Cross-Lingual Query-by-Example Spoken Term Detection: A Transformer-Based Approach
Abstract
Query-by-example spoken term detection (QbE-STD) is typically constrained by transcribed data scarcity and language specificity. This paper introduces a novel, language-agnostic QbE-STD model leveraging image processing techniques and transformer architecture. By employing a pre-trained XLSR-53 network for feature extraction and a Hough transform for detection, our model effectively searches for user-defined spoken terms within any audio file. Experimental results across four languages demonstrate significant performance gains (19-54%) over a CNN-based baseline. While processing time is improved compared to DTW, accuracy remains inferior. Notably, our model offers the advantage of accurately counting query term repetitions within the target audio.
Cite
@article{arxiv.2410.04091,
title = {Cross-Lingual Query-by-Example Spoken Term Detection: A Transformer-Based Approach},
author = {Allahdadi Fatemeh and Mahdian Toroghi Rahil and Zareian Hassan},
journal= {arXiv preprint arXiv:2410.04091},
year = {2024}
}