EP0119835A1
Expired / Lapsed
EP
⏱️ 0.0 years remaining
Speech recognition system based on word state duration and/or weight
Assignee: Scott Technologies Inc
•
Filing: 1984-03-15
•
Granted: —
•
Expires: 2004-03-15
Abstract
Abstract A low cost, speaker independent, limited vocabulary, word recognizing microcomputer (17). The microcomputer divides each spoken word into a series of word states, determines the length of each state and classifies each state as fricative, vowel-like, or silent. The incoming speech pattern, in the form of an array of classified word states and an array of associated word lengths is then compared sequentially with a series of templates, defining the limited vocabulary stored in the microcomputer's memory (19). Each stored template includes an array of classified states, an array of lengths corresponding to the states and an array of weights corresponding to at least some of the states. The state classifications ofthe incoming state array are compared sequentially with the template states of each template state array and where the classifications match, an error score is generated based on the difference in lengths between the template state lengths and the word state lengths. The weights are used to permit recognition of a spoken word as a template word even when the array of states representing the spoken word is not identical to an array of states in any of the template words.
Inventors
- 👤 Myron H. Hitchcock
Bibliographic Details
Application Number
—
Jurisdiction
EP
Data Source
google_patents
Last Synchronized
2026-11-08 20:28
CPC Classifications
G; G10; G10L; G10L15/00; G10L15/02; G10L15/08; G10L15/10; G10L25/00; G10L25/48; G10L25/51
Claims Analysis
Total Claims: 16
Claim 1
Independent Claim
157 words
Speech recognition apparatus characterised by: a circuit which divides incoming speech into an array of classified speech states and measures the length of each speech state to form a corresponding incoming state length array; a memory (19) storing at least one template representing a vocabulary word of the speech recognition apparatus, said template including a first array of classified word states and a second array composed of a sequence of values indicating the length of at least some of the corresponding stored word states; a circuit which acts to compare the classifications of the incoming speech state array with the classifications of the template state array; and a circuit which generates a measure of the degree of matching between the incoming speech state array and the template word state array, said measure being a function of the differences in length between the incoming speech states and the corresponding template word states of states that have matching classifications.
Claim 2
Dependent Claim
157 words
Speech recognition apparatus characterised by: a circuit which divides incoming speech into an array of classified speech states and measures the length of each speech state to form a corresponding incoming state length array; a memory (19) storing at least one template representing a vocabulary word of the speech recognition apparatus, said template including a first array of classified word states and a second array composed of a sequence of values indicating the length of at least some of the corresponding stored word states; a circuit which acts to compare the classifications of the incoming speech state array with the classifications of the template state array; and a circuit which generates a measure of the degree of matching between the incoming speech state array and the template word state array, said measure being a function of the differences in length between the incoming speech states and the corresponding template word states of states that have matching classifications.
Claim 3
Dependent Claim
77 words
Apparatus as claimed in claim 1, further characterised by a circuit which compares the generated measure of the degree of matching between the template and the incoming set of arrays with a threshold level to determine whether or not to recognize the incoming speech as the vocabulary word corresponding to the template; and an output signal generator (23) for indicating that the set of incoming speech arrays has been recognized as the word corresponding to the template.
Claim 4
Dependent Claim
83 words
Apparatus as claimed in claim 1, characterised in that said memory (19) stores a plurality of templates each one representing a vocabulary word; wherein the comparison circuit acts to sequentially compare the set of incoming arrays with each of the templates and the measure generating circuit acts to generate a measure of the degree of matching for each template; and wherein said apparatus further comprises a circuit for determining which of said templates is the closest match to the incoming set of arrays.
Claim 5
Dependent Claim
83 words
Apparatus as claimed in claim 3, further characterised by a circuit which compares the degree of matching measure for the template which has been determined to be the closest match with a threshold level to determine whether or not to recognize the incoming speech as the vocabulary word corresponding to the template; and an output signal generator (23) for indicating that the set of incoming speech arrays has been recognized as the word corresponding to the template determined to be the closest match.
