Talk Dirty Dance AI Prompt (Copy-Ready) — and How to Skip the Prompt Entirely
The Talk Dirty dance AI prompt is Seadanse's copy-ready 259-word edit prompt for the real 15-second concert clip, or skip it with 1 photo for $9.90.
The Seadanse Team18 min readThe team behind Seadanse. We run the models we write about, and we publish the specs, prices and limits the marketing leaves out.
TL;DR — The Talk Dirty dance AI prompt is a 259-word video edit prompt for the real 15-second concert clip. It puts the person in your photo in the lead's place.
Everything else stays: the dancers in varsity jackets, the flame columns, the spotlight, the microphone and the giant letters on the stage screen.
On Talk Dirty Dance AI for TikTok and Reels, the prompt's built in, so you add 1 photo. One 15-second 720p video is $9.90.
Since October 17, 2026, TikTok has filled with AI versions of one Talk Dirty dance clip. On TikTok, tutorials for the trend hand you a CapCut template instead of a prompt. That's a ready-made edit in the CapCut app with a slot for your photo.
The trend's most-shared tutorial on TikTok with 70,600 plays and 650 shares walks you through copying profile links and opening templates. If you've tried that route, you know the Talk Dirty dance CapCut template sends you through several extra steps.
People get stuck or get asked to pay along the way. Under that tutorial, one viewer wrote "What video on your profile. Can’t find video" as a comment. Under the same account's most-played post, another left "not free people" as a comment.
In our own count across TikTok, YouTube, Bilibili, X and Reddit, nobody has posted a working prompt for this trend. And none of the 66 TikTok comments with words that we read asks for one.
Here's the exact prompt our page runs, all 259 words of it. We'll break down what each part does. Then we'll show you the habits that keep it on track, and how to skip the prompt entirely.
What Is the Talk Dirty Dance AI Prompt?
The Talk Dirty dance AI prompt is the written instruction a video tool reads when it edits a clip. It tells the tool who to swap out on the concert stage and what to leave alone.
Here it is, word for word, as our page runs it.
Edit the video. Replace the bearded man with long curly hair in the navy-blue suit, light-blue shirt and pink-red tie, holding a small white microphone, who dances at the centre of the stage in front of the row of dancers, with whoever is in @Image 1 (a person, or a cartoon or game character, or a pet when @Image 1 shows one): their face, hair, body and whole outfit come from @Image 1 only, and they stay exactly the same in every second from the first frame to the last, through every zoom, blackout, spotlight, colour change of the stage light and flame burst. They do not wear the navy-blue suit, the light-blue shirt, the pink-red tie or the small lapel pin, and do not have the man's full beard or his long curly hair, dark or blond, unless @Image 1 shows them. Keep everything else exactly unchanged: every dance step and movement, the timing, the small white microphone in one hand, the phone camera filming from the crowd just below the front edge of the stage with its zooms and its framing, the dancers in yellow-and-black and red-and-black varsity jackets with wide black trousers, the dancer with long dreadlocks, the dancer in the blue cap and everyone else on stage with their own faces, the flame columns shooting up at the front of the stage, the white spotlight, the round floor lights and the blue, white and orange stage light, the drum riser and the giant white letters on the screen behind the stage. No captions or subtitles.In that text, @Image 1 is your photo, the first picture you attach. The video it edits is the 15-second concert clip, vertical 9:16.
It isn't a Dirty Dancing prompt, since those recreate the lift from the 1987 film, a completely different trend.
This is what the prompt gave back in our own run, with the clip's live sound laid under it.
