Audiobooks: how they are made, sold and borrowed
The audiobook went from a niche product on cassette to a format many people use every day. How a recording gets made, who earns what, how libraries lend them, and the arrival of synthetic voices.
From talking books to phones
Recorded books are almost as old as sound recording, and their first serious purpose was access. In the United States, the service now run by the Library of Congress as the National Library Service for the Blind and Print Disabled was created in 1931, and it began distributing recorded "talking books" a few years later. Similar services exist in many countries, and for readers who cannot use print they remain essential.
Commercial audiobooks followed on vinyl, cassette and CD, usually abridged to fit on a manageable number of discs and sold at high prices. Digital downloads removed the physical limit, which is one reason unabridged recordings became the norm. The smartphone did the rest: once everyone carried a player, a dedicated app and a pair of headphones, listening to a book on the commute, at the gym or while cooking became ordinary.
Industry groups such as the Audio Publishers Association in the United States run regular surveys of listeners and publishers, and those surveys have reported sustained growth in audiobook listening over many years. As with all industry statistics, it is worth reading how a survey was done before quoting its numbers.
How an audiobook is made
A professionally produced audiobook involves more people than most listeners imagine.
- The rights holder, usually the publisher or, for a self-published book, the author, decides to make an audio edition. Audio rights are often negotiated separately from print and e-book rights in a publishing contract, and an agent may sell them to a specialist audio publisher.
- Casting. The narrator matters enormously. Listeners often choose or abandon a book because of the voice. Some authors narrate their own work, which tends to suit memoir and personal non-fiction; fiction is more often read by professional narrators, some of whom are actors.
- Preparation. A good narrator reads the whole book first, notes characters and accents, and researches pronunciations of names and places.
- Recording. The book is recorded in a booth, often with a director or producer following the text to catch misreadings and keep the performance consistent.
- Editing, proofing and mastering. Engineers remove mistakes and noises, a proofer listens against the text, and the files are mastered to the technical specifications retailers require.
Narrators are commonly paid either a fee per finished hour of audio, or a share of the royalties, or a combination. Because a long novel can run to many hours, a full-cast or high-profile production represents a real investment, which is why not every book gets an audio edition.
Self-published audiobooks
Independent authors can produce audiobooks too. The main routes are:
- Marketplaces that match authors with narrators, such as Amazon's ACX, where a rights holder posts an audition script and a narrator or producer records the book for a fee, a royalty share, or both. ACX distributes to Audible and Amazon among other outlets.
- Independent distributors that take finished audio files and distribute them to many retailers and library platforms, often without exclusivity.
- Hiring a producer or studio directly and then choosing distribution separately.
The trade-offs mirror those in self-publishing generally: exclusivity with one large retailer can bring higher royalty rates or more visibility, while non-exclusive distribution reaches more stores and libraries. Read the term and rights clauses carefully before signing anything that grants exclusive audio distribution.
How audiobooks are sold
Audiobooks reach listeners through several models, and the same book may be available through all of them:
- Credit-based memberships, where a monthly fee buys a credit that can be exchanged for a title of the listener's choice, plus discounts on others.
- Retail purchase of individual titles, at prices that are often higher than for the printed book, reflecting the cost of production.
- Listening hours within subscriptions, where a music or reading subscription includes a monthly allowance of audiobook listening.
- Unlimited or catalogue subscriptions, common in some European markets.
How authors and publishers are paid differs across these models, and the move toward subscription listening has caused the same arguments over per-use payments that music streaming caused for musicians. If you are an author, check how each outlet calculates royalties.
Borrowing audiobooks from libraries
Most public libraries now lend digital audiobooks through apps such as Libby, using the same kinds of licences that govern library e-books. Those licences can be expensive and time-limited, so popular titles often have long waiting lists. Our article on libraries and e-books explains how the licensing works. For readers with print disabilities, national accessible-reading services offer free recorded books in formats designed for them.
Synthetic narration
The newest change in audio is synthetic voices. Text-to-speech has existed for decades, mostly as an accessibility tool, but voices generated by machine learning now sound far more natural. Several large retailers and distributors have introduced programmes that let publishers and authors produce audiobooks narrated by synthetic voices, at a small fraction of the cost of a human recording.
The arguments on each side are real.
- For: many books will never earn enough to pay for a human narrator, so synthetic narration makes audio editions possible for backlist titles, niche non-fiction, academic works and small presses. That widens access, including for listeners who depend on audio.
- Against: narration is a performance, not just pronunciation, and synthetic voices still struggle with emotion, comic timing, dialogue and unusual names. Narrators worry about their livelihoods, and about their own voices being used to train or clone synthetic ones without consent or payment, which has become a subject of contract negotiation in the industry.
For listeners, the practical advice is to check the listing. Retailers that carry synthetic narration generally label it, and samples let you hear the voice before buying. For more on the wider questions, see AI and books.
Does listening count as reading?
This argument never seems to end, and it is mostly about status rather than substance. Listening and reading are different activities: listening is paced by the narrator, makes it harder to skim or reread, and shares attention with whatever else you are doing. Reading print allows you to set your own pace and look back. Studies that compare comprehension after listening and after reading have produced mixed results, and where differences appear they tend to depend on the difficulty of the material and on whether the reader can control the pace and go back over the text. Our article on reading on screens versus paper discusses the comparable research on screens.
But for many people the choice is not between listening and reading; it is between listening and not encountering the book at all. Audiobooks have brought long-form writing to people who would not otherwise have time for it, and to people for whom print is difficult or impossible. That is a gain for books, whatever one calls it.
Sources
- Library of Congress, National Library Service for the Blind and Print Disabled (history and service).
- Audio Publishers Association (industry surveys).
- ACX (Audiobook Creation Exchange).
- OverDrive, Libby.