Add two clear portraits
Choose a separate portrait for each consenting adult. Put the person who should open the exchange in slot one and the person who should answer in slot two. JPG, PNG, and WebP are accepted, up to 10 MB each.
AI rap video template
Rap Booth Duo
Rap Booth Duo is a short recording-booth exchange made from two authorized adult portraits and one theme. The planned 10-second vertical video includes original English rap lines and an instrumental beat. Watch the actual H3 sample below; paid generation is paused while delivery is tested.
Real H3 sample10 sec · 9:16 · 768P
This is an actual H3 output made with two reference portraits and an original instrumental. Turn on sound to hear the beat beneath the voices. New faces, timing, and movement can produce different results.
Create your scene
Add your two performers
Put one person in each photo slot and write a short theme. You can preview the photos now. Sign in and confirm the displayed credit price when generation opens.
Planned AI models: MiniMax-M3.1-Flash for lyrics, MiniMax music-3.0 for the instrumental beat, and MiniMax-H3 for video.
Sign in before choosing photos if you plan to generate a video. Local photo previews are not kept through login. Sign in
Person one01
Person two02
Photo previews stay in this browser. This preview does not upload your photos.
Cost per videoNot set yet
Video generation is being prepared. It will open after the AI service, private storage, and delivery checks are complete.
Rap Booth DuoTwo portraits. One theme. Original lyrics and a beat in one video.
Three steps
How to make a Rap Booth Duo video
You provide two photos and a theme. The workflow prepares the lines and instrumental before H3 renders the duet; no rap recording is needed from you.
Describe your rap theme
In 3–160 characters, name a single occasion and a specific detail: a birthday surprise, a team win, or a reunion joke. The system plans three brief original English lines from that direction.
Generate the beat and video
When generation opens, confirm the credit charge. A new instrumental is made first and passed to MiniMax H3 with both portraits. The requested result is one vertical video with the beat under the rap voices.
Make the inputs count
Choose the two performers and their story
Rap Booth Duo gives each portrait a role and uses the theme to shape a very short exchange. Decide who starts, choose images where each person is easy to recognize, and describe one moment they can share. Those choices matter more here than writing a long prompt full of camera directions.
Decide who starts the verse
Person one starts on the left, person two replies on the right, and both join the final line. If the birthday guest should answer, put their photo in slot two. Choose the order yourself; the theme cannot tell the model who should speak first.
The order is an instruction to H3, not a frame-by-frame guarantee. The generated staging can vary, and ten seconds allow only one compact exchange rather than long separate verses. Check the finished clip to see whether the roles and timing match your intention.
Choose photos you may use
Use a clear portrait with a visible face in each slot. This is easier to identify than a distant group shot, though no photo guarantees likeness. The form accepts JPG, PNG, or WebP up to 10 MB each and is designed for two different adults.
Both people must agree to appear, and you need permission to use their images. Selected photos stay in your browser during this preview; sign-in or reload loses them. When generation opens, submitted files go to private project storage and the AI services used for the task. Sign in before choosing files.
Write a theme the clip can use
A Rap Booth Duo theme gives the pair something to react to: ‘Two sisters celebrate finishing their first marathon’ or ‘Friends laugh about getting lost on a rainy road trip.’ Each fits one scene. Several unrelated memories are harder to express in three short English lines.
The field accepts 3–160 characters. A theme guides original lines but cannot lock exact wording or pronunciation. There is no custom lyric, song upload, voice-cloning, or artist-style input. If exact words or a specific voice are essential, this Rap Booth Duo form cannot provide them.
What you get
What a Rap Booth Duo video includes
The requested scene keeps two performers together in one booth: person one opens, person two replies, and both join the ending. The instrumental is supplied to H3 as an audio reference and is meant to remain audible in the same MP4 as the voices. You receive one finished video, not separate music and vocal tracks.
The sample above shows one completed render. Generated likeness, lip sync, gestures, line timing, and music balance may differ with new photos and themes. Watch and listen to your own result before sharing it; the sample is a way to judge the format, not a guarantee of identical output.
- Length
- 10 seconds
- Format
- 9:16 · 768P
- Soundtrack
- Original beat + AI rap voices
Watch before you choose
How to judge the real Rap Booth Duo sample
The player above contains one finished H3 output from two portraits and an original instrumental. Play it with sound on. Evaluate the things that matter to your idea: whether the pair reads clearly, whether you can hear both rap and beat, and whether ten seconds gives the exchange enough room.
Check the two-person scene
Look for two performers sharing one booth, an opening, a reply, and a joint ending. The sample shows the intended framing and movement. A story needing several locations or outfit changes will not fit this short format.
A sample cannot prove how new faces will appear. Facial detail, hands, and mouth movement may vary. Judge Rap Booth Duo as one shared social clip, not an exact recreation of a filmed performance.
Hear the voices and instrumental together
Listen for the beat beneath the rap, not just isolated voices. The instrumental is created first and passed to H3 with the images. In the sample, check that words remain understandable while the rhythm stays audible.
New renders may differ in delivery, timing, and mix. There is no BPM or separate volume control, and the MP4 does not include editable stems. Rap Booth Duo cannot use your own track or let you control every bar.
Decide whether the duration works
Ten seconds allow a setup, answer, and finish. A birthday greeting or friendly challenge may fit; a biography or full song will not. Choose one idea the pair can express together.
The sample shows one video's pacing. Your theme may change the lyric rhythm and gestures. This preview cannot guarantee an exact phrase, move, or identity detail in a new result.
Rap Booth Duo explained
Rap Booth Duo FAQ
Can I use one group photo for Rap Booth Duo?+
The form expects one separate portrait in each person slot. A group photo does not clearly tell the model which face should take each role. Use two images of different consenting adults, and check that you may submit both.
Does Rap Booth Duo copy the people's real voices?+
No voice recording is requested. The photos guide the visual performers, while the video model generates the rap voices. This template does not offer voice cloning or a way to match either person's natural voice.
Are photos uploaded when I preview the Rap Booth Duo form?+
No. Selecting files now creates browser-only previews. When generation opens, submitting a task will send the images to private project storage and the AI model services needed to make the video.
Can I add my own song or exact lyrics?+
Not in this version. Rap Booth Duo accepts portraits and a theme, then plans original lines and an instrumental. It has no song upload or exact-lyric field, and it cannot be used to request a copy of a named performer.
Can I generate and download Rap Booth Duo now?+
You can watch the real sample and preview your selected photos locally. Generation and checkout are paused for everyone until service and delivery checks are complete.
What does Rap Booth Duo cost, and what if generation fails?+
The planned price is 299 credits for one 10-second, 768P video. You must confirm the displayed charge before submission. A confirmed failure restores that task's credits. Processing time is not guaranteed.
Your duet starts with a theme