Should an Audio Drama Have a Narrator?

Short answer: an audio drama needs a narrator only when the story depends on something dialogue and sound cannot carry, such as a prose voice worth keeping, a big jump in time, or a single point of view. If your scenes can be understood from voices and sound alone, cut the narrator. Most indie productions land in between: a narrator for the seams between scenes, and a full cast inside them.

This is the biggest structural choice in audio fiction, and it is worth making on purpose rather than by default. Below is how to decide, what goes wrong on each side, and how to build the hybrid most productions actually use. If you are still choosing between an audiobook and an audio drama in the first place, start with audio drama vs audiobook; this post assumes you have already picked drama.

The question to ask first: what would the listener miss without one?

Take a scene from your script and delete every narrator line. Then read what is left out loud, or better, listen to it. Ask one question: what does the listener no longer know?

• If the answer is "nothing", the narrator was repeating what the dialogue and sound already said. Leave it out.
• If the answer is "where we are" or "how much time passed", you have a transition problem. That can be solved with a narrator line, but often it can be solved with sound instead.
• If the answer is "what the main character is thinking" or "the tone of the book", you have a narrator problem that sound cannot solve. Keep a narrator.

Run the test on three or four scenes, not one. A pattern will show up quickly, and it usually points to one of the three answers below.

When a narrator earns its place

A narrator is not a crutch. In some productions it is the most important voice in the cast.

Adapting prose you want to keep. If you are turning a novel into audio and the writing style is part of why people read it, stripping the narration throws away the thing readers came for. A first-person novel with a strong narrative voice is the clearest case: the narrator is the character, and removing them removes the book.

Big time jumps. Sound can move a listener from a kitchen to a street in a second. It is much worse at "eleven years later." A single narrator line handles a jump that would otherwise take an awkward exchange of dialogue like "It has been eleven years, hasn't it?" which nobody says in real life.

A story told through one point of view. If the tension comes from what one character notices, suspects, or hides, the listener needs access to that head. Dialogue only shows what people say out loud. A narrator, or a character narrating in their own voice, gives you the inside.

Framing devices. Anthologies, found-footage style stories, and tales told by someone in the story itself often need a voice that stands outside the scenes. That voice is doing structural work, not description.

When a narrator kills the drama

The most common failure in first audio dramas is not a missing narrator. It is a narrator describing what the listener has already heard.

If the door creaks, the narrator does not need to say "the door creaked open." If a character is crying, the performance carries it and "she said through tears" only slows the scene down. Every narrator line that repeats the sound design costs you twice: it breaks the moment, and it teaches the listener that the sound does not matter because someone will explain it anyway.

The second failure is the old radio-play feel, where a narrator walks the listener through every scene in the same even tone. It can be charming in a period piece. In most modern productions it makes listeners drift, because the narrator becomes the loudest voice and the characters become illustrations of what the narrator said.

A good warning sign: if you find yourself writing narrator lines that start with "Meanwhile" or "Moments later" in most scenes, the narrator is doing scene-transition work that sound can usually do better. The mechanics of that, silence, ambience changes, and cues at the seam, are covered in how to use sound for scene transitions in an audio drama. This post will not repeat them; the point here is that if sound can carry the transition, the narrator line should go.

The hybrid most indie productions actually use

Between "narrator everywhere" and "no narrator at all" sits the shape most independent audio dramas settle on:

• The narrator speaks at the seams. One or two lines when a scene opens somewhere new, when time jumps, or when a chapter begins. Then the narrator goes quiet.
• Inside a scene, the cast takes over. Dialogue, performance, and sound carry everything from the first line of the scene to the last.
• The narrator never explains a sound. If the listener can hear it, the narrator does not say it.

In practice this means going through your script and sorting each narrator line into one of two piles: lines that open or close a scene, and lines inside a scene. Keep most of the first pile. Cut most of the second, or turn it into sound or dialogue. A line inside a scene survives only if it tells the listener something they cannot hear, like a thought or a detail no one would say out loud.

The result sounds like a drama, not a reading, but the listener never gets lost.

Casting the narrator so it does not collide with a character

Once you keep a narrator, even a quiet one, the casting matters more than people expect. The listener has no visual cue for who is speaking, so two similar voices blur together, and the narrator is the one voice that appears across every scene.

• Make the narrator audibly different from every main character. Different pitch range, different pace, different texture. Not just a different preset that sounds similar.
• Watch the first-person trap. If the narrator is also a character, as in a first-person adaptation, use one voice for both. Two voices for the same person is confusing.
• Keep the narrator calmer than the cast. The characters carry the emotion inside scenes; the narrator carries orientation. A narrator who performs as hard as the cast competes with them.
• Audition against real lines. Test each candidate narrator voice on the actual scene-opening lines from your script, right before a character speaks, so you hear the handoff.

If your production also has an announcer who introduces episodes, that is a separate role from the narrator; how to make a radio drama with AI voices covers the announcer. For the broader question of one voice versus a full cast, see single voice vs full cast.

How to try both versions of the same scene

The fastest way to settle this is to hear it. In AudioProducer you paste a chapter or import an EPUB, and Auto-Assign Characters tags every line by speaker, with the narrator as its own speaker alongside the named characters. You cast a voice for each character and a separate voice for the narrator from the Voices page, or clone your own voice for the narrator if you want to read it yourself. Auto-Assign Sounds places music, ambience, and sound effects as a starting point you can adjust.

From there, make two versions of one scene: one with the narrator lines intact, and one with them cut down to the seams. Add a longer pause where a narrator line used to be, so the sound has room to do the work. Generate both and download the MP3s. Listen to them back to back, away from the script. Most writers know within a minute which one is the drama.

AudioProducer makes the audio; where you publish it is up to you. The free tier includes 1,200 words per month, which is enough to test a scene both ways. Start a free project and hear your scene with and without a narrator. For more guides on audio drama, casting, and sound, browse all articles.