VSEGAN: Visual Speech Enhancement Generative Adversarial Network.

Xinmeng Xu Yang Wang Dongxiang Xu Yiyuan Peng Cong Zhang Jie Jia Binbin Chen

Published in: ICASSP (2022)

Keyphrases

speech enhancement
noisy environments
visual information
signal to noise ratio
visual features
noise reduction
image processing
background noise
vocal tract