Query engine for processing voice based queries including semantic decoding
First Claim
Patent Images
1. A speech query recognition system comprising:
- a speech recognition engine for generating recognized words taken from an articulated speech utterance; and
a natural language engine configured for processing said recognized words to generate at least two different types of search predicates for said articulated speech utterance;
wherein said search predicates correspond to logical operators to be satisfied by a potential recognition match;
a query formulation engine adapted to convert said recognized words and said search predicates into a structured query suitable for locating a set of one or more corresponding recognized matches for said articulated speech utterance; and
said natural language engine further being configured for processing said set of one or more corresponding recognized matches to determine a final match for said articulated speech utterance using both semantic decoding and statistical based processing performed on said recognized words;
wherein said semantic decoding is performed on entire word sentences contained in said articulated speech utterance to determine semantic variants of said word sentences in said one or more corresponding recognized matches, said semantic decoding being based on a term frequency calculation, which term frequency calculation is based on calculating a lexical distance between each word in said recognized words with each word of one or more topic query entries using a lexical dictionary.
5 Assignments
0 Petitions
Accused Products
Abstract
An intelligent query system for processing voiced-based queries is disclosed. This distributed client-server system, typically implemented on an intranet or over the Internet accepts a user'"'"'s queries at his/her computer, PDA or workstation using a speech input interface. After converting the user'"'"'s query from speech to text, a natural language engine, a database processor and a full-text SQL database is implemented to find a single answer that best matches the user'"'"'s query. Both statistical and semantic decoding are used to assist and improve the performance of the query recognition.
885 Citations
28 Claims
-
1. A speech query recognition system comprising:
-
a speech recognition engine for generating recognized words taken from an articulated speech utterance; and a natural language engine configured for processing said recognized words to generate at least two different types of search predicates for said articulated speech utterance; wherein said search predicates correspond to logical operators to be satisfied by a potential recognition match; a query formulation engine adapted to convert said recognized words and said search predicates into a structured query suitable for locating a set of one or more corresponding recognized matches for said articulated speech utterance; and said natural language engine further being configured for processing said set of one or more corresponding recognized matches to determine a final match for said articulated speech utterance using both semantic decoding and statistical based processing performed on said recognized words; wherein said semantic decoding is performed on entire word sentences contained in said articulated speech utterance to determine semantic variants of said word sentences in said one or more corresponding recognized matches, said semantic decoding being based on a term frequency calculation, which term frequency calculation is based on calculating a lexical distance between each word in said recognized words with each word of one or more topic query entries using a lexical dictionary. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17)
-
-
18. A method of recognizing a speech query comprising the steps of:
-
(a) recognizing text in an articulated speech utterance; and (b) processing said recognized text to generate at least two different types of search predicates for said articulate speech utterance; wherein said search predicates correspond to logical operators to be satisfied by a potential recognition match; (c) generating a query to identify a potential match for said speech utterance, said query being based on said recognized text and said search predicates; (d) determining a final match for said speech utterance by comparing any potential matches identified by said query with said articulated speech utterance; wherein both semantic decoding and statistical based processing operations are used to determine said final match; further wherein said semantic decoding is performed on entire word sentences contained in said articulated speech utterance to determine semantic variants of said word sentences in said potential matches, and is based on a combined metric that includes term frequency, semantic coverage, and semantic distance, the semantic decoding using a lexical dictionary. - View Dependent Claims (19, 20, 21, 22, 23, 24, 25, 26, 27)
-
-
28. A method for recognizing a speech query, comprising:
-
recognizing text in an articulated speech utterance; processing said recognized text to generate at least two different types of search predicates for said articulate speech utterance, the search predicates corresponding to logical operators to be satisfied by a potential recognition match; generating a query to identify a potential match for the speech utterance, said query being based on the recognized text and the search predicates; and determining a final match for said speech utterance by comparing any potential matches identified by the query with the articulated speech utterance, the comparing including comparing word phrases in the potential matches identified by the query with word phrases in the articulated speech utterance, and searching for a best semantic match between the potential matches identified by the query and the articulated speech utterance by calculating a combined semantic metric including term frequency, semantic coverage, and semantic distance using a lexical dictionary.
-
Specification