[HTML][HTML] A Comprehensive Review on Discriminant Analysis for Addressing Challenges of Class-Level Limitations, Small Sample Size, and Robustness

L Qu, Y Pei - Processes, 2024 - mdpi.com
The classical linear discriminant analysis (LDA) algorithm has three primary drawbacks, ie,
small sample size problem, sensitivity to noise and outliers, and inability to deal with multi …

[書籍][B] Machine learning for speaker recognition

MW Mak, JT Chien - 2020 - books.google.com
This book will help readers understand fundamental and advanced statistical models and
deep learning models for robust speaker recognition and domain adaptation. This useful …

Mixture of PLDA for noise robust i-vector speaker verification

MW Mak, X Pang, JT Chien - IEEE/ACM Transactions on Audio …, 2015 - ieeexplore.ieee.org
In real-world environments, noisy utterances with variable noise levels are recorded and
then converted to i-vectors for cosine distance or PLDA scoring. This paper investigates the …

Cancellable speech template via random binary orthogonal matrices projection hashing

KY Chee, Z **, D Cai, M Li, WS Yap, YL Lai, BM Goi - Pattern Recognition, 2018 - Elsevier
The increasing advancement of mobile technology explosively popularizes the mobile
devices (eg iPhone, iPad). A large number of mobile devices provide great convenience and …

Spatial map** Zataria multiflora using different machine-learning algorithms

M Edalat, E Dastres, E Jahangiri, G Moayedi, A Zamani… - Catena, 2022 - Elsevier
Understanding the relationships between environmental factors that influence the
distribution of medicinal plants is crucial to identifying suitable habitats for them. Studies …

DNN-driven mixture of PLDA for robust speaker verification

N Li, MW Mak, JT Chien - IEEE/ACM Transactions on Audio …, 2017 - ieeexplore.ieee.org
The mismatch between enrollment and test utterances due to different types of variabilities is
a great challenge in speaker verification. Based on the observation that the SNR-level …

[PDF][PDF] Analysis of Methods and Techniques Used for Speaker Identification, Recognition, and Verification: A Study on Quarter-Century Research Outcomes

TS Mohammed, KM Aljebory, MAA Rasheed… - Iraqi Journal of …, 2021 - iasj.net
The theories and applications of speaker identification, recognition, and verification are
among the well-established fields. Many publications and advances in the relevant products …

[PDF][PDF] From adaptive score normalization to adaptive data normalization for speaker verification systems

S Cumani, S Sarni - Proceedings of the 24th INTERSPEECH …, 2023 - isca-archive.org
Abstract Domain and trial-dependent mismatch between training and evaluation data can
severely affect the performance of speaker verification systems, and are usually addressed …

Use of Neumann series decomposition to fit the Weighted Euclidean distance and Inner product scoring models in automatic speaker recognition

M Djellab, N Mehallegue, A Achi - Pattern Recognition Letters, 2019 - Elsevier
It is commonly agreed that the main performance indicators in real life biometric applications
are: accuracy, calibration quality and computation cost in the case of large databases use …

SNR-invariant PLDA with multiple speaker subspaces

N Li, MW Mak - … Conference on Acoustics, Speech and Signal …, 2016 - ieeexplore.ieee.org
To deal with the mismatch between the enrollment and test utterances caused by noise with
different signal-to-noise ratios (SNR), we have recently proposed an SNR-invariant PLDA …