Linear dynamical models in speech synthesis | IEEE Conference Publication | IEEE Xplore