End-to-End User-Defined Keyword Spotting using Shifted Delta Coefficients

TL;DR

This paper introduces the use of shifted delta coefficients (SDC) for user-defined keyword spotting, demonstrating improved accuracy over traditional features like MFCC by capturing temporal speech dynamics.

Contribution

The study proposes a novel application of SDC features in an end-to-end system for user-defined keyword spotting, outperforming existing methods on multiple datasets.

Findings

SDC features outperform MFCC in keyword spotting accuracy.

The approach improves AUC by 8.32% and EER by 8.69% on challenging datasets.

The method surpasses state-of-the-art UDKWS techniques.

Abstract

Identifying user-defined keywords is crucial for personalizing interactions with smart devices. Previous approaches of user-defined keyword spotting (UDKWS) have relied on short-term spectral features such as mel frequency cepstral coefficients (MFCC) to detect the spoken keyword. However, these features may face challenges in accurately identifying closely related pronunciation of audio-text pairs, due to their limited capability in capturing the temporal dynamics of the speech signal. To address this challenge, we propose to use shifted delta coefficients (SDC) which help in capturing pronunciation variability (transition between connecting phonemes) by incorporating long-term temporal information. The performance of the SDC feature is compared with various baseline features across four different datasets using a cross-attention based end-to-end system. Additionally, various configurations of SDC are explored to find the suitable temporal context for the UDKWS task. The experimental results reveal that the SDC feature outperforms the MFCC baseline feature, exhibiting an improvement of 8.32% in area under the curve (AUC) and 8.69% in terms of equal error rate (EER) on the challenging Libriphrase-hard dataset. Moreover, the proposed approach demonstrated superior performance when compared to state-of-the-art UDKWS techniques.