Dorian
Audiobook baritone for literary fiction and slow reveals
Workflow
Turn a Story into Audio
Hearing your own story read aloud is the oldest editing tool there is — it exposes every clumsy sentence instantly. Doing it with a proper narrator and a cast turns that editing tool into something you can publish.
Why this works here
It is the best proofreading pass available
Repeated words, run-on sentences, dialogue that no human would say — all of it becomes obvious the moment it is spoken. Writers have read work aloud for centuries for exactly this reason; the only new thing is not having to do the reading yourself.
Characters become distinguishable
A story with four speaking parts read by one narrator forces the listener to track who is talking. Cast them separately and that work disappears, which is the single biggest quality difference between amateur and professional audio fiction.
It works from a draft
You do not need a finished manuscript. Paste a chapter, hear it, change it, hear it again. The loop is fast enough to be part of writing rather than something you do at the end.
It publishes
The same project that helped you edit produces the file you put on a podcast feed, a video, or an audiobook distributor. There is no second production process.
Voices that suit this kind of work
Wren
Bedtime-story narrator, slow and warm, never wakes anyone up
Hale
Documentary narrator with a hand on the listener's shoulder
Sable
Noir detective narrating her own bad decisions
Kestrel
Low, weary swordswoman who has already lost this war once
Orrin
Old man on a porch who remembers the flood
Hazel
Mum voice: patient, then suddenly not
Solène
Continental accent, unhurried, very expensive sounding
How to do it
- Start with one chapter
Not the whole book. A chapter is small enough to iterate on and large enough to reveal whether the narrator works.
- Cast the narrator on your worst paragraph
Audition on the dense one with the difficult names, not on the opening line. That is where narrators fail.
- Let it find the dialogue
Quoted speech with attribution becomes character assignments automatically. Check them, fix what is wrong, move on.
- Do the pronunciation pass
Names, places, invented words. Fix each once; the fix applies to the whole project and to every later chapter.
- Listen at normal speed
Do not skim. The problems you are looking for — pacing, repetition, a scene that drags — only appear in real time.
- Export and publish
Mixed MP3 for a feed, stems if you want music, SRT if you are making a video version.
Reading aloud as an editing tool
Every writing guide recommends reading your work aloud, and almost nobody does it, because reading 5,000 words aloud is tiring and you start skimming by page three. The advice is nevertheless correct: the ear catches things the eye has been trained to skip. Repeated sentence openings, a paragraph with no variation in rhythm, a line of dialogue that no person would say — all of these are invisible on the page and glaring in audio.
Having the text read to you removes the fatigue problem and adds something else: an unfamiliar voice. Reading your own work in your own voice, you unconsciously perform it the way you intended. A neutral narrator does not, which means you hear what is actually on the page rather than what you meant.
Narration and dialogue are different problems
Prose narration wants an even, unobtrusive voice that can hold attention across hours. Dialogue wants distinct, characterful voices that are immediately separable. These requirements conflict, which is why single-narrator audiobooks are a compromise — the narrator is chosen for endurance and then asked to do character work they were not cast for.
Splitting the two lets you optimise both. Cast the narrator for stamina and clarity; cast the characters for distinctiveness. The listener then gets the best version of each, and long dialogue scenes stop being the part where attention drifts.
The one rule is separation: the narrator must not sound like any character. If your narrator and your protagonist occupy the same pitch band, the audience will spend the whole book unsure whether they are hearing description or speech.
Short stories are the sweet spot
A short story is between fifteen and sixty minutes of audio, which fits inside a modest plan, can be checked in one sitting, and is a complete publishable unit. It is also a format with almost no commercial audio production, because the economics of hiring a narrator for a 5,000-word story have never worked.
This makes short fiction the most immediately practical use here. Literary magazines have started publishing audio versions, podcast feeds of short fiction are a growing format, and individual writers can now produce a listenable version of every story they publish for the cost of an afternoon.
