6.Acknowledgement

This work was supported by the National Natural Science-Foundation of China （No.61175016，61304250），Key Fund projects of 61233009，and financial support from CASS Innovation Project “Articulatory model for pronunciation training”.