Psychosis speech recognition algorithm based on deep embedded sparse stacked autoencoder and manifold ensemble_Journal of Biomedical Engineering

Authors：

ZHANG Yi ^1,2,3 , QIN Xiaolin ^1,2 , LIN Yuan ⁴ ,  LI Yongming ^4,5 , WANG Pin ⁴ , ZHANG Zuwei ³ , LI Xiaofei ³

1. Chengdu Institute of Computer Application, Chinese Academy of Sciences, Chengdu 610041, P.R.China;
2. University of Chinese Academy of Sciences, Beijing 100049, P.R.China;
3. Chongqing Acoustic-Optic-Electronic Co. Ltd, China Electronics Technology Group, Chongqing 401332, P.R.China;
4. School of Microelectornics and Communication Engineering, Chongqing University, Chongqing 400044, P.R.China;
5. Chongqing mental health center, Chongqing 400020, P.R.China;

Corresponding author：

LI Yongming, Email: yongmingli@cqu.edu.cn

Keywords：

psychosis speech recognition; embedded hybrid features sparse stacked autoencoder; L1 regularization; feature fusion; manifold ensemble learning

DOI：

10.7507/1001-5515.202010050

Video：

Export PDF Favorites Scan Get Citation

Abstract Full text Figures/Tables Video References Cited by

Speech feature learning is the core and key of speech recognition method for mental illness. Deep feature learning can automatically extract speech features, but it is limited by the problem of small samples. Traditional feature extraction (original features) can avoid the impact of small samples, but it relies heavily on experience and is poorly adaptive. To solve this problem, this paper proposes a deep embedded hybrid feature sparse stack autoencoder manifold ensemble algorithm. Firstly, based on the prior knowledge, the psychotic speech features are extracted, and the original features are constructed. Secondly, the original features are embedded in the sparse stack autoencoder (deep network), and the output of the hidden layer is filtered to enhance the complementarity between the deep features and the original features. Third, the L1 regularization feature selection mechanism is designed to compress the dimensions of the mixed feature set composed of deep features and original features. Finally, a weighted local preserving projection algorithm and an ensemble learning mechanism are designed, and a manifold projection classifier ensemble model is constructed, which further improves the classification stability of feature fusion under small samples. In addition, this paper designs a medium-to-large-scale psychotic speech collection program for the first time, collects and constructs a large-scale Chinese psychotic speech database for the verification of psychotic speech recognition algorithms. The experimental results show that the main innovation of the algorithm is effective, and the classification accuracy is better than other representative algorithms, and the maximum improvement is 3.3%. In conclusion, this paper proposes a new method of psychotic speech recognition based on embedded mixed sparse stack autoencoder and manifold ensemble, which effectively improves the recognition rate of psychotic speech.

Citation： ZHANG Yi, QIN Xiaolin, LIN Yuan, LI Yongming, WANG Pin, ZHANG Zuwei, LI Xiaofei. Psychosis speech recognition algorithm based on deep embedded sparse stacked autoencoder and manifold ensemble. Journal of Biomedical Engineering, 2021, 38(4): 655-662. doi: 10.7507/1001-5515.202010050 Copy

