Abstract: A multichannel digital speech synthesizer comprises a pulse generator storing periodic and aperiodic excitation signals to be processed in a lattice filter according to weighting parameters, such as gain and reflection coefficients, transmitted from a computer via a control unit and a plurality of input modules assigned to respective output channels. Each input module includes a resettable counter for timing the emissions of periodic or aperiodic excitation signals, to generate a voiced or an unvoiced speech element, and for requesting a new set of parameters from the computer upon detecting the end of a validity interval for a current set of parameters; the module further comprises a pair of buffer memories alternating in reading and writing operations under the control of the counter to ensure a continuous flow of parameter sets to the filter.

Abstract: To convert a sample of an amplitude-modulated voltage wave into binary pulses, the sample is simultaneously compared in two coders with two sets of threshold voltages of positive and negative polarity, respectively. Either one of the two coders yields in unbroken succession of pulses, having the same polarity as the sample, whereas the other coder produces a combination of both positive and negative pulses. Following inversion of the pulses issuing from one coder, and upon suppression of the pulses of one polarity (e.g. negative), the two pulse trains are combined into a series of bit groups each representing the absolute amplitude value of a respective sample. The final step in the processing of any sample is the generation of a sign bit in accordance with the polarity of the unipolar pulse train emitted by one of the coders; this sign bit is then transposed to the beginning of the corresponding group of amplitude bits.