Relational Proxy Loss for Audio-Text based Keyword Spotting

TL;DR

This paper introduces Relational Proxy Loss (RPL), a novel loss function that leverages the structural relations within acoustic and text embedding spaces to improve audio-text based keyword spotting performance.

Contribution

The study proposes RPL, which considers relational structures in embeddings, enhancing existing deep metric learning methods for audio-text keyword spotting.

Findings

Improved accuracy on WSJ corpus

Enhanced embedding relational structure

Outperforms traditional proxy-based losses

Abstract

In recent years, there has been an increasing focus on user convenience, leading to increased interest in text-based keyword enrollment systems for keyword spotting (KWS). Since the system utilizes text input during the enrollment phase and audio input during actual usage, we call this task audio-text based KWS. To enable this task, both acoustic and text encoders are typically trained using deep metric learning loss functions, such as triplet- and proxy-based losses. This study aims to improve existing methods by leveraging the structural relations within acoustic embeddings and within text embeddings. Unlike previous studies that only compare acoustic and text embeddings on a point-to-point basis, our approach focuses on the relational structures within the embedding space by introducing the concept of Relational Proxy Loss (RPL). By incorporating RPL, we demonstrated improved performance on the Wall Street Journal (WSJ) corpus.