기계학습에 의한 후두 장애음성 식별기의 성능 비교
Performance comparison on vocal cords disordered voice discrimination via machine learning methods
0
조회
0
다운로드
2
인용
0
참고문헌
키워드
초록
This paper studies how to improve the identification rate of laryngeal disability speech data by convolutional neural network (CNN) and machine learning ensemble learning methods. In general, the number of laryngeal dysfunction speech data is small, so even if identifiers are constructed by statistical methods, the phenomenon caused by overfitting depending on the training method can lead to a decrease the identification rate when exposed to external data. In this work, we try to combine results derived from CNN models and machine learning models with various accuracy in a multi-voting manner to ensure improved classification efficiency compared to the original trained models. The Pusan National University Hospital (PNUH) dataset was used to train and validate algorithms. The dataset contains normal voice and voice data of benign and malignant tumors. In the experiment, an attempt was made to distinguish between normal and benign tumors and malignant tumors. As a result of the experiment, the random forest method was found to be the best ensemble method and showed an identification rate of 85%.
Abstract
This paper studies how to improve the identification rate of laryngeal disability speech data by convolutional neural network (CNN) and machine learning ensemble learning methods. In general, the number of laryngeal dysfunction speech data is small, so even if identifiers are constructed by statistical methods, the phenomenon caused by overfitting depending on the training method can lead to a decrease the identification rate when exposed to external data. In this work, we try to combine results derived from CNN models and machine learning models with various accuracy in a multi-voting manner to ensure improved classification efficiency compared to the original trained models. The Pusan National University Hospital (PNUH) dataset was used to train and validate algorithms. The dataset contains normal voice and voice data of benign and malignant tumors. In the experiment, an attempt was made to distinguish between normal and benign tumors and malignant tumors. As a result of the experiment, the random forest method was found to be the best ensemble method and showed an identification rate of 85%.
참고문헌 (25)
- Aicha, A. B., "Noninvasive detection of potentially precancerous lesions of vocal fold based on glottal wave signal and SVM approaches", Procedia Computer Science, 126, pp.586~595 (2018)
- Al-Nasheri, A., "An investigation of multidimensional voice program parameters in three different databases for voice pathology detection and classification", Journal of Voice, 31(1), pp.113.e9~113.e18 (2017)
- Bezdek, J. C., "Fuzzy models and algorithms for pattern recognition and image processing", pp.442~490 (2005), Springer
- Fang, S. H., "Detection of pathological voice using cepstrum vectors : A deep learning approach", Journal of Voice, 33(5), pp.634~641 (2019)
- Hegde, S., "A survey on machine learning approaches for automatic detection of voice disorders", Journal of Voice, 33(6), pp.947.e11~947.e33 (2019)
- 전병욱, "도로교통 이머징 리스크 탐지를 위한 AutoML과 CNN 기반 소프트 보팅 앙상블 분류 모델", 융합정보논문지, 11(7), pp.14~20 (2021), 중소기업융합학회[DOI]
- Jo, C., "Screening of pathological voice from ARS using neural networks", Proceedings of the Models and Analysis of Vocal Emissions for Biomedical Applications (MAVEBA) 2nd International Workshop, pp.241~245 (2001)
- 정희철, "대용량 이미지넷 인식을 위한 CNN 기반 Weighted 앙상블 기법", 대한임베디드공학회논문지, 15(4), pp.197~204 (2020), 대한임베디드공학회[DOI]
- Kim, H. B., "Convolutional neural network classifies pathological voice change in laryngeal cancer with high accuracy", Journal of Clinical Medicine, 9(11), pp.3415~ (2020)
- Ko, H., "Pneumonia detection with weighted voting ensemble of CNN models", Proceedings of the 2019 2nd International Conference on Artificial Intelligence and Big Data (ICAIBD), pp.306~310 (2019)
- Lee, J. Y., "Experimental evaluation of deep learning methods for an intelligent pathological voice detection system using the Saarbruecken voice database", Applied Sciences, 11(15), pp.7149~ (2021)
- "Librosa: Audio and music processing in Python" (2021)[URL]
- Liu, F., "Multi-classifier decision-level fusion classification of workpiece surface defects based on a convolutional neural network", Symmetry, 12(5), pp.867~ (2020)
- Lv, X., "A new method for region-based majority voting CNNs for very high resolution image classification", Remote Sensing, 10(12), pp.1946~ (2018)
- Massachusetts Eye and Ear Infirmary, "Voice disorders database, version. 1. 03(CD-ROM)", Kay Elemetrics, ~ (1994)
- Morvant, E., "Majority vote of diverse classifiers for late fusion", Proceedings of the Joint IAPR International Workshops on Statistical Techniques in Pattern Recognition (SPR) and Structural and Syntactic Pattern Recognition (SSPR), pp.20~ (2014)
- Roy, S., "Pathological voice classification using deep learning", Proceedings of the 20191st International Conference on Advances in Science, Engineering and Robotics Technology (ICASERT) (2019)
- Ruta, D., "An overview of classifier fusion methods", Computing and Information Systems, 7(1), pp.1~10 (2000)
- "Saarbruecken Voice Database" (2020)[URL]
- Saldanha, J. C., "Vocal fold pathology assessment using mel-frequency cepstral coefficients and linear predictive cepstral coefficients features", Journal of Medical Imaging and Health Informatics, 4(2), pp.168~173 (2014)
- "Ensemble methods" (2022)[URL]
- Su, Y., "Environment sound classification using a two-stream CNN based on decision"level fusion", Sensors, 19(7), pp.1733~ (2019)
- Szmurlo, R., "Deep CNN ensemble for recognition of face images", Proceedings of the 2021 22nd International Conference on Computational Problems of Electrical Engineering (CPEE), pp.1~4 (2021)
- "Tensorflow" (2021)[URL]
- Wu, H., "Convolutional neural networks for pathological voice detection", Proceedings of the 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pp.1~4 (2018)