Efficient Acoustic Echo Suppression with Condition-Aware Training

TL;DR

This paper introduces an improved convolutional recurrent network for acoustic echo suppression that reduces complexity, enhances performance in double-talk scenarios, and incorporates condition-aware training to balance echo suppression and speech preservation.

Contribution

The paper presents a more efficient CRN architecture with better performance and introduces condition-aware training methods for acoustic echo suppression.

Findings

Reduced model size and computational complexity.

Outperforms baseline architectures in double-talk conditions.

Effective control of echo suppression and speech preservation trade-off.

Abstract

The topic of deep acoustic echo control (DAEC) has seen many approaches with various model topologies in recent years. Convolutional recurrent networks (CRNs), consisting of a convolutional encoder and decoder encompassing a recurrent bottleneck, are repeatedly employed due to their ability to preserve nearend speech even in double-talk (DT) condition. However, past architectures are either computationally complex or trade off smaller model sizes with a decrease in performance. We propose an improved CRN topology which, compared to other realizations of this class of architectures, not only saves parameters and computational complexity, but also shows improved performance in DT, outperforming both baseline architectures FCRN and CRUSE. Striving for a condition-aware training, we also demonstrate the importance of a high proportion of double-talk and the missing value of nearend-only speech in DAEC training data. Finally, we show how to control the trade-off between aggressive echo suppression and near-end speech preservation by fine-tuning with condition-aware component loss functions.