E2E-AEC: Implementing an end-to-end neural network learning approach for acoustic echo cancellation

TL;DR

This paper introduces a novel end-to-end neural network approach for acoustic echo cancellation that operates without traditional methods, utilizing progressive learning, knowledge transfer, attention optimization, and voice activity detection to improve echo suppression and speech quality.

Contribution

The paper presents a new neural network-based E2E-AEC method that eliminates reliance on traditional linear techniques and enhances performance through innovative training and attention strategies.

Findings

Effective echo suppression demonstrated on public datasets

Improved speech quality with voice activity detection

Enhanced time alignment via optimized attention mechanism

Abstract

We propose a novel neural network-based end-to-end acoustic echo cancellation (E2E-AEC) method capable of streaming inference, which operates effectively without reliance on traditional linear AEC (LAEC) techniques and time delay estimation. Our approach includes several key strategies: First, we introduce and refine progressive learning to gradually enhance echo suppression. Second, our model employs knowledge transfer by initializing with a pre-trained LAECbased model, harnessing the insights gained from LAEC training. Third, we optimize the attention mechanism with a loss function applied on attention weights to achieve precise time alignment between the reference and microphone signals. Lastly, we incorporate voice activity detection to enhance speech quality and improve echo removal by masking the network output when near-end speech is absent. The effectiveness of our approach is validated through experiments conducted on public datasets.