Adaptive Noise Resilient Keyword Spotting Using One-Shot Learning

TL;DR

This paper introduces a low-resource, one-shot learning method for noise adaptation in keyword spotting systems, significantly improving accuracy in noisy environments while maintaining suitability for embedded devices.

Contribution

It presents a novel, lightweight approach for continuous noise adaptation of pretrained KWS models using only one-shot learning and one epoch, suitable for resource-constrained devices.

Findings

Adapted models outperform pretrained models across all tested noise scenarios.

Accuracy improvements range from 4.9% to 46.0% at low SNRs.

Method is effective and lightweight for real-world deployment.

Abstract

Keyword spotting (KWS) is a key component of smart devices, enabling efficient and intuitive audio interaction. However, standard KWS systems deployed on embedded devices often suffer performance degradation under real-world operating conditions. Resilient KWS systems address this issue by enabling dynamic adaptation, with applications such as adding or replacing keywords, adjusting to specific users, and improving noise robustness. However, deploying resilient, standalone KWS systems with low latency on resource-constrained devices remains challenging due to limited memory and computational resources. This study proposes a low computational approach for continuous noise adaptation of pretrained neural networks used for KWS classification, requiring only 1-shot learning and one epoch. The proposed method was assessed using two pretrained models and three real-world noise sources at signal-to-noise ratios (SNRs) ranging from 24 to -3 dB. The adapted models consistently outperformed the pretrained models across all scenarios, especially at SNR $\leq$ 18 dB, achieving accuracy improvements of 4.9% to 46.0%. These results highlight the efficacy of the proposed methodology while being lightweight enough for deployment on resource-constrained devices.