Efficient audio-visual information fusion using encoding pace synchronization for Audio-Visual Speech Separation

Open Webpage

Xinmeng Xu, Weiping Tu, Yuhong Yang

Published: 01 Jan 2025, Last Modified: 03 Mar 2025Inf. Fusion 2025EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: Highlights•Propose an Encoding Pace Synchronization Network for AVSS.•Allowing information to be encoded at paces of audio and visual modalities.•Synchronizing encoding paces of audio and visual modalities.•Preserving the distinct characteristics of each modality.