Converting prose is the step most format guides leave out. A full-cast audiobook is an audiobook in which a narrator reads the narration and each character speaks their own lines in a separate voice, and it needs no conversion at all; AudioForge is a production tool that turns a manuscript into a multi-voice audiobook, with a distinct voice for every character.

Below: the parts of a script, the main conventions, a worked conversion of a short passage, and when you need a drama script at all.

What does an audio drama script look like?

An audio drama script looks like a play broken into cues, where every cue is a speech, a sound or a piece of music.

Radio playwright and director Tony Palermo's radio play format guide at RuyaSonic shows the old-time-radio style he uses:

• Scene heading: a film-style "slug," his example being SCENE ONE - EXT. SPANISH GALLEON - DAY (1720).
• Numbered cues: every speech, sound and music cue in his sample script has a number.
• Speeches: the character's name in capitals, a colon, then the line. A delivery note, a short instruction on how to say the line, goes in parentheses, such as (LAUGHING) or (DISTANT).
• Sound and music cues: each on its own line, labeled SOUND: or MUSIC:.

The UK's Wireless Theatre Company gives similar advice in its guide to formatting an audio drama script: number each new speech cue, starting from 1 on each page, mark sound effects clearly even though they need no number, start each new scene on a new page, and never split a speech across two pages.

What are the main audio drama script format conventions?

The main audio drama script conventions help actors and the editor find things fast: capitals for names and cues, numbers for order, and short notes for how a line is spoken.

The mic notes come from BBC radio drama practice, as described in Final Draft's guide to formatting a podcast script: (OFF) for speaking away from the microphone, (CLOSE) for close to it, and (D) for a distorted voice on a phone or radio.

Styles vary between producers. Pick one and use it the same way on every page.

How do you turn a prose chapter into a script?

You turn a prose chapter into an audio drama script by sorting every sentence into one of three jobs: something a listener hears, something a character says, or something only a narrator can tell.

Here is a short passage of prose:

Three days earlier, the letter had arrived. Now rain drummed on the tavern shutters. Mara slid into the corner booth, keeping her hood up. "You're late," Tobin said, not looking up from his cards. She dropped the sealed letter on the table. "And you're still cheating."

And the same passage as an audio drama script:

SCENE TWO - INT. TAVERN - NIGHT

1. NARRATOR: Three days after the letter came, Mara went to find Tobin.
2. SOUND: RAIN ON SHUTTERS. TAVERN MURMUR UNDER.
3. SOUND: A BENCH CREAKS AS SOMEONE SITS.
4. TOBIN: (NOT LOOKING UP) You're late, Mara.
5. SOUND: PAPER SLAPS ON WOOD.
6. MARA: (LOW) And you're still cheating, Tobin.
7. SOUND: CARDS SHUFFLING STOPS.

Five changes did the work:

1. Setting became sound. The rain and the tavern are now cues, not description.
2. Tags disappeared. "Tobin said" is gone because the listener hears Tobin's voice.
3. Names moved into dialogue. The hood and the booth are lost, so the characters name each other to tell the listener who is there.
4. Action became a sound. The dropped letter is a slap of paper; the cards stopping says Tobin is listening now.
5. Only the time jump stayed with a narrator. Sound and dialogue cannot say "three days later" on their own.

Emily Inkpen's guide to converting prose to script makes the same points: there is no point describing visuals, and repeating characters' names in dialogue is a hallmark of audio. For a whole chapter, work scene by scene with the same three jobs.

Do you need an audio drama script, or a full-cast audiobook?

You need an audio drama script only if you want to rewrite your book as a performance; a full-cast audiobook keeps your prose and needs no rewrite.

An audio drama script and a full-cast script handle the same passage differently:

For a full-cast audiobook, AudioForge builds the script for you. It parses a chapter into a screenplay-style script with every line tagged by speaker and delivery, flags lines it was unsure about, and lets you change any speaker, delivery or word before you generate audio. Turning a chapter into a full-cast audiobook walks through the steps, and dialogue tags that cast cleanly shows how to write prose that parses well.

The bottom line

An audio drama script uses scene headings, cues, character names in capitals and marked sound and music cues, and converting a chapter means turning narration into sound, dialogue or a short narrator line. If you would rather keep your prose as written, a full-cast audiobook needs no rewrite. To see your own chapter as a script, upload it to AudioForge, free during early access, and check how each line is tagged, following the help center.

Both formats give listeners two distinct voices for the leads. AudioForge is a production tool that turns a manuscript into a multi-voice audiobook, with a distinct voice for every character, which is a third way to get there.

Below: what each format means, how they compare, what two narrators cost and how ACX handles them, and how to choose from your manuscript.

What is dual narration?

Dual narration is a format in which two narrators trade off by chapter or point of view. Audible Blog writers define it that way in an article on dual and duet narration, posted to ACX's blog in August 2025.

A romance told in alternating chapters, for example, gives the heroine's chapters to one narrator and the hero's to another. In most dual productions, each narrator reads everything in their own chapters, including the other lead's lines, so the heroine's narrator also voices the hero when he speaks in her chapters.

Some listeners find that the weak spot: a narrator voicing the opposite lead can sound strained, and the character sounds different depending on whose chapter it is.

What is duet narration?

Duet narration is a format in which two narrators perform the main characters' voices throughout the story, often alternating during dialogue. The same Audible article uses that definition.

In a duet, the hero always sounds like the hero. His narrator voices his lines in every chapter, and the heroine's narrator voices hers, so a conversation becomes two performers speaking to each other, whether they record together or separately and are edited together.

Duet takes more coordination. Each narrator needs to know how the other plays every shared character, and their lines have to be joined into one scene without seams.

How do dual and duet narration compare?

Dual and duet narration differ in who voices the dialogue: dual splits the book by chapter, duet splits it by character.

A full cast, or multicast, is a production in which every speaking role has its own voice. Our post on why AI audiobooks sound flat explains why distinct voices matter most in crowded scenes.

What do two narrators cost, and how does ACX handle them?

Two narrators raise the cost or the coordination, and ACX contracts support one rights holder and one producer, according to ACX's help page dated June 2025.

PFH, short for per finished hour, is the rate a narrator charges for each hour of the final, edited audiobook. Author's Republic's guide to per-finished-hour rates, revised October 2025, puts beginners at about $100 PFH, experienced narrators at about $200 and experts at $300 to $1,000 or more. A 90,000-word novel runs about 9.7 finished hours at the 9,300 words per hour that guide uses.

For that book, one narrator at $200 PFH costs about $1,940 at those October 2025 rates. With two narrators, the total depends on how they charge:

• Each paid for the hours they voice: a dual split near half and half can come close to the single-narrator total, about $1,940.
• Each charging a full-book rate or a minimum: the total can approach double, about $3,880 at $200 PFH each.
• Duet: the extra coordination may show up in the quote, but no published rate sets a premium.

Ask each narrator to quote for your exact format; no published rate covers how two narrators split a fee.

On ACX, a producer is the narrator or studio that records the audiobook. Pay for Production is an ACX deal in which you pay the narrator a one-time PFH fee, and Royalty Share is an ACX deal in which the narrator takes a share of your royalty instead.

ACX's help page on finding a narrator says royalties can only be split with the one narrator or producer on the contract. It suggests hiring a producer or studio that works with a roster of narrators, or contracting the second narrator outside ACX and paying both on a Pay for Production basis rather than Royalty Share. The Audible article adds a third route: pay the narrator you hire through ACX the combined rate for both. For how ACX's deals work, see what is ACX.

Which one does your book need?

An audiobook's narration format, whether single, dual, duet or full cast, depends more on the manuscript's point-of-view (POV) structure than on its genre.

