Cascade Multi-Channel Noise Reduction and Acoustic Feedback Cancellation
Santiago Ruiz, Toon van Waterschoot, Marc Moonen · ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · 2022
Acoustic feedback and noise are common problems that corrupt microphone signals and affect the performance of speech and audio signal processing applications and devices. In this paper, a cascade noise reduction (NR) and acoustic feedback cancellation (AFC) algorithm is presented for speech applications where a multi-channel Wiener filter (MWF) based NR is applied first followed by a single-channel prediction-error method (PEM) based adaptive feedback cancellation stage. It is shown that by using a rank-2 estimate of the speech correlation matrix in the NR stage it is possible to obtain a good feedback path estimate for the reference microphone in the AFC stage. Closed-loop simulations with M microphones and 1 loudspeaker are presented using both an M-channel rank-1 and an (M + 1)-channel rank-2 MWF and it is shown that for the considered input signal-to-noise ratios the proposed algorithm increases the added stable gain (ASG) of the system.