Speech synthesiser using formant vocoder principle - has digital store generating sinusoidal waveform with positive and negative correction facility

摘要

Voice synthesis based on the formant vocoder concept uses sinusoidal waveforms combined to generated tuned sounds that correspond to the frequency spectrum characteristic of speech. A digital memory is used to store the positive half wave digitised sine wave values as eight bit numbers. The memory is addressed over a multiplexer at a frequency of 8 kHz. Addition stages ensure that the formant amplitude characteristics are generated with the correct logarithmic form. A correction stage uses information held in a store to allow positive or negative phase shift correction.