Feature Extraction for Spoofing Detection in Automatic Speaker Verification Systems

Please enable javascript in your browser.

Adiban, Mohammad | 2017

562 Viewed

Type of Document: M.Sc. Thesis
Language: Farsi
Document No: 50103 (19)
University: Sharif University of Technology
Department: Computer Engineering
Advisor(s): Sameti, Hossein
Abstract:
Automatic speaker verification systems are becoming increasingly epidemic by the development of technology. The use of these systems in applications such as smart phones is becoming increasingly popular as a user password, and the security of these systems against spoofing attacks has always been a concern. In this research, the purpose is to comprehensively investigate all types of possible spoofing attacks on speaker verification systems and then provide a countermeasure system against such attacks. The countermeasure system presented in this report, extracts features from speech signals that can make the difference between the genuine and spoofed speech spectrum more prominent by analysing the nature of spoofed attacks. Accordingly, the work presented in this study carried out on two standard datasets and focused on spectral properties based on the magnitude and phase of the speech signal and also by training different classifiers, have been able to improve the results obtained from the best baseline system about 34% and 58% in terms of equal error rate in the first and second dataset, respectively
Keywords:
Feature Extraction ; Speaker Verification ; Ownership Authentication ; Spoofed Speech ; Spooling Attacks ; Countermeasure

No TOC