Deve Loping a Speech Production Mathematical Model in Terms of the Z-transform for Speech Recogniton Implementation in the Computer