To make a Bulin 47 AI video, you take the viral 16-second freestyle and put someone else in the rapper's place: the red lights, the crowd filming on their phones, the portable speaker and the dembow beat stay, and you, a friend, your grandma or your dog deliver the bars. The fastest way is one photo and a ready-made template. The hands-on way is a reference-to-video model, the original clip and a careful prompt.
Here's what most guides skip: the Bulin 47 AI trend only works if the lips still match the freestyle. The joke is that someone unexpected is saying "ya todos me conocen, soy Bulin 47". That decides which tool you use, which photo you pick and even which sound you post with.
We built a Bulin 47 AI video generator for this trend. To build it, we went through the original 16 seconds frame by frame, and that breakdown is in this guide too.
How to make a Bulin 47 AI video: 4 steps
- Pick one clear photo of one person or one pet, face visible, waist up, in good light.
- Drop it into a Bulin 47 AI template such as our Bulin 47 AI trend generator, choose 480p, 720p or 1080p and generate. The cost in credits is shown before anything is charged.
- Download the MP4. You get a vertical 16-second Bulin 47 AI video with the original freestyle audio on it and no watermark.
- Post it to TikTok, Reels or Shorts with #bulin47 and #damehombro.
The rest of this guide explains each step, the do-it-yourself route, which sound to post with, and how to fix a Bulin 47 AI video that comes out wrong.
Where the Bulin 47 freestyle comes from
Around October 2, 2026, creator Kevin Wai walked up to Dominican rapper Bulin 47 outside a restaurant in Madrid with a portable speaker playing a dembow beat. Bulin 47, real name Hancel Teodoro Vargas Reynoso, answered with an improvised freestyle, laughing and bouncing his shoulders while the crowd filmed. The line everyone quotes is "No ando en película ni ando en paquete, ya todos me conocen, soy Bulin 47", roughly "I'm not putting on a show and I'm not faking it, everyone knows me, I'm Bulin 47".
The clip passed 50 million plays across platforms within days, and MARCA reported that the original Instagram post alone had passed 65 million views. The two then turned the moment into a studio single, "Dame Hombro", released on October 7 with producer Breyco En Producidera.
The AI remixes came right behind it. Messi, Ronaldo, Mbappé and Haaland were put into the freestyle with AI, and an AI clip of Benjamin Netanyahu doing the moves got a reaction from Netanyahu himself. That's the Bulin 47 AI trend: same clip, new face. If you're here for the AI version, keep reading.
The 16-second freestyle, second by second
Every Bulin 47 AI video copies the same take, so it pays to know what happens in it. This is what we logged when we built our template:
| Time | What happens |
|---|---|
| 0:00–0:01 | The handheld camera pushes through the crowd at a red-lit entrance. The rapper walks in; a black speaker is raised in the foreground. |
| 0:01–0:03 | He stops and turns to the speaker, held up by a man in a beige hoodie with his back to the camera. Phones go up. |
| 0:03–0:08 | The freestyle. Medium shot, talking straight at the speaker, both hands gesturing, phone screens filming him in the foreground. |
| 0:08–0:10 | He leans back, one hand up, and starts laughing. |
| 0:10–0:13 | A big open-mouthed laugh, hand by his face, shoulders going: the "dame hombro" bounce. |
| 0:13–0:16 | He leans into the crowd, half behind the beige hoodie, still laughing and bouncing as the crowd cheers. |
Three things in this table decide how your Bulin 47 AI video turns out:
- The face is never a close-up. It sits mid-frame in a medium shot, so a soft or tiny face in your photo has very little to work with.
- It keeps getting covered. The speaker, the phones and the beige hoodie block the performer again and again. Each time he reappears, the AI has to bring your face back, and that's where weak results drift.
- Everything is red. The whole venue is lit red, so skin tones shift. A photo in plain, even light gives the model the true colours to start from.
Method 1: Make a Bulin 47 AI video from one photo
A one-photo template is how most people join the Bulin 47 AI trend. The clip, the crowd, the camera and the audio are already loaded; you only add the person.
Here's how it works in our Bulin 47 AI video maker:
- Upload one photo. Use the checklist below.
- Pick a resolution. The model, the 9:16 frame and the 16-second length are locked to match the trend. 480p costs 500 credits, 720p 1,500 and 1080p 3,100. Try a new photo at 480p and only re-run the keepers in higher quality.
