Global-Local Convolution with Spiking Neural Networks for Energy-efficient Keyword Spotting
This work addresses energy efficiency for keyword spotting systems on resource-constrained edge devices, presenting an incremental improvement in model design.
The paper tackles the problem of energy-efficient keyword spotting on edge devices by proposing a lightweight spiking neural network model with global-local convolution and bottleneck modules, achieving competitive performance among SNN-based models with fewer parameters on the Google Speech Commands Dataset.
Thanks to Deep Neural Networks (DNNs), the accuracy of Keyword Spotting (KWS) has made substantial progress. However, as KWS systems are usually implemented on edge devices, energy efficiency becomes a critical requirement besides performance. Here, we take advantage of spiking neural networks' energy efficiency and propose an end-to-end lightweight KWS model. The model consists of two innovative modules: 1) Global-Local Spiking Convolution (GLSC) module and 2) Bottleneck-PLIF module. Compared to the hand-crafted feature extraction methods, the GLSC module achieves speech feature extraction that is sparser, more energy-efficient, and yields better performance. The Bottleneck-PLIF module further processes the signals from GLSC with the aim to achieve higher accuracy with fewer parameters. Extensive experiments are conducted on the Google Speech Commands Dataset (V1 and V2). The results show our method achieves competitive performance among SNN-based KWS models with fewer parameters.