How to turn a novel passage into narration and character voices

REAL ENGLISH AUDIO

Listen: The letter in the attic

A real generated English passage with narration and one character voice. Narration uses Tim; Elena uses Dacey. A 0.45-second pause separates each segment. No music or cloned voice is included.

Generated 2026-10-26 · 17.27 seconds · Real speech-service output · No generation request is made when you play this file.

Source text for this example

Original fiction passage:

Elena found the letter beneath a loose floorboard. The paper was cold, though the attic was warm.
"This is my handwriting," she whispered.
Outside, the clock struck thirteen. Elena had never heard the bell before.

Reviewed role-labelled adaptation used for the audio:

Narrator: Elena found the letter beneath a loose floorboard. The paper was cold, though the attic was warm.
Elena: This is my handwriting.
Narrator: Outside, the clock struck thirteen. Elena had never heard the bell before.

These are original example texts. The audio uses the role-labelled version shown above.

1. Start with a short, self-contained passage

Use a passage that introduces its speakers clearly. The example contains narration, one quotation and a return to narration. Paste it into the script input and parse it. Before choosing voices, check whether the text has been divided into useful spoken segments.

Long chapters create more opportunities for ambiguous attribution. Work scene by scene while you establish the cast. This also makes it easier to review pronunciation and regenerate a small part when you change the text.

2. Verify the distinction between narration and speech

The first and final segments belong to the narrator. The quotation belongs to Elena. If the parser assigns that quotation incorrectly, change the role yourself or rewrite the input in explicit Character: dialogue format. Automated suggestions do not replace reading the passage.

Decide how to handle dialogue tags such as “she whispered.” You may keep them in the narration or omit them from an adapted script. Make that editorial choice deliberately, rather than accidentally speaking the same direction twice.

3. Cast the narrator and the character separately

This sample uses Tim for narration and Dacey for Elena. The narration requests a storytelling delivery. Elena uses a natural delivery so the text remains easy to understand. The word “whispered” in the original passage is not proof that a generated voice will produce an appropriate whisper automatically.

If you enable Smart casting, review its initial suggestions. It uses local rules to infer clues about roles and delivery. It can miss context, misread age or leave a line neutral. Correct those choices in the director workspace before generating the complete scene.

4. Check continuity before creating the next scene

Listen for a clean return to the narrator after the quotation. Keep the same narrator voice, pronunciation choices and character names in later scenes. A short cast list outside the text can help you keep a longer project consistent.

Export the finished scene and save its project settings. Check the plan limits and credit estimate before generating longer text. The small example here demonstrates a workflow; it does not establish the cost or quality of a complete audiobook.

Theater: generate, play and follow the script

Expand a dialogue line to choose its voice, emotion and speed. Its action changes from Generate to Play and Stop. While audio plays, the source script highlights the corresponding sentence and scrolls to it; long list summaries wrap instead of truncating the text.

Use More actions to rename a role, duplicate, move, delete or regenerate a line. Changing dialogue or delivery requires a new generation; cached replay does not apply the new settings automatically. Review the displayed credit estimate before confirming.

Writing prompts for AI narration, dialogue and game voices

Treat these prompts as writing briefs for preparing your script, not commands to paste into spoken dialogue. Use the existing emotion, speed and pause controls for delivery. A role label such as Narrator: or Merchant: identifies the speaker and is not spoken as part of the line.

Narration brief: write a calm 20-second scene in two short sentences, with no stage directions. Dialogue brief: write six alternating lines for two characters with consistent role labels. Game brief: write one greeting, one warning and one farewell for a fantasy merchant; choose an available character preset, then preview each line.

For SRT video dubbing, ask for short, speakable lines that preserve the meaning and names within your existing timestamps. Count timing by listening: shortening the wording is often better than extreme speed. Review recognition errors and overlapping timestamps before generating the full video.

Before you export

• Narration is not assigned to a dialogue character.
• Quotation boundaries match the intended spoken text.
• Dialogue tags are handled consistently.
• Smart casting suggestions have been reviewed.

Questions about this workflow