1. One point of view throughout: a single narrator, or a full cast if the dialogue is busy.
2. Two POV characters in alternating chapters, with quiet, introspective chapters: dual.
3. Two POV characters whose scenes run on back-and-forth dialogue: duet.
4. Three or more POV characters, or a large ensemble: a full cast.

A quick test: open a scene where both leads talk. If the heroine's narrator voicing the hero would sound wrong for half the page, you need duet or a cast. If one narrator could carry it, dual is enough.

If you produce with AI voices, the choice changes shape. In AudioForge you cast a voice for the narrator and for each character, and casting is saved per book, so both leads keep their own voices in every chapter. You review the script, fix the speaker of any line, and regenerate only that line; the steps are in the help center. AI-narrated audio cannot go to Audible through ACX, which requires a human narrator as of November 2026, while Spotify and Kobo accept it with a disclosure. If Audible is your main market, human dual or duet narration is the better choice.

The bottom line

Dual narration splits chapters between two narrators by point of view, and duet narration gives each lead their own narrator for every line they speak. Choose dual for alternating-POV chapters, duet for banter-led scenes, and a full cast when more than two characters drive the dialogue. AudioForge is free during early access, so you can upload one chapter, cast both leads and hear them as separate voices before paying a narrator, following the help center.

The answer depends on who does the work: you, a hired narrator, or AI voices. AudioForge is a production tool that turns a manuscript into a multi-voice audiobook, with a distinct voice for every character.

Below: how long a book is in audio, what the hours go into, a timeline for each route, and what slows a release down.

How long will your audiobook be?

An audiobook of a 90,000-word novel runs about 9.7 finished hours. A finished hour is one hour of the final, edited audiobook a listener hears, and it is the unit narrators and ACX use to measure work and price. ACX is Audible's self-publishing service, the route independent authors use to reach Audible.

ACX's page on how long a narrated audiobook will be says most performers narrate about 9,300 words per hour. Divide your word count by 9,300 to estimate your own book:

• 60,000 words: about 6.5 finished hours
• 90,000 words: about 9.7 finished hours
• 120,000 words: about 12.9 finished hours

Every estimate below starts from that number.

How many hours of work goes into one finished hour?

One finished hour of an audiobook takes about 4 to 6.2 hours of work, according to two ACX help pages.

Recording, editing and reviewing one finished hour takes about 6.2 hours, or about 60 hours for a 90,000-word novel, according to ACX's page on producing and recording your audiobook (dated August 2025):

ACX's page on managing offers (dated September 2025) gives a lower figure for working narrators: about four hours per finished hour, two reading and two editing, which is about 39 hours for the same book. The first page also says new narrators generally take 5 to 7 hours just to record one finished hour, so a first book can run well past 60 hours. Other guides, such as the one cited in our post on how to create an audiobook, put self-narration at 3 to 5 hours per finished hour; ACX's own figures are the more conservative.

Mastering is the final pass that brings every file to the store's technical standard. ACX's audio submission requirements are specific as of November 2026 (page dated April 2026): average loudness (RMS) between -23 dB and -18 dB, peaks no higher than -3 dB, one chapter per file and no file longer than 120 minutes.

How long does each route take for a 90,000-word novel?

Each way of producing an audiobook of a 90,000-word novel spends its time in a different place: narrating yourself costs your hours, hiring a narrator costs calendar weeks, and AI production moves the work to checking and listening.

The AI route has one limit the others do not: as of November 2026, ACX requires a human narrator unless it has authorized otherwise, so AI audio cannot go to Audible through ACX. If Audible is your main market, a human narrator is the better choice.

The hire-a-narrator steps come from ACX's own rules. ACX's audition page keeps a title open until you accept an offer or 30 days pass. A producer is ACX's term for the narrator or studio that records the book.

Under ACX's production standard terms, you aim to send the manuscript within 3 business days of an accepted offer and to review the 15-minute checkpoint, a first sample you approve before the rest is recorded, within 4 business days. You then aim to review the finished audiobook within 10 business days, after which ACX treats it as approved. ACX's checkpoint page says a producer can usually deliver that sample within a day or two of starting.

In AudioForge, generation runs in the background, so your time goes into the parts that need a person: checking who speaks each line, choosing a voice for each character and listening back. The steps are in the help center, and fixing one line means regenerating only that line, not the chapter.

What slows an audiobook down after recording?

Review and store approval slow an audiobook down after recording, and an author cannot hurry either one.

• ACX's quality check. ACX's audio review page gives no fixed time as of November 2026 and says to contact ACX if a title has been in review for more than a few days. Its last published target, in a 2021 announcement, was sale within 10 business days for titles that meet requirements. ACX's selling page adds that Apple Books may take up to two more business days.
• Distributors. Authors Republic says in its October 2025 self-publishing guide that an audiobook can appear as soon as five days after distributor approval but may take up to 60 days.
• Files that fail the standard. A chapter that misses the loudness or peak limits goes back for another edit and another check. Measuring each file before you upload saves a round trip.

Stores that accept AI narration are listed in where you can publish an AI-narrated audiobook.

How to record an audiobook yourself: how do you plan the time?

Self-narration time for an audiobook is your word count divided by 9,300, multiplied by your hours per finished hour, divided by the hours you can record each week.

1. Measure your book. Divide the word count by 9,300 for finished hours.
2. Multiply by your rate. Use 6.2 hours per finished hour as a floor for a first book, and about 4 once you are practiced.
3. Divide by your weekly hours. At 10 hours a week, a 60-hour job is about six weeks.
4. Record a test chapter first. Time the recording and the editing separately, then redo the sum with your real numbers.

If the total is more time than you can spare from writing, the same post on creating an audiobook sets out what hiring a narrator costs.

The bottom line

Recording an audiobook takes about 4 to 6.2 hours of work per finished hour, or roughly 39 to 60 hours for a 90,000-word novel, and review and store approval add days to weeks after that. Narrating it yourself costs your time, hiring a narrator costs calendar weeks, and AI production turns the job into checking and listening. AudioForge is free during early access, so add one chapter and compare two voices for your lead character, following the steps in the help center.

Dramatized and full-cast labels blur on store pages, which matters when you decide what to make of your own book. AudioForge is a production tool that turns a manuscript into a multi-voice audiobook, with a distinct voice for every character.

Below: what each term means, how the formats compare, what they cost, and how an independent author can make a full-cast version.

What does "dramatized audiobook" mean?

A dramatized audiobook is a performance adapted from a book, with several voices, music and sound effects. Audible's Dramatized Adaptation page describes the format as translating written works into theatrical or screenplay performances.

Two things set a dramatization apart from a standard audiobook:

• The text can change. An unabridged audiobook is every word of the book; an adaptation may cut scenes, turn narration into dialogue or sound, and drop descriptions where a sound effect does the work.
• The production adds a soundtrack. GraphicAudio, a publisher known for the format, sums up its productions on its home page as "Full Cast. Cinematic Music. Sound Effects."

Some dramatizations are sold as a separate edition next to the single-narrator version, so readers can choose.

What is a full-cast audiobook?

A full-cast audiobook is the text of a book performed by a cast of actors rather than a single narrator. Book Riot's 2020 comparison of audio dramas and full-cast audiobooks puts it as the text broken down for a cast.

The narrator still reads the narration, and each character speaks their own lines. The text is usually not rewritten, and Book Riot notes that most full-cast audiobooks are not abridged. A full cast is also called a multicast. Some full-cast productions add light music or sound effects, which is where the line with dramatization blurs.

