Classification of Sleep Sounds Using MFCC Features and Adaboost Ensemble Learning Method

Year 2023, Volume: IDAP-2023 : International Artificial Intelligence and Data Processing Symposium Issue: IDAP-2023, 31 - 36, 18.10.2023

Ensar Arif Sağbaş

https://doi.org/10.53070/bbd.1347221

Abstract

A regular and quality night's sleep is vital in human life. Sleep quality has a great impact on the daily lives of people and those around them. Many people today suffer from sleep disorders. Such disorders affect daily life and can impair mental health. This study proposes an approach using an ensemble learning method for the automatic classification of sleep sounds. In the study, a dataset containing 7 different sleep sounds was used. First of all, MFCC features were extracted from the audio files. Afterward, the extracted features were classified by known methods such as logistic regression, support vector machine, kNN, and random forest, which are frequently used in sound classification. In order to increase classification success, the approach of using these base classifiers together with the Adaboost ensemble learning method was proposed. An increase in classification success was observed with the proposed approach. The most successful result was obtained from the Adaboost+Random forest method with 96.439%.

Keywords

Adaboost, ensemble learning, random forest, MFCC, sleep sounds

MFCC Öznitelikleri ve Adaboost Topluluk Öğrenme Yöntemi Kullanılarak Uyku Seslerinin Sınıflandırılması

Abstract

Düzenli ve kaliteli bir gece uykusu insan hayatında hayati önem taşımaktadır. Uyku kalitesi, insanların ve çevrelerindekilerin günlük yaşamları üzerinde büyük bir etkiye sahiptir. Günümüzde birçok insan uyku bozuklukları konusunda sıkıntı çekmektedir. Bu tarz rahatsızlıklar günlük hayatı etkilemekte ve akıl sağlığını bozabilmektedir. Bu çalışma uyku seslerinin otomatik olarak sınıflandırılması için topluluk öğrenme yöntemini kullanan bir yaklaşım önermektedir. Çalışmada 7 farklı uyku sesini içeren bir veri kümesinden faydalanılmıştır. Öncelikli olarak ses dosyalarından MFCC öznitelikleri çıkartılmıştır. Sonrasında çıkartılan öznitelikler ses sınıflandırılmasında sıklıkla kullanılan lojistik regresyon, destek vektör makinesi, kNN ve rastgele orman gibi bilinen yöntemlerle sınıflandırılmıştır. Sınıflandırma başarısını artırmak amacı ile bu temel sınıflandırıcılar Adaboost topluluk öğrenme yöntemi ile birlikte kullanılması yaklaşımı önerilmiştir. Önerilen yaklaşım ile sınıflandırma başarısında artış gözlemlenmiştir. En başarılı sonuç %96.439 ile Adaboost+Rastgele orman yönteminden elde edilmiştir.

