Neural Kalman Filters for Acoustic Echo Cancellation

TL;DR

This paper explores enhancing frequency-domain adaptive Kalman filters for acoustic echo cancellation with deep neural networks, aiming to improve convergence speed, echo suppression, and speech preservation in challenging conditions.

Contribution

It introduces neural Kalman filter variants that outperform traditional FDKF in echo cancellation and speech preservation, supported by a comprehensive comparison of DNN-based extensions.

Findings

Neural Kalman filters achieve faster convergence.

They provide better echo cancellation performance.

They preserve near-end speech more effectively.

Abstract

Kalman filtering is a powerful approach to adaptive filtering for various problems in signal processing. The frequency-domain adaptive Kalman filter (FDKF), based on the concept of the acoustic state space, provides a unifying solution to the adaptive filter update and the related stepsize control. It was conceived for the problem of acoustic echo cancellation and, as such, is frequently applied in hands-free systems. This article motivates and briefly recapitulates the linear FDKF and investigates how it can be further supported by deep neural networks (DNNs) in various ways, specifically to overcome the challenges and limitations related to the usually required estimation of process and observation noise covariances for the Kalman filter. While the mere FDKF comes with very low computational complexity, its neural Kalman filter variants may deliver faster (re)convergence, better echo cancellation, and even exceed the FDKF in its excellent double-talk near-end speech preservation both under linear and nonlinear loudspeaker conditions. To provide a synopsis of the state of the art, this article contributes a comparison of a range of DNN-based extensions of FDKF in the same training framework and using the same data.