A two microphone-based approach for speech enhancement in adverse environments

Kai Li, Yanmeng Guo, Qiang Fu, Yonghong Yan

Abstract: In this paper, we present a two microphone-based speech enhancement algorithm using spatial information-based spectral amplitude estimation in adverse environments. A target presence probability estimator based on Bayes detection rules using both phase difference and magnitude squared coherence is proposed for soft decision. Experimental results show that the proposed algorithm achieves better noise suppression and lower signal distortion over the comparative dual-channel algorithms.