Yapay Zeka Algoritmaları Kullanılarak Sesle Cinsiyet Tespiti

Abstract

Bilim ve teknolojideki gelişmeler sonucunda sağlıktan eğitime, ticaretten sosyal hayatımıza kadar tüm yaşam alanlarımız dijital ortama taşınmıştır. Bu süreçle birlikte insan gibi düşünen ve hareket eden sistemler oluşturmak amacıyla geliştirilmiş yapay zeka kavramı da hayatımızın her alanında kullanılmaya başlanmıştır. Bu çalışmada, yapay zeka algoritmaları kullanılarak ses verilerinin incelenmesiyle cinsiyet belirlemeyi hedefleyen bir algoritma geliştirilmiştir. Cinsiyet tespitine yönelik olarak yapılan Bu tespit sayesinde sosyal mühendislik gibi çeşitli alanlarda ve dolandırıcılık, kişi tespiti, reklam yatırımları gibi siber güvenlik alanlarında önemli katkılar sağlanması hedeflenmiştir. Uygulama geliştirilirken, çeşitli yapay zeka algoritmaları için tamamen açık kaynak kodlu R yazılımı kullanılmıştır. Bu sayede yukarıda bahsedilen güvenlik önlemlerinin yüksek maliyetli sistemler yerine düşük maliyetli önlemler alınmasına ve pazarlama gibi alanlarda satış rakamlarının artırılmasına çözüm aranmıştır. Ayrıca çalışmada yapay zeka algoritması olarak denetimli öğrenme kullanılmıştır. Çalışmanın yapay zeka analiz sonuçları, ses verileri aracılığıyla kişinin cinsiyetinin çok başarılı oranlarda belirlenebildiğini göstermiştir.

Keywords

• Yapay Zeka
• Cinsiyet Tanıma
• Siber Güvenlik
• Ses Analizi
• Özellik Çıkarma
• Denetimli Öğrenme
• Gradyan Artırma Makinesi

Gender Detection Via Voice Using Artificial Intelligence Algorithms

Abstract

As a result of the developments in science and technology, all our living spaces, from health, education, and trade to our social life, have been moved to the digital environment. With this process, artificial intelligence, which is the ultimate goal of creating systems that think and act like human beings, has started to be used in all areas of our lives. This study focuses on gender determination by using artificial intelligence algorithms on voice data. Thanks to this determination, significant contributions will be made in various fields such as social engineering and cyber security such as fraud, person detection, and advertising investments. The study used a completely open-source R application for various artificial intelligence algorithms. In this way, a solution has been provided to take the security as mentioned above measures with low cost instead of high-cost systems and increase the sales figures in areas such as marketing. In the study, supervised learning artificial intelligence algorithms were examined. The artificial intelligence analysis results of the study have shown that the gender of the person could be determined at very successful rates through the voice data.

Keywords

• Artificial Intelligence
• Gender Recognition
• Cyber Security
• Voice Analysis
• Feature Extraction
• Supervised Learning
• Gradient Boosting Machine

References

1. A. Ross, S. Banerjee, and A. Chowdhury, "Security in smart cities: A brief review of digital forensic schemes for biometric data," Pattern Recognition Letters, vol. 138, pp. 346-354, 2020. Doi: 10.1016/j.patrec.2020.07.009
2. M. A. Ferrag, L. Maglaras, A. Derhab, and H. Janicke, "Authentication schemes for smart mobile devices: Threat models, countermeasures, and open research issues," Telecommunication Systems, vol. 73, no. 2, pp. 317-348, 2020. Doi: 10.1007/s11235-019-00612-5
3. A. Abozaid, A. Haggag, H. Kasban, and M. Eltokhy, "Multimodal biometric scheme for human authentication technique based on voice and face recognition fusion," Multimedia tools and applications, vol. 78, no. 12, pp. 16345-16361, 2019. Doi: 10.1007/s11042-018-7012-3
4. S. Tanwar, M. S. Obaidat, S. Tyagi, and N. Kumar, "Online signature-based biometric recognition," in Biometric-based physical and cybersecurity systems: Springer, 2019, pp. 255-285. Doi: 10.1007/978-3-319-98734-7_10
5. J. Chu, “Artificial intelligence model detects asymptomatic Covid-19 infections through cellphone-recorded coughs”. Available: https://news. mit. edu/2020/covid-19-cough-cellphone-detection [Access Date: 08.04. 2022].
6. P. Gupta, S. Goel, and A. Purwar, "A stacked technique for gender recognition through voice," in 2018 Eleventh International Conference on Contemporary Computing (IC3), 2018: IEEE, pp. 1-3.
7. F. Ertam, "An effective gender recognition approach using voice data via deeper LSTM networks," Applied Acoustics, vol. 156, pp. 351-358, 2019. Doi: 10.1016/j.apacoust.2019.07.033
8. M. Buyukyilmaz and A. O. Cibikdiken, "Voice gender recognition using deep learning," in 2016 International Conference on Modeling, Simulation and Optimization Technologies and Applications (MSOTA2016), 2016: Atlantis Press, pp. 409-411.

