1. Choose clear photos
Use one person per image, with their face visible and well lit. A front-facing portrait or a waist-up photo gives the face more detail than a distant full-body shot.
- JPG, PNG or WebP, up to 10 MB per image.
- At least 240 pixels on each side. Larger, sharp photos work better.
- Avoid group shots, heavy blur, sunglasses and covered faces.
- Use photos of yourself or other adults who agree to appear in an AI video.
2. Assign the five roles
Five people gives each performer a separate photo: Lead, Backup 1, Backup 2, Backup 3 and Sax. Add the person you want at the microphone to Lead; place the backing performers in the next three slots.
Solo uses one photo for all five roles. You can switch between Five people and Solo without losing the photos already selected for either mode. Click a filled slot to replace its photo, or use its remove button.
3. Set quality and sound
Draft selects 480p. Standard selects 720p for more detail. Both use the same five-role arrangement and vertical 9:16 format.
Sound is on by default. Turn the Sound switch off if you want a silent video. The example player has its own volume control; muting the example does not change your video settings.
4. Review before paying
Confirm that everyone pictured is an adult and has agreed to this use. Select Create my video to review the cast, quality, sound and total before continuing to a payment provider.
The order review shows the payment methods currently available. You confirm payment with Stripe or PayPal. If checkout is unavailable, the page keeps your draft and does not start a payment.
Return to your work
Photos and settings are saved in this browser. Wait for the saved indicator before closing the tab. Open My workspace to return to your draft or review a saved order.
Browser data does not sync between devices. Private browsing, clearing site data or changing browsers can remove access to a draft. Keep the original photos until you have finished.