How the Talk Dirty Dance AI Prompt Is Built
The Talk Dirty dance AI prompt is built in three parts. There's who to replace, what comes from your photo (and what never carries over), and what stays.
| Part | What the prompt says (word for word) | What it does for you |
|---|---|---|
| Who to replace | "the bearded man with long curly hair in the navy-blue suit … who dances at the centre of the stage in front of the row of dancers" | Points the edit at the lead, the one man in a suit, and not at the dancers |
| What comes from your photo | "their face, hair, body and whole outfit come from @Image 1 only" | You lead the stage in your own clothes |
| What never carries over | "They do not wear the navy-blue suit, the light-blue shirt, the pink-red tie or the small lapel pin" | None of his suit or tie ends up on you |
| The moves and the mic | "every dance step and movement, the timing, the small white microphone in one hand" | Every step lands on the same beat, with the mic in your hand |
| The crew | "the dancers in yellow-and-black and red-and-black varsity jackets with wide black trousers" | The dancers around you stay the clip's own |
| The stage | "the flame columns shooting up at the front of the stage, the white spotlight, the round floor lights" | The flames and the lights hit on the same seconds |
| The camera | "the phone camera filming from the crowd just below the front edge of the stage with its zooms and its framing" | It still looks like a fan filmed it from the crowd |
| The last line | "No captions or subtitles." | No subtitles get added, and the letters on the stage screen stay |
The "what stays" part is the longest, taking up 112 of the 259 words. That detail matters because every dancer, light and camera move it names stays as it is in the clip.
You'll also notice the prompt doesn't name a person, a song, a tour or a creator.
How Do You Use the Talk Dirty Dance AI Prompt?
To use the Talk Dirty dance AI prompt, you'll need a video tool that edits a clip you upload. You'll also need the 15-second concert clip and one photo.
- The tool: You'll need a video tool that edits a video you upload. It takes one video and one picture together, and it changes the video while keeping what you don't ask it to change.
- The clip: You'll need the 15-second concert clip, vertical 9:16 (the tall, vertical frame TikTok and Reels use), as the video to edit.
- @Image 1: You attach your photo as the first picture. Use one person, character or pet, with the face clear, facing the camera, in good light. Full body works best.
- The prompt: You'll paste all 259 words as they are into the tool's instruction box.
- The sound: The edit comes back without sound, so you'll lay the clip's own live sound under it from the first second.
We ran this exact prompt on October 21, 2026. For @Image 1, we used a picture of an AI-made man (a made-up adult, not a real person).
He's clean-shaven, with short bleached hair and dark faded sides. He wears an oversized white T-shirt, a silver chain, black cargo trousers and white sneakers. The picture was a character sheet showing a close-up of his face next to four full-body views.
The run passed on its first take, with no change to the prompt. Here's what came back:
- One lead: He leads the stage in every frame, and there's only one of him throughout the entire run.
- The old look: No frame shows the suit, the tie, the beard or the curly hair.
- The mic: The small white microphone stays in his hand.
- The flames: The flames burst right on time.
- The screen: The letters on the screen stayed.
- The sound: The live sound went under it from the first second.
We checked it second by second, frame by frame around each light change, and with close-ups at 13 moments.
If you'd rather not set all that up, our Talk Dirty dance AI generator runs the same prompt from one photo.
Habits That Keep the Talk Dirty Dance AI Prompt on Track
Seven habits keep the Talk Dirty dance AI prompt on track, and you can see each one in the prompt above.
Name the Lead by What You See
The prompt describes the lead only by what's visible on the screen: "the bearded man with long curly hair in the navy-blue suit, light-blue shirt and pink-red tie, holding a small white microphone, who dances at the centre of the stage in front of the row of dancers". He's the only man on stage in a suit, while every backup dancer wears a varsity jacket.
Copies of the clip online don't all show the same lead. So describe the lead in your own copy by clothes and hair. That keeps the edit targeted.
Naming his place at the centre in front of the row of dancers tells the edit which person to change.
Say What Never Carries Over
The prompt lists what must not carry over: "They do not wear the navy-blue suit, the light-blue shirt, the pink-red tie or the small lapel pin, and do not have the man's full beard or his long curly hair, dark or blond, unless @Image 1 shows them." The words "dark or blond" cover both looks of his hair. It looks dark before the spotlight switches on at 2.4 seconds, and blond after it.
