Speech-to-Gender Recognition Based on Machine Learning Algorithms

Abstract

Speech recognition has several application areas such as human machine interaction, classification of phone calls by gender, voice tagging, STT, etc. Predicting gender from audio signals is a problem that is easy for humans to solve, difficult to solve by a computer. In this study, a model based on MFCC and classification with machine learning is proposed for gender estimation from Turkish voice signals. Within the scope of the study, 58 different series and films were examined and a new original dataset was created with 894 audio recordings consisting of 5 sec sections taken from them. Mel-frequency cepstral coefficients (MFCC) and spectrogram, which are frequently used in the literature, were used for feature extraction from audio data. The results were first evaluated separately using two features in one way. A hybrid feature vector was then created using two feature vectors. Different machine learning algorithms (LR, DT, RF, XGB etc.) were tested in the classification process and it was seen that the best accuracy was achieved in the hybrid model and logistic regression with 89%. Recall, precision and f-score values were obtained as 86.8%, 92% and 89.3%, respectively. The obtained test results revealed that the proposed model, together with the hybrid feature vector used, the original dataset and the classifier based on machine learning, showed classification success in terms of accuracy and was a stable and robust model.

Keywords

• Gender Recognition
• Machine Learning
• MFCC
• spectrogram
• logistic regression
• Turkish

References

1. R. S. Arslan and N. Barışçı, “Development of output correction methodology for long short term memory-based speech recognition,” Sustainability, , cilt 11(15), 2019.
2. R. S. Arslan and N. Barışçı, “A detailed survey of Turkish automatic speech recognition,” Turkish journal of electrical engineering and computer science, pp. 3253-3269, 2020.
3. H. Erokyar, “Age and Gender Recognition for Speech Applications based on Support Vector Machines,” Florida, 2014.
4. A. Oğuz, “Ses Sinyallerinden Yaş Grubu ve Cinsiyet Bilgisinin Tahmin Edilmesi,” Siirt, 2018.
5. S. Hızlısoy and Z. Tüfekçi, “Noise robust speech recogniton using parallel model compensation and voice activity detection methods,” 2015 5th international conference on electronics, devices, systems, and applications(ICEDSA), pp. 1-4, 2016.
6. S. Hızlısoy and R. S. Arslan, “Text independent speaker recognition based on MFCC and machine learning,” Selcuk University Journal of Engineering Sciences, no. 20(3), pp. 73-78, 2021.
7. S. Hızlısoy, S. Yıldırım and Z. Tüfekçi, “Music emotional recognition using convolutional long short term memory deep neural networks,” Engineering science and technology, an international journal, no. 24(3), pp. 760-767, 2021.
8. A. Tursunov, Mustaqeem, J. Y. Choeh and S. Kwon, “Age and Gender Recognition Using a Convolutional Neural Network with a Specially Designed Multi-Attention Module through Speech Spectrograms,” Sensors, 09 2021.

