A curvilinear tongue articulatory model
暂无分享,去创建一个
The construction of articulatory models from medical images of the vocal tract, especially X-ray images, relies on the application of an articulatory grid before deriving deformation modes via some factor analysis method. One difficulty faced with the classical semi-polar grid is that some tongue contours do not intersect the grid giving rise to incomplete input vectors, and consequently poor tongue modeling in the front part of the mouth cavity which plays an important role in the articulation of many consonants. First, this paper describes preparation of data, i.e. drawing or tracking articulator contours, compensation for head movements. Then it presents two coordinate systems used to represent and model tongue contours. The first is an adaptive polar grid whose center is a landmark attached to the mandible and the second consists of using curvilinear coordinates. Both offer the advantage of capturing the forward movement of the tongue apex better than other models. Furthermore, the curvilinear model enables any tongue shape, particularly those presenting a sublingual cavity, to be modeled correctly.
[1] Julie Fontecave Jallon,et al. A semi-automatic method for extracting vocal tract movements from X-ray films , 2009, Speech Commun..
[2] Yves Laprie,et al. Modeling the articulatory space using a hypercube codebook for acoustic-to-articulatory inversion. , 2005, The Journal of the Acoustical Society of America.
[3] Shinji Maeda,et al. Compensatory Articulation During Speech: Evidence from the Analysis and Synthesis of Vocal-Tract Shapes Using an Articulatory Model , 1990 .