An audio drama is a story scripted specifically to be listened to, the definition Book Riot uses, with no narrator reading prose. Scenes are carried by dialogue, music and sound effects. Audio drama is the best-known form of audio fiction, stories written for the ear rather than adapted from a printed book. If you want to write one from your own chapters, our guide to audio drama script format shows the layout.

How do the formats compare?

Single-narrator, dual or duet, full-cast, dramatized and audio-drama productions differ in two ways: how many voices perform, and whether the book's text stays as written.

Dual narration splits chapters between two narrators, and duet gives each lead their own narrator; our post on dual vs duet narration covers that choice.

What can go wrong in a full-cast or dramatized audiobook?

A full-cast or dramatized audiobook can go wrong in three ways, and each one is a production decision you control.

I grew up reading sci-fi, and I've been a space-opera fan since reading Dune about 15 years ago. Space opera and epic fantasy carry big casts and invented names, which is where these three risks show most:

• Effects drowning the voices. Keep music and sound effects light and place them at scene changes, not under every line.
• Uneven volume between characters. Every voice should sit at the same level, so a quiet character is not lost.
• Names that change between voices. Every performer has to say your invented names the same way.

Keeping the text whole also avoids a complaint listeners raise about some dramatizations: that scenes from the book are missing.

What does a full-cast production cost?

A full-cast production with human actors costs far more than a single narrator, because every role is another performer. Author Media's July 2026 guide to the five tiers of audiobooks says a full cast of actors can cost as much as $10,000, and it lists AI full-cast production as a cheaper tier.

ACX, Audible's self-publishing service, adds a practical limit. Its help page on finding a narrator, dated June 2025, says ACX contracts support one rights holder and one producer, the narrator or studio that records the book. A cast on ACX therefore comes through one producer, such as a studio with a roster of narrators.

How can an independent author make a full-cast audiobook?

An independent author can make a full-cast audiobook by producing it with AI voices and casting one voice per character. AudioForge does this in four steps:

1. Add a chapter. Create your book in My Books, then paste or upload the chapter.
2. Parse it into a script. AudioForge turns the chapter into a script line by line and tags every line with who speaks it and how it should be delivered. Lines it was unsure about are flagged so you assign the speaker yourself, and you can edit any line, so the text stays as you wrote it unless you change it.
3. Cast your voices. Choose a voice from the Voice Library for the narrator and each character. Casting is saved per book, so recurring characters keep their voice in every chapter.
4. Generate and export. AudioForge stitches the lines, with your pauses, into one MP3 per chapter.

For a dramatized feel, place sound effects from the built-in library where a scene needs them and use the Soften control to tame harsh ones. A delivery cue is a bracketed stage direction such as [whispers], placed before the words it applies to. A per-book pronunciation glossary keeps invented names the same across the cast, and a line that lands wrong can be fixed and regenerated alone.

AI narration also decides where you can sell. As of November 2026, Spotify and Kobo accept it with a disclosure, while ACX requires a human narrator; where to publish an AI-narrated audiobook has each store's rule.

The bottom line

A dramatized audiobook adapts a book into a performance with a cast, music and sound effects, while a full-cast audiobook keeps the text and gives every character a voice. A full-cast, unabridged version gives readers every word with a cast of voices and no rescripting cost. To hear your own characters as a cast, upload one chapter to AudioForge, free during early access, and cast your two busiest speakers, following the help center.

AudioForge is a production tool that turns a manuscript into a multi-voice audiobook, with a distinct voice for every character. It is one of the options below, and not the right one for every book.

Below: the kinds of generator, how the main options compare as of October 2026, which suits which book, and what the work involves.

What kinds of AI audiobook generators are there?

AI audiobook generators come in three kinds: tools built into a store, general text-to-speech apps that export audio, and production tools made for multi-voice fiction.

• Store tools turn your book into an audiobook sold in that store, often for free. Spotify, Google Play Books, Apple Books and Amazon each have one.
• General text-to-speech apps read any text aloud and export an audio file. They are flexible, but leave chapter files, store specs and disclosure to you.
• Multi-voice production tools such as AudioForge parse a chapter into a script, assign each line to a speaker and give each character a voice you choose.

How do the main options compare?

As of October 2026, Spotify, Google Play Books, Apple Books, Amazon virtual voice and AudioForge are all free to start, so the deciding factor is who owns the audio and where it can sell.

Each store tool is the better choice in its own store. Spotify's tool publishes most titles to Spotify within 72 hours, Google Play already offers character voices, and Amazon's virtual voice puts an AI-narrated edition on Audible, which no outside tool can do through ACX today. Amazon's ownership terms are covered in what authors give up with Audible virtual voice.

Which generator suits your book?

The right AI audiobook generator depends on how much dialogue your book has and where your readers buy.

• One store matters most: use that store's tool. It handles the format and the disclosure for you.
• Quiet, prose-led book: a single good voice may be all it needs, so a store tool or a text-to-speech app can work.
• Dialogue-heavy fiction or a long serial: choose a tool with a voice per character, so listeners can tell speakers apart by ear. Our post on why AI audiobooks sound flat explains why this matters.
• You want to sell in several places: choose a generator that leaves the files with you.

If you start from an EPUB, turning an EPUB into an audiobook walks through each store tool's rules in more detail.

What does making an AI audiobook involve?

Making an AI audiobook involves a review pass, not a recording session: the generator does the reading, and you check its decisions.

1. Add the manuscript. Store tools usually take an EPUB; production tools such as AudioForge take pasted text or a document file.
2. Check names. Spotify's tool and Google Play both offer a pronunciation step, and some multi-voice tools keep a pronunciation glossary.
3. Check speakers. In multi-voice tools, a line given to the wrong speaker gets the wrong voice.
4. Listen and fix. Preview the busiest chapter first and correct the lines that sound wrong.
5. Disclose and publish. Label the narration as synthetic wherever the store asks.

No generator removes the review. Expect to listen to every chapter before release.

Can you sell an AI-generated audiobook on Audible?

An AI-generated audiobook made outside Amazon cannot be sold on Audible through ACX as of October 2026. ACX's submission requirements say audiobooks must be narrated by a human unless otherwise authorized, and Audible says it is working to accept text-to-speech content later, with no date. For the stores that do accept AI narration, see where you can publish an AI-narrated audiobook.

The bottom line

To make an audiobook with an AI audiobook generator, pick the tool by its voices, its stores and its ownership terms, then plan for a careful listen before you publish. Store tools are the fastest route into a single store; a multi-voice tool suits dialogue-heavy fiction you want to sell widely. To hear your own characters cast with distinct voices, paste one chapter into AudioForge and follow the four steps in the help center.

Below: whether any ebook can be converted, which tools take an EPUB, what each lets you do with the audio afterward, how to prepare the file, and how long the audiobook will run.

Can you turn any EPUB into an audiobook?

An EPUB can become an audiobook you sell only if you own its audio rights. Reading apps can read any EPUB aloud for your own listening, but a store needs the rights holder to publish it.

The rule appears in the stores' own terms. Google Play Books' auto-narration policy says you must own the audiobook rights to any ebook you select, and ACX asks that you hold the audio rights before you claim a title, according to its What is ACX page. If you signed a publishing deal, check whether the publisher holds the audio rights before you start.

Which tools turn an EPUB into an audiobook?

Spotify, Google Play Books and Apple Books each turn an ebook file into an AI-narrated audiobook, and Amazon does the same from a Kindle ebook; the deciding factor is what you can do with the audio once it exists.

Sources: Spotify's Audiobook Creation Tools help and tool page, Google's policy and character voices guide, Apple's digital narration page, and KDP's virtual voice help and beta terms.

