RESOURCE EFFICIENT DEEP EIGENVECTOR BEAMFORMING

Zohrer, M; Pfeifenberger, L; Schindler, G; Froning, H; Pernkopf, F

Zohrer, M (reprint author), Graz Univ Technol, Signal Proc & Speech Commun Lab, Graz, Austria.

2018 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2018; (): 3354

Abstract

We propose binary neural networks (BNNs) for acoustic beamforming. This makes the speech enhancement approach resource efficient and applicable for em......

Full Text Link