Plug-and-Play Co-Occurring Face Attention for Robust Audio-Visual Speaker Extraction

Open in new window