What to expect, honestly
The result is a clear, well-paced, consistently cast reading. It is not an interpretation. A great human narrator makes decisions about subtext on every page, and those decisions are the reason people follow particular narrators from book to book. Nothing here does that.
What you get instead is availability. Every story you write can have an audio version, immediately, at no marginal cost, revisable in minutes. For most writers that trade is straightforwardly worth making, and it is worth making with clear eyes about which half of the trade is which.
Adapting prose rather than reading it
There is a step between a manuscript and a good audio version that most people skip: adaptation. Prose written for the eye contains things that do not survive being spoken — long parenthetical asides, sentences with three nested clauses, paragraphs that rely on the reader's ability to glance back.
Adapting does not mean rewriting the story. It usually means splitting a handful of sentences, removing two or three parentheticals, and converting a passage of internal reflection into something the narrator states plainly. Twenty minutes per chapter, and the audio version stops feeling like a reading of a book and starts feeling like a piece of audio.
The one thing worth adding rather than removing is attribution. Prose drops "she said" as soon as an exchange establishes rhythm, which readers follow effortlessly and listeners lose after about four lines. Putting a few tags back costs the text almost nothing and saves the listener real effort.
Publishing a short story as audio
The distribution question is simpler than people expect. A short story is fifteen to sixty minutes, which is a podcast episode. Setting up a feed takes an afternoon, and a feed of your own short fiction is a genuinely useful thing to own, independent of any platform.
The alternative is posting the file alongside the text wherever the story already lives — a personal site, a newsletter, a literary magazine's page. Readers who prefer audio will use it and the rest will ignore it, which is exactly the right outcome for an option rather than a replacement.
Other workflows
Voice categories to cast from
What it costs
Free
$0 / month
Enough to finish a short scene and hear what your script sounds like cast.
- 10,000 characters every month — the quota resets, it is not a one-time trial
- 2 cast members per project
- 5 minutes of finished audio
- Exports carry a short CastDub watermark
Creator
$12 / month
For one person turning their own stories, fanfic or scripts into finished audio.
- 5 cast members per project
- 60 minutes of finished audio a month
- No watermark, commercial license included
- 1 voice clone slot
- SRT subtitle export
Studio
$29 / month
For serialised work: long-running dramas, audiobooks, dubbing pipelines.
- Unlimited cast members per project
- 200 minutes of finished audio a month
- Line-by-line emotion control
- Multitrack and stem export (one WAV per character)
- 3 voice clone slots
Team
$69 / month
For a studio where a writer, a director and an editor touch the same project.
- 5 seats, shared projects and shared character cards
- 500 minutes of finished audio a month
- 10 voice clone slots
- Priority rendering queue
Frequently asked questions
Will this replace a human narrator?
For a bestselling novel with a budget, no — a great narrator does interpretive work that synthesis does not attempt. For a short story with no budget, a self-published novel, or a draft you want to hear, it is the difference between having audio and not having it. Those are the cases this serves.
Can I use it just for editing?
Yes, and a lot of people do. The free plan's five minutes a month is enough to check the passages you are unsure about, which is often exactly what you need.
What about poetry?
Poetry is the hardest case. Line breaks, deliberate ambiguity and rhythm are carried by a reader's interpretation, and synthesis renders the literal text. It can be useful for checking meter; it will not perform a poem.
How do I stop it mispronouncing a name?
Add a pronunciation override once. It applies everywhere that word appears in the project, including in later chapters, which is why doing this pass early matters.
Can I mix my own recorded narration with generated character voices?
Yes, and it is an effective hybrid: narrate yourself, generate the characters. Record in a reasonably dry room at a similar level and the two sit together without difficulty.
How long is a novel in audio?
Roughly one hour of audio per 9,000 words. A 90,000-word novel is about ten hours, which is more than any single month's allowance — plan it across months or edit for a shorter adaptation.
Cast your first scene tonight
Paste a script, let CastDub assign a voice to every character, and export a finished drama.
Create for freeNo credit card. Free plan renews every month.