9. A. Suppa et al., "Voice analysis with machine learning: one step closer to an objective diagnosis of essential tremor," Movement Disorders, vol. 36, no. 6, pp. 1401-1410, 2021. Doi: 10.1002/mds.28508
10. K. Schultebraucks, V. Yadav, and I. R. Galatzer-Levy, "Utilization of machine learning-based computer vision and voice analysis to derive digital biomarkers of cognitive functioning in trauma survivors," Digital biomarkers, vol. 5, no. 1, pp. 16-23, 2021. Doi:10.1159/000512394
11. C. Robotti et al., "Machine learning-based voice assessment for the detection of positive and recovered COVID-19 patients," Journal of Voice, 2021. Doi:10.1016/j.jvoice.2021.11.004.
12. C. Shimon, G. Shafat, I. Dangoor, and A. Ben-Shitrit, "Artificial intelligence enabled preliminary diagnosis for COVID-19 from voice cues and questionnaires," The Journal of the Acoustical Society of America, vol. 149, no. 2, pp. 1120-1124, 2021. Doi: 10.1121/10.0003434
13. F. T. Al-Dhief et al., "Voice pathology detection and classification by adopting online sequential extreme learning machine," IEEE Access, vol. 9, pp. 77293-77306, 2021. Doi: 10.1109/ACCESS.2021.3082565
14. D. Shin et al., "Detection of minor and major depression through voice as a biomarker using machine learning," Journal of Clinical Medicine, vol. 10, no. 14, p. 3046, 2021. Doi: 10.3390/jcm10143046
15. C. W. Espinola, J. C. Gomes, J. M. S. Pereira, and W. P. dos Santos, "Detection of major depressive disorder using vocal acoustic analysis and machine learning—an exploratory study," Research on Biomedical Engineering, vol. 37, no. 1, pp. 53-64, 2021. Doi: 10.1007/s42600-020-00100-9
16. J. Carrón, Y. Campos-Roca, M. Madruga, and C. J. Pérez, "A mobile-assisted voice condition analysis system for Parkinson’s disease: assessment of usability conditions," BioMedical Engineering OnLine, vol. 20, no. 1, pp. 1-24, 2021. Doi: 10.1186/s12938-021-00951-y
17. L. Moro-Velazquez, J. A. Gomez-Garcia, J. D. Arias-Londoño, N. Dehak, and J. I. Godino-Llorente, "Advances in Parkinson's disease detection and assessment using voice and speech: A review of the articulatory and phonatory aspects," Biomedical Signal Processing and Control, vol. 66, p. 102418, 2021. Doi: 10.1016/j.bspc.2021.102418
18. E. Majda-Zdancewicz, A. Potulska-Chromik, J. Jakubowski, M. Nojszewska, and A. Kostera-Pruszczyk, "Deep learning vs feature engineering in the assessment of voice signals for diagnosis in Parkinson’s disease," Bulletin of the Polish Academy of Sciences. Technical Sciences, vol. 69, no. 3, 2021. Doi: 10.24425/bpasts.2021.137347
19. M. Araya‐Salas and G. Smith‐Vidaurre, "warbleR: an R package to streamline analysis of animal acoustic signals," Methods in Ecology and Evolution, vol. 8, no. 2, pp. 184-191, 2017. Doi: 10.1111/2041-210X.12624
20. K. Becker, “Gender recognition by voice”. Available: https://www.kaggle.com/datasets/primaryobjects/voicegender [Access Date: 08.04. 2022].