A Vocal Tract Segmentation and Analysis over a European Portuguese MRI Database

Knowledge of the speech production mechanism is essential for the development of speech production models and theories. Magnetic Resonance Imaging delivers high quality imaging of soft tissue. To our knowledge, there are no complete systematic Magnetic Resonance Imaging studies of European Portuguese production. With this work, we intend to create a database of Magnetic Resonance Imaging images of European Portuguese and to implement segmentation techniques well known on the image processing field, but rarely used on the context of vocal tract segmentation. We present 2D vocal tract contours as well as their evaluation using a similarity metric, the Pratt Index. This metric allows not only to validate the segmentation technique, but also to infer on the differences among the sounds in study.

[1]  Shrikanth Narayanan,et al.  An approach to real-time magnetic resonance imaging for speech production. , 2003, The Journal of the Acoustical Society of America.

[2]  Investigation of coarticulation in continuous speech of Japanese , 2004 .

[3]  Didier Demolin,et al.  Segmentation of the airway from the surrounding tissues on magnetic resonance images: a comparative study , 1998, ICSLP.

[4]  Olov Engwall,et al.  Are static MRI measurements representative of dynamic speech? results from a comparative study using MRI, EPG and EMA , 2000, Interspeech.

[5]  Carlos Ferreira,et al.  Quantitative evaluation of a pulmonary contour segmentation algorithm in X-ray computed tomography images. , 2004, Academic radiology.

[6]  Diamantino Freitas,et al.  Morphological dynamic study of human vocal tract , 2006, CompIMAGE.

[7]  Kiyoshi Honda,et al.  A method of tooth superimposition on MRI data for accurate measurement of vocal tract shape and dimensions , 2004 .

[8]  Ian Middleton,et al.  Segmentation of magnetic resonance images using a combination of neural networks and active contour models. , 2004, Medical engineering & physics.

[9]  Guido Gerig,et al.  User-guided 3D active contour segmentation of anatomical structures: Significantly improved efficiency and reliability , 2006, NeuroImage.

[10]  Pierre Badin,et al.  Towards a 3D articulatory model of velum based on MRI and CT images , 2005 .

[11]  Shrikanth S. Narayanan,et al.  Semi-Automatic Processing of Real-time MR Image Sequences for Speech Production Studies , 2006 .

[12]  Rolf Adams,et al.  Seeded Region Growing , 1994, IEEE Trans. Pattern Anal. Mach. Intell..

[13]  Robert Sader,et al.  Dynamic near‐real‐time magnetic resonance imaging for analyzing the velopharyngeal closure in comparison with videofluoroscopy , 2004, Journal of magnetic resonance imaging : JMRI.