Query-by-Example Keyword Spotting system using Multi-head Attention and Softtriple Loss

TL;DR

This paper introduces a neural network for query-by-example keyword spotting that combines multi-head attention, a multi-layered GRU, and softtriple loss, demonstrating effective performance across multiple datasets.

Contribution

It presents a novel architecture integrating multi-head attention with softtriple loss for improved user-defined keyword spotting.

Findings

Effective on internal and public datasets

Outperforms baseline systems

Component ablation confirms architecture benefits

Abstract

This paper proposes a neural network architecture for tackling the query-by-example user-defined keyword spotting task. A multi-head attention module is added on top of a multi-layered GRU for effective feature extraction, and a normalized multi-head attention module is proposed for feature aggregation. We also adopt the softtriple loss - a combination of triplet loss and softmax loss - and showcase its effectiveness. We demonstrate the performance of our model on internal datasets with different languages and the public Hey-Snips dataset. We compare the performance of our model to a baseline system and conduct an ablation study to show the benefit of each component in our architecture. The proposed work shows solid performance while preserving simplicity.