现代电影技术Issue(2)：45-52,8.DOI:10.3969/j.issn.1673-3215.2025.02.007
基于Omniverse Audio2Face的语音驱动数字人三维面部动画生成技术应用研究
Research on the application of speech-driven digital human 3D facial anima-tion generation based on Omniverse Audio2Face
摘要
Abstract
To enhance the realism of 3D facial animation in digital humans and improve their interactive experience while exploring alternative solutions to traditional high-cost facial capture techniques,this study investigates speech-driven ap-proaches to digital human facial animation.Based on Omniverse Audio2Face,the proposed approach enables the anima-tion of Unreal Engine MetaHuman characters through voice input.Experimental evaluations demonstrate advantages in both accuracy and efficiency.This paper summarizes relevant optimization suggestions,providing insights for the applica-tion and development of digital human facial animation technologies.关键词
数字人/三维面部动画/语音驱动/人工智能Key words
Digital Human/3D Facial Animation/Speech-driven/Artificial Intelligence分类
信息技术与安全科学引用本文复制引用
苏文吉,鲁梦河..基于Omniverse Audio2Face的语音驱动数字人三维面部动画生成技术应用研究[J].现代电影技术,2025,(2):45-52,8.基金项目
北京电影学院人才队伍建设资助计划——创新团队项目"电影智能制作中的艺术与科学融合发展研究"(3040025002). （3040025002）