The words "unless @Image 1 shows them" mean that if your photo shows a beard or a suit, you keep yours.
This sentence does the exact job a negative prompt (a list of things the video must not show) does in other tools.
Hold the Look Through the Spotlight and the Blackouts
The prompt demands that your look stays "exactly the same in every second from the first frame to the last, through every zoom, blackout, spotlight, colour change of the stage light and flame burst." That instruction guards against the clip's intense lighting shifts.
Here's when the light changes in the clip:
- 1.0-1.2 seconds: The stage goes almost black.
- 1.2-1.8 seconds: It flashes white.
- 2.4 seconds: The white spotlight switches on.
- 3.5-6.2 seconds: It turns orange.
- 6.5-7.2 seconds: It shifts to dark blue.
- 11.2 seconds: It flashes white again.
In our run, the man from the photo stayed the same person right through the 2.4-second spotlight crouch and the 11.2-second flash. He stayed the same in both dark stretches too.
Keep the Microphone in the Lead's Hand
The microphone appears twice in the prompt to protect a tiny, key detail. It helps pick out the lead in the opening lines ("holding a small white microphone"), and it appears again in the keep list as "the small white microphone in one hand".
Because it isn't listed among the items that never carry over, the person from your photo holds it naturally. In our run, the small white microphone stayed firmly in his grip through every turn.
List the Crew and the Stage, Not the Steps
The prompt doesn't describe a single dance step. Instead, "every dance step and movement, the timing" sits in the keep list, so every move comes directly from the clip.
The prompt names the dancers in varsity jackets, "the dancer with long dreadlocks", "the dancer in the blue cap" and tells the tool to keep everyone "with their own faces". That protects the crew while flame columns burst along the stage. They go off at 3.5, 4.5, 5.5, 6.0, 8.2, 10.2, 13.0, 14.0 and 14.8 seconds.
The dancer with long dreadlocks stands right beside the lead at 2.5-5 seconds and 8-10 seconds. In our run, the dancer with long dreadlocks and the dancer in the blue cap kept their own looks.
Ask for No Captions, Not No Text
The final line of the prompt is "No captions or subtitles." It doesn't ban all text from the frame.
The giant white letters on the screen behind the stage are part of the scene, and that's why the prompt asks only for no captions or subtitles to keep them. The prompt calls them "the giant white letters on the screen" so the tool leaves the arena set piece alone while skipping unwanted subtitles.
In our run, the stage screen kept its own giant letters, and no captions were added.
Add the Sound After the Edit
The edit prompt only changes the picture. It doesn't contain any words about audio, and the video tool returns an edit without sound.
To get the full concert feel, you'll lay the clip's live sound under the video after the edit finishes. In the song "Talk Dirty", horns back the hook, and the clip's live sound keeps that music in time with the dance break.
On our page, the clip's own live sound goes under the finished video from the first second.
A 259-Word Talk Dirty Dance AI Prompt vs a One-Line Prompt
The one other prompt posted for the Talk Dirty dance AI trend is a single line in AI Photo Animator's how-to post. It typed the prompt in Chinese and gives this English line: "Replace the woman dancing in green in the video with the character and outfit from image 1. Keep everything else the same."
Its post says the prompt is short on purpose. Its clip was different from ours, as the post explains: "Ours was the 15-second stage solo, with the lead dancer in green."
Its own tip points out that when a prompt only asks to replace the person, "the model may pick a backup dancer," in the post's words.
| The one-line prompt | The 259-word prompt | What it means for you | |
|---|---|---|---|
| Who to replace | "the woman dancing in green" | The lead by his beard, curly hair, navy-blue suit, shirt, tie, microphone and his place at the centre | Each prompt names the lead in its own copy of the clip |
| What comes from your photo | "the character and outfit from image 1" | Face, hair, body and whole outfit, the same in every second | Your look is spelled out through every blackout and spotlight |
| What never carries over | Not named | The suit, shirt, tie, lapel pin, beard and curls | The lead's look is ruled out by name |
| What stays | "Keep everything else the same." | 14 things, from every step and the microphone to the drum riser and the screen letters | Each thing you want to keep has its own words |
| Text on screen | Not named | "No captions or subtitles." | No subtitles get added, and the stage letters stay |
Each prompt names the lead in its own copy of the clip, so match the who-to-replace part to the lead you see.