Claim 6
Dependent Claim
22 words
Speech recognition apparatus as claimed in claim 1, characterised in that each speech state is classified as either fricative, vowel-like, or silence.
Claim 7
Dependent Claim
35 words
Speech recognition apparatus as claimed in claim 1, characterised in that said comparing circuit sequentially compares the classifications of the states of the incoming word state array with the classifications of the template state array.
Claim 8
Dependent Claim
182 words
Speech recognition apparatus as claimed in claim 1, characterised in that each stored template further includes a third array having a series of weighting values, one value being assigned to each stored word state based on the importance of the word state to recognition of the word; wherein said measure is an error value that is a function both of the difference in length between incoming speech states and the corresponding template word states for states that have matching classifications as well as of the weights corresponding to states that do not have matching classifications; wherein said measure generating circuit includes: a circuit increasing the error value by the absolute value of the difference between the values of said duration arrays corresponding to the particular states being compared when the classifications of the states being compared are the same; and a circuit increasing the error value by an amount equal to the weighting value of said weighting array corresponding to the particular state of said template state array being compared when the classifications of the states being compared are not the same.
Claim 9
Dependent Claim
45 words
Speech recognition apparatus as claimed in claim 7, characterised in that said comparing circuit further includes a circuit comparing the last compared incoming speech state with the next state in the template state array when the last compared states have classifications which do not match.
Claim 10
Dependent Claim
41 words
Speech recognition apparatus as claimed in claim 7, characterised in that said measure generating circuit further includes a circuit designating the template with the lowest error value as the template which is the closest match to the incoming speech state array.
Claim 11
Dependent Claim
45 words
Speech recognition apparatus as claimed in claim 9, further characterised by a circuit which rejects matches that are too remote, and wherein said rejecting circuit rejects the closest matching template when the error value corresponding to that template is greater than a predetermined threshold value.
Claim 12
Dependent Claim
44 words
Speech recognition apparatus as claimed in claim 10 further characterised by a circuit recognizing the incoming spoken word state array as the word corresponding to the template selected as having the closest match provided this template has not been rejected by said rejecting means.
Claim 13
Dependent Claim
59 words
Apparatus as claimed in claim 1, characterised in that said measure is an error value and said measure generating circuit includes a circuit increasing the error value by the absolute value of the difference between the values of said durational arrays corresponding to the particular states being compared when the classifications of the states being compared are the same.
Claim 14
Independent Claim
129 words
A method for speech recognition characterised by: dividing incoming speech into an array of classified speech states; measuring the length of each speech state to form a corresponding incoming state length array; providing a template representing a stored vocabulary word, said template including a first array of classified word states and a second array including a sequence of values indicating the length of at least some of the corresponding word states; comparing the classifications of the incoming speech state array with the classifications of the template state array; and generating a measure of the degree of matching between said incoming speech state array and said template word state array, said measure being a function of the differences in length between said incoming speech states and matching template word states.
Claim 15
Independent Claim
129 words
A method for speech recognition characterised by: dividing incoming speech into an array of classified speech states; measuring the length of each speech state to form a corresponding incoming state length array; providing a template representing a stored vocabulary word, said template including a first array of classified word states and a second array including a sequence of values indicating the length of at least some of the corresponding word states; comparing the classifications of the incoming speech state array with the classifications of the template state array; and generating a measure of the degree of matching between said incoming speech state array and said template word state array, said measure being a function of the differences in length between said incoming speech states and matching template word states.
Claim 16
Dependent Claim
29 words
1 4 . A method as claimed in claim 13, characterised in that said dividing step includes classifying each of said speech states as either fricative, vowel-like or silent.
References Cited (Backward Citations)
No citation records indexed.
USPTO Annuity & Maintenance Windows
Maintenance schedule is calculated for granted US utility patents.