Filter Model of Speech
A [see page 17, model] of speech which emulates human speech using source-filter-modelling.
Essentially we have a sound source that we direct into a parameterised vocal-tract filter. The parameters of the filter emulate how the actual vocal tract articulators change to produce different noises. The sound source will be a impulse generator for voiced speech or a noise-source for unvoiced speech.
This model essentially emulates the human speech apparatus mathematically.