EP0033412A2
Expired / Lapsed
EP
⏱️ 0.0 years remaining
Method and apparatus for speech recognition
Assignee: Scott Instruments Corp
•
Filing: 1980-11-06
•
Granted: —
•
Expires: 2000-11-06
Abstract
Abstract Recognition of human speech is carried out by recording a template for each unit of-speech to produce a dictionary of recorded words and phrases. A given speech signal is converted in a similar manner to produce a template which is compared to the stored template to find the closest comparison. The word or phrase corresponding tothe identified template is produced and displayed to complete the recognition of the speech signal. The speech signal is transmitted through low and high pass filters (18, 20) to produce two separate frequency components. The frequency components are input to a comparator impulse generator (22, 34) which generates a pulse train at the rate of the frequency component. The pulse trains is integrated by filters (26, 38) to produce a DC signal proportional to the frequency of the frequency component. The frequency components are also rectified by rectifiers (24, 36) an envelope detector using low pass filters (30, 42) to produce amplitude signals corresponding to the envelope of the frequency components. The products and ratios of the pairs of frequency and amplitude signals are produced to generate a plurality of relational signals which comprise the templates corresponding to each speech signal. The products are produced by a product detector (50) and the ratios are produced by a ratio detector (52). The relational signals are converted from analog to digital format in a converter (60) and then recorded digitally by a computer (64). In building the dictionary each speech sample is submitted a number of times to produce an average template value together with the variants for each data point. The frequency signals are optionally produced by squaring each signal in a squaring circuit (748, 752) and counting the resulting outputs with a gated counter (750) and a timed counter (754). The digital counts thus produced are processed by the computer (64) which calculates the values for each of the frequency signals.
Inventors
- 👤 Brian Lee Scott
- 👤 Lee Henry Hardesty
Bibliographic Details
Application Number
—
Jurisdiction
EP
Data Source
google_patents
Last Synchronized
2026-10-26 10:44
CPC Classifications
G; G10; G10L; G10L15/00
Claims Analysis
Total Claims: 23
Claim 1
Independent Claim
57 words
A method for producing a template to represent an audio signal, comprising the steps of: extracting a plurality of signal components from said audio signal, producing one or more relational signals'each of which is proportional to a plurality of said signal components, and storing said relational signals to serve as said template which represents said audio signal.
Claim 2
Independent Claim
57 words
A method for producing a template to represent an audio signal, comprising the steps of: extracting a plurality of signal components from said audio signal, producing one or more relational signals'each of which is proportional to a plurality of said signal components, and storing said relational signals to serve as said template which represents said audio signal.
Claim 3
Dependent Claim
97 words
A method-according to Claim 1 wherein the step of extracting a plurality of signal components comprises: filtering said audio signal through filters, each of which passes a different spectral region of said audio signal, producing a frequency signal for each signal produced at the output of said filters, each said frequency signal proportional to the frequency of the output signal from the respective filter, and producing an amplitude signal for each of said signals produced at the output of said filters, each said amplitude signal proportional to the envelope of the output signal from the respective filter.
Claim 4
Dependent Claim
78 words
A method according to Claim 2 wherein the step of generating one or more relational signals comprises: producing a frequency product signal which is proportional to the product of said frequency signals, producing a frequency ratio signal which is proportional to the ratio of said frequency signals, producing an amplitude product signal which is proportional to the product of said amplitude signals, and producing an amplitude ratio signal which is proportional to the ratio of said amplitude signals.
Claim 5
Dependent Claim
48 words
A method according to Claim 1 wherein the step of generating one or more relational signals comprises: producing a first relational signal which is the product of two of said signal components, and producing a second relational signal which is the ratio of two of said signal components.
Claim 6
Dependent Claim
40 words
A method according to Claim 3 wherein the step of producing a frequency signal comprises detecting the zero crossovers of said frequency component signals, producing a pulse for each detected crossover and filtering said pulses to produce said frequency signals.
Claim 7
Dependent Claim
28 words
A method according to Claim 3 wherein the step of producing an amplitude signal comprises envelope detecting each of said frequency component signals to produce said amplitude signals.