Google's character voices stand out. Its tool can detect characters in English books in one click, and you then assign a voice to each one, with the default narrator reading everything else.

What can you do with the audio afterward?

As of October 2026, each EPUB tool leaves you with different rights over the finished audio, and this matters more than the voices.

• Spotify publishes the audiobook to Spotify once you submit it, most within 72 hours, and says Spotify for Authors is non-exclusive.
• Google Play Books lets you download the audiobook and sell it elsewhere, as long as it is also on sale on Google Play in every country where you sell it, at a price no higher than elsewhere.
• Apple Books says there are no restrictions on producing and distributing other versions of the audiobook.
• Amazon virtual voice keeps the audio file with Amazon under its beta terms, and if your ebook is in KDP Select you cannot sell audio versions elsewhere during each 90-day period. Our post on Audible virtual voice covers that trade in detail.

Wherever the audio ends up, label it honestly. The rules for each store are in where you can publish an AI-narrated audiobook.

How do you prepare an EPUB for narration?

Preparing an EPUB for narration means checking the things a listener hears but a reader skims: names, dialogue and structure.

1. Confirm the audio rights are yours, in writing if a publisher was ever involved.
2. List every invented name and term. Spotify's tool has a pronunciation review step before publishing, and Google Play lets you set pronunciations with phoneme tags.
3. Check who says each line. In a tool with character voices, a misattributed line puts the wrong voice in a character's mouth.
4. Check chapter breaks. Spotify and Google Play both work chapter by chapter, so make sure each chapter in your EPUB starts where you want it to.
5. Preview before you publish. Spotify offers chapter previews, so listen to the busiest chapter first.

How long will the audiobook be?

An audiobook runs about one finished hour for every 9,300 words, the average narration pace used in Author's Republic's guide to per-finished-hour rates. A finished hour is one hour of polished, edited audio.

At 9,300 words per finished hour, a 90,000-word novel runs about 9.7 hours, and each extra 10,000 words adds about an hour (count the words without front matter such as the table of contents):

AI voices do not all read at the same pace, so treat these as a planning range rather than a promise. Length matters when you set a price and when you plan how long the review pass will take.

The bottom line

To turn an EPUB into an audiobook you can sell, start with the audio rights, then pick the tool whose rules match where you want to sell: Spotify for a fast Spotify release, Google Play if you want character voices and files you can take elsewhere, or Amazon's virtual voice if Audible is all that matters.

If you would rather cast and direct every character yourself, AudioForge is a production tool that turns a manuscript into a multi-voice audiobook, with a distinct voice for every character, working from your manuscript as a .docx, .txt or PDF rather than the EPUB. Whichever you choose, check your audio rights first and then produce one chapter before committing the whole book.

ACX matters to you if Audible is where your readers buy, because it is the service Audible runs for independent rights holders. Below: who can use ACX, how a title goes from manuscript to sale, what its three deals cost and pay, and where AI narration stands.

Who can use ACX?

ACX is open to rights holders who live in the United States, the United Kingdom, Canada or Ireland, according to ACX's "What is ACX" page as of October 2026.

To sign up you need:

• A mailing address, a valid local tax ID and bank details in one of those four countries.
• The audio rights to the book.
• The book on sale in the Amazon Kindle Store.

ACX lists no other countries.

How do you make an audiobook on ACX?

Making an audiobook on ACX takes five steps: create a profile, claim your title, choose who narrates, produce the audio, and pass ACX's quality check.

1. Create a profile and add your payment details.
2. Claim your title from the Kindle Store and write the audiobook's description.
3. Choose a narrator. You can record it yourself or post the title for auditions and pick a producer. On ACX, a producer is the narrator or studio that records, edits and masters the audio.
4. Produce the audio to ACX's technical standard, one chapter per file.
5. Submit for review. ACX's quality team checks the files, and a title that misses the standard may be rejected or have its release delayed.

The standard is specific. ACX's audio submission requirements, dated April 15, 2026, ask for opening and closing credits, a retail sample of five minutes or less, files no longer than 120 minutes, volume between -23 dB and -18 dB RMS, peaks below -3 dB and 192 kbps or higher constant bit rate MP3 files.

What do ACX's three deals cost and pay?

ACX offers three production deals, Pay for Production, Royalty Share and Royalty Share Plus, and they differ in who pays upfront and whether you must stay exclusive.

PFH, short for per finished hour, is the rate a narrator charges for each hour of finished audio. Pay for Production is an ACX deal in which you pay the narrator a one-time fee at that rate. Royalty Share is an ACX deal in which the narrator takes no upfront fee and splits your royalty instead. Royalty Share Plus is an ACX deal in which the narrator takes a reduced upfront fee plus half your royalty. The figures come from ACX's royalty rules under its new model, as of October 2026.

Pay for Production and narrating it yourself cost money or time upfront and leave exclusivity to you, while Royalty Share and Royalty Share Plus cut the upfront cost in exchange for exclusivity and half the royalty:

Exclusive distribution means Audible sells your audiobook through Audible, Amazon and Apple Books only, according to ACX's distribution options page. Non-exclusive lets you sell it elsewhere too, at the lower rate. Either way, Audible has the right to distribute the audiobook for seven years.

From May 26, 2026, the new model applies to newly claimed titles, new users and existing titles that enroll. Titles still on the legacy model earn 40% exclusive and 25% non-exclusive, and 20% each under Royalty Share, until that model is discontinued at the end of 2026. To see what a narrator costs for a full novel, our guide on how to create an audiobook works through the numbers.

Is ACX part of Amazon?

ACX is part of the Amazon family through Audible, which runs it. The ACX site is copyrighted to Audible, Inc., and ACX and Audiobook Creation Exchange are listed as trademarks of Amazon.

That also answers the common question of whether ACX is legitimate. ACX is Audible's own service for rights holders and narrators, not a third-party marketplace.

Can you use AI narration on ACX?

ACX does not accept AI narration that Audible has not authorized. Its submission requirements say your audiobook must be narrated by a human unless otherwise authorized, and that unauthorized text-to-speech, AI or automated recordings are prohibited.

The same page says Audible is working to accept text-to-speech content from other publishers and creators, with timing to follow. Amazon's own virtual voice program is separate and does not go through ACX; we cover what it gives up in Audible virtual voice: what authors give up. Stores that do accept AI narration today are listed in where you can publish an AI-narrated audiobook.

The bottom line

ACX is how independent authors in the US, UK, Canada and Ireland put a human-narrated audiobook on Audible, Amazon and Apple Books, with a royalty of 25% to 50% under the new model as of October 2026, depending on the deal. If ACX's human-narration rule rules it out for your plans, AudioForge is a production tool that turns a manuscript into a multi-voice audiobook, with a distinct voice for every character, for the stores that accept AI narration. Either way, your next step is the same: read ACX's submission requirements before you commit to a narrator or a deal.

AudioForge is a production tool that turns a manuscript into a multi-voice audiobook, with a distinct voice for every character. It covers some of these steps differently from a studio, and this guide says where.

Below: the stages of audiobook production, the standard finished audio must meet, who does each step on each route, and how long it takes.

What are the stages of audiobook production?

Audiobook production has six stages: preparation, narration, editing, proofing, mastering and delivery.

1. Preparation. Read the manuscript for the ear, settle how every name is pronounced and decide who voices whom.
2. Narration. Record or generate every line, chapter by chapter.
3. Editing. Remove mistakes, breaths and noise, and record pickups, the short re-recordings that fix a misread line.
4. Proofing. Listen against the text to catch skipped words and misreads.
5. Mastering. Bring every file to consistent levels that meet the store's specification.
6. Delivery. Add opening and closing credits and a retail sample, then upload one file per chapter.

