How to Turn Fanfiction into an Audio Drama

Podfic — fanfiction recorded aloud — has existed for as long as fandom has had somewhere to host audio. It has always been limited by the same arithmetic: a 10,000-word fic is about an hour of finished audio and several hours of work, performed by one person doing every voice.

The demand was never the problem. Fanfiction is unusually dialogue-heavy and built around the dynamic between two or three specific characters, which is exactly what audio is good at. The bottleneck was production.

This is a guide to doing it with a cast instead of a single narrator — what changes, what gets harder, and the etiquette that keeps this welcome in fandom spaces.

Step one: pick the right fic

Not every fic works as audio, and picking badly is the most common reason a first attempt is disappointing.

Good candidates are dialogue-heavy, have a small cast, and depend on the relationship between two or three characters. Scenes rather than summary. Present-tense tension rather than retrospective reflection.

Difficult candidates include anything with heavy internal monologue, elaborate formatting jokes, chat-fic and text-message formats, and fics whose effect depends on how they look on the page. Epistolary work is genuinely hard; a fic told in emails is a structural problem, not a casting one.

Length matters more than people expect. Prose runs at roughly 150 words per minute of finished audio, so a 5,000-word chapter is about thirty-three minutes. Start with something under 3,000 words for your first attempt. You will learn more from finishing a short one than from abandoning a long one.

Step two: understand why prose is harder than script

A screenplay labels every speaker. Prose does not. It uses attribution tags — she said, he muttered — and then drops them once a rhythm is established, so a page of back-and-forth can run twenty lines with no names in it at all.

Human readers track this effortlessly. Automatic speaker detection does not, and this is where your attention goes.

The reliable trouble spots are:

• Long unattributed exchanges, especially two-character banter.
• Three-way conversations, where the inferred alternation breaks down.
• Characters referred to by epithet rather than name — "the taller man", "the blond" — which is a deeply ingrained fanfic habit.
• Dialogue interrupted by a paragraph of action, after which the attribution resets.

Scan the assignments before you render, not after. It takes a couple of minutes and it saves the frustration of hearing the wrong character say something important.

Step three: cast the narrator first

The narrator carries the prose, which in most fic is more than half the runtime. Get this wrong and nothing else matters.

Two rules. First, audition on your worst paragraph — the dense descriptive one, or the one with the invented names — not on the opening line. Second, listen to three full minutes, not three sentences. A voice that is charming for ten seconds can be unbearable for thirty minutes, and that only becomes apparent at length.

Then make sure the narrator is clearly separated from every character. If your narrator and your protagonist occupy the same pitch band, listeners will spend the whole fic unsure whether they are hearing description or speech. This is the single most common flaw in amateur audio fiction and it is entirely avoidable.

Step four: cast the characters for separation

Listeners identify speakers by pitch band and speaking rate, with no visual information at all. Two characters in the same band will blur, no matter how different their personalities are on the page.

The practical method: cast your two leads as far apart as the characterisation allows, then place everyone else in the gaps. If your fic has two characters who are both described as having low voices, one of them is going to have to move, and the story will survive it.

A useful discipline is the eyes-closed test. Render a page of dialogue, listen without looking at the script, and see whether you can follow who is speaking. If you cannot — and you wrote it — nobody can.

Step five: save character cards before chapter two

This is the step people skip, and skipping it is the single most common reason long podfic projects collapse.

A character card holds the voice, the pace, the default emotional register and any pronunciation fixes, under the character's name. When you open chapter twelve three months later, you load the card rather than trying to remember what you chose in chapter one.

Drift does not come from the technology. It comes from a human being re-picking a similar-but-different voice weeks later. Store the decision and the problem disappears.

While you are there, do the pronunciation pass. Every fandom has names that are guessed wrong — invented places, non-English names, terms from the source material. Fix each once, early, and it applies to every chapter that follows.

Step six: direct the beats, not the whole fic

The temptation is to mark every emotional line with an emotion. The result is uniform mush, because contrast is what makes an emotional moment land.

Set each character's resting register — guarded, eager, flat — as their default. Then mark only the specific lines where that register breaks. A confession, a threat, the moment someone stops pretending. In a 3,000-word chapter you should be directing perhaps five lines by hand.

This is also the single biggest quality difference between amateur and professional audio fiction, and it costs nothing but restraint.

Step seven: the etiquette

Fandom's norms around transformative work are informal but strongly held, and generated audio has made people justifiably nervous. Three habits keep you welcome.

Ask the author. If you are podficcing someone else's fic, message them. Many authors have a blanket transformative-works statement in their profile; where there is not one, a short request is normal courtesy and is almost always answered warmly. Credit them prominently wherever you post.

Disclose that the voices are generated. Communities that would happily listen to a generated podfic react badly to finding out afterwards, and the reaction is about honesty, not about technology. One line in the post description settles it.

Do not clone real people. This is the hard line, and it is the most common question on this topic. Cloning the voice of an actor who played a character is a publicity-rights problem and a policy violation at every reputable tool. It is also the thing rights holders actually pursue, and a corner of fandom that becomes known for it puts the tolerance that lets transformative work exist at risk for everyone.

Cast a voice with a similar register instead. It is legal, it works, and nobody minds.

Step eight: publish in chapters

Work in the units the fic is published in. A chapter is a publishable unit, a re-renderable unit and a listening-check unit all at once, and organising the project this way is what keeps a long work manageable.

Listen to each chapter end to end at normal speed before you post it. Do not skim. The problems you are looking for — a scene that drags, a repeated phrase, a voice that stops being distinguishable when two characters get emotional — only appear in real time.

What to expect

A clear, well-cast, consistently performed reading with a different voice for every character. That is genuinely better than most solo podfic for dialogue-heavy work, because a single reader doing four voices is a compromise listeners tolerate rather than enjoy.

What you will not get is interpretation. A skilled podficcer makes choices about subtext on every page, and those choices are why people follow particular podficcers. Nothing generated does that.

The honest framing is that this makes audio versions available for fics that would never have had one. That is a real gain, and it is worth being clear-eyed about what it is not.