Back to Publications
CVPR 2026 Findings 2025
TalkVid: A Large-Scale Diversified Dataset for Audio-Driven Talking Head Synthesis
Shunian Chen*, Hejin Huang*, Yexin Liu*, Zihan Ye, Pengcheng Chen, Chenghao Zhu, Michael Guan, Rongsheng Wang, Junying Chen, Benyou Wang
Abstract
A large-scale diversified dataset for audio-driven talking head synthesis.