Synthesising consonants

The source characteristics of consonants differ depending on the class of sound being produced. All unvoiced sounds use only the noise source nasals and approximants use only the periodic source whereas voiced obstruents (i.e. voice fricatives, affricates and stops) rise both sources. Approximants are generated in much the same way as vowels. Some consonants (such as [h]) can be synthesised in the same way as vowels that is, by sending a sound source through the oral cavity resonators. Most other consonants are. 

Hgure 13.2 Formant configurations (a) cascade or serial configuration and (b) parallel configuration. 

In real speech obstrnents have their frication source generated at a place of constriction, not by the glottis. Normally this place of constriction is caused by the tongue nearing the roof of the mouth. In these cases, some of the sound created at the place of constriction travels back towards the glottis, is reflected there and travels forwards, through the constriction and towards the lips. Hence the vocal tract is effectively split into separate front and rear cavities, which have separate resonance properties. 

Further refinement can be achieved by the use of zeros. These can be used to create anti-resonances, corresponding to a notch in the frequency response. Here the format-synthesis model again deviates from the aU-pole tube model, but recall that we adopted 

Synthesis techniques based on vocal-tract models 

---