A Digital Simulation of the Time‐Varying Vocal Tract

The Kelly‐Lochbaum model for the simulation of the vocal tract is based upon the solution of Webster's horn equation for a static tube. This model has been found to be adequate for the synthesis of steady vowels; however, for time‐varying vocal tract shapes, the model gives rise to discontinuities in pressure and volume velocity. These discontinuities were found to produce noise in the speech synthesized by the model. A solution of Webster's horn equation for time‐varying tubes was obtained by representing the vocal tract by a sequence of uniform cylindrical sections. Recursion equations for pressure and volume velocity at the junctions of adjacent sections were derived. Sentence‐length utterances synthesized by this model were found to be free from noticeable noise.