Should a prologue be read by a different narrator than the rest of the audiobook?

Only when the prologue is genuinely a different speaker, such as a first-person prologue from a character who is not your point-of-view character, a found document, or a scene set long before the main story. If the prologue is the same narrator shortly before chapter one, keep the same voice and separate the two with a longer pause instead. A voice change there tells the listener a new character has arrived, and then nobody arrives.

Do you read the word Prologue out loud in an audiobook?

Usually yes. It is the only label a listener gets, since they cannot see the heading on the page, and it costs about a second. In AudioProducer.ai there is no separate heading field that can be spoken or skipped independently of the body, so the narration is whatever text is in your project. If you want the word read, leave it in the text. If you do not, remove it before you build.

How do I make an epilogue use the same voice as the prologue ten hours later?

Tag the epilogue's lines to the same speaker you created for the prologue rather than choosing a similar-sounding voice again by ear. Reusing the speaker draws the identical voice with no drift, which is what makes the callback land. Re-auditioning voices at the end of a long project tends to produce two reads that are close enough to feel wrong and far enough apart for a listener to notice.

How to Narrate a Prologue and an Epilogue

A prologue is usually not chapter one wearing a different hat. It is often a different year, or a voice that never comes back after page four. On the page, a heading and a half-inch of white space tell the reader that before they read a word. Audio has no white space. Read the prologue in the same voice at the same pace as chapter one and the listener has no way to know they were moved. Here is how to cast and pace a prologue and an epilogue in an AI-narrated audiobook.

The prologue is a casting question

Before you think about pacing, decide who is reading. In AudioProducer.ai, narration is assigned by speaker: you tag lines in the editor to a speaker, and each speaker gets a voice you pick from the Voices page, where you can preview before committing. So a prologue in a separate voice comes down to a second speaker that you tag those lines to, then assign a voice to.

Auto-Assign Characters gives you a starting point and is worth running, but know how it fails here. It keys off dialogue and attribution, so it is good at finding people who speak and get named. A prologue narrated in an unnamed voice has neither, and it will almost always be swept into your default narrator with the rest of the prose. Plan on tagging those narration lines by hand.

When a second voice earns its place

A different voice works when the prologue is genuinely a different speaker. A first-person prologue from someone who is not your point-of-view character. A found document, a letter, a ship's log. A scene set two hundred years earlier with a narrator who never reappears. In each of those the voice change does the work the page break used to do, and the listener reads it correctly on the first sentence.

It backfires in one common case: the prologue is the same narrator, a few days before chapter one. A cold open. Swap voices there and you have told the listener someone new is speaking, so they spend twenty minutes waiting for that person to show up. Nobody arrives. Keep the same voice and separate with pacing instead. If your book runs two narrators throughout rather than for one opening section, that is a different structure, and dual point-of-view narration covers it.

The handoff into chapter one

Silence is the main transition tool audio has, and AudioProducer gives you pauses at more than one level: a project-wide default under Edit Project, a per-paragraph override, and an inline pause you can drop inside a line. For a prologue handoff the default is almost never long enough. A pause that reads as a page turn shrinks to a paragraph break when the listener is also absorbing a jump in time and speaker. Set a noticeably longer pause on the prologue's last paragraph than you use between ordinary chapters.

What AudioProducer will not do is score the transition. Auto-Assign Sounds places sound effects the text itself implies, so it will find a door or a thunderclap in your prose, but there is no way to lay a music bed under a scene change. If you want a musical sting between the prologue and chapter one, build the audiobook here, download the MP3, and add the bed in whatever audio editor you already use. AudioProducer exports MP3 and stops there. It does not distribute or upload your audiobook anywhere. More on the pacing side in pauses and dramatic timing, and on the structural side in splitting your manuscript into chapters.

The epilogue problem: a voice from ten hours ago

An epilogue that returns to the prologue's voice is a real payoff, and also where this gets fumbled. Ten hours of listening sit between the two. A voice heard for four minutes at the start has to be recognized instantly at the end, or the callback does not happen.

The mechanical answer is to reuse the same speaker tag rather than picking a similar-sounding voice again by ear. Tag the epilogue's lines to the speaker you already created for the prologue and it draws the identical voice, with no drift. Re-auditioning at the end of a long project is how you end up with two voices close enough to feel wrong and far enough apart to notice. The same discipline across a whole book is covered in keeping a character voice consistent.

If the epilogue is a jump forward for the same narrator, the voice stays and per-line dialogue emotion carries the change. An epilogue set years later usually wants a settled read rather than the tension the book ended on.

Does the word "Prologue" get read aloud?

Usually yes, for a practical reason: it is the only label the listener gets. A print reader sees the heading. A listener gets whatever the narrator says. Reading "Prologue" aloud costs one second and removes the question. Same for "Epilogue" at the other end.

One thing to know about AudioProducer here: there is no separate heading field that gets spoken or skipped independently of the body. The narration is the text in your project. If the word is in the text, it is read, and if you do not want it read, take it out before you build. Chapter heads work the same way, and chapter intros and outros goes deeper on what to say around one.

Frame stories, where the outer narrator returns

A frame story is the hardest version of this and the one that rewards planning most. An outer narrator opens the book, hands off to the inner story for three hundred pages, and returns at the end. Treat that outer narrator as a full speaker from day one. Preview its voice against the main narrator to be sure the two are clearly distinguishable rather than merely different, and tag every outer-frame passage to it, including mid-book interruptions where the frame breaks in. Where that narrator is unreliable, the casting stakes go up, and narrating an unreliable narrator is the closer read.

You can test all of this cheaply. The free tier covers 1,200 words with no card, enough for most prologues plus a page of chapter one, which is exactly the seam that is hard to judge from the manuscript. Build the prologue, cast a second speaker, set the handoff pause, and listen to the join. Paid plans start from $39.99/month for the whole book. If you want a cloned voice for a frame narrator, cloning requires your own material or documented consent from the voice owner.

The other non-prose material around the edges of a book follows the same logic. Narrating footnotes and endnotes covers what to keep, what to cut, and what to move to a companion document.