Motion Analysis And Synthesis System Of Pronunciation

In this paper, we construct a three-dimensional visualization system for Chinese phoneme by combine the advantage of ultrasound data and three-dimensional physiological model together, which is expected to provide intuitive and accurate visualization of articulator movements. Our visualization system consists of two modules, one is the three-dimensional articulatory postures which is obtained from a physiological model simulations, and the other is two-dimensional Chinese ultrasound imagine data.At present, the pronunciation visualization for Chinese studies are inadequate, especially the three-dimensional model visualization, therefore pronunciation visualization system plays an important role in language learning, especially second language and speech therapy.We analyzed configurations of 30 Chinese phonemes based on an ultrasound database. Then, in order to cover most of the articulatory space, nearly 200,000 muscle activation patterns were designed and used to drive the physiological model.In this paper, we extract the tongue surface from ultrasound data manually, the three-dimensional model simulation database were built by using Supercomputer in Tianjin University Supercomputer Center. With the development of speech analysis and observation technology, the pronunciation visualization is becoming to be easily observed and easily presented.