萬 2.1 幻想肖像-音頻驅動-KJ

“Wan2_1_FantasyPortrait” can be understood as: generating lip movements strictly aligned with the audio, augmented by facial expressions and subtle head motion. Built on Wan2.1’s video generation capabilities and combined with audio‑driven lip‑sync, it enables “single portrait image + audio → lip‑synced digital human video.” Note: quality may degrade with large side views, occlusions, or rapid head movements.