Author's Republic lists the same work inside a narrator's per-finished-hour rate: reading the script, recording, pickups and revisions, editing, proofing, mastering and uploading retail-ready files.

What standard does finished audio have to meet?

Finished audio has to meet the technical standard of the store you deliver to, and ACX publishes a detailed one.

ACX's audio submission requirements, dated April 15, 2026, set eight checks, from volume and peaks to credits and a retail sample, and every file must pass all of them:

ACX also requires human narration unless otherwise authorized, so AI-narrated audio goes to other stores. Which ones accept it is in where you can publish an AI-narrated audiobook.

Who does each step on each route?

Narrating it yourself puts every stage on you, a hired narrator usually covers editing, proofing and mastering in the PFH rate, and AI production with AudioForge leaves you checking the script and listening; the deciding factor is how much of the work you want to pay for.

PFH, short for per finished hour, is what a narrator charges for each hour of finished audio. Our ACX guide explains the deals you can strike with a narrator there.

If you want help with the production itself, AudioForge's Author Help Program offers a done-for-you option where our team produces the audiobook and you review and approve it. Places are limited.

How long does audiobook production take?

Audiobook production takes several hours of work for every finished hour of audio, and estimates range from three to six or more hours.

A finished hour is one hour of polished, edited audio, and Author's Republic uses an average narration pace of about 9,300 words per finished hour. For a 90,000-word novel of about 9.7 finished hours:

• A hired narrator needs six or more hours per finished hour, according to Author's Republic's guide, revised October 2025: about 58 hours or more, at $100 to $1,000 or more per finished hour, so roughly $970 to $9,700 or more.
• Narrating it yourself takes three to five hours per finished hour, according to ScribeCount's July 2026 guide: about 29 to 48 hours, plus $300 to $700 of equipment.
• AI production moves the hours from recording to review. You still listen to every chapter, roughly 9.7 hours of listening for this book, plus the time to fix lines.

The full cost comparison by route is in how to create an audiobook.

The bottom line

Audiobook production is six stages of work, from preparation to delivery, and the route you pick decides how many of them land on you. A hired narrator usually covers editing, proofing and mastering in the per-finished-hour rate, while narrating it yourself or producing with AI puts the review in your hands. To see the AI route on your own book, produce one chapter in AudioForge, free during early access, following the steps in the help center.

AudioForge is a production tool that turns a manuscript into a multi-voice audiobook, with a distinct voice for every character. We think it is the better route for authors who care about owning their audio, and this post sets out why, including where virtual voice still makes sense.

Below: why authors hesitate, who owns what, what each route pays, and how the two compare.

Why are authors wary of Audible virtual voice?

Authors are wary of Audible virtual voice because Amazon, not the author, owns the audio file it creates. Many want to know who controls the audio edition before they make one.

Author Larada Horner-Miller put it plainly when weighing virtual voice for her own book: "I don't want to give up ownership to the audiobook." She quoted the line from Amazon's terms that gave her pause, that Amazon will "retain ownership in the AVV (Audiobook with Virtual Voice) Edition file" (her post, June 2024).

That worry matters most to authors who hope to sell their audio rights one day. A publisher buying those rights will want to know what audio editions of the book already exist and who controls them.

What do you keep, and what does Amazon keep?

With Audible virtual voice, you keep the copyright in your book and Amazon keeps the audio file.

The virtual voice beta terms, last updated on October 21, 2026, say you "retain all ownership rights" in your book, and that Amazon will "retain ownership in the AVV Edition file". You grant Amazon a nonexclusive, irrevocable license, limited to the virtual voice edition, to create it and make it available through Amazon's services for the term of the agreement.

Three practical points follow:

• The audio stays with Amazon. Amazon's pages describe no way to download the file or sell it in other stores.
• Amazon can remove it too. You can unpublish the audiobook in KDP, but the terms also let Amazon remove the edition from sale at any time, and the file stays Amazon's either way.
• KDP Select adds exclusivity. If your eBook is enrolled, Amazon gets the exclusive right to sell the virtual voice edition, and you cannot sell or distribute the book in audio, or let anyone else do so, during each 90-day period of exclusivity. Kindle Unlimited became Kindle+ in October 2026, as Written Word Media reported, and KDP's Kindle+ page says the edition is added to Kindle+ automatically and cannot be taken out on its own: you unpublish the audiobook or unenroll the eBook.

What does each Amazon route pay you?

As of October 2026, Amazon pays a 40% royalty on individual sales of a virtual voice audiobook, and ACX's new royalty model, in place since May 26, 2026, pays 30% to 50%. ACX is Audible's self-publishing service, where authors hire a human narrator or narrate the book themselves.

KDP's pricing and royalties page says virtual voice audiobooks earn a 40% royalty on individual sales, at a list price you set between $3.99 and $14.99, and that Amazon may sell at a different price. If your eBook is in KDP Select, Kindle+ listening is paid from the KDP Select Global Fund, and a page earns once per reader whether it is read or heard first. Audible Plus listening is paid from a separate fund.

Under ACX's current distribution options, new royalty rates are 50% if you distribute exclusively through Audible, Amazon and Apple Books, and 30% if you do not. Audible has the right to distribute your audiobook for seven years.

Virtual voice pays 40% of individual sales, ACX exclusive 50% and ACX non-exclusive 30%, and the deciding factor is whether you accept exclusivity:

Each program calculates royalties on its own basis, so check its royalty page before comparing figures. ACX titles on the legacy model earn 40% and 25% until that model ends at the end of 2026, according to ACX's royalty page.

How does AudioForge compare?

With AudioForge you own the audio files, choose where to sell them and keep all of your royalties.

As a production tool, not a publisher, AudioForge leaves your text, your characters, your audio files and your royalties with you. It is free for a limited time while it grows with its first authors, and future pricing will be pay-as-you-go by the word, not a share of your sales. The work is four steps: paste or upload a chapter, check the script, cast a voice for each character, then generate and export the chapter as an MP3.

Audible virtual voice and AudioForge differ most on who owns the audio file:

Because you hold the files, you can release them your own way: a chapter a week to patrons, a free first chapter on YouTube, or the full book on your site. The steps from manuscript to MP3 are in the help center, and each store's rules for AI narration are in where you can publish an AI-narrated audiobook.

When is virtual voice still the better choice?

Audible virtual voice is the better choice when Audible is the only store that matters to you and speed matters more than control.

If your readers buy almost entirely on Audible in the US, your book suits Amazon's voices, and you have no plans to sell audio rights, virtual voice is the least work for an Audible listing. The same holds if your eBook already earns from Kindle+ page reads and you want listeners in that subscription too. Producing your own AI audio cannot get you onto Audible: ACX requires human narration unless it has authorized otherwise. If Audible is not essential, turning an EPUB into an audiobook compares the other stores' own AI tools.

The bottom line

Audible virtual voice is quick and costs nothing upfront, but Amazon owns the audio file, pays you a 40% royalty and can remove the edition from sale at any time. If you want to own your audiobook, sell it where you choose and keep your options open for a publishing deal, produce it yourself. With AudioForge the files and the royalties are 100% yours: produce your first chapter free today, following the four steps in the help center.

AudioForge is a production tool that turns a manuscript into a multi-voice audiobook, with a distinct voice for every character. It is one of these routes, and not always the right one, so this guide sets out what each route costs and where each one wins.

Below: the four routes side by side, what each costs for a typical novel, how Audible's virtual voice works, and how to produce a multi-voice audiobook step by step.

What are your options for creating an audiobook?

