English

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

Computation and Language 2021-05-11 v2 Machine Learning

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.

Keywords

Cite

@article{arxiv.2102.07061,
  title  = {Query-by-Example Keyword Spotting system using Multi-head Attention and Softtriple Loss},
  author = {Jinmiao Huang and Waseem Gharbieh and Han Suk Shim and Eugene Kim},
  journal= {arXiv preprint arXiv:2102.07061},
  year   = {2021}
}

Comments

Accepted by ICASSP 2021

R2 v1 2026-06-23T23:08:19.150Z