Audio is the fastest growing format, and most authors skip it.
An audiobook is not a recording of someone reading your book. It is a separate production with its own standards, and the gap between those two things is where most self-made audiobooks fail.
A meaningful share of readers will never open your book. They will listen to it while driving, walking, or doing something else entirely. If there is no audio edition, those people are not a lost sale so much as an audience that never knew the book existed.
Audio is also the format authors most often postpone, usually because the production process is opaque. Here is what actually happens, and what separates an audiobook a retailer will accept from a recording it will reject.
Narration is not reading aloud
This is the misconception that produces most failed self-recorded audiobooks. Reading aloud is what you do to a child at bedtime. Narration is a performance with sustained consistency across many hours.
A narrator holds pace, tone, and energy steady across sessions recorded on different days, in different moods, possibly weeks apart. They differentiate speakers without slipping into caricature. They make a paragraph of exposition listenable when the reader cannot skim it. And they do all of that while producing audio clean enough to pass a technical review.
What actually happens in production
The process has more stages than most authors expect, and the recording itself is the shortest of them.
- Manuscript preparation. The text is marked up for pronunciation, names, emphasis, and anything ambiguous when spoken. Numbers, abbreviations, and foreign words all need decisions made in advance.
- Voice selection. Matching a voice to the book. A memoir, a thriller, and a business book want genuinely different qualities.
- Recording. In a treated room, in sessions, with the narrator re-recording lines as they go.
- Editing. Removing breaths, mouth noise, false starts, and page turns. This takes considerably longer than the recording did.
- Mastering. Bringing the audio to the technical specification retailers require.
- Quality control. A full listen against the manuscript to catch skipped or misread words.
That final pass is the one most often skipped by cheap production, and it is the one listeners notice. A dropped sentence in chapter nine is the kind of error that generates a refund.
The technical specification, in plain terms
Retailers reject audio on measurable grounds. These are the numbers that decide acceptance, and they are worth knowing even if someone else is handling the production.
| Requirement | Typical standard | What goes wrong |
|---|---|---|
| Format | 192kbps or higher MP3, 44.1kHz | Low bitrate audio sounds thin and is rejected |
| Channels | Mono is standard for narration | Stereo doubles file size for no benefit |
| Peak level | Below −3dB | Clipping is audible and cannot be repaired |
| Noise floor | Below −60dB | Room hum and air conditioning fail this |
| File structure | One file per chapter | A single long file is rejected outright |
| Room tone | Brief silence at head and tail | Abrupt starts sound unprofessional |
The noise floor requirement is the one that defeats home recording. A quiet room is not the same as a treated room, and a laptop microphone in a bedroom will not reach −60dB no matter how still you sit.
Choosing a voice
Voice selection is the decision with the largest effect on how the finished audiobook is received, and it is usually made too quickly. The instinct is to pick the most pleasant voice available. The better question is which voice suits the book.
A thriller wants restraint and control, because tension comes from what the narrator withholds. Business and self-development want warmth and authority, close to how a good speaker sounds from a stage. Literary fiction usually wants a narrator who gets out of the way entirely. Memoir wants intimacy, as though the story is being told to one person rather than performed to a room.
Two practical considerations get overlooked. First, accent and regional voice should match the book's setting and narrator, and a mismatch is distracting in a way listeners find hard to articulate but consistently mention. Second, if you are writing a series, the same narrator should carry all of it, so availability across future books is worth confirming before you commit to the first one.
Should you narrate it yourself?
Sometimes yes. Memoir and personal-story nonfiction genuinely benefit from the author's own voice, and listeners forgive a lot when the voice is the point. Novels and business books usually do not.
Narrate it yourself only if all of these are true
- The book is memoir or first-person nonfiction where your voice is part of the value
- You can access a treated recording space, not simply a quiet room
- You can commit to consistent sessions across days without your energy visibly shifting
- You are comfortable re-recording the same paragraph repeatedly without frustration
- You have someone else doing the editing, mastering, and quality control
If more than one of those is false, a professional narrator produces a better result for less total effort. Our audiobook production service covers narration through mastering and quality control.