1.	赖虹宇, 冯静雯, 王毅, 等. 抑郁症和精神分裂症患者静息态脑电信号的分类研究. 生物医学工程学杂志, 2019, 36(6): 916-923.
2.	Wolkin A, Sanfilipo M, Wolf A P, et al. Negative symptoms and hypofrontality in chronic schizophrenia. Arch Gen Psychiatry, 1992, 49(12): 959-965.
3.	Tsanas A, Little M A, Mcsharry P E, et al. Novel speech signal processing algorithms for high-accuracy classification of Parkinson's disease. IEEE Trans Biomed Eng, 2012, 59(5): 1264-1271.
4.	Liu Zhenyu, Kang Huanyu, Feng Lei, et al. Speech pause time: a potential biomarker for depression detection//2017 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). Kansas: IEEE, 2017: 2020-2025.
5.	Stasak B, Epps J, Lawson A. Analysis of phonetic markedness and gestural effort measures for acoustic speech-based depression classification//2017 Seventh International Conference on Affective Computing and Intelligent Interaction Workshops and Demos (ACIIW), San Antonio: IEEE Computer Society, 2017: 165-170.
6.	田德财, 张新卿. 交互式语音应答系统在抑郁障碍筛查中的应用. 中国神经精神疾病杂志, 2010, 36(11): 695-697.
7.	张静, 潘忠德, 桂超, 等. 精神分裂症患者语音信号的临床分析. 上海精神医学, 2016, 28(2): 95-102.
8.	潘玮, 汪静莹, 刘天俐, 等. 基于语音的抑郁症识别. 科学通报, 2018, 63(20): 2081-2092.
9.	Jiang Haihua, Hu Bin, Liu Zhenyu, et al. Investigation of different speech types and emotions for detecting depression using different classifiers. Speech Communication, 2017, 90: 39-46.
10.	田熙燕, 徐君鹏, 杜留锋. 基于语谱图和卷积神经网络的语音情感识别. 河南科技学院学报: 自然科学版, 2017, 45(2): 62-68.
11.	He Lang, Cao Cui. Automated depression analysis using convolutional neural networks from speech. J Biomed Inform, 2018, 83: 103-111.
12.	Zheng L, Wang S, Tian L, et al. Query-adaptive late fusion for image search and person re-identification//2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston: IEEE, 2015: 1741–1750.
13.	Zhang X, Zhang H, Zhang Y, et al. Deep fusion of multiple semantic cues for complex event recognition. IEEE Trans Image Process, 2016, 25(3): 1033-1046.
14.	Liu Ziqiong, Wang Shengjin, Zheng Liang, et al. Robust ImageGraph: rank-level feature fusion for image search. IEEE Transactions on Image Processing, 2017, 26(7): 3128-3141.
15.	Qi Yu, Wang Yueming, Zheng Xiaoxiang, et al. Robust feature learning by stacked autoencoder with maximum correntropy criterion//2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Florence: IEEE, 2014: 6716-6720.
16.	Yuan Xiaofeng, Huang Biao, Wang Yalin, et al. Deep learning-based feature representation and its application for soft sensor modeling with variable-wise weighted SAE. IEEE Trans Industr Inform, 2018, 14(7): 3235-3243.
17.	Lv Fei, Han Min, Qiu Tie. Remote sensing image classification based on ensemble extreme learning machine with stacked autoencoder. IEEE Access, 2017, 5: 9021-9031.
18.	Tao Chao, Pan Hongbo, Li Yansheng, et al. Unsupervised spectral–spatial feature learning with stacked sparse autoencoder for hyperspectral imagery classification. IEEE Geoscience and Remote Sensing Letters, 2015, 12(12): 2438-2442.
19.	Jiang Guoqian, He Haibo, Xie Ping, et al. Stacked multilevel-denoising autoencoders: a new representation learning approach for wind turbine gearbox fault diagnosis. IEEE Trans Instrum Meas, 2017, 66(9): 2391-2402.
20.	Dai Jie, Hui Song, Sheng Gehao, et al. Cleaning method for status monitoring data of power equipment based on stacked denoising autoencoders. IEEE Access, 2017, 5: 22863-22870.
21.	Lei You, Yuan Wang, Wang Hongpeng, et al. A skin segmentation algorithm based on stacked autoencoders. IEEE Trans Multimedia, 2017, 19(4): 740-749.
22.	Qi Yumei, Shen Changqing, Wang Dong, et al. Stacked sparse autoencoder-based deep network for fault diagnosis of rotating machinery. IEEE Access, 2017, 5(99): 15066-15079.
23.	Sheehan D V, Lecrubier Y, Sheehan K H, et al. The Mini-international neuropsychiatric interview (M. I. N. I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry, 1998, 59(20): 22-33.
24.	Dsm A A, Association A, Association A P, et al. Diagnostic and statistical manual of mental disorders. Psychiatry Res, 2000, 25(2): 1-4.
25.	de Jong N H, Wempe T. Praat script to detect syllable nuclei and measure speech rate automatically. Behav Res Methods, 2009, 41(2): 385-390.
26.	Sakar B E, Isenkul M E, Sakar C O, et al. Collection and analysis of a Parkinson speech dataset with multiple types of sound recordings. IEEE J Biomed Health Inform, 2013, 17(4): 828-834.
27.	Hinton G E, Osindero S, Teh Y W. A fast learning algorithm for deep belief nets. Neural Comput, 2006, 18(7): 1527-1554.
28.	Liu Yuchuan, Tan Xiaoheng, Li Yongming, et al. Weighted local discriminant preservation projection ensemble algorithm with embedded micro-noise. IEEE Access, 2019, 7: 143814-143828.
29.	Lin Jianhai, Lu KAI, Sun Yuxuan. a novel relief feature selection algorithm based on mean-variance model. System Simulation Technology, 2013, 8(16): 3921-3929.
30.	Donoho D, Jin J. Higher criticism thresholding: optimal feature selection when useful features are rare and weak. Proceedings of the national academy of sciences, 2008, 105(39): 14790-14795.
31.	Jolliffe I T. Principal component analysis. Journal of Marketing Research, 2002, 87(4): 513.
32.	Li C H, Kuo B C, Lin C T. LDA-Based clustering algorithm and its application to an unsupervised feature extraction. IEEE Transactions on Fuzzy Systems, 2011, 19(1): 152-163.
33.	Mulfari D, Meoni G, Marini M, et al. Machine learning assistive application for users with speech disorders. Applied Soft Computing, 2021, 103: 107147.

