Clustering Pathologic Voice with Kohonen SOM and Hierarchical Clustering

The main purpose of clustering voice pathologies is the attempt to form large groups of subjects with similar pathologies to be used with Deep-Learning. This paper focuses on applying Kohonen's Self-Organizing Maps and Hierarchical Clustering to investigate how these methods behave in the clustering procedure of voice samples by means of the parameters absolute jitter, relative jitter, absolute shimmer, relative shimmer, HNR, NHR and Autocorrelation. For this, a comparison is made between the speech samples of the Control group of subjects, the Hyper-functional Dysphonia and Vocal Folds Paralysis pathologies groups of subjects. As a result, the dataset was divided in two clusters, with no distinction between the pre-defined groups of pathologies. The result is aligned with previous result using statistical analysis.

[1]  Anil K. Jain,et al.  Data clustering: a review , 1999, CSUR.

[2]  L. Infante,et al.  Hierarchical Clustering , 2020, International Encyclopedia of Statistical Science.

[3]  João Paulo Teixeira,et al.  Clustering of Voice Pathologies based on Sustained Voice Parameters , 2020, BIOSIGNALS.

[4]  João Paulo Teixeira,et al.  Parameters for Vocal Acoustic Analysis - Cured Database , 2019, CENTERIS/ProjMAN/HCist.

[5]  J. Perkell,et al.  Aerodynamic and acoustic voice measurements of patients with vocal nodules: variation in baseline and changes across voice therapy. , 2003, Journal of voice : official journal of the Voice Foundation.

[6]  Felipe Teixeira,et al.  Harmonic to Noise Ratio Measurement - Selection of Window and Length , 2018, CENTERIS/ProjMAN/HCist.

[7]  L. Finger,et al.  Medidas vocais acústicas de mulheres sem queixas de voz e com laringe normal , 2009 .

[8]  Samad EJ Golzari,et al.  Vocal Cord Paralysis and its Etiologies: A Prospective Study , 2014, Journal of cardiovascular and thoracic research.

[9]  M. V. Velzen,et al.  Self-organizing maps , 2007 .

[10]  João Paulo Teixeira,et al.  Vocal Acoustic Analysis - Classification of Dysphonic Voices with Artificial Neural Networks , 2017, CENTERIS/ProjMAN/HCist.

[11]  João Paulo Ramos Teixeira,et al.  Acoustic Analysis of Chronic Laryngitis - Statistical Analysis of Sustained Speech Parameters , 2018, BIOSIGNALS.

[12]  A. C. D. Felippe,et al.  Normatização de medidas acústicas para vozes normais , 2006 .

[13]  João Paulo Teixeira,et al.  Jitter, Shimmer and HNR Classification within Gender, Tones and Vowels in Healthy Voices , 2014 .

[14]  M. Fawcus.  Voice Disorders and their Management , 1991, Springer US.

[15]  S. Hyakin,et al.  Neural Networks: A Comprehensive Foundation , 1994 .

[16]  Felipe Teixeira,et al.  Long Short Term Memory on Chronic Laryngitis Classification , 2018 .

[17]  João Paulo Teixeira,et al.  Vocal Acoustic Analysis: ANN Versos SVM in Classification of Dysphonic Voices and Vocal Cords Paralysis , 2020, Int. J. E Health Medical Commun..

[18]  Eduardo Lleida,et al.  Voice Pathology Detection on the Saarbrücken Voice Database with Calibration and Fusion of Scores Using MultiFocal Toolkit , 2012, IberSPEECH.

[19]  João Paulo Teixeira,et al.  Algorithm for Jitter and Shimmer Measurement in Pathologic Voices , 2016 .

[20]  Pavel Berkhin,et al.  A Survey of Clustering Data Mining Techniques , 2006, Grouping Multidimensional Data.

[21]  Ryszard Tadeusiewicz,et al.  Acoustic analysis assessment in speech pathology detection , 2015, Int. J. Appl. Math. Comput. Sci..

[22]  João Paulo Teixeira,et al.  Deep-learning in Identification of Vocal Pathologies , 2020, BIOSIGNALS.

[23]  T. Kandoğan,et al.  Effectiveness of voice therapy in hyperfunctional dysphonia in adult patients. , 2009, Kulak burun bogaz ihtisas dergisi : KBB = Journal of ear, nose, and throat.

[24]  João Paulo Teixeira,et al.  Acoustic Analysis of Vocal Dysphonia , 2015, CENTERIS/ProjMAN/HCist.

[25]  Felipe Teixeira,et al.  Classification of Control/Pathologic Subjects with Support Vector Machines , 2018 .

[26]  Max A. Little,et al.  Nonlinear speech analysis algorithms mapped to a standard metric achieve clinically useful quantification of average Parkinson's disease symptom severity , 2011, Journal of The Royal Society Interface.