Low complexity online convolutional beamforming

TL;DR

This paper introduces a low complexity online convolutional beamformer using a Kalman filter-based affine projection algorithm, achieving significant computational efficiency and improved performance in dereverberation and noise reduction tasks.

Contribution

It presents a novel low complexity adaptive filtering method for convolutional beamforming, reducing computational load while maintaining or improving performance.

Findings

Significantly lower computational complexity than existing methods

Slightly better performance on the REVERB challenge dataset

Effective for joint dereverberation and noise reduction

Abstract

Convolutional beamformers integrate the multichannel linear prediction model into beamformers, which provide good performance and optimality for joint dereverberation and noise reduction tasks. While longer filters are required to model long reverberation times, the computational burden of current online solutions grows fast with the filter length and number of microphones. In this work, we propose a low complexity convolutional beamformer using a Kalman filter derived affine projection algorithm to solve the adaptive filtering problem. The proposed solution is several orders of magnitude less complex than comparable existing solutions while slightly outperforming them on the REVERB challenge dataset.