1. 赖虹宇, 冯静雯, 王毅, 等. 抑郁症和精神分裂症患者静息态脑电信号的分类研究. 生物医学工程学杂志, 2019, 36(6): 916-923.
2. Wolkin A, Sanfilipo M, Wolf A P, et al. Negative symptoms and hypofrontality in chronic schizophrenia. Arch Gen Psychiatry, 1992, 49(12): 959-965.
3. Tsanas A, Little M A, Mcsharry P E, et al. Novel speech signal processing algorithms for high-accuracy classification of Parkinson's disease. IEEE Trans Biomed Eng, 2012, 59(5): 1264-1271.
4. Liu Zhenyu, Kang Huanyu, Feng Lei, et al. Speech pause time: a potential biomarker for depression detection//2017 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). Kansas: IEEE, 2017: 2020-2025.
5. Stasak B, Epps J, Lawson A. Analysis of phonetic markedness and gestural effort measures for acoustic speech-based depression classification//2017 Seventh International Conference on Affective Computing and Intelligent Interaction Workshops and Demos (ACIIW), San Antonio: IEEE Computer Society, 2017: 165-170.
6. 田德财, 张新卿. 交互式语音应答系统在抑郁障碍筛查中的应用. 中国神经精神疾病杂志, 2010, 36(11): 695-697.
7. 张静, 潘忠德, 桂超, 等. 精神分裂症患者语音信号的临床分析. 上海精神医学, 2016, 28(2): 95-102.
8. 潘玮, 汪静莹, 刘天俐, 等. 基于语音的抑郁症识别. 科学通报, 2018, 63(20): 2081-2092.
9. Jiang Haihua, Hu Bin, Liu Zhenyu, et al. Investigation of different speech types and emotions for detecting depression using different classifiers. Speech Communication, 2017, 90: 39-46.
10. 田熙燕, 徐君鹏, 杜留锋. 基于语谱图和卷积神经网络的语音情感识别. 河南科技学院学报: 自然科学版, 2017, 45(2): 62-68.
11. He Lang, Cao Cui. Automated depression analysis using convolutional neural networks from speech. J Biomed Inform, 2018, 83: 103-111.
12. Zheng L, Wang S, Tian L, et al. Query-adaptive late fusion for image search and person re-identification//2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston: IEEE, 2015: 1741–1750.
13. Zhang X, Zhang H, Zhang Y, et al. Deep fusion of multiple semantic cues for complex event recognition. IEEE Trans Image Process, 2016, 25(3): 1033-1046.
14. Liu Ziqiong, Wang Shengjin, Zheng Liang, et al. Robust ImageGraph: rank-level feature fusion for image search. IEEE Transactions on Image Processing, 2017, 26(7): 3128-3141.
15. Qi Yu, Wang Yueming, Zheng Xiaoxiang, et al. Robust feature learning by stacked autoencoder with maximum correntropy criterion//2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Florence: IEEE, 2014: 6716-6720.
16. Yuan Xiaofeng, Huang Biao, Wang Yalin, et al. Deep learning-based feature representation and its application for soft sensor modeling with variable-wise weighted SAE. IEEE Trans Industr Inform, 2018, 14(7): 3235-3243.
17. Lv Fei, Han Min, Qiu Tie. Remote sensing image classification based on ensemble extreme learning machine with stacked autoencoder. IEEE Access, 2017, 5: 9021-9031.
18. Tao Chao, Pan Hongbo, Li Yansheng, et al. Unsupervised spectral–spatial feature learning with stacked sparse autoencoder for hyperspectral imagery classification. IEEE Geoscience and Remote Sensing Letters, 2015, 12(12): 2438-2442.
