Sparse Binarization for Fast Keyword Spotting

TL;DR

This paper introduces a sparse input representation and linear classifier for keyword spotting, achieving four times faster performance, improved accuracy, and robustness in noisy environments on edge devices.

Contribution

The paper presents a novel sparse binarization approach for KWS that significantly enhances speed and robustness over previous models.

Findings

Model is 4x faster than previous state-of-the-art.

Achieves better accuracy and robustness in noisy conditions.

Suitable for deployment on resource-constrained edge devices.

Abstract

With the increasing prevalence of voice-activated devices and applications, keyword spotting (KWS) models enable users to interact with technology hands-free, enhancing convenience and accessibility in various contexts. Deploying KWS models on edge devices, such as smartphones and embedded systems, offers significant benefits for real-time applications, privacy, and bandwidth efficiency. However, these devices often possess limited computational power and memory. This necessitates optimizing neural network models for efficiency without significantly compromising their accuracy. To address these challenges, we propose a novel keyword-spotting model based on sparse input representation followed by a linear classifier. The model is four times faster than the previous state-of-the-art edge device-compatible model with better accuracy. We show that our method is also more robust in noisy environments while being fast. Our code is available at: https://github.com/jsvir/sparknet.