Data-Centric Optimization of Enrollment Selection in Speaker Identification

Long-Quoc Le, Long-Quoc Le, Minh-Nhut Ngo, Minh-Nhut Ngo

2025

Abstract

In this paper, we introduce a novel method for optimizing enrollment selection in speaker identification systems, with a particular focus on low-resource languages. Unlike traditional approaches that rely on random enrollment samples, our method systematically analyzes pair-wise similarities between enrollment utterances to eliminate poor-quality samples often impacted by noise or adverse environments. By retaining only high-quality and representative utterances, we ensure a more robust speaker profile. This innovative approach, applied to the Vietnam-Celeb dataset using the state-of-the-art ECAPA-TDNN model, delivers substantial performance improvements. Our method boosts accuracy from 73.38% in bad scenarios to 93.62% and increases the F1-score from 72.91% to 95.48%, demonstrating the effectiveness of focusing on quality-driven enrollment selection even in low-resource contexts.

Paper Citation

in Harvard Style

Le L. and Ngo M. (2025). Data-Centric Optimization of Enrollment Selection in Speaker Identification. In Proceedings of the 14th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM; ISBN 978-989-758-730-6, SciTePress, pages 344-351. DOI: 10.5220/0013256800003905

in Bibtex Style

@conference{icpram25,
author={Long-Quoc Le and Minh-Nhut Ngo},
title={Data-Centric Optimization of Enrollment Selection in Speaker Identification},
booktitle={Proceedings of the 14th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM},
year={2025},
pages={344-351},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013256800003905},
isbn={978-989-758-730-6},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 14th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM
TI - Data-Centric Optimization of Enrollment Selection in Speaker Identification
SN - 978-989-758-730-6
AU - Le L.
AU - Ngo M.
PY - 2025
SP - 344
EP - 351
DO - 10.5220/0013256800003905
PB - SciTePress