Exploiting the directional coherence function for multichannel source extraction

Shan Liang, Guanjun Li, Shuai Nie, Zhanlei Yang, Wenju Liu, Jianhua Tao

Abstract: Highlights•We propose a directional coherence function for detecting the desired speech.•We apply the convex programming to design spatial filters.•We propose a new desired speech presence probability (DSPP).•The DSPP informed GSC is used for extracting the desired speech.