Speech synthesis integrated circuit device

Claim 1

A system for generating synthetic human speech comprising: (a) first memory means for storing a plurality of digital speech values including digital speech values representative of digital filter coefficients; (b) second memory means for storing digital data including at least one selected address location of said first memory means; (c) controller means coupled to said first and second memory means for accessing selected portions of said plurality of digital speech values including said digital speech values representative of digital filter coefficients in said first memory means in response to said selected address location in said second memory means; (d) digital filter speech synthesis means coupled to said controller means comprising: (i) an excitation generator for generating excitation signals; (ii) a single multiplier circuit coupled to said first memory means and to said excitation generator and having a selective feedback multiplex network interconnecting the output and the input thereof in a closed feedback loop for selectively multiplying said excitation signals with said accessed digital speech values representative of digital filter coefficients in a multiplexing multiply operation; and (iii) digital-to-analog converter means for converting the output of said single multiplier circuit to analog signals representative of human speech; and (e) audio means coupled to said digital-to-analog converter means for converting said analog signals into audible sounds.

Claim 2

A system for generating synthetic human speech comprising: (a) first memory means for storing a plurality of digital speech values including digital speech values representative of digital filter coefficients; (b) second memory means for storing digital data including at least one selected address location of said first memory means; (c) controller means coupled to said first and second memory means for accessing selected portions of said plurality of digital speech values including said digital speech values representative of digital filter coefficients in said first memory means in response to said selected address location in said second memory means; (d) digital filter speech synthesis means coupled to said controller means comprising: (i) an excitation generator for generating excitation signals; (ii) a single multiplier circuit coupled to said first memory means and to said excitation generator and having a selective feedback multiplex network interconnecting the output and the input thereof in a closed feedback loop for selectively multiplying said excitation signals with said accessed digital speech values representative of digital filter coefficients in a multiplexing multiply operation; and (iii) digital-to-analog converter means for converting the output of said single multiplier circuit to analog signals representative of human speech; and (e) audio means coupled to said digital-to-analog converter means for converting said analog signals into audible sounds.

Claim 10

A system for generating synthetic human speech comprising: (a) first memory means for storing a plurality of digital speech values including digital speech values representative of digital filter coefficients; (b) second memory means for storing digital data including at least one selected address location of said first memory means; (c) digital control means coupled to said first and second memory means for accessing sequences of digital speech values including said digital speech values representative of digital filter coefficients in said first memory means in response to said selected address location in said second memory means; (d) digital filter speech synthesis means for generating analog signals representative of human speech, said digital filter speech synthesis means comprising: (i) filter excitation generator means coupled to said first memory means and responsive to selected ones of said sequences of digital speech values for generating digital excitation signals; (ii) a single multiplier circuit coupled to said first memory means and to said filter excitation generator means and having a selective feedback multiplex network interconnecting the output and the input thereof in a closed feedback loop for selectively multiplying said accessed digital speech values representative of digital filter coefficients with said digital excitation signals in a multiplexing multiply operation; (iii) digital-to-analog converter means for converting the output of said multiplier circuit to analog signals representative of human speech; and (e) audio means coupled to said digital-to-analog converter means for converting said analog signals into audible sounds.

Claim 11

A system for generating synthetic human speech comprising: (a) first memory means for storing a plurality of digital speech values including digital speech values representative of digital filter coefficients; (b) second memory means for storing digital data including at least one selected address location of said first memory means; (c) digital control means coupled to said first and second memory means for accessing sequences of digital speech values including said digital speech values representative of digital filter coefficients in said first memory means in response to said selected address location in said second memory means; (d) digital filter speech synthesis means for generating analog signals representative of human speech, said digital filter speech synthesis means comprising: (i) filter excitation generator means coupled to said first memory means and responsive to selected ones of said sequences of digital speech values for generating digital excitation signals; (ii) a single multiplier circuit coupled to said first memory means and to said filter excitation generator means and having a selective feedback multiplex network interconnecting the output and the input thereof in a closed feedback loop for selectively multiplying said accessed digital speech values representative of digital filter coefficients with said digital excitation signals in a multiplexing multiply operation; (iii) digital-to-analog converter means for converting the output of said multiplier circuit to analog signals representative of human speech; and (e) audio means coupled to said digital-to-analog converter means for converting said analog signals into audible sounds.

Claim 20

A system for generating synthetic human speech comprising: (a) first memory means for storing a plurality of digital speech values indicative of pitch, amplitude, and digital filter coefficients; (b) second memory means for storing digital data including at least one selected address location of said first memory means; (c) controller means coupled to said first and second memory means for accessing selected portions of said plurality of digital speech values in response to said selected address location in said second memory means; (d) operator input means for receiving inputs from an operator and coupled to said controller means for directing the operation of said controller means; (e) digital filter speech synthesis means coupled to said controller means comprising: (i) an excitation generator for generating excitation signals; (ii) a single multiplier circuit having first and second inputs and being coupled to said first memory means via said first input and to said excitation generator via said second input for selectively multiplying said excitation signals with said accessed digital speech values indicative of amplitude and digital filter coefficients; (iii) a selective feedback network operatively interconnected between the output of said single multiplier circuit and said second input thereof in a closed feedback loop including an adder wherein excitation signals and intermediate products of previously accessed digital speech values are combined prior to reintroduction to said single multiplier circuit via said second input thereof in a repetitive sequence for a further multiplying operation with succeeding digital speech values indicative of amplitude and digital filter coefficients in a multiplexing multiply operation; and (iv) digital-to-analog converter means for converting the output of said single multiplier circuit to analog signals representative of human speech; and (f) audio means coupled to said digital-to-analog converter means for converting said analog signals into audible sounds.

Claim 21

A system for generating synthetic human speech comprising: (a) first memory means for storing a plurality of digital speech values indicative of pitch, amplitude, and digital filter coefficients; (b) second memory means for storing digital data including at least one selected address location of said first memory means; (c) controller means coupled to said first and second memory means for accessing selected portions of said plurality of digital speech values in response to said selected address location in said second memory means; (d) operator input means for receiving inputs from an operator and coupled to said controller means for directing the operation of said controller means; (e) digital filter speech synthesis means coupled to said controller means comprising: (i) an excitation generator for generating excitation signals; (ii) a single multiplier circuit having first and second inputs and being coupled to said first memory means via said first input and to said excitation generator via said second input for selectively multiplying said excitation signals with said accessed digital speech values indicative of amplitude and digital filter coefficients; (iii) a selective feedback network operatively interconnected between the output of said single multiplier circuit and said second input thereof in a closed feedback loop including an adder wherein excitation signals and intermediate products of previously accessed digital speech values are combined prior to reintroduction to said single multiplier circuit via said second input thereof in a repetitive sequence for a further multiplying operation with succeeding digital speech values indicative of amplitude and digital filter coefficients in a multiplexing multiply operation; and (iv) digital-to-analog converter means for converting the output of said single multiplier circuit to analog signals representative of human speech; and (f) audio means coupled to said digital-to-analog converter means for converting said analog signals into audible sounds.

Claim 27

The system according to either of claims 1 or 18, wherein the digital speech values as stored in said first memory means are in a predetermined coded format as coded digital speech values, and further including decoder means coupled between said first memory means and said controller means for decoding the coded digital speech values accessed by said controller means.

Claim 29

The system according to any of claims 1, 9, or 18, further including means operably associated with said second memory means for incrementing the address location contained therein.

Claim 31

The system according to any of claims 2, 10, or 19, wherein said second memory means comprises a second non-volatile read only memory separate from the first non-volatile read only memory comprising said first memory means.