Turn a reference image and a speech or music track into a cinematic moving portrait, powered by Wan2.2 S2V-14B.
A clear, front-facing image works best. The audio length sets the clip length.
Generation can take several minutes. Uploaded media is sent to the hosted provider only to create this clip.
Your finished clip will appear here with the original audio included.
Powered by Wan-AI/Wan2.2-S2V-14B · Apache 2.0