Papers by Redouane Benhammoud
The main concern of this paper is the clinical assessment of disordered voices using an automatic... more The main concern of this paper is the clinical assessment of disordered voices using an automatic classification method based on a hybrid hidden Marcov model (HMM) and support vector machine (SVM). We investigate the effectiveness of the Mel-frequency cepstral coefficients (MFCC), log-frequency power coefficients (LFPC) and linear predictive cepstral coefficients (LPCC) as acoustic features for the classifier. The efficiency of the hybrid HMM-SVM model is tested on a concatenation of two Dutch sentences spoken by 28 normophonic speakers and 223 pathological speakers with several levels of dysphonia. The performance of the hybrid HMM-SVM classifier in terms of classification accuracy is compared with that of the conventional HMM and SVM classifiers when used separately. The highest two and three categories classification accuracies obtained by the hybrid HMM-SVM classifier are 97.35 and 92.01%, respectively, while the highest classification accuracies obtained by the two and three class classification with the HMMs are respectively 89.16 and 80.69%. The highest two and three class classification accuracies obtained by the SVM classifier with a radial basis function (RBF) kernel are respectively 77.36 and 75.15%, while the results obtained by the SVM classifier with a linear kernel are respectively 75.22 and 71.33% for the classification into two and three categories.
Automatic Classification of Disordered Voices Based on a Hybrid HMM-SVM Model
Journal of Communications Technology and Electronics, 2021
Automatic Classification of Disordered Voices Based on a Hybrid HMM-SVM Model, 2022
Abstract—The main concern of this paper is the clinical assessment of disordered voices using an ... more Abstract—The main concern of this paper is the clinical assessment of disordered voices using an automatic classification method based on a hybrid hidden Marcov model (HMM) and support vector machine (SVM).
We investigate the effectiveness of the Mel-frequency cepstral coefficients (MFCC), log-frequency power coefficients (LFPC) and linear predictive cepstral coefficients (LPCC) as acoustic features for the classifier.
The efficiency of the hybrid HMM-SVM model is tested on a concatenation of two Dutch sentences spoken by 28 normophonic speakers and 223 pathological speakers with several levels of dysphonia. The performance
of the hybrid HMM-SVM classifier in terms of classification accuracy is compared with that of the conventional HMM and SVM classifiers when used separately. The highest two and three categories classification
accuracies obtained by the hybrid HMM-SVM classifier are 97.35 and 92.01%, respectively, while the highest classification accuracies obtained by the two and three class classification with the HMMs are respectively 89.16 and 80.69%. The highest two and three class classification accuracies obtained by the SVM classifier with a radial basis function (RBF) kernel are respectively 77.36 and 75.15%, while the results obtained by the SVM classifier with a linear kernel are respectively 75.22 and 71.33% for the classification into two and three categories.
Automatic Classification of Disordered Voices with Hidden Markov Models
2018 International Conference on Signal, Image, Vision and their Applications (SIVA), 2018
The auditory-based method is commonly used in the assessment of voice disorders. This method is s... more The auditory-based method is commonly used in the assessment of voice disorders. This method is subjective in the sense that the evaluation result depends on the listener and a great deal of expertise is required to obtain reproducible evaluations. Automatic assessment of disordered voices provides a support for clinicians to establish a better assessment of the disorder. The main concern of this article is the clinical evaluation of disordered voices by using an automatic classification method based on Hidden Markov Models (HMMs). The HMM classification technique is based on a learning theory that uses Baum-Welch algorithm, a special case of the EM algorithm. The descriptors that are used as inputs to the HMM classifier are the cepstral MFCC coefficients. The performances of the classification method that is supervised, in terms of classification accuracy are investigated using a database consisting of a concatenation of two Dutch-speaking sentences produced by 28 normophonic and 223 dysphonic speakers.
2018 International Conference on Signal, Image, Vision and their Applications (SIVA), Guelma, Algeria, IEEE, 2018
The auditory-based method is commonly used in the
assessment of voice disorders. This method is s... more The auditory-based method is commonly used in the
assessment of voice disorders. This method is subjective in the
sense that the evaluation result depends on the listener and a
great deal of expertise is required to obtain reproducible
evaluations. Automatic assessment of disordered voices provides
a support for clinicians to establish a better assessment of the
disorder. The main concern of this article is the clinical
evaluation of disordered voices by using an automatic
classification method based on Hidden Markov Models (HMMs).
The HMM classification technique is based on a learning theory
that uses Baum-Welch algorithm, a special case of the EM
algorithm. The descriptors that are used as inputs to the HMM
classifier are the cepstral MFCC coefficients. The performances
of the classification method that is supervised, in terms of
classification accuracy are investigated using a database
consisting of a concatenation of two Dutch-speaking sentences
produced by 28 normophonic and 223 dysphonic speakers
Uploads
Papers by Redouane Benhammoud
We investigate the effectiveness of the Mel-frequency cepstral coefficients (MFCC), log-frequency power coefficients (LFPC) and linear predictive cepstral coefficients (LPCC) as acoustic features for the classifier.
The efficiency of the hybrid HMM-SVM model is tested on a concatenation of two Dutch sentences spoken by 28 normophonic speakers and 223 pathological speakers with several levels of dysphonia. The performance
of the hybrid HMM-SVM classifier in terms of classification accuracy is compared with that of the conventional HMM and SVM classifiers when used separately. The highest two and three categories classification
accuracies obtained by the hybrid HMM-SVM classifier are 97.35 and 92.01%, respectively, while the highest classification accuracies obtained by the two and three class classification with the HMMs are respectively 89.16 and 80.69%. The highest two and three class classification accuracies obtained by the SVM classifier with a radial basis function (RBF) kernel are respectively 77.36 and 75.15%, while the results obtained by the SVM classifier with a linear kernel are respectively 75.22 and 71.33% for the classification into two and three categories.
assessment of voice disorders. This method is subjective in the
sense that the evaluation result depends on the listener and a
great deal of expertise is required to obtain reproducible
evaluations. Automatic assessment of disordered voices provides
a support for clinicians to establish a better assessment of the
disorder. The main concern of this article is the clinical
evaluation of disordered voices by using an automatic
classification method based on Hidden Markov Models (HMMs).
The HMM classification technique is based on a learning theory
that uses Baum-Welch algorithm, a special case of the EM
algorithm. The descriptors that are used as inputs to the HMM
classifier are the cepstral MFCC coefficients. The performances
of the classification method that is supervised, in terms of
classification accuracy are investigated using a database
consisting of a concatenation of two Dutch-speaking sentences
produced by 28 normophonic and 223 dysphonic speakers