College of Graduate Studies: Theses & Dissertations

Term of Award

Spring 2026

Degree Name

Master of Science, Information Technology

Document Type and Release Option

Thesis (open access)

Copyright Statement / License for Reuse

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.

Department

<--Please Select Department-->

Committee Chair

Kim Jongyeop

Committee Member 1

Hayden Wimmer

Committee Member 2

Lei Chen

Abstract

This study develops and evaluates a machine learning and deep learning-based voice authentication system for secure identity verification. As traditional authentication methods such as passwords, PINs, and security tokens continue to face challenges, including identity theft, forgetting, and unauthorized access, voice biometrics offers a more secure, convenient, and user-friendly alternative, especially for remote, hands-free, and accessibility-focused applications. The study adopts a closed-set speaker identification framework, where the system determines the most likely speaker from a predefined group of enrolled users. A structured methodology is implemented, beginning with audio preprocessing and feature extraction. Key acoustic features, including Mel-Frequency Cepstral Coefficients (MFCCs), spectral centroid, spectral rolloff, zero-crossing rate, and root square energy, are extracted and combined into fixed-length feature vectors. These features are used to train and evaluate multiple machine learning models, including classical approaches such as Support Vector Machines and Random Forests, as well as deep learning models such as Convolutional Neural Networks and recurrent architecture. This study demonstrates that voice can serve as a secure and effective biometric password while emphasizing the need for authentication systems that are not only intelligent, but also reliable, inclusive, and resilient against emerging cybersecurity threats such as voice cloning and deepfake

attacks, identity theft, and financial fraud. The results contribute to the advancement of secure, cost-effective, and human-centered biometric authentication solutions, with significant implications for cybersecurity, fraud prevention, digital identity protection, and improved accessibility for individuals with disabilities. To ensure a comprehensive evaluation, the study goes beyond standard accuracy metrics by incorporating cross-validation, receiver operating characteristic analysis, and statistical significance testing. In addition, a robust analysis is conducted under various noise conditions, including Gaussian noise, signal-to-noise ratio variations, quantization effects, and feature dropout. This provides insight into model performance in realistic and potentially adversarial environments.

OCLC Number

1588662260

Research Data and Supplementary Material

Yes

Download

Share

COinS