INTERACTIVE SYSTEM, SERVER AND CONTROL METHOD THEREOF

• US 20150161204A1
• Filed: 10/30/2014
• Published: 06/11/2015
• Est. Priority Date: 12/11/2013
• Status: Active Application

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: KIM, Kyoung-min
• Granted Patent: US 10,255,321 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24522 Translation of natural language queries to structured queries
• CPC Class Codes: G06F 16/3329 Natural language query formulation or dialogue systems
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications

Abstract

An interactive system, an first server and a control method thereof are provided. According to a method for controlling the interactive system, the first server includes storing query data generated by inputting a keyword to a pre-analyzed utterance pattern by an first server, receiving a user'"'"'s voice, converting the user'"'"'s voice to voice data and transmitting the voice data to the first server by a display apparatus and in response to the voice data being received, performing a search regarding the user'"'"'s voice by being the first server using the stored query data.

Claims

1. A method of controlling an interactive system, the method comprising:

storing query data generated by inputting a keyword to a pre-analyzed utterance pattern by an first server;

receiving a user'"'"'s voice, converting the user'"'"'s voice to voice data and transmitting the voice data to the first server by a display apparatus; and

in response to the voice data being received, performing a search related to the user'"'"'s voice by using the stored query data by the first server.

2. The method as claimed in claim 1, wherein the performing a search comprises:

in response to data which corresponds to a user'"'"'s voice received from the display apparatus being consistent with the query data stored in a cache memory of the server, performing a search with the consistent query data; and

in response to data which corresponds to a user'"'"'s voice received from the display apparatus not being consistent with the query data stored in a cache memory of the server, performing natural language processing related to data which corresponds to the received user'"'"'s voice and performing a search according to a result of the natural language processing.

3. The method as claimed in claim 1, wherein the utterance pattern comprises collecting the user'"'"'s voice, parsing and analyzing the user'"'"'s voice.

4. The method as claimed in claim 3, wherein the storing comprises:

collecting the user'"'"'s voice, analyzing the utterance pattern by parsing the user'"'"'s voice and generating the query data by inputting the keyword by an second server; and

receiving the query data from the second server and storing the query data in a cache memory by the server.

5. The method as claimed in claim 4, wherein the second server performs analyzing the utterance pattern and generating the query data in response to a number of the received user'"'"'s voice being received less than a predetermined number of times.

6. The method as claimed in claim 1, wherein the keyword is a word related to contents generated within a predetermined period of times or contents searched more than a predetermined number of times.

7. The method as claimed in claim 1, wherein the server further comprises;

deleting the query data which is not used for the search within a predetermined period of time.

8. An interactive system comprising:

a display apparatus configured to receive a user'"'"'s voice, convert the user'"'"'s voice to voice data and transmit the voice data to the first server; and

an first server configured to store generated query data by inputting a keyword to a pre-analyzed utterance pattern, and when the voice data is received, perform a search regarding the user'"'"'s voice by using the stored query data.

9. The interactive system as claimed in claim 8, wherein the first server comprises:

in response to the voice data being consistent with the query data stored in a cache memory, performing a search with the consistent query data; and

in response to the voice data not being consistent with the query data stored in the cache memory, performing a natural language processing related to the voice data, and performing a search according to the result of the natural language processing.

10. The interactive system as claimed in claim 8, wherein the second server comprises collecting the user'"'"'s voice, parsing the user'"'"'s voice and analyzing an utterance pattern.

11. The interactive system as claimed in claim 8, further comprising an second server;

wherein the second server configured to collect the user'"'"'s voice, analyze the utterance pattern by parsing the user'"'"'s voice, generate the query data by inputting the keyword and transmitting the query data to the first server; and

wherein the first server is configured to receive the query data from the second server and store the received query data in a cache memory.

12. The interactive system as claimed in claim 11, wherein the second server further comprises analyzing the utterance pattern and generating the query data in response to a number of the received user'"'"'s voice being less than a predetermined number.

13. The interactive system as claimed in claim 11, wherein the keyword is a word related to contents generated within a predetermined period of time or contents searched more than a predetermined number of times.

14. The method as claimed in claim 11, wherein the first server comprises deleting the query data which is not used for the search within a predetermined period of time.

17. The interactive system of claim 8, wherein the second server is configured to compare the stored query data by inputting a keyword to a pre-analyzed utterance pattern, and in response to the voice data being received, perform a search related to the user'"'"'s voice by using the stored query data.

18. The interactive system of claim 17, wherein the second server is further configured to analyze the utterance pattern and generate the query data in response to a number of the received user'"'"'s voice being less than a predetermined number.

19. The interactive system of claim 17, wherein the keyword is a word related to contents generated within a predetermined period of time or contents searched more than a predetermined number of times.

20. The method of claim 11, wherein the first server is configured to delete the query data which is not used for the search within a predetermined period of time.

15. A method for controlling an first server comprising:

storing query data which is generated by inputting a keyword to a pre-analyzed utterance pattern;

in response to data which corresponds to the user'"'"'s voice being received, determining whether the data is consistent with the stored query data; and

in response to data which corresponds to the received user'"'"'s voice being consistent with the stored query data, performing a search with the stored query data, and in response to data which corresponds to the received user'"'"'s voice not being consistent with the stored query data, performing natural language processing with data which corresponds to the received user'"'"'s voice and performing a search according to the result of the natural language processing.

16. An first server comprising:

a storage configured to store query data generated by inputting a keyword to a pre-analyzed utterance pattern;

a communicator; and

a controller configured to perform a search with the stored query data in response to data which corresponds to the received user'"'"'s voice being consistent with the stored query data, upon data corresponding to a user'"'"'s voice being received through the communicator, by determining whether the received voice data is consistent with the stored query data, and perform natural language processing and perform a search according to the result of the natural-language processing in response to data not being consistent with the received user'"'"'s voice.