19. Jiang Guoqian, He Haibo, Xie Ping, et al. Stacked multilevel-denoising autoencoders: a new representation learning approach for wind turbine gearbox fault diagnosis. IEEE Trans Instrum Meas, 2017, 66(9): 2391-2402.
20. Dai Jie, Hui Song, Sheng Gehao, et al. Cleaning method for status monitoring data of power equipment based on stacked denoising autoencoders. IEEE Access, 2017, 5: 22863-22870.
21. Lei You, Yuan Wang, Wang Hongpeng, et al. A skin segmentation algorithm based on stacked autoencoders. IEEE Trans Multimedia, 2017, 19(4): 740-749.
22. Qi Yumei, Shen Changqing, Wang Dong, et al. Stacked sparse autoencoder-based deep network for fault diagnosis of rotating machinery. IEEE Access, 2017, 5(99): 15066-15079.
23. Sheehan D V, Lecrubier Y, Sheehan K H, et al. The Mini-international neuropsychiatric interview (M. I. N. I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry, 1998, 59(20): 22-33.
24. Dsm A A, Association A, Association A P, et al. Diagnostic and statistical manual of mental disorders. Psychiatry Res, 2000, 25(2): 1-4.
25. de Jong N H, Wempe T. Praat script to detect syllable nuclei and measure speech rate automatically. Behav Res Methods, 2009, 41(2): 385-390.
26. Sakar B E, Isenkul M E, Sakar C O, et al. Collection and analysis of a Parkinson speech dataset with multiple types of sound recordings. IEEE J Biomed Health Inform, 2013, 17(4): 828-834.
27. Hinton G E, Osindero S, Teh Y W. A fast learning algorithm for deep belief nets. Neural Comput, 2006, 18(7): 1527-1554.
28. Liu Yuchuan, Tan Xiaoheng, Li Yongming, et al. Weighted local discriminant preservation projection ensemble algorithm with embedded micro-noise. IEEE Access, 2019, 7: 143814-143828.
29. Lin Jianhai, Lu KAI, Sun Yuxuan. a novel relief feature selection algorithm based on mean-variance model. System Simulation Technology, 2013, 8(16): 3921-3929.
30. Donoho D, Jin J. Higher criticism thresholding: optimal feature selection when useful features are rare and weak. Proceedings of the national academy of sciences, 2008, 105(39): 14790-14795.
31. Jolliffe I T. Principal component analysis. Journal of Marketing Research, 2002, 87(4): 513.
32. Li C H, Kuo B C, Lin C T. LDA-Based clustering algorithm and its application to an unsupervised feature extraction. IEEE Transactions on Fuzzy Systems, 2011, 19(1): 152-163.
33. Mulfari D, Meoni G, Marini M, et al. Machine learning assistive application for users with speech disorders. Applied Soft Computing, 2021, 103: 107147.

Journal of Biomedical Engineering

Psychosis speech recognition algorithm based on deep embedded sparse stacked autoencoder and manifold ensemble

Abstract Full text Figures/Tables Video References Cited by

Previous Article

Next Article

Format

Content