Simultaneous Denoising, Dereverberation, and Source Separation Using a Unified Convolutional Beamformer

This article investigates the applicability of a Weighted Power minimization Distortionless response convolutional beamformer (WPD) to simultaneous denoising, dereverberation, and source separation. The WPD is a recently proposed maximum likelihood (ML) convo-lutional beamformer that performs denoising and dereverberation simultaneously by unifying a Weighted Prediction Error dereverberation method (WPE) and a Minimum Power Distortionless Response beamformer (MPDR). In this paper, we extend the application of the WPD not only to simultaneous denoising and dereverberation, but also to source separation. For this purpose, we introduce a source parameter estimation unit that estimates the steering vectors and the time-varying powers of all the sources from noisy reverberant sound mixtures, and integrate the unit with the WPD. We experimentally confirm the effectiveness of the integrated method in terms of objective speech enhancement measures and Automatic Speech Recognition (ASR) performance.