Ant Group 10thOpen SourceNamed EchoMimic new projects that can bePortrait facial featuresandAudioCome help the character"Lip Sync”, combining facial landmarks and audio content to generate more stable and natural videos.
The project has high stability and naturalness. By fusing the features of audio and facial landmarks (key facial features and structures, usually located at the eyes, nose, mouth, etc.), it can generateMore in line with real facial movements and expression changesVideo.
It supports the use of audio or facial landmarks to generate portrait videos, and also supports combining audio and portrait photos to create a "lip sync" effect. It is reported that it supports multiple languages (including Mandarin Chinese and English) and multiple styles, and can also handleSingAnd other scenes.
Attached related links:
-
Project address:https://badtobest.github.io/echomimic.html