A comparison of different support vector machine kernels for artificial speech detection

Tan, Choon Beng; Hijazi, Mohd Hanafi Ahmad; Nohuddin, Puteri N. E.

doi:10.12928/telkomnika.v21i1.24259

Cited by 4 publications

(3 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recent advancements in artificial speech detection have significantly contributed to fortifying the security of ASV systems, which are susceptible to spoofing attacks involving artificial speech. Notably, one recent study has explored different Support Vector Machine (SVM) kernels for artificial speech detection [2]. This recent work focused on evaluating SVM kernels in conjunction with a diverse set of fused features.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

A Comparative Evaluation on Data Transformation Approach for Artificial Speech Detection

Tan,

Ahmad Hijazi

2024

ITM Web Conf.

Self Cite

View full text Add to dashboard Cite

The rise of voice biometrics has transformed user authentication and offered enhanced security and convenience while phasing out less secure methods. Despite these advancements, Automatic Speaker Verification (ASV) systems remain vulnerable to spoofing, particularly with artificial speech generated swiftly using advanced speech synthesis and voice conversion algorithms. A recent data transformation technique achieved an impressive Equal Error Rate (EER) of 1.42% on the ASVspoof 2019 Logical Access Dataset. While this approach predominantly relies on Support Vector Machine (SVM) as the backend classifier for artificial speech detection, it is vital to explore a broader range of classifiers to enhance resilience. This paper addresses this research gap by systematically assessing classifier efficacy in artificial speech detection. The objectives are twofold: first, to evaluate various classifiers, not limited to SVM, and identify those best suited for artificial speech detection; second, to compare this approach's performance with existing methods. The evaluation demonstrated SVM-Polynomial as the top-performing classifier, surpassing the end-to-end learning approach. This work contributes to a deeper understanding of classifier efficacy and equips researchers and practitioners with a diversified toolkit for building robust ASV spoofing detection systems.

show abstract

Section: Related Workmentioning

confidence: 99%

“…One notable approach that has garnered attention in addressing this vulnerability employs data transformation techniques for artificial speech detection [2]. This approach has exhibited impressive results, achieving a remarkably low Equal Error Rate (EER) of 1.42% on the ASVspoof 2019 Logical Access Dataset [3].…”

Section: Introductionmentioning

confidence: 99%

A Comparative Evaluation on Data Transformation Approach for Artificial Speech Detection

Tan,

Ahmad Hijazi

2024

ITM Web Conf.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Demand from a wide range of users has led to a fast increase in the number of urban vehicles [12]- [14]. Due to the current application environment's need for high-speed transmission and minimal delay, conventional vehicular ad hoc networks (VANETs) based on 4G networks fall short [15]- [19].…”

Section: Introductionmentioning

confidence: 99%

Authentication with privacy-preserving scheme for 5G-enabled vehicular networks

Hamdan,

Makhlouf,

Mnif

2024

Bulletin EEI

View full text Add to dashboard Cite

The 5G-enabled vehicular network is an innovative technology that has promise for intelligent transportation systems. It enables the transmitting of messages about traffic that deliver the most recent information on congestion, road conditions, and driving surroundings. The communication channel used by vehicle networks is inherently open, which unfortunately exposes the system to privacy and security concerns. To solve the problems of deploying a safe vehicular network, some academics have put forth plans. However, a number of current methods have significant computational or communication overhead costs. To solve this problem, an efficient and secure authentication with a privacypreserving (ES-APP) scheme established elliptic curve encryption is introduced. With the proposed ES-APP, the data signed and verified for vehicle-to-vehicle and vehicle-to-infrastructure modes in the 5G-based vehicular network are more effective. The ES-APP scheme’s goal is to meet the criteria for the security and privacy of the 5G-enabled automotive network. Ultimately, this work discusses the critical survey of the existing studies and the expected outcome for the ESAPP scheme and further works in the 5G-enabled vehicular network.

show abstract

Unmasking the Truth: A Deep Learning Approach to Detecting Deepfake Audio Through MFCC Features

Altalahin,

AlZu'bi,

Alqudah

et al. 2023

2023 International Conference on Information Technology (ICIT)

View full text Add to dashboard Cite

A comparison of different support vector machine kernels for artificial speech detection

Cited by 4 publications

References 26 publications

A Comparative Evaluation on Data Transformation Approach for Artificial Speech Detection

A Comparative Evaluation on Data Transformation Approach for Artificial Speech Detection

Authentication with privacy-preserving scheme for 5G-enabled vehicular networks

Unmasking the Truth: A Deep Learning Approach to Detecting Deepfake Audio Through MFCC Features

Contact Info

Product

Resources

About