Portrait & audio preparation checklist
Prepare the right portrait mode and a short audio input before opening a paid generation request.
1. Keep both faces readable
Use sharp, front-facing or gently angled adult portraits. The eyes, nose and mouth should be visible. Choose even light instead of bright backlight. For Solo, choose a single clear person. Avoid tiny faces in a crowd when the performer needs to remain recognizable.
2. Count files and performers separately
One person uses one adult portrait. Where verified support is available, Two portraits uses one adult in each of two files; Group photo uses one shared image. Only enabled modes can be submitted. A visible concept illustration does not verify group consistency.
3. Prepare short MP3 or WAV audio
For Use my audio, prepare an authorized MP3 or WAV of at most 30 seconds. It must cover the chosen video duration. You can listen locally after selection, but this is not a timeline editor and it does not cut, re-time or beat-sync a complete song. For Original rap, prepare a concise topic instead of an audio file.
4. Respect the person in the frame
Only upload adults who have given permission. Do not use a public figure's identity, a child's photo or someone else's private image. A better technical photo never replaces consent. Keep your original files and review our privacy policy before uploading sensitive material.