Claim 8
Dependent Claim
77 words
A method according to Claim 1 wherein the extracted signal components are spectral component signals of said audio signal, said extracting step including quantizing said spectral component signals into sequential units, and wherein said relational signals are produced-corresponding to each spectral component signal wherein the relational signal comprises a sequence of data points, each data point comprising the ratio of corresponding sequential unit in serial order to the first sequential unit, said relational signals comprising said templates.
Claim 9
Dependent Claim
29 words
A method according to Claim 7 wherein the step of quantizing said spectral component signals comprises digitizing said spectral component signals to produce digital words representing each sequential unit.
Claim 10
Dependent Claim
45 words
A method according to Claim 8 wherein the step of producing a relational signal comprises sequentially dividing each of said digital word making up a spectral component signal by the first of said digital words, said relational signal comprising the resulting sequence of digitized words.
Claim 11
Dependent Claim
130 words
A method according to Claim 1, 2 or 3 which is carried out for producing a generalized template representing a speech pattern signal which is repeated a number of times comprising the steps of: extracting a plurality of signal components from each said speech pattern signal, generating one or more relational signals for each said speech pattern, each of said relational signals proportional to a plurality of said signal components derived from one of said speech pattern signals, averaging the corresponding relational signals of said speech pattern signals to produce an average relational signal, measuring the deviation of said relational signals from said average relational signal to produce a deviation signal for each of said relational signals, and storing each of said average relational signals and the corresponding deviation signals.
Claim 12
Independent Claim
216 words
A method according to any one of Claims 1, 2 or 3 wherein the extracted signal components are at least two frequency component signals of said audio signal and including-the steps of:- providing a first of said frequency component signals to the input terminal of a gated counter, providing a second of said frequency component signals to the gate terminal of said counter to cause the first frequency component signal to be counted when the gate of the counter is activated by said second frequency component signal, the first frequency component signal count thus produced being proportional to the ratio of the frequency of first frequency component signal to the frequency of the second frequency component signal, counting the second frequency component signal to produce a second frequency signal proportional to the frequency of the second frequency component signal, producing a first frequency signal proportional to the frequency of said first frequency component signal by multiplying said first frequency component signal count by said second frequency signal, and producing an amplitude signal from each of said frequency component signals, said amplitude signals proportional to the envelopes of the respective frequency -component signal, said relational signals being produced such that each is proportional to a plurality of signals selected from said frequency signals and said amplitude signals.
Claim 13
Independent Claim
20 words
A method according to any one of the preceding claims including the step of digitizing said relational signals before storing.
Claim 14
Independent Claim
51 words
A method according to any one of the preceding claims including the steps of:- comparing the relational signal template of said given speech signal with the templates representing a dictionary of speech signals to find the closest comparison, and identifying the given speech signal with the closest comparison stored speech signal.
Claim 15
Dependent Claim
46 words
A method according to Claim 13 including the steps of:- storing the words represented by each of said speech signals included in said dictionary, and displaying the word represented by the identified speech signal from said dictionary, the displayed word corresponding to the given speech signal.
Claim 16
Dependent Claim
93 words
A method according to Claim 13 wherein the steps of extracting a plurality of signal components and producing one or more relational signals are carried out for each of a plurality of utterances for each speech signal in said dictionary, the utterances for each speech signal not identical, averaging said relational signals to produce said templates, recording the deviations of said relational signals for said utterances for said averaged relational signals, and utilizing said deviations to determine the closest comparison of the template for said given signal to the temples in said library.
Claim 17
Dependent Claim
78 words
Apparatus for producing a template to represent an audio signal comprising:- means for producing a plurality of signal components derived from said audio signal, mathematical operational means connected to receive said signal-components for generating proportional relationships for combinations of said signal components to produce one or more relational signals each of which is proportional to a plurality of said signal components, and means for storing said relational signals to serve as said template which represents said audio signal.
