Method and system for providing a voice interface
First Claim
Patent Images
1. A method comprising:
- determining weights for attributes by ranking the attributes based on user interactions with a user terminal;
storing the weights for the attributes in a memory of the user terminal;
processing, on the user terminal after the storing, a voice input in a first analysis, wherein the first analysis includes identifying one of the attributes;
identifying, as a result of the first analysis, one domain of a plurality of domains, wherein the identifying the one domain includes retrieving the stored weight for the identified attribute and identifying the one domain based on the identified attribute and the retrieved weight;
processing, on the user terminal, the voice input in a second analysis using specialized information of the one domain, wherein each of the plurality of domains comprises different respective specialized information; and
outputting as synthesized speech a response resulting from the second analysis.
2 Assignments
0 Petitions
Accused Products
Abstract
A classifier voice interface of a user terminal may receive a query, may parse the query to identify an attribute, and may process the query to select a first domain-specific voice interface of a plurality of domain-specific voice interfaces based on the attribute, wherein each of the domain-specific voice interfaces comprises specialized information to process queries of different types. The classifier voice interface may further instruct the first domain-specific voice interface to process the query.
31 Citations
24 Claims
-
1. A method comprising:
-
determining weights for attributes by ranking the attributes based on user interactions with a user terminal; storing the weights for the attributes in a memory of the user terminal; processing, on the user terminal after the storing, a voice input in a first analysis, wherein the first analysis includes identifying one of the attributes; identifying, as a result of the first analysis, one domain of a plurality of domains, wherein the identifying the one domain includes retrieving the stored weight for the identified attribute and identifying the one domain based on the identified attribute and the retrieved weight; processing, on the user terminal, the voice input in a second analysis using specialized information of the one domain, wherein each of the plurality of domains comprises different respective specialized information; and outputting as synthesized speech a response resulting from the second analysis. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16)
-
-
17. One or more non-transitory computer readable media storing computer-executable instructions which, when executed by a processor, cause the processor to:
-
determine weights for attributes by ranking the attributes based on user interactions with a user terminal; store the weights for the attributes in a memory; process, after the weights are stored in the memory, a voice input in a first analysis, wherein the first analysis includes identifying one of the attributes; identify, as a result of the first analysis, one domain of a plurality of domains wherein the identifying the one domain includes retrieving the stored weight for the identified attribute and identifying the one domain based on the identified attribute and the retrieved weight; process the voice input in a second analysis using specialized information of the one domain, wherein each of the plurality of domains comprises different respective specialized information; and cause a response resulting from the second analysis to be output as synthesized speech. - View Dependent Claims (18, 19, 20)
-
-
21. An apparatus comprising:
-
a processor; and memory including computer readable instructions, wherein the memory including the computer readable instructions are configured to, with the processor, cause the apparatus to; determine weights for attributes by ranking the attributes based on user interactions with the apparatus; store the weights for the attributes in the memory; process, after the weights are stored in the memory, a voice input in a first analysis, wherein the first analysis includes identifying one of the attributes; identify, as a result of the first analysis, one domain of a plurality of domains wherein the identifying the one domain includes retrieving the stored weight for the identified attribute and identifying the one domain based on the identified attribute and the retrieved weight; process the voice input in a second analysis using specialized information of the one domain, wherein each of the plurality of domains comprises different respective specialized information; and cause a response resulting from the second analysis to be output as synthesized speech. - View Dependent Claims (22)
-
-
23. An apparatus comprising:
-
means for determining weights for attributes by ranking the attributes based on user interactions with a user terminal; means for storing the weights for the attributes in a memory; means for processing, after the weights are stored in the memory, a voice input in a first analysis, wherein the first analysis includes identifying one of the attribute; means for identifying, as a result of the first analysis, one domain of a plurality of domains wherein the identifying the one domain includes retrieving the stored weight for the identified attribute and identifying the one domain based on the identified attribute and the retrieved weight; means for processing the voice input in a second analysis using specialized information of the one domain, wherein each of the plurality of domains comprises different respective specialized information; and means for outputting as synthesized speech a response resulting from the second analysis.
-
-
24. A method comprising:
-
determining weights for attributes by ranking the attributes based on user interactions with a user terminal; storing the weights for the attributes in a memory of the user terminal; performing, on the user terminal after the storing, a first analysis of a voice input, wherein the first analysis includes identifying one of the attributes; identifying, as a result of the first analysis, one domain of a plurality of domains, wherein the identifying the one domain includes retrieving the stored weight for the identified attribute and identifying the one domain based on the identified attribute and the retrieved weight; performing, based on the one domain, a second analysis on the user terminal of the voice input using specialized information of the one domain, wherein each of the plurality of domains comprises different respective specialized information; and outputting as synthesized speech a response resulting from the second analysis, wherein the identified attribute, the specialized information, or both the identified attribute and the specialized information are based on user data stored on the user terminal prior to the first analysis.
-
Specification