Jianwu DANG and Kiyoshi HONDA
A Physiological Model of a Dynamic Vocal Tract for Speech Production
Abstract:A physiological articulatory model has been developed to simulate the dynamic actions of the speech organs. This model consists of the tongue, mandible, hyoid bone, and the outer wall of the vocal tract in three dimensions. The soft tissue of the tongue is outlined by a midsagittal plane and two parasagittal planes with 1-cm intervals. The mandible and hyoid bone are modeled to yield rotation and translation motion. The palatal and pharyngeal walls are modeled as a rigid structure. Thus, the proposed dynamic model of the vocal tract consists of soft and hard components with the rigid outer wall of the vocal tract. The frameworks of the rigid organs and muscle geometry of the model are extracted from volumetric MR images of a male speaker. All the soft tissue and rigid organs are modeled using mass-points and two types of connections: viscoelastic springs for connective tissue, and rigid beams for bony organs. The muscle definitions are based on an improved Hill's model. This study also develops a model's control method to generate tongue and jaw movements of the model by using generalized muscle activation signals that are similar to integrated EMG signals. The collisions of the tongue and the outer wall of the vocal tract were also considered in the model. The model produces plausible dynamic patterns of the tongue-jaw complex with relatively short computational time in comparison with the finite element method because the displacements of both the soft tissue and rigid organs are computed with the same algorithm.