Preparing your manuscript for narration
A manuscript that reads perfectly on the page can be awkward to speak, and the problems are predictable. Catching them before recording is far cheaper than fixing them after.
What to resolve before a narrator opens the file
- Pronunciation of every proper noun. Names of people, places, brands, and anything invented. Supply a list with phonetic spellings rather than leaving it to guesswork.
- Numbers and dates. Is 1984 nineteen eighty-four or one thousand nine hundred eighty-four? Is 1.5M read as one point five million or a million and a half? Decide once and note it.
- Abbreviations and acronyms. Some are spoken as letters, some as words, and some are read out in full.
- Foreign words and phrases. Confirm the pronunciation you want, particularly for Spanish, which listeners will notice if it is mangled.
- Visual-only elements. Tables, charts, footnotes, and URLs do not survive being read aloud and need a spoken equivalent or removal.
- Scene breaks and section shifts. These are visual on the page and need an audible signal in narration.
- Anything that relies on spelling. Wordplay and puns that work on the page can disappear entirely in audio.
Footnotes deserve particular attention. In print they are optional and easy to skip. Read aloud, they interrupt the sentence they belong to. The usual solutions are to fold essential ones into the main text and drop the rest, and that is a decision only the author can make.
What listeners actually complain about
Audiobook reviews are unusually specific, and the complaints cluster tightly. Almost none of them concern the writing.
- Inconsistent volume between chapters, which forces the listener to keep adjusting and is the single most common technical complaint
- Audible edits, where the tone or room sound changes mid-paragraph because two sessions were spliced
- Mouth noise and breath, which is unnoticeable in conversation and grating over eight hours through headphones
- Character voices that drift, so a character sounds different in chapter twelve than in chapter two
- Pace that never varies, which flattens tension and makes long stretches hard to follow
- Mispronounced names, especially of real places, which readers who know the place find genuinely distracting
Every item on that list is a production issue rather than a writing issue, which is precisely why audio is a separate discipline. A well-written book with poor production reviews as a bad audiobook, and that review sits on the same product page as the print edition.
Where the finished audiobook is sold
The files are prepared for Audible, Apple Books, and Google Books, alongside your ebook and print editions on Kindle and Barnes & Noble. Distribution breadth matters in audio more than in print, because listening habits are strongly tied to whichever ecosystem someone already uses. A listener with an Apple subscription is not going to switch platforms for one title.
What it costs and how long it takes
Audiobook production is $0.045 per word of your manuscript. Pricing per word rather than per finished hour means you know the cost before production begins, instead of receiving an estimate that moves once the narrator has recorded.
Turnaround is about 2 weeks from payment and receipt of your final manuscript. It must be final. Changing text after recording means re-recording, and matching a re-recorded line to surrounding audio is harder than recording the chapter again.
You can see the exact figure for your word count in the audiobook calculator, and the rate is published on the pricing page before you talk to anyone. Barker Publishing House has been producing audiobooks since 2018, and you keep 100 percent of your rights and royalties.
One sequencing point worth flagging: audio should come after editing and after the text is locked, for the same reason formatting does. If you are planning audio alongside other services, bundling reduces the total by 40 percent.
Audiobook questions, answered.
Ready to reach the readers who never open a book?
Talk to a real publisher first, or see your exact price in the calculator.
More on getting your book taken seriously.
Where to sell your audiobook
Audible, Apple Books, Google Books and libraries compared, and why listening habits are locked to ecosystems.
How audiobooks are priced
Per finished hour versus per word, why one shifts the risk to you, royalty share explained, and what it actually costs.
How long does an audiobook take to produce?
Why finished hours are a fraction of studio hours, what each stage consumes, and how to plan a date that holds.