Improved Vocal Effort Transfer Vector Estimation for Vocal Effort-Robust Speaker Verification
Abstract
Despite the maturity of modern speaker verification technology, its performance still significantly degrades when facing non-neutrally-phonated (e.g., shouted and whispered) speech. To address this issue, in this paper, we propose a new speaker embedding compensation method based on a minimum mean square error (MMSE) estimator. This method models the joint distribution of the vocal effort transfer vector and non-neutrally-phonated embedding spaces and operates in a principal component analysis domain to cope with non-neutrally-phonated speech data scarcity. Experiments are carried out using a cutting-edge speaker verification system integrating a powerful self-supervised pre-trained model for speech representation. In comparison with a state-of-the-art embedding compensation method, the proposed MMSE estimator yields superior and competitive equal error rate results when tackling shouted and whispered speech, respectively.
Keywords
Cite
@article{arxiv.2305.02147,
title = {Improved Vocal Effort Transfer Vector Estimation for Vocal Effort-Robust Speaker Verification},
author = {Iván López-Espejo and Santi Prieto and Alfonso Ortega and Eduardo Lleida},
journal= {arXiv preprint arXiv:2305.02147},
year = {2023}
}