Keywords

Adaboost, topluluk öğrenme, rastgele orman, MFCC, uyku sesleri

References

• Adesuyi, T. A., Kim, B. M., & Kim, J. (2022). Snoring sound classification using 1D-CNN model based on multi-feature extraction. International Journal of Fuzzy Logic and Intelligent Systems, 22(1), 1-10.
• Akbal, E., & Tuncer, T. (2021). FusedTSNet: an automated nocturnal sleep sound classification method based on a fused textural and statistical feature generation network. Applied Acoustics, 171, 107559.
• Akyol, S., Yildirim, M., & Alatas, B. (2023). Multi-feature fusion and improved BO and IGWO metaheuristics based models for automatically diagnosing the sleep disorders from sleep sounds. Computers in Biology and Medicine, 157, 106768.
• Ayvaz, U., Gürüler, H., Khan, F., Ahmed, N., Whangbo, T., & Bobomirzaevich, A. (2022). Automatic speaker recognition using mel-frequency cepstral coefficients through machine learning. CMC-Computers Materials & Continua, 71(3).
• Ben-Israel, N., Tarasiuk, A., & Zigel, Y. (2010, August). Nocturnal sound analysis for the diagnosis of obstructive sleep apnea. In 2010 Annual International Conference of the IEEE Engineering in Medicine and Biology, (pp. 6146-6149). IEEE.
• Chattu, V. K., Manzar, M. D., Kumary, S., Burman, D., Spence, D. W., & Pandi-Perumal, S. R. (2018, December). The global problem of insufficient sleep and its serious public health implications. In Healthcare, (Vol. 7, No. 1, p. 1). MDPI.
• Chen, J., Dang, X., & Li, M. (2022, April). Heart Sound Classification Method based on Ensemble Learning. In 2022 7th International Conference on Intelligent Computing and Signal Processing, (pp. 8-13). IEEE.
• Christofferson, K., Chen, X., Wang, Z., Mariakakis, A., & Wang, Y. (2022, March). Sleep Sound Classification Using ANC-Enabled Earbuds. In 2022 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, (pp. 397-402). IEEE.
• Davis, S., & Mermelstein, P. (1980). Comparison of parametric representations for monosyllabic word recognition in continuously spoken sentences. IEEE transactions on acoustics, speech, and signal processing, 28(4), 357-366.
• Freund, Y., & Schapire, R. E. (1997). A decision-theoretic generalization of on-line learning and an application to boosting. Journal of computer and system sciences, 55(1), 119-139.
• Freund, Y., Schapire, R., & Abe, N. (1999). A short introduction to boosting. Journal-Japanese Society For Artificial Intelligence, 14(771-780), 1612.
• Hajipour, F., Jozani, M. J., & Moussavi, Z. (2020). A comparison of regularized logistic regression and random forest machine learning models for daytime diagnosis of obstructive sleep apnea. Medical & Biological Engineering & Computing, 58, 2517-2529.
• Kılıç, E., & Erdamar, A. (2020). Destek vektör makineleri kullanarak uyku seslerinin çoklu sınıflandırılması. Journal of the Institute of Science and Technology, 10(4), 2474-2485.
• Kim, J. W., Cho, S. W., Lee, K., & Shin, J. Y. (2019). Correlation analysis of sleep breathing sound and polysomnographic features. ERJ Open Research, 5 : Suppl. 3, P55
• Martinez, J., Perez, H., Escamilla, E., & Suzuki, M. M. (2012, February). Speaker recognition using Mel frequency Cepstral Coefficients (MFCC) and Vector quantization (VQ) techniques. In Conielecomp 2012, 22nd International conference on electrical communications and computers (pp. 248-251). IEEE.
• McFee, B., Raffel, C., Liang, D., Ellis, D. P., McVicar, M., Battenberg, E., & Nieto, O. (2015, July). librosa: Audio and music signal analysis in python. In Proceedings of the 14th python in science conference (Vol. 8, pp. 18-25).
• Mohammed, E. A., Keyhani, M., Sanati-Nezhad, A., Hejazi, S. H., & Far, B. H. (2021). An ensemble learning approach to digital corona virus preliminary screening from cough sounds. Scientific Reports, 11(1), 1-11.
• Otsuka, Y., Kaneita, Y., Itani, O., Nakagome, S., Jike, M., & Ohida, T. (2017). Relationship between stress coping and sleep disorders among the general Japanese population: a nationwide representative survey. Sleep medicine, 37, 38-45.
• Sillaparaya, A., Bhatranand, A., Sudthongkong, C., Chamnongthai, K., & Jiraraksopakun, Y. (2022, November). Obstructive Sleep Apnea Classification Using Snore Sounds Based on Deep Learning. In 2022 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (pp. 1152-1155). IEEE.
• Vankara, J., Lavanya Devi, G. PAELC: Predictive Analysis by Ensemble Learning and Classification heart disease detection using beat sound. Int J Speech Technol, 23, 31–43 (2020). https://doi.org/10.1007/s10772-020-09670-6
• Yıldırım, M. (2022). MFCC Yöntemi ve Önerilen Derin Model ile Çevresel Seslerin Otomatik Olarak Sınıflandırılması. Fırat Üniversitesi Mühendislik Bilimleri Dergisi, 34(1), 449-457.
• Zhao, S., Zhang, Y., Xu, H., & Han, T. (2019). Ensemble classification based on feature selection for environmental sound recognition. Mathematical Problems in Engineering, 2019.