9. E. Çolakoğlu, S. Hızlısoy ve A. Recep Sinan, “Konuşmadan duygu tanıma üzerine detaylı bir inceleme: özellikler ve sınıflandırma metodları,” Avrupa bilim ve teknoloji dergisi, pp. 471-483, 2021.
10. A. Recep Sinan ve N. Barışçı, “Farklı optimizasyon tekniklerinin bağlantıcı zamansal sınıflandırma kullanılan uçtan uca Türkçe konuşma tanıma sistemlerine etkisi,” 2nd International Symposium on Multidisciplinary Studies and Innovative Technologies(ISMSIT), pp. 19-21, 10 2018.
11. A. Pahwa and G. Aggarwal, “Speech Feature Extraction for Gender Recognition,” I.J. Image, Graphics and Signal Processing,, pp. 17-25, 9 2016.
12. F. Ertam, “An effective gender recognition approach using voice data via deeper LSTM networks,” Applied Acoustics, pp. 351-358, 08 2019.
13. A. Oğuz, “Ses sinyallerinden yaş grubu ve cinsiyet bilgisinin tahmin edilmesi” Siirt Üniversitei Fen Bilimleri Enstitüsü, Siirt, 2018.
14. S. Levitan, T. Mishra and S. Bangalore, “Automatic identification of gender from speech,” Speech Prosody 2016, Boston, USA, 2016.
15. Ö. Eskidere ve F. Ertaş, “Mel Frekansı Kepstrum Katsayılarındaki Değişimlerin Konuşmacı Tanımaya Etkisi,” Uludağ Üniversitesi Mühendislik-Mimarlık Fakültesi Dergisi Cilt 14, Sayı 2, 2009.
16. R. S. Alkhawaldeh, “DGR: Gender Recognition of Human Speech Using One-Dimensional Conventional Neural Network,” Scientific Programming, pp. 1-12, 12 2019.
17. A. Alan ve M. Karabatak, “Veri Seti - Sınıflandırma İlişkisinde Performansa Etki Eden Faktörlerin Değerlendirilmesi,” Fırat Üniversitesi Müh. Bil. Dergisi , cilt 32(2), no. 531-540, pp. 531-540, 8 2020.
18. M. M. Nasef, A. M. Sauber and . M. M. Nabil, “Voice gender recognition under unconstrained environments using self-attention,” Applied Acoustics, 11 2020.
19. Y. S. Taspinar, M. M. Saritas, İ. Cinar and M. Koklu, “Gender Determination Using Voice Data,” International Journal of Applied Mathematics, Electronics and Computers, 11 2020.
20. A. Sadek, I. Shariful and H. Alamgir , “Gender Recognition System Using Speech Signal,” International Journal of Computer Science, Engineering and Information Technology, pp. Vol.2, No.1, 02 2012.
21. B. Zhong, Y. Liang, J. Wu, B. Quan, C. Li, W. Wang, J. Zhang and Z. Li, “Gender Recognition of Speech based on Decision Tree Model,” %1 içinde Proceedings of the 3rd International Conference on Computer Engineering, Information Science & Application Technology, Chongqing, China, 2019.
22. E. Yücesoy and V. V. Nabiyev, “Gender Identification Of A Speaker From Voice Source,” %1 içinde 21st Signal Processing and Communications Applications Conference, Haspolat, Turkey, 2013.
23. M. A. Uddin, R. K. Pathan, H. Sayem and M. Biswas, “Gender and region detection from human voice using the three-layer feature extraction method with 1D CNN,” Journal of Information and Telecommunication, pp. 27-42, 08 2021.
24. B. Jena, A. Mohanty and S. K. Mohanty, “Gender Recognition of Speech Signal using KNN and SVM,” %1 içinde International Conference on IoT based Control Networks and Intelligent Systems, Kottayam, Kerala,India, 2020.
25. E. Yücesoy ve V. V. Nabiyev, “Konuşmacı yaş ve cinsiyetinin GKM süpervektörlerine dayalı bir DVM sınıflandırıcısı ile belirlenmesi,” Journal of the Faculty of Engineering and Architecture of Gazi University, 09 2016.
26. J. Thangaiyan, K. Vinothkumar and A. Vijayaselvi, “Automatic Gender Identification in Speech Recognition by Genetic Algorithm,” Applied Mathematics & Information Sciences, pp. 907-913, 05 2017.
27. F. Kiani, M. A. Kutlugün ve M. Y. Çakır, “Derin Sinir Ağları ile Konuşma Tespiti ve Cinsiyet Tahmini,” %1 içinde 22. Türkiye’de Internet Konferansı, İstanbul, 2017.
28. E. Yücesoy, “Konuşmacının Yaş ve Cinsiyetine Göre Sınıflandırılmasında DVM Çekirdeğinin Etkisi,” El-Cezerî Fen ve Mühendislik Dergisi, pp. 970-982, 05 2020.
29. S. KARASARTOVA, “Metinden Bağımsız Konuşmacı Tanıma Sistemlerinin İncelenmesi ve Gerçekleştirilmesi,” Ankara, 2011.
30. Ö. Eskidere and F. Ertaş, “The Effects of Filter Frequency Scale Variability On Speaker Identification Performance,” Journal of Engineering and Natural Sciences, pp. 197-207, 09 2009.
31. İ. TÜRKER, “Ses Si̇nyallerinin Graf Tabanlı Temsillerinin Yapay Zekâ Yöntemleri İle Sınıflandırılması,” Karabük, 2022.
32. “wikipedia,” [Online]. Available: https://en.wikipedia.org/wiki/Confusion_matrix.
33. G. Öğündür, “Model Seçimi-K Fold Cross Validation,” 13 01 2020. [Online]. Available: https://medium.com/@gulcanogundur/model-se%C3%A7imi-k-fold-cross-validation-4635b61f143c. [Access: 12 2022].
34. Ö. Eskidere and F. Ertaş, “The effects of filter frequency scale variability on speaker identification performance” Journal of Engineering and Natural Sciences Mühendislik ve Fen Bilimleri Dergisi, 9 2009.
35. S. Aksu, “Ses sinyallerinin graf tabanlı temsillerinin yapay zeka yöntemleri ile sınıflandırılması “ Karabük Üniversitesi, Karabük, 2022.