Lexical tree pre-filtering in speech recognition
First Claim
1. A method of using lexical tree pre-filtering to identify sets of words for use in recognizing speech, the method comprising:
- representing a vocabulary of words using a lexical tree;
identifying a first subset of the vocabulary that may correspond to speech spoken beginning at a first time by propagating through the lexical tree information about the speech spoken beginning at the first time;
identifying a second subset of the vocabulary that may correspond to speech spoken beginning at a second time by propagating through the lexical tree information about the speech spoken beginning at the second time; and
recognizing words included in the speech by comparing speech spoken beginning at the first time with words from the first subset of the vocabulary and speech spoken beginning at the second time with words from the second subset of the vocabulary;
wherein a state of the lexical tree is not reset between the first and second identifying steps so that the state of the lexical tree at initiation of the second identifying step corresponds to the state of the lexical tree at conclusion of the first identifying step.
8 Assignments
0 Petitions
Accused Products
Abstract
A speech recognition technique uses lexical tree pre-filtering to obtain lists of words for use in performing speech recognition. The lexical tree pre-filtering includes representing a vocabulary of words using a lexical tree and identifying a first subset of the vocabulary that may correspond to speech spoken beginning at a first time by propagating through the lexical tree information about the speech spoken beginning at the first time. A second subset of the vocabulary that may correspond to speech spoken beginning at a second time is identified by propagating through the lexical tree information about the speech spoken beginning at the second time. Words included in the speech are recognized by comparing speech spoken beginning at the first time with words from the first subset of the vocabulary and speech spoken beginning at the second time with words from the second subset of the vocabulary. The state of the lexical tree is not reset between identifying the first and second subsets.
238 Citations
30 Claims
-
1. A method of using lexical tree pre-filtering to identify sets of words for use in recognizing speech, the method comprising:
-
representing a vocabulary of words using a lexical tree; identifying a first subset of the vocabulary that may correspond to speech spoken beginning at a first time by propagating through the lexical tree information about the speech spoken beginning at the first time; identifying a second subset of the vocabulary that may correspond to speech spoken beginning at a second time by propagating through the lexical tree information about the speech spoken beginning at the second time; and recognizing words included in the speech by comparing speech spoken beginning at the first time with words from the first subset of the vocabulary and speech spoken beginning at the second time with words from the second subset of the vocabulary; wherein a state of the lexical tree is not reset between the first and second identifying steps so that the state of the lexical tree at initiation of the second identifying step corresponds to the state of the lexical tree at conclusion of the first identifying step. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26)
-
-
27. An apparatus for using lexical tree prefiltering to identify sets of words for use in recognizing speech, the apparatus comprising:
-
means for representing a vocabulary of words using a lexical tree; means for identifying a first subset of the vocabulary that may correspond to speech spoken beginning at a first time by propagating through the lexical tree information about the speech spoken beginning at the first time; means for identifying a second subset of the vocabulary that may correspond to speech spoken beginning at a second time by propagating through the lexical tree information about the speech spoken beginning at the second time; and means for recognizing words included in the speech by comparing speech spoken beginning at the first time with words from the first subset of the vocabulary and speech spoken beginning at the second time with words from the second subset of the vocabulary; wherein a state of the lexical tree is not reset between identifying the first and second subsets of the vocabulary so that the state of the lexical tree at initiation of identifying the second subset corresponds to the state of the lexical tree at conclusion of identifying the first subset.
-
-
28. A speech recognition system that uses lexical tree pre-filtering to identify sets of words for use in recognizing speech, the system comprising:
-
an input device configured to convert speech into a digital signal, and a processor configured to; represent a vocabulary of words using a lexical tree; identify a first subset of the vocabulary that may correspond to speech spoken beginning at a first time by propagating through the lexical tree information about the speech spoken beginning at the first time; identify a second subset of the vocabulary that may correspond to speech spoken beginning at a second time by propagating through the lexical tree information about the speech spoken beginning at the second time; and recognize words included in the speech by comparing a portion of the digital signal corresponding to speech spoken beginning at the first time with words from the first subset of the vocabulary and a portion of the digital signal corresponding to speech spoken beginning at the second time with words from the second subset of the vocabulary; wherein the processor is further configured to maintain the state of the lexical tree between identifying the first and second subsets of the vocabulary so that the state of the lexical tree at initiation of identifying the second subset corresponds to the state of the lexical tree at conclusion of identifying the first subset.
-
-
29. Computer software, residing on a computer-readable storage medium, comprising instructions for causing a computer to:
-
represent a vocabulary of words using a lexical tree; identify a first subset of the vocabulary that may correspond to speech spoken beginning at a first time by propagating through the lexical tree information about the speech spoken beginning at the first time; identify a second subset of the vocabulary that may correspond to speech spoken beginning at a second time by propagating through the lexical tree information about the speech spoken beginning at the second time; recognize words included in the speech by comparing speech spoken beginning at the first time with words from the first subset of the vocabulary and speech spoken beginning at the second time with words from the second subset of the vocabulary; and maintain the lexical tree between identifying the first and second subsets of the vocabulary so that the state of the lexical tree at initiation of identifying the second subset corresponds to the state of the lexical tree at conclusion of identifying the first subset.
-
-
30. A method of using lexical tree pre-filtering to identify sets of words for use in recognizing speech, the method comprising:
-
representing a vocabulary of words using a lexical tree; identifying a first subset of words from the vocabulary, the first subset including words that may correspond to speech spoken beginning at a first time, by propagating through the lexical tree information about the speech spoken beginning at the first time; identifying a second subset of words from the vocabulary, the second subset including words that may correspond to speech spoken beginning at a second time, by propagating through the lexical tree information about the speech spoken beginning at the second time; recognizing words included in the speech by comparing speech spoken beginning at the first time with words from the first subset of the vocabulary and speech spoken beginning at the second time with words from the second subset of the vocabulary; and maintaining a state of the lexical tree between the first and second identifying steps so that the state of the lexical tree at initiation of the second identifying step corresponds to the state of the lexical tree at conclusion of the first identifying step.
-
Specification