病态嗓音的识别与研究  

study and recognition of pathological voice

作　　者：陈承义[1] 高俊芬[2] 

机构地区：[1]柳州铁道职业技术学院,广西柳州545007 [2]广西师范大学,广西桂林541004

出　　处：《计算机工程与应用》2013年第7期123-125,共3页Computer Engineering and Applications

基　　金：广西自然科学基金(No.2010GXNSFA013128)

摘　　要：通过分析嗓音的发音机理,提取正常与病态嗓音的传统声学参数:基频、共振峰、Mel倒谱系数(MFCC),以及非线性特征参数:计盒维数与截距,作为病态嗓音识别的特征矢量集。应用高斯混合模型(GMM)对156例正常嗓音与146例病态嗓音进行建模与识别。结果表明:非线性特征参数计盒维数与截距能很好地区分正常与病态嗓音,它们与传统声学参数基频和共振峰的组合,能够取得92.60%的识别率。By analyzing the mechanism of pronunciation, normal and pathological voice of traditional acoustic parameters, fun-damental frequency, formant, Mel Frequency Cepstrum Coefficient（MFCC）, and non-linear feature parameters：box-counting dimension and intercept, are extracted as feature vectors of recognition of pathological voice. 156 normal voice samples and 146 pathological voice samples are recognized based on Gaussian Mixture Model （GMM）. The results show that the nonlinear fea- ture parameters of box-counting dimension and intercept can well distinguish between normal and pathological voice. The com- bination of box-counting dimension, intercept and the traditional acoustic parameters-fundamental frequency and forrnant can achieve a better recognition rate of 92.60%.

关 键 词：高斯混合模型 病态嗓音 计盒维数 截距 

分 类 号：TP39[自动化与计算机技术—计算机应用技术]