- Generate. Your photo and the original freestyle go to Seedance 2.5 as a reference-to-video job. Only the performer changes: the crowd, the venue and the camera stay exactly as filmed, and the freestyle audio is laid back underneath afterwards, so the lips line up with the words. It takes a few minutes, and you can leave the page in the meantime.
- Download. The Bulin 47 AI rap video lands in My Creations as an MP4.
Your outfit comes from your photo, not from Bulin 47. Wear a blazer in the picture and you'll freestyle in a blazer. No glasses in your photo means no tinted glasses in the video, either.
Photo checklist for a Bulin 47 AI video
- One subject. Crop group photos down to one person or one pet first.
- Face the camera or turn slightly. A side profile gives the model half a face.
- Waist up. The clip never shows much below the waist, and your hands are in almost every shot.
- Sharp and well lit. Daylight or bright indoor light. The red venue will tint you anyway; don't start from a dark photo.
- Nothing over the face. No sunglasses, masks or hands in the way.
- No heavy filters. Beauty filters and stylised edits make your face harder to keep through the occlusions.
Pets, grandmas and the coworker in a blazer
The Bulin 47 AI videos that travel furthest star someone you'd never expect to freestyle. A dog works: it stands where the rapper stood, with the crowd and the phones pointed at it, keeping its own fur and markings. Use a photo where the pet is alone and its face is in focus; if someone is holding it, they get left out.
The same goes for a grandmother with her handbag on her arm, or the friend who signs every email "best regards". Just use photos of people who are in on the joke.
Method 2: Recast the clip yourself with a reference-to-video model
If you want to build a Bulin 47 AI video yourself, maybe for a different cut of the freestyle or a cartoon character, you'll need the source clip, a reference image and a model that can edit the clip rather than invent a new scene. This takes more setup and a few takes.
Pick the right kind of tool first:
- Motion-control tools such as Kling Motion Control animate your picture with the clip's moves. The background comes from your image, so you lose the red-lit crowd unless your image already has it. Kling also follows a single performer, and in this clip the biggest body in frame is often the beige hoodie in the foreground, not the rapper. Crop the reference so the performer is clearly the main subject.
- Reference-to-video models such as Seedance 2.5 take the clip and your image together, and can keep the crowd, the venue and the camera while swapping one person. That's the closer match to what you see in the trend.
Then:
- Trim the source. A short, clean stretch of the freestyle with the face visible. Use only footage you're allowed to use.
- Prepare one reference image of your person or character, front-on and waist up.
- Lock down what stays. The prompt has to name the one person who changes and protect everyone else. A starting point:
Replace only the bald, bearded man in tinted glasses and a fur-collared
leather jacket in @Video1 with the person in @Image1. Copy his motion,
gestures, laughter, shoulder bounce and lip movement frame for frame.
Keep everyone else, the speaker, the phones, the red lighting, the
venue and the handheld camera exactly as filmed. Face, hair and outfit
come from @Image1 only; no beard or glasses from the original.
Keep the identity stable every time he reappears from behind the crowd.
- Generate three to five takes. Check the frames right after he reappears from behind the speaker; that's where faces slip.
- Put the audio back if the model exports silence, and slide it until the mouth matches the words.
One-photo template vs. doing it yourself
| One-photo template | Recast it yourself | |
|---|---|---|
| What you need | One photo | The clip, a reference image, a prompt |
| Crowd and venue | Kept as filmed | Kept only with a model that edits the clip |
| Audio | Freestyle audio included | Often silent; re-sync by hand |
| Works with pets | Yes | Hit or miss |
| Takes per good result | Usually one | Three to five |
| Best for | Posting today | Custom cuts, cartoon characters |
Our take: if you want a Bulin 47 AI video of a person or a pet you know, start from one photo. Go the DIY route only when you need a cut of the freestyle the templates don't use, or a character instead of a real photo.
Freestyle audio or "Dame Hombro": which Bulin 47 rap to post with
The trend now has two audios: the street freestyle recorded on the night, crowd noise and all, and the studio single "Dame Hombro". They are not interchangeable.
The AI copies the mouth movement of the freestyle. Put the studio track under it and your star is mouthing one set of words while the speakers play another. For a Bulin 47 AI video, keep the freestyle audio as the main sound; the lip match is what sells it.
If you also want the post to show up on a sound page, here's what to do on TikTok:
- Upload the video with its own audio on.
