Frequency & Channel Attention Network for Small Footprint Noisy Spoken Keyword Spotting

TL;DR

This paper introduces FCA-Net, a lightweight CNN with a novel attention module and training strategy, significantly improving noisy keyword spotting accuracy while maintaining a small model size.

Contribution

The paper presents FCA-Net, a new CNN architecture with a channel and frequency-specific attention module and curriculum training, enhancing noise robustness in small-footprint KWS systems.

Findings

Outperforms state-of-the-art small-footprint KWS in noisy environments

Uses a novel attention module for channel and frequency-specific feature weighting

Employs curriculum-based multi-condition training for robustness

Abstract

In this paper, we aim to improve the robustness of Keyword Spotting (KWS) systems in noisy environments while keeping a small memory footprint. We propose a new convolutional neural network (CNN) called FCA-Net, which combines mixer unit-based feature interaction with a two-dimensional convolution-based attention module. First, we introduce and compare lightweight attention methods to enhance noise robustness in CNN. Then, we propose an attention module that creates fine-grained attention weights to capture channel and frequency-specific information, boosting the model's ability to handle noisy conditions. By combining the mixer unit-based feature interaction with the attention module, we enhance performance. Additionally, we use a curriculum-based multi-condition training strategy. Our experiments show that our system outperforms current state-of-the-art solutions for small-footprint KWS in noisy environments, making it reliable for real-world use.