Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper addresses the challenge of computational complexity in acoustic echo control (AEC) by employing knowledge distillation (KD) to enhance the efficiency of machine-learned models. The authors introduce the CGGN16 student model, which achieves significantly lower near-end speech distortion while operating at just 2% of the computational complexity of its teacher model. Notably, this approach allows the CGGN16 to outperform both a more complex model trained on ground-truth labels and other recent AEC architectures, demonstrating a substantial improvement in performance without the typical trade-offs associated with model scaling.
Achieving superior acoustic echo control with a model that operates at just 2% of the computational cost of its more complex counterpart is a game-changer for real-time applications.
In recent years, many efforts have been made to supersede classical acoustic echo control (AEC) algorithms with more powerful machine-learned approaches. While surpassing the performance of well-established adaptive filters is very much possible, a remaining challenge is computational complexity. Popular architectures, such as convolutional recurrent networks (CRNs), are by multiple orders of magnitude computationally more expensive than classical signal processing solutions. Scaling down such models is usually straight-forward, but it comes at the cost of a notably reduced performance. We show - to the author's knowledge for the first time in AEC - how these performance drops can be successfully alleviated to a large degree by employing an effective knowledge distillation (KD) process, enabling more potent efficient AEC. Our proposed CGGN16 student AEC models show significantly less near-end speech distortion at only 2% of its teacher's computational complexity, surpass the overall performance of a six times more complex model trained on ground-truth labels, and outperform other AEC-focused architectures from recent literature.