基于非线性动力学和高斯混合模型/支持向量机的病态嗓音识别与研究  

Recognition and Study of Pathological Voice Based on Nonlinear Dynamics Using Gaussian Mixture Model/Support Vector Machine

作　　者：高俊芬[1] 胡维平[1] 

机构地区：[1]广西师范大学电子工程学院,桂林541004

出　　处：《生物医学工程学杂志》2012年第4期750-753,759,共5页Journal of Biomedical Engineering

基　　金：广西自然科学基金资助项目(2010GXNSFA013128);广西高校重点实验室项目资助

摘　　要：传统的病态嗓音的识别研究中,通常采用线性分析技术分析嗓音的特性,将嗓音产生过程用一个经典的线性模型来近似,然而,这样却忽略了嗓音产生过程中的非线性特性。本文基于非线性动力学的分析方法,定量分析并提取了嗓音的7维非线性特征——Hurst参数、时间延迟、第二阶Rényi熵、香农熵、关联维、Kolmogorov熵(K熵)、最大Lyapunov指数。实验结果表明,非线性动力学的方法能够弥补传统分析方法的不足,较好分析正常与病态嗓音;应用高斯混合模型(GMM)和支持向量机(SVM)的模式识别方法,分别对测试集39例正常嗓音和36例病态嗓音进行识别,均得到较好的识别率,分别为97.22%和97.30%。In the traditional identification of pathological voice, linear analysis techniques are usually used to analyze the characteristics of voice, and the linear classical model is often considered to be approximate to of the real voice production process. However, this must have ignored the nonlinear characteristics of voice in the actual generation process. In the paper, based on the nonlinear dynamics analysis method, the pathological voice is analyzed quantita- tively and 7-dimensional nonlinear features, Hurst parameter, time delay, the second-order R6nyi entropy, Shannon entropy, correlation dimension, Kolmogorov entropy and the largest Lyapunov exponent are extracted. The experi- mental results showed that the method of nonlinear dynamics could compensate the deficiencies of the traditional methods, and could analyze normal and pathological voice well. Gaussian mixture model （GMM） and support vector machine （SVM） methods for pattern recognition were used to discriminate the test set including 39 cases of normal and 36 cases of pathological voice, and a better recognition rate is received, 97.22%and 97.30% for each of the mentioned normal and pathological cases, respectively.

关 键 词：非线性动力学 高斯混合模型 支持向量机 混沌理论 病态嗓音 

分 类 号：TN912.34[电子电信—通信与信息系统]