Claim 18
Dependent Claim
78 words
Apparatus for producing a template to represent an audio signal comprising:- means for producing a plurality of signal components derived from said audio signal, mathematical operational means connected to receive said signal-components for generating proportional relationships for combinations of said signal components to produce one or more relational signals each of which is proportional to a plurality of said signal components, and means for storing said relational signals to serve as said template which represents said audio signal.
Claim 19
Dependent Claim
114 words
Apparatus according to Claim 16 wherein said means for producing a plurality of signal components comprises: filter means for filtering said audio signal to produce a plurality of output signals, each output signal corresponding to a different spectral region of said audio signal, means for detecting the zero crossings for each of said output signals and generating a pulse.train with a pulse occurring at each of said zero crossovers, means for converting said pulse trains into frequency signals each of which is proportional to the frequency of one of said output signals, and means for envelope detecting each of said output signals to produce an amplitude signal corresponding to each of said output signals.
Claim 20
Dependent Claim
121 words
Apparatus according to claim 16 or 17 wherein said mathematical operational means comprises: multiplication means connected to receive said frequency signals and said amplitude signals for producing a frequency product signal which is the product of said frequency signals and for producing an amplitude product signal which is the product of said amplitude signals, and division-means converted to receive said frequency signals and said amplitude signals for producing a frequency ratio signal which is the ratio of one of said frequency signals to another of said frequency signals and for producing an amplitude ratio signal which is the ratio of one of said amplitude signals to another of said amplitude signals, said product and said ratio signals comprising said relational signals.
Claim 21
Dependent Claim
256 words
Apparatus according to Claim 16 wherein the means for producing a plurality of signal components comprises: filter means connected to receive said audio signal for passing a first frequency component thereof, filter means connected to receive said audio signal for passing a second frequency component thereof, said first component being higher than said second component, means connected to receive said frequency components for producing a frequency signal for each frequency component, each said frequency signal proportional to the predominant frequency of the corresponding frequency component, means connected to receive said first frequency component for producing an amplitude signal corresponding to the envelope of said first frequency component, means connected to receive said second frequency component for producing an amplitude signal corresponding to the envelope of said second frequency signal, and wherein the mathematical operational means comprises:- multiplication means connected to receive said frequency and said amplitude signals for producing a frequency product signal which is proportional to the product of said frequency signals and for producing an amplitude product signal which is proportional to the product of said amplitude signals, and division means connected to receive said frequency and amplitude signals for producing a frequency ratio signal which is proportional to the ratio of one of said frequency signals to another of said frequency signals and for producing an amplitude ratio signal which is proportional to the ratio of one of said amplitude signals-to another of said amplitude signals, said product and ratio signals being stored to serve as said template to represent said audio signal.
Claim 22
Dependent Claim
73 words
Apparatus according to Claim 19 wherein said means connected to receive said frequency components for producing a frequency signal for each frequency component comprises:- means for detecting the zero crossings of each of said frequency components, means for generating a pulse for each of said detected zero crossings to produce a pulse train for each of said frequency components, and means for integrating each of said pulse trains to produce said frequency signals.
Claim 23
Dependent Claim
176 words
Apparatus according to Claim 19 wherein said means connected to receive said frequency components for producing a frequency signal for each frequency component comprises:- a first squaring circuit connected to receive said high frequency component and produce a pulsed signal corresponding in frequency to said high frequency component, a second squaring circuit connected to receive said low frequency component and produce a pulsed signal corresponding in frequency to said low frequency component, a gated counter connected to receive and count said pulsed signal from said first squaring circuit, the gate of said counter connected to be operated by said pulsed signal from said received squaring circuit, said gated counter producing a digitized signal proportional to the ratio of said frequency signals, a clocked-counter connected to receive and count said pulsed signal from said second squaring circuit, said clocked counter producing a digitized signal proportional to the frequency of said low frequency signal, and means for multiplying said digitized ratio signal by said digitized frequency signal to produce a digitized signal proportional to said high frequency component.
References Cited (Backward Citations)
No citation records indexed.
USPTO Annuity & Maintenance Windows
Maintenance schedule is calculated for granted US utility patents.