Fourier Descriptor Analysis and Unification of Voice Range Profile Contours: Method and Applications

• Published in: Journal of speech, language, and hearing research Vol. 54; no. 3; pp. 755 - 776
• Main Authors: Pabon, Peter, Ternström, Sten, Lamarche, Anick
• Format: Journal Article
• Language: English
• Published: United States American Speech-Language-Hearing Association (ASHA) 01.06.2011; American Speech-Language-Hearing Association
• Subjects: Adult; Child; Comparative Analysis; contour averaging; Female; Fourier Analysis; Fourier descriptors; HUMANIORA och RELIGIONSVETENSKAP; HUMANITIES and RELIGION; Humans; Languages and linguistics; Linguistic subjects; Linguistics; Lingvistik; Lingvistikämnen; Male; Methods; Models, Biological; norms; Observations; Phonation - physiology; phonetogram; Profiles; Properties; Sound Spectrography - methods; Speech Pathology; Språkvetenskap; Statistical Analysis; Studies; Voice; Voice - physiology; voice range profile; Netherlands
• ISSN: 1092-4388; 1558-9102; 1558-9102
• DOI: 10.1044/1092-4388(2010/08-0222)

Purpose: To describe a method for unified description, statistical modeling, and comparison of voice range profile (VRP) contours, even from diverse sources. Method: A morphologic modeling technique, which is based on Fourier descriptors (FDs), is applied to the VRP contour. The technique, which essentially involves resampling of the curve of the contour, is assessed and also is compared to density-based VRP averaging methods that use the overlap count. Results: VRP contours can be usefully described and compared using FDs. The method also permits the visualization of the local covariation along the contour average. For example, the FD-based analysis shows that the population variance for ensembles of VRP contours is usually smallest at the upper left part of the VRP. To illustrate the method's advantages and possible further application, graphs are given that compare the averaged contours from different authors and recording devices--for normal, trained, and untrained male and female voices as well as for child voices. Conclusions: The proposed technique allows any VRP shape to be brought to the same uniform base. On this uniform base, VRP contours or contour elements coming from a variety of sources may be placed within the same graph for comparison and for statistical analysis.