Finite Element Model of the Human Phonation Process