Ofer Schwartz

TL;DR: A novel algorithm to simultaneously suppress early reflections, late reverberation and ambient noise is presented, and a multi-microphone minimum mean square error estimator is used to obtain a spatially filtered version of the early speech component.

TL;DR: It is shown that the proposed correction method can decrease the speech distortion without significantly affecting the reverberation reduction, and a common weakness of spatial coherence-based estimators is their performance in high direct-to-diffuse ratio conditions.

TL;DR: This work forms the localization task as a maximum likelihood (ML) parameter estimation problem, and solves it by utilizing the expectation-maximization (EM) procedure, and proposes to adapt two recursive EM (REM) variants based on Titterington's scheme.

TL;DR: In this paper, a novel algorithm to simultaneously suppress early reflections, late reverberation, and ambient noise is presented and it is shown that significant improvement is obtained and that the proposed algorithm outperforms baseline single-channel and multichannel dereverberation algorithms, as well as a state-of-the-art multich channel derever beration algorithm.

TL;DR: A novel multichannel estimator is derived for the power spectral densities of the reverberation and the speech suitable also for noisy environments and is shown to provide improved performance measures as compared with the competing estimators.