The classification of pathological voice from healthy voice was studied based upon 27 acoustic features derived from a single sound signal of vowel /a:/. First, the feature space was transferred to reduce the data dim...The classification of pathological voice from healthy voice was studied based upon 27 acoustic features derived from a single sound signal of vowel /a:/. First, the feature space was transferred to reduce the data dimension by principle component analysis (PCA). Then the voice samples were classified according to the reduced PCA parameters by support vector machine (SVM) using radial basis function (RBF) as a kernel function. Meanwhile, by changing the ratio of opposite class samples, the accuracy under different features combinations was tested. Experimental data were provided by the voice database of Massachusetts Eye and Ear Infirmary (MEEI) in which 216 vowel /a:/ samples were collected from subjects of healthy and pathological cases, and tested with 5 fold cross-validation method. The result shows the positive rate of pathological voices was improved from 92% to 98% through the PCA method. STD, Fatr, Tasm, NHR, SEG, and PER are pathology sensitive features in illness detection. Using these sensitive features the accuracy of detection of pathological voice from healthy voice can reach 97%.展开更多
<strong></strong><strong>Objective(s):</strong> The aim of this study is to explore if there is a correlation between the typical voice classification and the oropharyngeal and laryngeal morpho...<strong></strong><strong>Objective(s):</strong> The aim of this study is to explore if there is a correlation between the typical voice classification and the oropharyngeal and laryngeal morphology, using video laryngeal stroboscopy and cervical posterior-anterior radiography on professional singers in Greece. <strong>Methods:</strong> 55 professional singers (28 females: 7 sopranos, 12 mezzo-sopranos, and 9 contraltos;27 males: 8 tenors, 12 baritones and 7 basses) were recruited for this study. All participants underwent stroboscopic and cervical posterior-anterior radiographic imaging of their oral pharyngeal and laryngeal area. Additionally, the voice classification and features (e.g., height, weight) of individuals were correlated statistically. <strong>Results:</strong> Statistically significant correlations were observed between the VC of the participants with the Phonetic Area (PA) (r = −0.451, p = 0.001) and the VC with the Oral-pharyngeal Cavity (OPC) area (r = −0.402, p = 0.001) in the total sample. Specifically, in male singers, the PA and VC correlation was r = −0.319, p = 0.047, and the VC and OPC area was r = −0.328, p = 0.044. Likewise, in female singers, the PA area and VC and PA were r = −0.336, p = 0.041 and the OPC area and VC were r = −0.344, p = 0.039. The analysis confirmed no correlations between VC and height and body weight. <strong>Conclusions:</strong> The cervical posteroanterior radiography in conjunction with laryngeal stroboscopy provided new morphometric correlations of the VC of professional singers with their Oropharyngeal and Laryngeal Anatomy.展开更多
Unvoiced/voiced classification of speech is a challenging problem especially under conditions of low signal-to-noise ratio or the non-white-stationary noise environment. To solve this problem, an algorithm for speech ...Unvoiced/voiced classification of speech is a challenging problem especially under conditions of low signal-to-noise ratio or the non-white-stationary noise environment. To solve this problem, an algorithm for speech classification, and a technique for the estimation of palrwise magnitude frequency in voiced speech am proposed. By using third order spectrum of speech signal to remove noise, in this algorithm the least spectrum difference to get refined pitch and the max harmonic number is given. And this algorithm utilizes spectral envelope to estimate signal-to-noise ratio of speech harmonics. Speech classification, voicing probability, and harmonic parameters of the voiced frame can be obtained. Simulation results indicate that the proposed algorithm, under complicated background noise, especially Gaussian noise, can effectively classify speech in high accuracy for voicing probability and the voiced parameters.展开更多
文摘The classification of pathological voice from healthy voice was studied based upon 27 acoustic features derived from a single sound signal of vowel /a:/. First, the feature space was transferred to reduce the data dimension by principle component analysis (PCA). Then the voice samples were classified according to the reduced PCA parameters by support vector machine (SVM) using radial basis function (RBF) as a kernel function. Meanwhile, by changing the ratio of opposite class samples, the accuracy under different features combinations was tested. Experimental data were provided by the voice database of Massachusetts Eye and Ear Infirmary (MEEI) in which 216 vowel /a:/ samples were collected from subjects of healthy and pathological cases, and tested with 5 fold cross-validation method. The result shows the positive rate of pathological voices was improved from 92% to 98% through the PCA method. STD, Fatr, Tasm, NHR, SEG, and PER are pathology sensitive features in illness detection. Using these sensitive features the accuracy of detection of pathological voice from healthy voice can reach 97%.
文摘<strong></strong><strong>Objective(s):</strong> The aim of this study is to explore if there is a correlation between the typical voice classification and the oropharyngeal and laryngeal morphology, using video laryngeal stroboscopy and cervical posterior-anterior radiography on professional singers in Greece. <strong>Methods:</strong> 55 professional singers (28 females: 7 sopranos, 12 mezzo-sopranos, and 9 contraltos;27 males: 8 tenors, 12 baritones and 7 basses) were recruited for this study. All participants underwent stroboscopic and cervical posterior-anterior radiographic imaging of their oral pharyngeal and laryngeal area. Additionally, the voice classification and features (e.g., height, weight) of individuals were correlated statistically. <strong>Results:</strong> Statistically significant correlations were observed between the VC of the participants with the Phonetic Area (PA) (r = −0.451, p = 0.001) and the VC with the Oral-pharyngeal Cavity (OPC) area (r = −0.402, p = 0.001) in the total sample. Specifically, in male singers, the PA and VC correlation was r = −0.319, p = 0.047, and the VC and OPC area was r = −0.328, p = 0.044. Likewise, in female singers, the PA area and VC and PA were r = −0.336, p = 0.041 and the OPC area and VC were r = −0.344, p = 0.039. The analysis confirmed no correlations between VC and height and body weight. <strong>Conclusions:</strong> The cervical posteroanterior radiography in conjunction with laryngeal stroboscopy provided new morphometric correlations of the VC of professional singers with their Oropharyngeal and Laryngeal Anatomy.
文摘Unvoiced/voiced classification of speech is a challenging problem especially under conditions of low signal-to-noise ratio or the non-white-stationary noise environment. To solve this problem, an algorithm for speech classification, and a technique for the estimation of palrwise magnitude frequency in voiced speech am proposed. By using third order spectrum of speech signal to remove noise, in this algorithm the least spectrum difference to get refined pitch and the max harmonic number is given. And this algorithm utilizes spectral envelope to estimate signal-to-noise ratio of speech harmonics. Speech classification, voicing probability, and harmonic parameters of the voiced frame can be obtained. Simulation results indicate that the proposed algorithm, under complicated background noise, especially Gaussian noise, can effectively classify speech in high accuracy for voicing probability and the voiced parameters.
