Joint Neural AEC and Beamforming with Double-Talk Detection

TL;DR

This paper introduces a deep learning-based joint AEC and beamforming model that effectively suppresses echoes, noise, and nonlinear distortions in full-duplex communication, improving speech quality and recognition.

Contribution

It presents a novel end-to-end neural network combining AEC, beamforming, and double-talk detection for enhanced echo and noise suppression.

Findings

Outperforms existing multi-channel AEC systems in speech recognition.

Achieves higher speech quality in noisy and echoic environments.

Effectively handles nonlinear distortions from audio devices.

Abstract

Acoustic echo cancellation (AEC) in full-duplex communication systems eliminates acoustic feedback. However, nonlinear distortions induced by audio devices, background noise, reverberation, and double-talk reduce the efficiency of conventional AEC systems. Several hybrid AEC models were proposed to address this, which use deep learning models to suppress residual echo from standard adaptive filtering. This paper proposes deep learning-based joint AEC and beamforming model (JAECBF) building on our previous self-attentive recurrent neural network (RNN) beamformer. The proposed network consists of two modules: (i) multi-channel neural-AEC, and (ii) joint AEC-RNN beamformer with a double-talk detection (DTD) that computes time-frequency (T-F) beamforming weights. We train the proposed model in an end-to-end approach to eliminate background noise and echoes from far-end audio devices, which include nonlinear distortions. From experimental evaluations, we find the proposed network outperforms other multi-channel AEC and denoising systems in terms of speech recognition rate and overall speech quality.