Dialogue Polisher
Paste your dialogue and get it back sharper: distinct voices per character, subtext instead of on-the-nose lines, rhythms that sound like speech. The difference between dialogue readers skim and dialogue they hear.
Rewrites flat dialogue with subtext, distinct voices, and rhythm — no more 'as you know, Bob'.
You are a dialogue editor for prestige television. I will paste a dialogue scene. Rewrite it following these rules: 1) Every character gets a distinct voice — no two characters should sound interchangeable, 2) Add subtext: characters must want something they don't say outright, 3) Cut 30% of the words — especially greetings, names used unnaturally, and on-the-nose exposition, 4) Vary rhythm: fragments for tension, longer lines for control, 5) Keep my plot beats intact. Return the rewrite plus a bullet list of what you changed and why. Here is the scene:
See what this prompt creates
What this prompt does
Bad dialogue tells; good dialogue reveals while pretending not to. This prompt polishes your exchanges on three levels: voice distinction (no two characters should sound interchangeable), subtext (what's meant vs. what's said), and rhythm (the beats, interruptions, and silences of real speech). It keeps your story intact and upgrades how it sounds.
How to get the best result
Give it character voice notes first: “Mara is terse and dry; Dev rambles when nervous.” The prompt then polishes toward those voices instead of toward generic “good writing.” Include the scene's emotional context — the same words land differently in a fight vs. a reconciliation, and the polish should know which it is.
Common mistakes to avoid
Over-polishing into uniformity: if every character suddenly sounds like a witty screenwriter, you've lost voice distinction. The goal is sharper versions of their voices, not one voice. Also, don't polish away dialect and roughness that characterize — clean isn't always better.
Try these variations
Run “subtext mode” on key confrontations: the prompt rewrites lines so characters talk around the real subject, like people actually do. For building the characters behind the voices, the character interview prompt deepens them first. More in the writing prompts collection.
How to use this prompt
- Copy it — hit the copy button above to grab the raw prompt text.
- Fill the brackets — replace anything in [BRACKETS] with your own details.
- Paste into your AI tool — works with ChatGPT, Claude, Gemini for text prompts; Midjourney, DALL-E or Stable Diffusion for image prompts; Runway, Pika or Sora for video.
- Iterate — tweak one bracket at a time until the output is exactly right.
Frequently asked questions
How do I keep characters sounding different from each other?
Give each a voice profile: vocabulary level, sentence length, verbal tics, what they avoid saying. The prompt polishes each character's lines toward their profile. Test it: cover the dialogue tags — if you can't tell who's speaking, the voices need more separation.
What is subtext in dialogue?
The meaning underneath the words: a character saying “fine” while meaning “leave me alone.” Real people rarely say exactly what they mean, especially under emotion. The prompt's subtext mode rewrites on-the-nose lines into layered ones.
Should dialogue use perfect grammar?
No — real speech is fragmented, interrupted, and ungrammatical. The prompt preserves natural roughness while cutting the accidental kind (repetition, filler with no purpose). Read polished dialogue aloud: if you stumble, it still needs work.
Can it help with accents and dialects?
Lightly, yes — suggestive spelling and rhythm rather than phonetic overload. Heavy dialect on the page exhausts readers; the prompt aims for flavor (word choice, cadence) over transcription.
