Search Results for author: Xulin Fan

Found 2 papers, 0 papers with code

Dual-Path Cross-Modal Attention for better Audio-Visual Speech Extraction

no code implementations9 Jul 2022 Zhongweiyang Xu, Xulin Fan, Mark Hasegawa-Johnson

Most current research upsamples the visual features along the time dimension so that audio and video features are able to align in time.

Speech Extraction

Cannot find the paper you are looking for? You can Submit a new open access paper.