Improving Voice Spoofing Detection Through Extensive Analysis of Multicepstral Feature Reduction
Voice biometric systems play a critical role in numerous security applications, including electronic device authentication, banking transaction verification, and confidential communications. Despite their widespread utility, these systems are increasingly targeted by sophisticated spoofing attacks t...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-08-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/25/15/4821 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849406296983339008 |
|---|---|
| author | Leonardo Mendes de Souza Rodrigo Capobianco Guido Rodrigo Colnago Contreras Monique Simplicio Viana Marcelo Adriano dos Santos Bongarti |
| author_facet | Leonardo Mendes de Souza Rodrigo Capobianco Guido Rodrigo Colnago Contreras Monique Simplicio Viana Marcelo Adriano dos Santos Bongarti |
| author_sort | Leonardo Mendes de Souza |
| collection | DOAJ |
| description | Voice biometric systems play a critical role in numerous security applications, including electronic device authentication, banking transaction verification, and confidential communications. Despite their widespread utility, these systems are increasingly targeted by sophisticated spoofing attacks that leverage advanced artificial intelligence techniques to generate realistic synthetic speech. Addressing the vulnerabilities inherent to voice-based authentication systems has thus become both urgent and essential. This study proposes a novel experimental analysis that extensively explores various dimensionality reduction strategies in conjunction with supervised machine learning models to effectively identify spoofed voice signals. Our framework involves extracting multicepstral features followed by the application of diverse dimensionality reduction methods, such as Principal Component Analysis (PCA), Truncated Singular Value Decomposition (SVD), statistical feature selection (ANOVA F-value, Mutual Information), Recursive Feature Elimination (RFE), regularization-based LASSO selection, Random Forest feature importance, and Permutation Importance techniques. Empirical evaluation using the ASVSpoof 2017 v2.0 dataset measures the classification performance with the Equal Error Rate (EER) metric, achieving values of approximately 10%. Our comparative analysis demonstrates significant performance gains when dimensionality reduction methods are applied, underscoring their value in enhancing the security and effectiveness of voice biometric verification systems against emerging spoofing threats. |
| format | Article |
| id | doaj-art-74bb17226a2f44ffb3736df862df4d3b |
| institution | Kabale University |
| issn | 1424-8220 |
| language | English |
| publishDate | 2025-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj-art-74bb17226a2f44ffb3736df862df4d3b2025-08-20T03:36:26ZengMDPI AGSensors1424-82202025-08-012515482110.3390/s25154821Improving Voice Spoofing Detection Through Extensive Analysis of Multicepstral Feature ReductionLeonardo Mendes de Souza0Rodrigo Capobianco Guido1Rodrigo Colnago Contreras2Monique Simplicio Viana3Marcelo Adriano dos Santos Bongarti4Department of Computer Science and Statistics, Institute of Biosciences, Letters and Exact Sciences, São Paulo State University, São José do Rio Preto 15054-000, SP, BrazilDepartment of Computer Science and Statistics, Institute of Biosciences, Letters and Exact Sciences, São Paulo State University, São José do Rio Preto 15054-000, SP, BrazilDepartment of Science and Technology, Federal University of São Paulo, São José dos Campos 12247-014, SP, BrazilDepartment of Computing, Federal University of São Carlos, São Carlos 13565-905, SP, BrazilWeierstraß Institute for Applied Analysis and Stochastics, 10117 Berlin, GermanyVoice biometric systems play a critical role in numerous security applications, including electronic device authentication, banking transaction verification, and confidential communications. Despite their widespread utility, these systems are increasingly targeted by sophisticated spoofing attacks that leverage advanced artificial intelligence techniques to generate realistic synthetic speech. Addressing the vulnerabilities inherent to voice-based authentication systems has thus become both urgent and essential. This study proposes a novel experimental analysis that extensively explores various dimensionality reduction strategies in conjunction with supervised machine learning models to effectively identify spoofed voice signals. Our framework involves extracting multicepstral features followed by the application of diverse dimensionality reduction methods, such as Principal Component Analysis (PCA), Truncated Singular Value Decomposition (SVD), statistical feature selection (ANOVA F-value, Mutual Information), Recursive Feature Elimination (RFE), regularization-based LASSO selection, Random Forest feature importance, and Permutation Importance techniques. Empirical evaluation using the ASVSpoof 2017 v2.0 dataset measures the classification performance with the Equal Error Rate (EER) metric, achieving values of approximately 10%. Our comparative analysis demonstrates significant performance gains when dimensionality reduction methods are applied, underscoring their value in enhancing the security and effectiveness of voice biometric verification systems against emerging spoofing threats.https://www.mdpi.com/1424-8220/25/15/4821spoofing detectiondimensionality reductionpattern recognitioncepstral analysismachine learning |
| spellingShingle | Leonardo Mendes de Souza Rodrigo Capobianco Guido Rodrigo Colnago Contreras Monique Simplicio Viana Marcelo Adriano dos Santos Bongarti Improving Voice Spoofing Detection Through Extensive Analysis of Multicepstral Feature Reduction Sensors spoofing detection dimensionality reduction pattern recognition cepstral analysis machine learning |
| title | Improving Voice Spoofing Detection Through Extensive Analysis of Multicepstral Feature Reduction |
| title_full | Improving Voice Spoofing Detection Through Extensive Analysis of Multicepstral Feature Reduction |
| title_fullStr | Improving Voice Spoofing Detection Through Extensive Analysis of Multicepstral Feature Reduction |
| title_full_unstemmed | Improving Voice Spoofing Detection Through Extensive Analysis of Multicepstral Feature Reduction |
| title_short | Improving Voice Spoofing Detection Through Extensive Analysis of Multicepstral Feature Reduction |
| title_sort | improving voice spoofing detection through extensive analysis of multicepstral feature reduction |
| topic | spoofing detection dimensionality reduction pattern recognition cepstral analysis machine learning |
| url | https://www.mdpi.com/1424-8220/25/15/4821 |
| work_keys_str_mv | AT leonardomendesdesouza improvingvoicespoofingdetectionthroughextensiveanalysisofmulticepstralfeaturereduction AT rodrigocapobiancoguido improvingvoicespoofingdetectionthroughextensiveanalysisofmulticepstralfeaturereduction AT rodrigocolnagocontreras improvingvoicespoofingdetectionthroughextensiveanalysisofmulticepstralfeaturereduction AT moniquesimplicioviana improvingvoicespoofingdetectionthroughextensiveanalysisofmulticepstralfeaturereduction AT marceloadrianodossantosbongarti improvingvoicespoofingdetectionthroughextensiveanalysisofmulticepstralfeaturereduction |