables, images and layout tricks do not translate to audio. Describe in words what the listener needs.
- Chapter headings. Decide whether you want the title read out. Titles and chapter breaks get longer pauses than ordinary paragraphs.
Step 2: Choose single narrator or a cast
A single narrator reads everything in one voice. It is simple, consistent and quick. A cast of voices gives the narrator one voice and each character another, so dialogue sounds like a conversation. A cast takes more work: you need to mark who speaks each line, and your characters' voices need to be different enough to tell apart. Start with a single narrator, and move to a cast if your book is heavy on dialogue and you are willing to do the marking.
Step 3: Listen to a sample before you commit
Always hear a sample of each voice before you export a whole book. Listen for pace, for how the voice handles questions and exclamations, and for names. If a voice does not suit your genre, try another. Choosing the voice is the single most important decision for how the finished book feels.
Step 4: Choose the file format and layout
Audiobook stores mostly want MP3 at a fixed quality, because it is small and widely supported. WAV is uncompressed, so the files are far larger, but it is useful if you plan to edit the audio further. You also choose between one file for the whole book and one file per chapter. Many stores expect separate chapter files, so check what yours asks for. Also decide whether to include a short sample file, which many stores want for the listing page.
Step 5: Listen and check
Do not assume the files are fine. For each file, listen to the opening, a section from the middle, and the ending. Check that chapter titles are read as intended, that pauses feel natural, that no passage is skipped or repeated, and that names sound acceptable. Fix any problems in the manuscript and re-export only the affected chapters if you made one file per chapter.
Step 6: Check the store's rules before you upload
Policies on synthesized narration differ between audiobook stores and distributors, they have changed over time, and some stores run their own programs for synthetic voices. Some require disclosure, some restrict it to certain programs, and some do not accept it at all. Before you upload anywhere, read that store's current content and audio requirements, and if you are asked whether the narration is synthesized, say so. Do not rely on a blog post, including this one, for a store's current rules: go to the store's own publisher pages.
Common mistakes
| Mistake | What happens | Fix |
|---|---|---|
| Exporting before the text is final | You have to regenerate the whole audio after every edit | Edit first; export last |
| Not listening to a sample | Wrong voice for the genre, or names mangled | Hear every voice and a passage with your hardest names |
| Skipping the listening check | Errors reach listeners and reviews | Listen to the start, middle and end of each file |
| Ignoring store rules | Rejected upload or a book pulled later | Read the current policy and disclose synthesized narration where required |
| A cast of voices that sound alike | Confusing dialogue | Pick contrasting voices and test them in one scene |
| Choosing WAV when the store wants MP3 | Huge files and rejected formats | Check the store's file specification first |
Do it in Quillen
Quillen has an audiobook export that reads your book with voices that run on your computer. It handles the pauses between paragraphs, headings, scene breaks and chapters for you.
- Finish editing your book, then click Export in the top bar and choose Audiobook.
- Choose a Reading style. Single narrator means one voice reads the whole book. Cast of voices gives the narrator a voice and each character one; see our guide to casting voices when you are ready for that. File choices are covered in how to export audiobook files.
- Pick the Voice from the drop-down, and press the play button beside it to hear a sample. Try several. Your choice is remembered for the book.
- Look at the preview. The center of the window shows your chapter as it will be read, and the Chapters list on the right lets you tick which chapters to include. Use Previous and Next to move between chapters.
- Set up the Files. Tick Begin with the title and author if you want an opening announcement. Choose One file for the whole book or One file per chapter (into a new folder). Choose MP3 (192 kbps, 44.1 kHz, mono, which is what audiobook stores ask for) or WAV (uncompressed, much larger). Tick Also save a 2-minute sample if you want a listing sample.
- Read the estimate. The bottom of the left panel shows the audio length and a reminder that making it takes about as long as it lasts, and longer on slower computers. In our test, 96 words came to roughly 36 seconds of audio.
- Read the notice above it, which reminds you that not every audiobook store accepts books read by an AI voice and that the rules change.
- Click Make audiobook and choose where to save. Quillen shows its progress. Keep your computer awake until it finishes.
- Listen to the results using the checks above.
One more thing worth knowing: Quillen reads your manuscript text as written and the window has no pronunciation editor. If a name sounds wrong in the sample, the practical fix is to make a copy of the book for audio and spell the name phonetically there, leaving your real manuscript untouched.
Checklist
- Manuscript edited and final.
- Numbers, abbreviations and unusual names checked in a sample.
- Single narrator or cast chosen, and voices tested.
- File format and layout match the store's specification.
- Start, middle and end of every file listened to.
- Current store policy on synthesized voices read, and disclosure made where needed.
- Original manuscript backed up.
Common questions
Can I make an audiobook without hiring a narrator?
Yes. A synthesized voice can read your manuscript. It is quicker and cheaper than a human narrator, though less expressive, and not every store accepts it. Quillen's audiobook export reads your book with voices that run on your own computer.
What file format should an audiobook be?
Most audiobook stores ask for MP3 at a fixed quality. Quillen exports MP3 at 192 kbps, 44.1 kHz, mono, or uncompressed WAV. Check your store's current file specification, including whether it wants one file per chapter.
How long does it take to make an audiobook in Quillen?
About as long as the audiobook lasts, and longer on slower computers. Quillen shows an estimate before you start. A 9-hour book is a long job, so leave your computer running.
Do audiobook stores accept AI-narrated books?
It depends on the store, and the rules change. Some require disclosure, some restrict synthesized narration to specific programs and some do not accept it. Read each store's current policy before uploading, and disclose synthesized narration if asked.
Is my manuscript uploaded anywhere when I make an audiobook in Quillen?
No. The voices run on your own computer, so the text and the audio stay on your machine until you choose to upload the files yourself.