You lunge into the lens and spit a grime verse in a dark, packed crowd, finger jabbing at the camera. One photo of you is all it takes.
Updated : Page published.


Front-facing, well-lit selfies work best. One face per photo.
By generating you confirm you have permission to use the people in your photos. Terms
The check one check two filter is a CapCut AI trend from October 2026 in which a person raps a short grime verse credited to JME straight into a handheld phone camera in a dark crowd; videosi makes it from one photo of you.
Check One Check Two AI is made with Seedance 2.0 / 2.5 (you pick the tier) by ByteDance. The prompt behind it has 5 parts, and the timing is what makes or breaks it.
Replace the young man with short brown hair, light stubble and a black leather jacket in Video 1 with the person from Image 1.…

Face to the camera, head and shoulders in frame, even light. The top you wear in the photo is the one you rap in.
9:16 for TikTok, Reels and Shorts, or 1:1 and 16:9. There is nothing else to set: the venue, crowd and camera move are built in.
Press Generate and the 8-second clip is rendered for you, an MP4 with no watermark and the check one check two sound already on it.
Post it as it is or swap the sound in your app, and label the clip as AI-generated.
Updated · videosi is an independent tool. It is not affiliated with TikTok, Instagram, YouTube, or with the owners of any trend, song, game or brand named on its pages.
The check one check two filter is a CapCut AI template that went around TikTok in late October 2026. You give it a photo and it returns a short, shaky front-camera clip: the person stands in a dark, crowded venue with a friend behind them, lunges into an extreme close-up and delivers a grime verse straight into the lens, jabbing a finger at the camera, raising their eyebrows and pulling faces on the beat.
The earliest posts we found are dated 19 October 2026, when one account posted AI versions starring British TV presenters and a streamer; its two most-liked posts reached about 100,000 and 76,000 likes. Within a week dozens of accounts were posting tutorials and footballer, politician and sitcom versions, and the CapCut template behind it had about 14,600 uses by 27 October.
The sound is an original sound that posts credit to the London grime MC JME, opening with the mic-check line "check one, check two". Tutorials call it the JME dyslexic freestyle; we have not confirmed the exact track.
This page makes your own version from one photo, with no celebrity in it. You are the one rapping into the lens; the sound is added to the 8-second clip, and your mouth moves with the rhythm rather than following every word.
Medium handheld shot in a dark venue: you stare into the lens, mouth open, with a friend standing just behind your shoulder.
You step into the phone and the frame snaps into an extreme selfie close-up, the crowd turning into blurred faces and cyan lights.
Tight on your face, head bobbing, eyebrows up, chin out: the verse starts.
A hand comes up and jabs toward the lens, blurring across the frame on the beat.
You keep rapping, dip your head and look down, then raise a hand and pull a grimace.
Eyes half closed, head swaying, still a few centimetres from the camera as the clip ends.
The clip is an extreme selfie close-up, so a photo where your face fills a good part of the frame, looking straight at the camera, works best.
A neutral face or a big grin both work. Avoid a hand or a drink near your mouth, which the model may keep in shot.
Puffer jackets, leather, hoodies and football shirts all suit the night-crowd look and read well in low light.
Clear glasses stay on. Sunglasses, caps pulled low and masks hide exactly what the close-up is about.
The check one check two filter is a CapCut AI template, first posted on TikTok on 19 October 2026, that turns one photo into a 10-second clip of that person shouting a UK grime verse straight into a phone camera in a dark crowd. The sound is credited to the rapper JME. videosi makes an 8-second version of it from one photo of you, with the sound on it.
Upload one front-facing photo of yourself to videosi's check one check two filter, pick 9:16, 1:1 or 16:9 and press Generate. You get an 8-second MP4 of you rapping into the lens in a dark crowd, with the check one check two sound already on it and no watermark. There is no prompt to write and no app to install.
The check one check two clips on TikTok come from shared CapCut templates (one had about 14,600 uses by 27 October 2026), not a live TikTok camera filter, and those templates change or disappear often. videosi is independent and not affiliated with TikTok, CapCut or JME; its check one check two filter works in the browser from one photo.
Yes. videosi (videosi.io) is a web check one check two AI generator: one photo in, an 8-second rap-to-camera clip out, with the trend sound on it. The camera move, venue and prompt are built in. A standard 8-second video (Standard model, 720p) costs 7 credits and has no watermark.
The check one check two filter on videosi is not free: a standard 8-second video (Standard model, 720p) is 7 credits, and better models or higher resolutions cost more. Credit packs start at $6.99 for 10 credits and never expire, monthly plans start at $29.90, and a failed or blocked render returns its credits.
videosi makes the check one check two video with ByteDance's Seedance (Seedance 2.0 Mini by default; Fast, Pro and Seedance 2.5 Ultra are one click away), guided by an 8-second motion reference clip and a five-part prompt. The full prompt is built in rather than published. Typed prompts alone usually lose the selfie lunge and the finger jab at 3 seconds.
Upload a photo and press Generate. 7 credits, refunded if it fails.
One email when new trends land on videosi, about once a week. Unsubscribe in one click.
Add videosi as a preferred source and our trend pages show up first in your Top Stories and AI answers on Google.
No subscription needed · Ready in about 3 minutes · Failed renders are free
No. The videosi check one check two video has the sound on it and your mouth moves with the rhythm, but it does not follow the words exactly; it is a hyped performance, not a word-for-word lip-sync. That holds for all 8 seconds. If you want exact mouth shapes, the CapCut template is the closer match.
The check one check two trend uses an 'original sound' on TikTok that posts credit to the London grime MC JME; tutorials call it the JME dyslexic freestyle. As of 27 October 2026 we have not confirmed the exact track or release. The videosi clip comes with that sound already on it, 8 seconds long.
"Check one, check two" is the phrase MCs and sound engineers say to test a microphone before a set. In the October 2026 TikTok trend it opens a short grime verse that people put in the mouths of friends, TV presenters and footballers with an AI template. videosi lets you star in it from one photo of you.
As of 27 October 2026 the check one check two trend is just starting to spread: the first template posts we found are dated 19 October 2026, the two most-liked have about 100,000 and 76,000 likes, and new tutorials appeared every day of the following week. videosi turns it into an 8-second clip of you.
Usually it is the photo: sunglasses, a cap, a side angle or a dark, grainy shot. For the videosi check one check two video, use one front-facing photo with your face filling much of the frame and even light. The 8-second clip is an extreme close-up, so likeness is not guaranteed, and it drifts most during the finger jab at about 3 seconds.
Not a celebrity: videosi does not accept photos of celebrities, public figures or minors, so the TV presenter and footballer versions on TikTok cannot be made here. A friend is fine if they are an adult and agreed to it. Photos and results are deleted within 30 days, and the 8-second clip should be labelled as AI-generated when posted.
You can try, but the check one check two look depends on a motion reference clip that most chat apps do not take, and these tools often refuse rap performances by real people or add text. videosi runs Seedance with the reference and a five-part prompt built in, for 7 credits per 8-second standard video, with the sound added.