A hardware vocal source simulator

Speech synthesis has been performed utilizing different devices, Roughly speaking, these devices are composed of a vocal tract transfer function simulator and a glottal waveform simulator. Of course, this description is very crude and in actual realizations some features which belong to one part may be taken into account by the simulation of the other one.For instance, in mathematical analyses, the zeroes of the glottal wave are treated as belonging to the vocal tract transfer function. The result is good for analysis - synthesis experiments but improving the quality of synthetized speech using physiological features is impossible.