You can create an audiobook in four ways: narrate it yourself, hire a narrator on ACX, use Audible virtual voice, or produce it with AudioForge. The deciding question is who performs the book.

No route is best for every book. A quiet literary novel and a 300-chapter LitRPG serial need different things.

Stores are also starting to build AI narration tools of their own. As of October 2026, Spotify's Audiobook Creation Tools turn an EPUB into an audiobook read by a digital voice you choose and publish it to Spotify, for US authors writing in English, up to five audiobook projects per account. Spotify's tool page says the first five are free. Spotify's help page describes choosing one digital voice for the book, and virtual voice lets you change voice per chapter; neither page describes casting a separate voice for each character.

What does each route cost?

An audiobook of a 90,000-word novel costs anywhere from nothing upfront to nearly $10,000 or more, depending on who performs it.

Length drives the price. A finished hour is one hour of the final, edited audiobook, and narrators charge per finished hour. The common estimate is about 9,300 words per finished hour, so a 90,000-word novel comes to roughly 9.7 finished hours.

For that example book, a hired narrator costs about $970 to $9,700 or more, narrating it yourself costs $300 to $700 plus your time, and virtual voice and AudioForge cost $0 upfront; the deciding factor is who performs:

The narrator rates come from Author's Republic's guide to per-finished-hour rates, and the self-narration figures from ScribeCount's guide to narrating your own audiobook. AudioForge's pricing, when it arrives, will be pay-as-you-go by the word with no subscription. ACX's own time estimates for recording, editing and review are set out in how long it takes to record an audiobook.

You can also avoid an upfront narrator fee on ACX with a Royalty Share deal. Royalty Share is an ACX contract in which the narrator takes no payment upfront and splits net royalties with you instead, which requires exclusive distribution through ACX, according to ACX's royalty page. The same page sets the royalty: since May 26, 2026, titles in ACX's new model earn 50% for exclusive distribution and 30% for non-exclusive, while legacy titles earn 40% and 25% until that model ends at the end of 2026. Under Royalty Share the exclusive royalty is split equally, so you and the narrator each get 25% in the new model or 20% in the legacy one.

If Audible is where your readers buy, ACX requires human narration, either yours or a narrator's, unless ACX has authorized otherwise. Who handles editing, proofing and mastering on each route is set out in what audiobook production involves.

How does Audible's virtual voice work?

Audible virtual voice is Amazon's program that turns an eligible Kindle eBook into a computer-narrated audiobook sold on Audible.

According to KDP's help pages, you choose from Amazon's voices, can set a different voice for each chapter, and preview and edit the narration before publishing. The audiobook is clearly labeled as virtual voice, Amazon's computer-generated narration, priced between $3.99 and $14.99, and pays a 40% royalty on individual sales.

Creating one is free during the beta. The limits are set out in the eligibility rules: the program is invite-only, distribution is US only, and the eBook must already be live on KDP. Kindle Unlimited became Kindle+ in October 2026, as Written Word Media reported, and if the eBook is in KDP Select the audiobook joins Kindle+ automatically, according to KDP's Kindle+ page. Amazon also keeps ownership of the audio file, which we cover in what authors give up with Audible virtual voice.

How do you produce a multi-voice audiobook?

To produce a multi-voice audiobook with AudioForge, you turn each chapter into a script, cast a voice for every character, then generate and export the audio. The four steps are the same ones shown in the help center:

1. Add a chapter. Paste or upload it. Each chapter becomes its own audio file.
2. Check the script. Every line is tagged with who speaks it and how it should be delivered. Lines the parser could not attribute are flagged so you assign the speaker yourself.
3. Cast your voices. Pick a voice for the narrator and each character. Casting is saved for the whole book, so a character sounds the same in every chapter.
4. Generate and export. The lines are stitched, with your pauses, into one MP3 per chapter. If a line sounds wrong, fix it and regenerate only that line.

You own the finished audio and can sell it wherever AI narration is accepted. Check each store's rules in where you can publish an AI-narrated audiobook before you choose. If your readers are mainly on Audible, remember that ACX requires human narration and prohibits unauthorized AI recordings.

The bottom line

Choose the route by who should perform your book and where your readers buy. A human narrator through ACX is the path to a standard Audible release and costs roughly $1,000 to $10,000 or more for a novel, while Audible's virtual voice is the low-effort option for US Kindle authors invited to its beta. If you want a distinct voice for every character and files you own, produce your first chapter with AudioForge for free today, following the steps in the help center, and hear your own book before you commit.

AudioForge is a production tool that turns a manuscript into a multi-voice audiobook, with a distinct voice for every character. That makes this question personal for us: every author who uses it hands over the thing they care about most.

Below: why the worry is reasonable, what happens to your text when you produce with us, what to check in any tool's terms, and what you own at the end.

Why do authors worry about AI training?

Because text that goes into a training set can shape what a model writes for everyone, and it cannot be taken back out.

Authors have followed a long public argument about books being used to build AI models without their say. Serial fiction writers feel it more than most, since their chapters are already public on reading sites. Turning a book into audio should not open one more door you did not choose to open.

The worry is not about AI voices as such. It is about where your words go after you press upload, and whether you can get them back.

What happens to your text when you produce with us?

It is used for one job, producing your audio, and nothing else.

When you add a chapter, it is read into a script, with every line tagged by who speaks it and how it should be delivered. When you generate the chapter, each line is voiced and the pieces are stitched into one file. You can follow both steps in the help center.

To do that work, your text is sent to the AI providers we run on. It is sent under API terms that exclude training, so your book is processed to make your audio and is not used to improve anyone's model.

You also stay in control of what is stored:

• You can delete your text from our servers at any time.
• If you sign a traditional deal, one click permanently deletes everything, text and audio.

The full details are on the IP policy page.

What should you check in any AI audio tool's terms?

Four things: training, ownership, deletion and rights. The table below is a quick checklist you can use on any tool, including ours.

If a tool's terms are vague on any of these, ask its support team before you upload a chapter. A clear answer in writing is worth more than a reassuring line on a sales page.

Who owns the audiobook you make?

You own all of it: the text, the characters, the audio files and the royalties.

We do not publish or distribute your book. You download each finished chapter as an MP3 and put it wherever you choose, whether that is a store or your own site. Stores set their own rules for AI narration, and some ask for a disclosure, so check where you can publish an AI-narrated audiobook before you choose.

Because you own the files outright, you can also move them later. Nothing about how you produced them ties a finished audiobook to us.

The bottom line

Whether your manuscript trains AI depends entirely on the terms of the tool you use, so read them for training, ownership, deletion and rights. With AudioForge your text is used only to make your audio, under terms that exclude training, and you can delete it at any time. You keep every right to the book and the audio you produce.

AudioForge is a production tool that turns a manuscript into a multi-voice audiobook, with a distinct voice for every character. So we follow this research closely, including the parts that cut against AI.

Below: what listener surveys say, what happened when listeners did not know they were hearing AI, where a human narrator still wins, and how to give your audiobook the best chance.

What do listener surveys say about AI narration?

Listener surveys show stated interest in AI narration is falling: 61% of listeners said they were willing to try an AI-narrated audiobook in 2026, down from 70% a year earlier.

That figure comes from the Audio Publishers Association's 2026 Consumer Survey, run by Edison Research in February 2026 among 1,706 spoken-word listeners. The same release found that only 16% of audiobook listeners had ever heard an AI-voiced title, and that AI titles made up 0.03% of audiobook sales revenue in 2025.

Put those together and most listeners are answering about an idea, not about a recording they have heard. That leaves room for a good production to change their mind.

What happens when listeners don't know it is AI?

