Speech based learning/training system using semantic decoding
First Claim
1. An interactive system adapted for responding to speech-based queries concerning topics addressed by such interactive system, the system comprising:
- a query file for storing a plurality of topic query entries, each topic query entry including a query relating to one or more of the topics covered by the speech-based interactive system; and
an answer file for storing a plurality of topic answer entries, each topic answer entry including an answer to one or more of said plurality of topic query entries, such that each topic query entry has at least one associated topic answer entry; and
a speech recognition system for generating recognized speech data from partially processed speech data associated with a speech-based query concerning one of said topics, said partially processed speech data being received from a remote speech capturing system; and
said speech recognition system further cooperating with a natural language engine which generates recognized speech sentence data corresponding to said speech-based query using both semantic decoding and statistical based processing performed on said recognized speech utterance data;
a query formulation system for converting said recognized speech data into a search query suitable for identifying a topic query entry corresponding to said speech-based query, and for locating at least one topic answer entry best matching said speech-based query.
5 Assignments
0 Petitions
Accused Products
Abstract
An intelligent query system for processing voiced-based queries is disclosed, which uses a combination of both statistical and semantic based processing to identify the question posed by the user by understanding the meaning of the user'"'"'s utterance. Based on identifying the meaning of the utterance, the system selects a single answer that best matches the user'"'"'s query. The answer that is paired to this single question is then retrieved and presented to the user. The system, as implemented, accepts environmental variables selected by the user and is scalable to provide answers to a variety and quantity of user-initiated queries.
455 Citations
28 Claims
-
1. An interactive system adapted for responding to speech-based queries concerning topics addressed by such interactive system, the system comprising:
-
a query file for storing a plurality of topic query entries, each topic query entry including a query relating to one or more of the topics covered by the speech-based interactive system; and
an answer file for storing a plurality of topic answer entries, each topic answer entry including an answer to one or more of said plurality of topic query entries, such that each topic query entry has at least one associated topic answer entry; and
a speech recognition system for generating recognized speech data from partially processed speech data associated with a speech-based query concerning one of said topics, said partially processed speech data being received from a remote speech capturing system; and
said speech recognition system further cooperating with a natural language engine which generates recognized speech sentence data corresponding to said speech-based query using both semantic decoding and statistical based processing performed on said recognized speech utterance data;
a query formulation system for converting said recognized speech data into a search query suitable for identifying a topic query entry corresponding to said speech-based query, and for locating at least one topic answer entry best matching said speech-based query. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10)
-
-
11. A method of implementing a speech-based interactive query system, including the steps of:
-
(a) storing a plurality of topic query entries, each topic query entry including a query relating to one or more of topics covered by the speech-based interactive query system; and
(b) storing a plurality of topic answer entries, each topic answer entry including an answer to one or more of said plurality of topic query entries, such that each topic query entry has at least one associated topic answer entry; and
(c) generating recognized speech utterance data associated with a speech-based query concerning one of said topics, such that said recognized speech utterance data is generated by partial recognition processing of said speech-based query by a first signal processing routine executing at a first computing device, and then completing recognition of said speech-based query through processing performed by a second signal processing routine executing at a second computing device; and
(d) converting said recognized speech utterance data with a natural language process into recognized speech sentence data, said recognized speech data being used by a search query suitable for identifying a topic query entry corresponding to said speech-based query; and
(e) locating at least one topic answer entry best matching said speech-based query;
wherein said natural language process includes both semantic decoding and statistical based processing operations performed on said recognized speech utterance data. - View Dependent Claims (12, 13, 14, 15, 16, 17)
-
-
18. A method of performing semantic decoding of a user question presented in a speech utterance to a networked online natural language processing system to determine a meaning of the user question, the method comprising the steps of:
-
(a) receiving a set of words forming the user question at the networked online natural language processing system;
wherein the user question can be associated with one or more known queries, and which known queries have a meaning that is understood by the networked online natural language processing system;
(b) determining a frequency of occurrence of one or more terms identified in the user question and in each of said one or more known queries;
(c) determining a percentage of terms which appear in the user question that also appear in each of said one or more known queries;
(d) measuring a semantic similarity between said set of words in the user question, and a corresponding set of words in each of said one or more known queries;
(e) combining the results of steps (b) through (d) to identify which one of said one or more known queries is to be used in identifying the meaning of the user question. - View Dependent Claims (19, 20, 21, 22, 23, 24, 25, 27, 28)
-
-
26. A method of populating a natural language speech lattice with semantically variant questions, the method comprising the steps of:
-
(a) receiving a user question;
(b) dividing the user question into a plurality of words corresponding to the user question;
(c) determining synonyms for each word in said plurality of words;
(d) formulating a random set of questions based on said synonyms;
(e) performing semantic decoding on said random set of questions, to identify a disambiguated set of questions;
(f) storing said set of disambiguated questions in a speech recognition lattice;
wherein said set of disambiguated questions correspond to semantic variants of questions that can be posed to a natural language speech engine.
-
Specification