Quick answer
Starrd's They Say I'm 2 Much template turns one photo into the last 30 seconds of ATM Danny's '2 Much' music video, with you as the rapper. You rap to a high fisheye camera outside a lit-up donut shop at night, holding an open box of donuts in some shots, across 15 shots cut on the beat of the 105 BPM track. Your mouth raps along to the original vocal in the rap shots and stays closed in the donut-box shots, just like the original. You keep your own face, hair and outfit, and pets work too, standing upright with the box in their paws. The motion comes from the original video as a depth map, so ATM Danny never appears. 30 seconds, vertical 9:16, Seedance 2.5, 480p, 250 credits (about $12.50). Contains explicit language.
One photo in, and out comes you as the rapper in ATM Danny's "2 Much" video: a lit-up donut shop at night, a box of donuts, a fisheye lens, every jump cut on the beat and every line lip-synced. Here it is:
They Say I'm 2 Much
One photo. You rap ATM Danny's '2 Much' outside a donut shop at night, lip-synced, every cut on the beat. 30 seconds.
What You Get
- The last 30 seconds of the video. The second half of the verse, through to "Go stupid".
- 15 shots, cut on the beat. Wides, mediums and close-ups from the same high fisheye angle, each cut landing where the original's does.
- Lip sync that switches on and off. You rap in the rap shots and keep your mouth closed in the donut-box shots, like the original.
- Your own outfit. Face, hair and the clothes from your photo, down to the shoes.
- The donut shop at night. The glowing sign on its pole, the bright window and an open box of donuts in your hands.
- 30 seconds, vertical 9:16, with the original track, ready for TikTok, Reels and Shorts.
What the Video Actually Is
ATM Danny posted the video for his song "2 Much" on October 21, 2026, captioned "THEY SAY I'M 2 MUCH". It's one rapper, one location and a lot of attitude: he raps straight into a fisheye lens mounted high above him outside a small donut shop at night, sometimes walking up to the camera, sometimes leaning right into it, sometimes holding a box of donuts like a trophy. The edit jumps between framings on almost every bar.
Why it suits AI: the whole look is one location, one prop and one camera trick, and the performance carries it. One photo puts you in front of that lens without memorising a verse or finding a donut shop that's open at midnight.
The 30 Seconds, Shot by Shot
The song runs at 105 BPM, so one bar lasts about 2.3 seconds, and each line starts one beat before the bar.
| Time | Shot | Mouth |
|---|---|---|
| 0:00 | Medium, holding the open box of donuts | Rapping |
| 2.0s | Wide, walking up to the camera | Rapping |
| 4.3s | Close-up, arms crossed, grinning | Rapping |
| 6.5s | Medium, both hands working | Rapping |
| 7.4s | Wide, waving with the box | Closed |
| 8.8s | Close-up, reaching for the lens | Rapping |
| 11.1s | Wide, swinging the box lid open | Closed |
| 13.4s | Wide, walking up and pointing | Rapping |
| 15.7s | Close-up, a quick head dip | Rapping |
| 16.5s | Medium, reaching for the lens | Rapping |
| 18.0s | Close-up, a chocolate donut held up to the lens | Rapping |
| 20.3s | Wide, hands on the beat | Rapping |
| 22.5s | Medium, both hands toward the lens | Rapping |
| 25.7s | Medium, picking at the box, then shading the eyes | Closed |
| 29.4s | Medium, hand by the head | Closed |
Your Look, His Video
- Your own face, hair and outfit. Nobody gets the original's cap, oversized tee or long shorts, and your build stays your own.
- Your own shoes. The wide shots show you head to toe, so the full outfit from your photo comes along. If your photo stops at the waist, the rest is filled in to match.
- His location. The donut shop, the glowing sign and the box of donuts, in every shot.
A full-length photo where you're standing and facing the camera works best, because the wide shots show your whole outfit. A clear, well-lit face matters most for the close-ups, where the lip sync is easiest to see.
The Fastest Way
They Say I'm 2 Much
One photo, 30 seconds, the original track, lip-synced and cut on the beat.
One photo, 250 credits (about $12.50), no prompt. It renders on Seedance 2.5 at 480p, vertical 9:16, 30 seconds.
Build It Yourself
1. Find the cuts and the tempo. Jump cuts in a one-location video fool scene-detection tools, because the background barely changes. Measure the frame-to-frame difference instead and look for spikes. Then fit a line through the song's beat times to get the exact tempo (here 105 BPM, a bar every 54.86 frames at 24 fps) and check the cuts against the bars.
2. Find when the mouth moves. Track the rapper's mouth frame by frame. In "2 Much" the donut-box shots are B-roll: the vocal keeps going but his mouth stays shut. Those are the windows your rapper must keep their mouth closed.
3. Turn the video into a depth map. Run it through a depth-estimation model (Depth Anything V2 works) and set the contrast per shot. Smooth the baseball cap down to a plain dome so your rapper doesn't inherit it, but leave the beard and jaw alone: a "hair" mask grabs the beard and flattens the mouth.
4. Make one storyboard sheet of six setups. The box shot, the wide walk-up, the arms-crossed close-up, the reach to the lens, the donut to the lens and the eye-shade ending, all generated from your photo at one donut shop, with a depth-map sheet of the original framings as a layout guide.
A close-up of the person from reference image 1 outside a small roadside donut shop at night, seen from a high camera through a wide fisheye lens, a big round glowing donut sign on a pole behind them: they hold a chocolate donut right up to the lens with an open pale-blue box of donuts in the other arm, rapping, mouth open mid-word, wearing exactly the clothes from their photo. Warm yellow shop light against teal shadows, a touch of grain. No text.
5. Drive it with the depth map and the song. Put the original vocal inside the depth video so the model hears what to lip-sync. Feed the sheet, your photo and the depth video to a model that copies motion from a reference video (we recommend Seedance 2.5), with a timed shot list and the mouth windows: when to rap, and when the mouth stays closed. Render it silent and lay the full track back on at the end.
Common Mistakes
Trusting scene detection. It finds one cut in this video. There are twenty.
Letting the mouth rap through the donut-box shots. Without closed-mouth windows, the model lip-syncs every second of the vocal. Say when the mouth is shut.
Driving with the raw video. The cap, the oversized tee and the original face come straight through. Use the depth map.
Rounding the tempo. A beat tracker's 103 or 107 BPM drifts off the cuts within a few bars. Fit the beat times and you get 105.
FAQ
How many photos? One.
How long is it? 30 seconds, vertical 9:16.
What does it cost? 250 credits, about $12.50.
Is it explicit? Yes, the original track is uncensored.
Related Reading
- How to Make the '58th Floor' AI Video: Anderson .Paak and Cordae's COLORS session, lip-synced with a friend
- How to Make the Clayface AI Video: rap Drake's "Solar Eclipse" to the camera in a nightclub VIP booth
- How to Make the Hotel Lobby AI Video (A COLORS SHOW): Quavo and Takeoff's "Hotel Lobby", starring you and a friend
- How to Make the What You Want Me To Do AI Video: the viral kitchen duet, one take, move for move
- Viral AI Video Trends 2026: every trend we've turned into a template this year
