NC AI's Varco SyncFace Generates Game Facial Animations From Audio Without Post-Processing
Jang's frank assessment that existing AI lip-sync tools still demand post-processing and QA, and that Varco SyncFace is designed to skip those steps, directly addresses the bottleneck keeping automated facial animation out of actual game production.
Reporting from 1 sources: 4Gamer.net.
At NDC26, NC AI head Jang Han-yong presented Varco SyncFace, a generative AI system that produces facial animations from voice input. The technology aims to eliminate the manual post-processing and QA that existing solutions still require. Jang argued that facial animation is the next frontier after graphics and body motion capture, but current tools degrade with game-like acting and reverb, forcing costly manual fixes. Varco SyncFace targets production-ready output that needs no further adjustment.
At Nexon's NDC26 conference, NC AI head Jang Han-yong presented Varco SyncFace, a generative AI system that creates facial animations from audio input. Jang argued that after graphics and body motion capture reached high quality, facial animation is the next expected improvement. He showed comparisons of game footage with and without the technology, calling the difference in perceived quality significant.
Jang reviewed existing solutions from NVIDIA, Epic Games, Speech Graphics, and JALI, but said each has weaknesses for game use. NVIDIA's tool degrades with intense acting and reverb. Epic's MetaHuman makes lips too smooth and fails to convey emotion. All require post-processing and QA before commercial use, adding time and cost. Varco SyncFace is built to output production-ready animation that needs no further manual work.
Synthesized by Yomimono from the 1 cited source below, including Japanese-language reporting where cited, then editorially reviewed before publishing.