When listeners don't know it is AI, they rate a multi-voice AI version higher than a single human narrator for character-driven stories.

The study was conducted by Edison Research at SSRS and released in July 2026 by Spoken, an AI audiobook company, so read it with that in mind. More than 1,000 US fiction listeners were split into two groups. One heard excerpts of a sci-fi thriller read by a professional human narrator, the other heard an unedited AI version with a distinct voice for each speaking character.

Two more results are worth knowing. After listeners were told which version was AI, purchase intent was 46% for the AI version and 49% for the human one, which the researchers call statistically comparable. And while only 31% said at the start that they were likely to listen to an AI audiobook, 65% said after hearing the excerpt that they were likely to listen to one narrated this way.

Where does a human narrator still win?

A human narrator still wins for exposition without multiple characters: the same study found listeners rated the human narrator higher there.

That fits how the two approaches work. A single skilled narrator carries long passages of description and reflection well. A cast of distinct voices earns its keep when several characters talk, and one voice for everyone is where AI narration tends to sound flat. We look at the causes in why AI audiobooks sound flat.

Quality will not win everyone. Some listeners and some reader communities object to AI in books on principle. If your readership leans that way, weigh it in the decision.

How do you give an AI-narrated audiobook the best chance?

To give an AI-narrated audiobook the best chance, make it sound like a production, be open about it, and let listeners hear it before they decide.

1. Give every character a voice. The study's advantage showed up in character-driven stories. In AudioForge, casting is saved for the whole book, so a character sounds the same in chapter forty as in chapter one.
2. Direct the lines that matter. A whispered threat and a shouted warning should not sound alike. Fix the delivery of a line and regenerate only that line, as described in the help center.
3. Say that it is AI. Stores that accept AI narration ask you to disclose it. Check each one's rule in where you can publish an AI-narrated audiobook.
4. Share a sample first. In the study, listeners who heard a multi-voice excerpt were far more open to it afterwards than the group as a whole had been at the start. A free first chapter lets your readers judge the recording, not the label.

The bottom line

Listeners are wary of AI narration as an idea, but many judge it on what they actually hear. In a large 2026 blind study, a multi-voice AI version was rated above a single human narrator for character-driven stories, while the human narrator was rated higher for exposition. If you choose AI, make it a real production and tell listeners what it is. You can cast every character and produce your first chapter free with AudioForge, then let your readers hear it.

None of this asks you to write differently for the page. The habits that make a scene easy to follow by ear are the same ones editors have always asked for. In audio, they decide whether a line reaches the right voice.

Below: why tags carry more weight in audio, the habits that make attribution clear, whether an existing manuscript needs changes, and how to check the cast before you generate anything.

Why do dialogue tags matter more in audio?

Because the speaker has to be decided before a line is voiced. In a full-cast audiobook each character has a voice of their own, so someone, or something, must work out who says every line.

AudioForge is a production tool that turns a manuscript into a multi-voice audiobook, with a distinct voice for every character. Its parser reads the prose and assigns a speaker to each line from the same clues a reader uses: tags, names, paragraph breaks. When those clues are missing, the line is marked Unsure and left for you to assign.

Clear attribution therefore saves you review time. The fewer lines the parser has to guess at, the fewer you have to check.

Which habits make the speaker clear?

Five small ones cover most scenes.

1. Name the speaker early in a scene. The first line or two should leave no doubt about who is present and who is talking.
2. Keep one speaker to a paragraph. A new paragraph signals a new speaker. Two people sharing a paragraph is the most common cause of a misattributed line.
3. Anchor long exchanges. After three or four untagged lines, add a tag or a beat. With three or more people in the room, tag almost every line.
4. Use names when pronouns collide. Two characters who are both "she" need their names more often than feels natural on the page.
5. Mark thoughts as thoughts. "She thought" or a consistent convention keeps inner voice apart from speech.

Here is the same exchange written two ways.

Do you need to rewrite your manuscript?

No. A manuscript written for readers usually works as it stands, and the fix for a misattributed line belongs in the script, not in the book.

When the parser assigns a line to the wrong character, you change the speaker on that line in the Director Studio. Your manuscript stays exactly as you wrote it. The help center shows where that happens in the workflow.

The habits above pay off most for chapters you have not written yet. If you publish a serial and plan to release audio alongside it, a tag here and a name there costs nothing while drafting and saves a correction later. A full audio drama is the one case that asks for a rewrite, with tags replaced by character cues, as our guide to audio drama script format shows.

How do you check the cast before generating audio?

Go to the uncertain lines first, then read one crowded scene. That finds most problems in a few minutes.

• Jump through the lines marked Unsure and assign each one.
• Pick the scene with the most speakers and skim it line by line. Crowded scenes are where attribution slips.
• Look at any exchange with no tags at all for a long stretch.
• Generate that one scene and listen. A wrong voice is obvious at once.

If a line still comes out in the wrong voice, change its speaker and regenerate it alone. Unchanged lines are reused, so a correction never means redoing the chapter. You can try this on a chapter of your own from the home page.

The bottom line

Write so that a listener always knows who is speaking: name the speaker early, give each speaker their own paragraph, and anchor long exchanges with a tag or a beat. Your existing manuscript does not need rewriting, because a misattributed line is fixed in the script. Check the uncertain lines and one crowded scene before you generate, and the cast will be right.

AudioForge is a production tool that turns a manuscript into a multi-voice audiobook, with a distinct voice for every character. The steps below are the same ones shown in the help center, with what to look for at each stage.

Below: what you need to start, how the script is made, how casting and direction work, and how the finished chapter comes out.

What do you need before you start?

A chapter of your manuscript and a free account. You create a book, then add text in whichever way suits you.

• Paste a chapter straight in.
• Upload a chapter as a .docx or .txt file.
• Import a whole manuscript as .docx, .txt or PDF. It is split into chapters, which you can rename, reorder, split or join before anything else happens.

Each chapter becomes its own audio file. That is why serials are a natural fit: you can produce a chapter as soon as it is written.

How does a chapter become a script?

The parser reads your prose and writes it out as a screenplay. Every line gets two labels: who speaks it, and how it should be delivered. A nervous aside and a shouted warning from the same character end up with different directions.

The parser is not always certain. When it cannot tell who said a line, the line carries an Unsure tag, and "Next uncertain line" takes you from one to the next. Checking those lines is the most valuable few minutes you will spend, because a wrong speaker is much easier to fix in the script than to notice in the audio.

How do you cast the voices?

You pick a voice for the narrator and for each character from the Voice Library. Casting is saved for the whole book, so a character who appears in chapter one sounds the same in chapter forty.

Two things help with a long series:

• Mark voices you like as favourites so you can find them again.
• Add invented names to the book's pronunciation glossary once, so they are said the same way every time.

If you are still deciding how far to go, what a dramatized audiobook is compares a full cast with a dramatized production.

How do you direct a line?

You edit it. In the Director Studio any line can be changed: the speaker, the delivery, the text itself, or the pause after it. What is in the script is what gets voiced, so a correction on the page is a correction in the audio.

You do not need to direct every line. Read through a tense scene, adjust the deliveries that matter, and leave the rest. Edits save automatically, and undo and redo work as you would expect.

How do you get the finished audio?

You generate the chapter and download one file. Generation runs in the background, so you can leave the page and a notification tells you when it is done.

Listen to the result. If a line sounds wrong, change it and regenerate. Only the lines you changed are produced again, and everything else is reused.

The bottom line

A full-cast chapter is four steps: parse, check, cast, generate. The parser does the heavy lifting and flags what it is unsure about, so your time goes into decisions rather than recording. Start with a single chapter, listen to it, and fix individual lines until it sounds the way you hear it. You can try it with one chapter for free during early access.

