Voice recognition system
First Claim
1. A text locating system comprising:
- means for storing an acoustic description of each of a plurality of vocabulary words;
means for representing a body of text having a sequence of words, said body of text being other than a list of said vocabulary words;
means for receiving an acoustic description of one or more utterances to be recognized;
recognition means for comparing the acoustic descriptions of the utterances to be recognized against acoustic descriptions of a recognition vocabulary comprised of one or more of said vocabulary words to select which of the words in said recognition vocabulary most probably correspond to said one or more utterances;
search-string means for receiving one or more vocabulary words determined by said recognition means as corresponding to said one or more utterances and for creating a search string out of them;
matching means for searching at least a portion of said body of text for a sub-sequence of one or more words matching said search string;
probability altering means for altering the probability that said recognition means will select a given vocabulary word as corresponding to a given utterance as a function of the frequency of occurrence of that vocabulary word in at least a portion of said body of text.
1 Assignment
0 Petitions
Accused Products
Abstract
A text locating system recognizes spoken utterances, uses the recognized words as a search string, and searches text for words matching that search string. The probability that a given vocabulary word is selected as a search word is altered both by limiting the recognizable vocabulary to words in the text to the searched, and by altering the probability that individual recognizable words will be selected as a function of the number of time they occur in that text. The system performs incremental searches by adding successively recognized words to the search string and searching for the next occurrence of the string in response to each such addition. The invention can be used in a text editing system which enables a user to switch between a dictation mode, which inserts recognized words into text, and a search mode, which uses them to search for new cursor locations. Broadly speaking, the invention provides a computer system which recognizes spoken words, which has a data structure representing words; which uses that data structure for a purpose other than speech recognition; and which alters the probability that a given vocabulary word will be recognized as a function of the frequency of that word in the data structure.
-
Citations
28 Claims
-
1. A text locating system comprising:
-
means for storing an acoustic description of each of a plurality of vocabulary words; means for representing a body of text having a sequence of words, said body of text being other than a list of said vocabulary words; means for receiving an acoustic description of one or more utterances to be recognized; recognition means for comparing the acoustic descriptions of the utterances to be recognized against acoustic descriptions of a recognition vocabulary comprised of one or more of said vocabulary words to select which of the words in said recognition vocabulary most probably correspond to said one or more utterances; search-string means for receiving one or more vocabulary words determined by said recognition means as corresponding to said one or more utterances and for creating a search string out of them; matching means for searching at least a portion of said body of text for a sub-sequence of one or more words matching said search string; probability altering means for altering the probability that said recognition means will select a given vocabulary word as corresponding to a given utterance as a function of the frequency of occurrence of that vocabulary word in at least a portion of said body of text. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8)
-
-
9. A text processing system comprising:
-
means for representing a body of text having a sequence of characters which form a sequence of words; means for performing editing functions on said body of text; means for storing an acoustic description of each of a plurality of vocabulary words; means for receiving an acoustic description of one or more utterances to be recognized; recognition means for comparing the acoustic descriptions of the utterances to be recognized against the acoustic descriptions of a recognition vocabulary comprised of one or more of said vocabulary words to select which of the words of said recognition vocabulary most probably correspond to said one or more utterances; search string means for receiving the one or more vocabulary words determined by said recognition means as corresponding to said one or more utterances and for creating a search string out of them; matching means for searching at least a portion of said body of text for an occurrence of a sub-sequence of one or more words matching said search string; cursor storing means for storing the location of a cursor, that is, a movable location in said body of text at which certain text processing functions can be performed; and cursor setting means, responsive to said string matching means, for setting the location stored in said cursor storing means to one which points to said matching sub-sequence of one or more words. - View Dependent Claims (10, 11, 12, 13)
-
-
14. A text processing system comprising:
-
means for representing a body of text having a sequence of characters which form a sequence of words; means for performing editing functions on said body of text; means for storing an acoustic description of each of a plurality of vocabulary words; means for receiving an acoustic description of one or more utterances to be recognized; recognition means for comparing the acoustic descriptions of the utterances to be recognized against the acoustic descriptions of a recognition vocabulary comprised of one or more of said vocabulary words to select which of the words of said recognition vocabulary most probably correspond to said one or more utterances; search-string means for receiving the one or more vocabulary words determined by said recognition means as corresponding to said one or more utterances and for creating a search string out of them; matching means for searching at least a portion of said body of text for an occurrence of a sub-sequence of one or more words matching said search string; block function means for performing a block function on a block of characters from said body of text; and means for causing said block function means to perform a block function on a block of characters defined by said matching sub-sequence of one or more words.
-
-
15. A computer system comprising:
-
means for storing an acoustic description of each of a plurality of vocabulary words; means for receiving an acoustic description of one or more utterances to be recognized; recognition means for comparing the acoustic descriptions of utterances to be recognized against the acoustic descriptions of a recognition vocabulary comprised of one or more of said vocabulary words to select which of the words in said recognitino vocabulary most probably corresponds to said one or more utterances; a data structure representing one or more words, said data structure being other than a list of said vocabulary words; means for using said data structure for a purpose independent both of indicating which words are vocabulary words and of indicating the probability that individual vocabulary words will be recognized as corresponding to a given utterance; probability altering means for determining which of said vocabulary words are represented by said data structure and for using that determination to alter the probability that each of said vocabulary words will be selected as corresponding to a given utterance. - View Dependent Claims (16, 17, 18, 19, 20, 21, 22)
-
-
23. A text locating system for locating a sequence of one or more words in a body of text comprising:
-
means for representing a body of human readable text as a sequence of individual machine readable characters corresponding to individual human readable characters in that text; speech recognition means for responding to the sound of a spoken word by producing a word symbol, identifying a human speakable word which said speech recognition means considers likely to correspond to said sound; search string means for representing a search string comprised of one or more machine readable characters and for responding to the word symbol produced by said speech recognition means by adding to said speech string a sequence of one or more of said machine readable characters corresponding to the spelling of the human speakable word identified by said word symbol; and string searching means for performing a character-by-character search of least a portion of said body of text for a sub-sequence of one or more of said machine readable characters matching the machine readable characters represented by said search string. - View Dependent Claims (24, 25, 26, 27, 28)
-
Specification