Neural Network Augmented Kalman Filter for Robust Acoustic Howling Suppression

TL;DR

This paper introduces a neural network-augmented Kalman filter that significantly improves acoustic howling suppression by refining reference signals and estimating covariance metrics, outperforming traditional methods in dynamic audio environments.

Contribution

The paper presents a novel integration of neural networks into Kalman filters specifically for acoustic howling suppression, enhancing adaptability and performance over existing standalone methods.

Findings

Improved suppression of acoustic howling compared to baseline methods

Effective neural network integration enhances Kalman filter adaptability

Experimental results validate superior performance in dynamic conditions

Abstract

Acoustic howling suppression (AHS) is a critical challenge in audio communication systems. In this paper, we propose a novel approach that leverages the power of neural networks (NN) to enhance the performance of traditional Kalman filter algorithms for AHS. Specifically, our method involves the integration of NN modules into the Kalman filter, enabling refining reference signal, a key factor in effective adaptive filtering, and estimating covariance metrics for the filter which are crucial for adaptability in dynamic conditions, thereby obtaining improved AHS performance. As a result, the proposed method achieves improved AHS performance compared to both standalone NN and Kalman filter methods. Experimental evaluations validate the effectiveness of our approach.