Towards articulatory speech synthesis with a dynamic 3D finite element tongue model

Abstract. We describe work towards articulatory speech synthesis driven by realistic 3D tissue and bone models. The vocal tract shape is modeled using a fast 3D finite element method (FEM) of a muscle-activated human tongue in conjunction with fixed rigid models of jaw, hyoid and palate connected to a deformable mesh representing the airway. Actuation of the tissue model deforms the airway providing a time-varying acoustic tube which is used for the synthesis of sound. We describe our initial validation of our models geometrically using magnetic resonance images and acoustically using articulatory configurations. 1