Support vector wavelet adaptation for pathological voice assessment

The presence of abnormalities in the vocal system affects the quality of the voice and changes its characteristics. Digital analysis of pathological voices can be an effective and non-invasive tool for the detection of such alterations. This paper proposes a wavelet-based method to distinguish between normal and disordered voices. Wavelet filter banks are used in conjunction with support vector machines, as feature extractors and classifiers, respectively. Orthogonal filter banks are implemented using a highly efficient structure known as "lattice" that parameterizes filter banks and produces a few parameters. The overall problem is to find these parameters such that perfect classification is achieved. To search for such parameters, a genetic algorithm with a fitness function corresponding to the classification result is applied. Simulation is done on the KAY database (a comprehensive database including 710 normal and pathological voice signals, developed by the Massachusetts Eye and Ear Infirmary Voice and Speech Lab), and one additional test set. It is observed that a genetic algorithm is able to find the filter bank parameters such that a 100% correct classification rate is achieved in classifying normal and pathological voices when the test is performed on both databases.

[1]  A. Geva,et al.  ECG feature extraction using optimal mother wavelet , 2000, 21st IEEE Convention of the Electrical and Electronic Engineers in Israel. Proceedings (Cat. No.00EX377).

[2]  R. Guido,et al.  Trying different wavelets on the search for voice disorders sorting , 2005, Proceedings of the Thirty-Seventh Southeastern Symposium on System Theory, 2005. SSST '05..

[3]  Chih-Jen Lin,et al.  A Practical Guide to Support Vector Classication , 2008 .

[4]  Truong Q. Nguyen,et al.  Wavelets and filter banks , 1996 .

[5]  Carlos Dias Maciel,et al.  Wavelet time-frequency analysis and least squares support vector machines for the identification of voice disorders , 2007, Comput. Biol. Medicine.

[6]  H. Wertzner,et al.  Analysis of fundamental frequency, jitter, shimmer and vocal intensity in children with phonological disorders , 2005, Brazilian journal of otorhinolaryngology.

[7]  Farshad Almasganj,et al.  Optimal selection of wavelet-packet-based features using genetic algorithm in pathological assessment of patients' speech signal with unilateral vocal fold paralysis , 2007, Comput. Biol. Medicine.

[8]  J. O. Wisbeck,et al.  Dysphonic voice classification using wavelet packet transform and artificial neural network , 2003, Proceedings of the 25th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (IEEE Cat. No.03CH37439).

[9]  M. Khezri,et al.  Introducing a New Multi-Wavelet Function Suitable for sEMG Signal to Identify Hand Motion Commands , 2007, 2007 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[10]  Dorothea Heiss-Czedik,et al.  An Introduction to Genetic Algorithms. , 1997, Artificial Life.

[11]  Marie-Françoise Lucas,et al.  Signal-dependent wavelets for electromyogram classification , 2005, Medical & Biological Engineering & Computing.

[12]  L. Salhi,et al.  Voice disorders classification using multilayer neural network , 2008, 2008 2nd International Conference on Signals, Circuits and Systems.

[13]  C. J. Long,et al.  Time-frequency dictionaries for improved discriminant feature extraction , 1997 .

[14]  Rodrigo Capobianco Guido,et al.  Discrete wavelet transform and support vector machine applied to pathological voice signals identification , 2005, Seventh IEEE International Symposium on Multimedia (ISM'05).

[15]  Babak Seyed Aghazadeh,et al.  Optimal feature selection for the assessment of vocal fold disorders , 2009, Comput. Biol. Medicine.

[16]  Marie-Françoise Lucas,et al.  Optimization of wavelets for classification of movement-related cortical potentials generated by variation of force-related parameters , 2007, Journal of Neuroscience Methods.

[17]  M. R. Raghuveer,et al.  Constructing MRAs from desired wavelet functions , 1994, Proceedings of 1994 28th Asilomar Conference on Signals, Systems and Computers.

[18]  Igor E. Kheidorov,et al.  Vocal fold pathology detection using modified wavelet-like features and support vector machines , 2007, 2007 15th European Signal Processing Conference.

[19]  Vladimir Vapnik,et al.  Statistical learning theory , 1998 .

[20]  Carlos Dias Maciel,et al.  Support vector machines and wavelets for voice disorder sorting , 2006, 2006 Proceeding of the Thirty-Eighth Southeastern Symposium on System Theory.

[21]  M.M. Elena,et al.  A new mother wavelet for fetal electrocardiography, to achieve optimal denoising and compressing results , 2006, 2006 Computers in Cardiology.

[22]  Raghuveer M. Rao,et al.  Algorithms for designing wavelets to match a specified signal , 2000, IEEE Trans. Signal Process..

[23]  Stéphane Mallat,et al.  Matching pursuits with time-frequency dictionaries , 1993, IEEE Trans. Signal Process..

[24]  Rajendra U Acharya,et al.  Classification and analysis of speech abnormalities , 2005 .

[25]  Joël M. H. Karel,et al.  Optimal discrete wavelet design for cardiac signal processing , 2005, 2005 IEEE Engineering in Medicine and Biology 27th Annual Conference.

[26]  Raghuveer M. Rao,et al.  Matched wavelets-their construction, and application to object detection , 1996, 1996 IEEE International Conference on Acoustics, Speech, and Signal Processing Conference Proceedings.

[27]  Marie-Françoise Lucas,et al.  Signal-dependent wavelets for electromyogram classification , 2006, Medical and Biological Engineering and Computing.

[28]  G. Steidl,et al.  Hybrid wavelet-support vector classification of waveforms , 2002 .

[29]  Pedro Gómez Vilda,et al.  Diagnosis of vocal and voice disorders by the speech signal , 2000, Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Networks. IJCNN 2000. Neural Computing: New Challenges and Perspectives for the New Millennium.

[30]  Elena L. Glassman,et al.  A wavelet-like filter based on neuron action potentials for analysis of human scalp electroencephalographs , 2005, IEEE Transactions on Biomedical Engineering.

[31]  Yiannos Manoli,et al.  Classification of Endocardial Electrograms Using Adapted Wavelet Packets and Neural Networks , 2004, Annals of Biomedical Engineering.

[32]  Gabriele Steidl,et al.  Efficient wavelet adaptation for hybrid wavelet-large margin classifiers , 2005, Pattern Recognit..

[33]  Cesar Ramos Rodrigues,et al.  ADAPTIVE ECG FILTERING AND QRS DETECTION USING ORTHOGONAL WAVELET TRANSFORM , .

[34]  C.J.A. Tollig,et al.  Wavelet neural network for classification of transient signals , 1997, Proceedings of the 1997 South African Symposium on Communications and Signal Processing. COMSIG '97.

[35]  César David Paredes Crovato,et al.  The Use of Wavelet Packet Transform and Artificial Neural Networks in Analysis and Classification of Dysphonic Voices , 2007, IEEE Transactions on Biomedical Engineering.

[36]  Marti A. Hearst.  Trends & Controversies: Support Vector Machines , 1998, IEEE Intell. Syst..