Upload your portrait to create videos with a lip-syncing AI spokesperson
0 / 0 avatars
💡 Set a default avatar for Autopilot generations.
CRITICAL SYSTEM REQUIREMENT
MANDATORY INITIALIZATION: The very first time you generate a video with a newly uploaded Avatar, it will physically take the cloud server roughly 15 minutes to mathematically map and cache your facial geometry. You MUST ensure your very first text script is 10 seconds or shorter! Once the 10-second video completes, your avatar is cached and you can safely generate long videos! (Note: If you upload and select a different Avatar photo, you must perform this 10-second caching run again for the new face!)
MAXIMUM LENGTH: Due to computational intensity, the absolute maximum avatar rendering length is 12 minutes per video. Scripts exceeding 12m will be automatically rejected.
📸Upload AvatarJPG/PNG (Photo) • MP4/MOV (Video) Max 30s. Video: 30fps only (1080p)
🧬
Clone Your Voice
Upload a 30-second voice sample to create your personal AI narrator
Loading clone status...
🎤
Drop your voice sample here
WAV format only • Max 20 MB • At least 10 seconds of clear speech
Processing voice clone...0%
Uploading audio sample...
✅
Voice Clone Active
Created: —
💡 Select "My Voice" in the narrator dropdown when creating video scripts to use your cloned voice.