Can You Make the Talk Dirty Dance Video Without a Prompt?
Yes, you can make the Talk Dirty dance video without a prompt on Seadanse. The prompt, the 15-second clip and its live sound are built in.
- Add one photo: You'll add one photo and keep the consent box ticked.
- Pick 480p, 720p or 1080p: You'll pick the quality, where every export is 15 seconds and vertical 9:16.
- Press Create: You press Create, and the price shows right on the button before you pay.
You'll pay once at checkout with no account and no subscription, and your video starts creating right on the page.
You can check Seadanse pricing for credit packs. The $9.90 pack of 500 credits covers a 720p video or a 480p video, and pack credits stay valid for 12 months.
The email from checkout brings a link that signs you in so you can open your video on any device.
When it finishes, you'll download a 15-second vertical MP4 with the clip's live sound and no watermark on exported video. If a video fails, you'll get its credits back automatically.
You can start on Seadanse's Talk Dirty Dance AI right in your browser.
We've also got a complete walkthrough on how to make the Talk Dirty dance AI video in three simple steps.
Talk Dirty Dance AI Prompt FAQ
Is this the prompt Seadanse actually uses?
Yes, this 259-word prompt is copied word for word from our page's code on October 21, 2026. It passed on its first take when we produced our sample video.
The prompts behind Charme Dance AI and Dime Dime Baby AI use the same three parts to keep their staging intact.
Which AI video tools accept the Talk Dirty dance AI prompt?
You'll need a tool that edits an uploaded video by taking one clip and one picture together. An image-to-video tool won't work because it creates a video from scratch and has no concert footage to keep.
You can compare different tools in our guide to the best Talk Dirty dance AI generators.
Do I need the original concert clip to use the prompt?
Yes, you'll need the original 15-second concert clip filmed from the crowd below the stage. The staging matches fan footage of the staging of Jason Derulo's The Last Dance World Tour from 2026. On Seadanse, the 15 seconds are already built in.
If you want to know more about the footage, read where the Talk Dirty dance AI clip comes from.
Does the Talk Dirty dance AI prompt add the song?
No, the edit prompt only changes the picture and returns a video without sound.
The track in the clip is "Talk Dirty" (feat. 2 Chainz), from Jason Derulo's 2013 song. On Seadanse, the clip's own live sound is laid under your video from the first second.
Can I write a Talk Dirty dance prompt for text-to-video instead?
You can write one, but a text prompt with no clip makes a brand-new stage with random dancers and different choreography. An edit prompt keeps the real concert arena, the real dancers and the real dance break.
On our page, that entire setup's ready to go with one photo.
Ready to Skip the Talk Dirty Dance AI Prompt?
You can run the 259-word prompt in an edit tool yourself. Or you can head to Seadanse and make a Talk Dirty dance AI video with a single picture.
Our tool keeps the dancers, the flame bursts, the live audio and the stage letters. It swaps in your face and clothes.
You can also browse our other AI video effects to turn pictures into viral clips in your browser.
By The Seadanse Team. We copied the prompt above from our Talk Dirty Dance AI page's code on October 21, 2026, the day we ran it.
Related Articles
- How to Make the Talk Dirty Dance AI Video From One Photo (3 Steps) — the presses on the page, from your photo to the download.
- What Is the Talk Dirty Dance AI Trend? (Real Stage, AI Lead) — the concert clip, how the AI versions began, and why so many people think they're real.
- Talk Dirty Dance CapCut Template: Which One It Is and Why It Fails — the two templates the tutorials link, and what trips people up.
- Best Talk Dirty Dance AI Generators (2026): 8 Tools Compared — the tools that put you on the concert stage, side by side.