- Tap Add sound (TikTok's sound guide covers the menus) and search "Bulin 47" or "Dame Hombro".
- Turn the added sound all the way down in the volume settings, and keep your original sound up.
Tag it #bulin47 and #damehombro, the two tags the trend actually lives under. TikTok also asks you to label realistic AI-generated content, so switch on the AI-generated label before you post.
Fixing your Bulin 47 AI video
Why does my face change halfway through?
It almost always happens right after the speaker or the beige hoodie passes in front. The model has to rebuild your face from your photo, and a small or soft face gives it little to go on. Use a sharper, front-on photo where your face fills more of the frame, and run it again.
Why do I have a beard or tinted glasses?
That's the original performer leaking through. It shows up most in fast, blurry frames. A clearer photo helps; in DIY tools, say outright in the prompt that the beard and glasses don't carry over.
Why did people in the crowd change?
Some models redraw the bystanders along with the performer. In a DIY prompt, protect the crowd explicitly. A template that only changes the one performer avoids it.
Why doesn't the mouth match the words?
Either the audio was swapped for "Dame Hombro", or a DIY tool exported silence and the freestyle was re-added out of sync. Use the freestyle audio and slide it until the first bar lines up.
Why is my skin red?
The whole venue is lit red, so some red cast is expected. If you look much redder than everyone around you, start from a photo with neutral light.
Why is the video blurry?
Check the resolution first. Treat 480p as a draft, then re-run the take you like at 720p or 1080p. If your face is only a few pixels wide in the photo, no setting can add the missing detail.
Bulin 47 AI video FAQ
What does "Dame Hombro" mean?
It's Spanish for "give me shoulder": a cue for the quick, rhythmic shoulder bounce at the heart of the freestyle. The song spells it out: "los hombros se mueven solos cuando estoy contento", my shoulders move on their own when I'm happy.
Is the Bulin 47 freestyle real or AI?
The original is real: Bulin 47 and Kevin Wai outside a restaurant in Madrid, filmed by the crowd. The versions with footballers, politicians, grandmas or dogs in his place are AI.
What app makes the Bulin 47 rap video?
There's no official app or TikTok filter for it. The viral Bulin 47 rap videos come from AI video tools: one-photo templates that already hold the clip (Method 1), or reference-to-video models where you bring the clip yourself (Method 2). A plain text-to-video generator can't reproduce the original take, because it invents a new scene from your words.
Can I make a Bulin 47 AI video of a celebrity?
The trend is full of them, but you're responsible for whose face you use. Platforms ask you to label realistic AI content, and putting a real person in a clip without their consent can break platform rules. A video of yourself, a friend who agreed or your pet is the safest bet.
How long is a Bulin 47 AI video?
Ours runs the full 16 seconds of the original take, in vertical 9:16.
Will there be a watermark?
Not from our generator: the downloaded MP4 is watermark-free.
Can I make a Bulin 47 AI video for free?
New accounts get free credits on sign-up and from the daily check-in, but a 16-second video needs more than that. The one-time Trial Pack covers one 480p Bulin 47 AI video with no subscription.
Make your Bulin 47 AI video
That's how to make a Bulin 47 AI video: pick a sharp, front-on photo, keep the freestyle audio so the lips still match, check the frames where the crowd covers the face, and post with #bulin47 and #damehombro.
If you'd rather skip the setup, give the Bulin 47 AI video generator one photo and it returns the whole freestyle, crowd and audio included. Browse more trends like it in our AI video effects, or see how to do the Drake Burning Bridges trend for another one-photo party scene.
Sources
- Improvisación de Bulin 47 alcanza más de 50 millones de reproducciones, Listín Diario
- Going Viral! Bulin 47 Improvised in Front of a Restaurant, MARCA
- Bulin 47 tras lanzar su nuevo tema 'Dame hombro', Listín Diario
- Freestyle de Bulin 47 se vuelve viral en redes con videos hechos con IA, NTelemicro
- Netanyahu reacciona a video suyo hecho con IA bailando "Dame hombro", Diario Libre
- Dame Hombro, single by Bulin 47, Kevin Wai and Breyco En Producidera, Spotify
- Dame Hombro lyrics, Musica.com
- Sounds, TikTok Help Center
- AI-generated content, TikTok Help Center
- Motion Control user guide, Kling
