What you need
One clear photo and enough credits for the resolution you want. The photo supplies the person in the prepared role. The template supplies the phone, skatepark, wardrobe, camera movement, original voice and surprise reaction.
There is no location, microphone or editing timeline to arrange. The clip already exists; the generation places a new identity inside it.
Step 1: choose a photo
Use a recent portrait where the face and head are easy to see. Keep one person in frame. Avoid sunglasses, masks, extreme angles and heavy filters that change the shape of the face.
The image should be between 300 and 6000 pixels on each side. Even daylight or soft indoor light is more useful than a dramatic backlight.
Step 2: confirm the template
Watch the prepared 10 second clip in the generator. It is a single continuous shot: the lead holds a phone, speaks into it, checks the screen, reacts with surprise and continues the skit.
The gray beanie, navy hoodie, olive shirt, blue jeans and skatepark background remain fixed. Your uploaded photo does not add its own pose, background or expression.
Step 3: choose the frame
Auto keeps the 9:16 source frame. That is the safest choice for TikTok, Reels and Shorts. The alternate ratios are available for other placements, but they change the output canvas rather than the source performance.
For portrait-led content, a second vertical render is more useful than forcing a wide crop onto the same file.
Step 4: choose resolution
480P uses 50 credits and is a reasonable low-cost test. 720P uses 100 credits and is the practical default for a finished social clip. 1080P uses 200 credits when detail matters.
The credit total appears before submission. Credits are charged when the paid task starts, and failed tasks refund unused credits automatically.
Step 5: generate
Submit the task after confirming the photo and settings. The custom preview player shows the template while the result is pending, then switches to the finished clip when it is ready.
Generation time depends on the provider. Keep the page or account history available so a completed task can be found later.
Step 6: review and download
Watch the full clip once before downloading. Check that the face remains readable, the phone stays in the correct position, the shot is continuous and the original voice aligns with the visible reaction.
Download the finished vertical file and label it as AI-generated where the platform requires disclosure. Only upload photos you own or have permission to use.
A quality checklist
Use one person, a clear face, soft light, the native vertical frame and a resolution that matches the final channel. Those choices solve most problems before the paid task begins.
The goal is not to build a more complicated scene. It is to keep the joke readable: one question, one blunt answer and one surprised reaction.
Ready to make your own version?
Upload one clear portrait and generate the millionaire in 30 days phone skit with the original voice and reaction intact.