Rules in this area change quickly. Every claim below links to the store's own page and reflects what that page said in October 2026, so follow the link before you plan a launch around it.

The question matters most if you produce the audio yourself. AudioForge is a production tool that turns a manuscript into a multi-voice audiobook, with a distinct voice for every character. You download the finished files and decide where they go, which means the stores' rules decide your options.

Below: the short answer in a table, what each store asks for, how to disclose the narration, and where else the audio can go.

Which stores accept AI narration?

As of October 2026, Spotify, Kobo and Google Play Books accept AI narration made elsewhere, Audible through ACX does not, and Apple Books publishes no rule; what decides it is whether the store has a written rule.

What do Spotify, Kobo and Google Play ask for?

Spotify, Kobo and Google Play Books all ask you to say plainly that the voice is synthetic.

Spotify's help page on digital voice narration says it accepts audiobooks narrated this way. When you upload through Spotify for Authors, you select the option "This audiobook uses digital voice narration", and Spotify adds a short sentence to the book's description so listeners know. One limit: Spotify says it does not currently share digitally narrated audiobooks with its referral partners.

Spotify also has a tool of its own. As of October 2026, its Audiobook Creation Tools turn an EPUB into an audiobook narrated by a digital voice you choose, for authors in the US writing in English, up to five projects per account. Those titles carry a "Narrated by Digital Voice" label on Spotify.

Kobo Writing Life's page on AI narration says Kobo accepts it too. You list the narrator as "Read by" one of Kobo's "Synthesised voice" options: male, female or unspecified.

As of October 2026, Google Play Books covers it in its Partner Center guide to adding a book. The narrator field offers Human narrator or Synthesized voice, and the page says to select Synthesized voice if you upload an AI-generated audiobook. You then pick unspecified, male, female or based on a real actor.

Does Audible accept AI narration?

Audible does not accept AI narration that you upload yourself. ACX is the service independent authors use to reach Audible, and its audio submission requirements state that a submitted audiobook must be narrated by a human unless otherwise authorized. Unauthorized text-to-speech, AI or automated recordings are prohibited.

That may change. The same page says Audible is working to accept text-to-speech content from other publishers and creators, and will share timing later. Until it does, treat Audible as closed for an audiobook you produced with AI voices. The rule is explicit, so do not try to slip one through.

Amazon's own AI route is separate. Its virtual voice audiobooks are made by Amazon from an eligible Kindle eBook. Kindle Unlimited became Kindle+ in October 2026, as Written Word Media reported, and KDP's Kindle+ help page says virtual voice audiobooks join it automatically, in the US only, when the eBook is in KDP Select. That route does not take audio made elsewhere; we cover what it costs you in what authors give up with Audible virtual voice.

What about Apple Books?

Apple Books sells AI-narrated audiobooks, but its published page describes only its own tool. Apple offers digital narration to independent authors and small publishers, produced from an ebook file.

The page states no rule for audio narrated with other AI voices. Audiobooks usually reach Apple Books through a distributor, and distributors set their own conditions. One large distributor, for example, accepts digital narration only from providers it has approved.

Before you count on Apple Books, ask the distributor or Apple's support in writing whether they will take your files and how they want the narration labeled.

How should you disclose AI narration?

Disclose AI narration in the place each store gives you, and in your own words as well.

1. Select the store's digital or synthetic narration option when it has one.
2. Credit the narrator as a synthesized voice, never under an invented human name.
3. Mention it in your description, so listeners who care can choose.

Never submit AI narration as human narrated. It is the one mistake a clear disclosure cannot fix afterward.

Where else can the audio go?

AI-narrated audio you own can go anywhere you control. With AudioForge the audio files are yours, as the IP policy sets out, so you can release chapters to patrons on Patreon, post them on YouTube, or sell the audiobook from your own site.

That suits serial fiction well: a chapter a week, in audio, for readers who already follow the story. Each chapter exports as its own file, and the help center shows the steps from manuscript to download. Stores can come later, once their rules are clear for your book.

The bottom line

Publish where the rule is written down. Today that means Spotify, Kobo and Google Play Books, each with a clear disclosure, plus any channel you run yourself. Audible stays closed to AI-narrated uploads through ACX until it says otherwise, and Apple Books needs a direct question before you rely on it. Decide on stores after you have heard a chapter: produce your first one free during early access, following the steps in the help center.

Ask authors what they think of AI audiobooks and you hear the same words: flat, one-dimensional, dull. Listeners notice within a minute, and the book is judged before the story has a chance. The verdict is deserved, but it is a verdict on one way of making audio, not on the idea.

Below: what a single voice cannot do, what changes with a cast, how the work is done, and how to judge the result yourself.

What can a single voice not do?

It cannot show who is speaking. On the page, a reader knows from the tags, the paragraph breaks and the rhythm of the scene. They hear each character differently in their head, and the narration in between has its own calmer register.

Hand all of that to one voice and the structure collapses. Two characters arguing become one person talking to themselves. The narrator's dry aside arrives with the same energy as the line before it. Thoughts are indistinguishable from speech. Every word is pronounced correctly, and the scene still does not land.

A human narrator solves this by acting: shifting pitch and pace between characters, dropping their voice for the narration, leaning into the tense lines. A text-to-speech voice cannot, because all it is given is words.

What changes when every character has a voice?

Three things change at once: who speaks, how they speak, and what you can fix.

• Every character gets a distinct voice. The listener no longer works out who is talking from context, because the voice tells them. The narrator has a voice of their own.
• Every line gets a direction. A line marked nervous is delivered nervously, and a line marked as a whisper is whispered. The same character can be furious in one paragraph and exhausted in the next.
• Every line can be fixed on its own. When a delivery misses, you change that line and regenerate it, not the chapter.

This is what audio drama producers have always done. What kept it out of reach for an independent author was the cost of a studio, a cast and an engineer for a sixty-chapter serial. Human productions often settle on two narrators instead, the trade-off set out in dual vs duet narration.

How is a full-cast audiobook made from a manuscript?

The chapter becomes a script first, and the script is what gets voiced. AudioForge is a production tool that turns a manuscript into a multi-voice audiobook, with a distinct voice for every character, and the work runs in four steps.

1. Add a chapter. Paste it or upload it. Each chapter becomes its own audio file.
2. Parse it into a script. Every line is tagged with who speaks it and how it should be delivered. Lines the parser was unsure about are flagged for you to assign.
3. Cast the voices. The narrator and each character get a voice, saved for the whole book, so a character sounds the same in chapter one and chapter forty.
4. Generate and export. The lines are voiced, stitched with the pauses you placed, and downloaded as one file.

The parsing step is the one single-voice tools skip, and it is where most of the difference comes from. The help center walks through each step, and our comparison of AI audiobook generators shows which tools offer a voice per character.

How do you know if it is good enough?

Listen before you decide. Scepticism about AI narration is reasonable, and an argument will not settle it. A sample will.

Play the full-cast sample on the home page and wait for the moment two characters speak to each other. Compare it with the last single-voice AI audiobook you gave up on.

Then try a chapter of your own. It will not be perfect on the first pass. The parser can be unsure who said a line, and a delivery can miss. That is why uncertain lines are flagged, and why a single line can be redone without touching the rest. During early access it costs nothing to find out, and you keep every right to the text and the audio.

The bottom line

Flat AI narration is a production choice, not a limit of the technology. One voice reading a whole chapter cannot tell characters apart or change its delivery, so it sounds dull. Give every character a voice and every line a direction, and the same book sounds like a performance. Hear it on your own chapter before you judge.