Interactive server and method for controlling the server

• US 9,837,082 B2
• Filed: 01/06/2015
• Issued: 12/05/2017
• Est. Priority Date: 02/18/2014
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: Kim, Kyung-duk
• Primary Examiner(s): Ky, Kevin
• Application Number:
  US14/590,630
• Publication Number:
  US 20150235643A1
• Time in Patent Office: 1,064 Days
• Field of Search: None
• CPC Class Codes: G06F 40/20 Natural language analysis semantic analysis of natural language G06F40/30
• CPC Class Codes: G10L 15/197 Probabilistic grammars, e.g. word n-grams
• CPC Class Codes: G10L 17/22 Interactive procedures; Man-machine interfaces
• CPC Class Codes: H04N 21/233 Processing of audio elementary streams monitoring, identification or recognition of audio in broadcast systems H04H60/58
• CPC Class Codes: H04N 21/42203 sound input device, e.g. microphone
• CPC Class Codes: H04N 21/47 End-user applications
• CPC Class Codes: H04N 21/6582 Data stored in the client, e.g. viewing habits, hardware capabilities, credit card number arrangements where receivers interact with the broadcast H04H20/38

Abstract

An interactive server and a controlling method thereof are provided. The method of controlling an interactive server includes receiving data corresponding to a user voice from a user terminal, determining an utterance type of the user voice, in response to determining that the utterance type of the user voice is an integrated utterance type, generating a search response and a chatting response in response to the user voice and generating an integrated response by combining the generated search response and the generated chatting response, and transmitting the generated integrated response to the user terminal.

Claims

1. A method of controlling an interactive server for providing a response according to an utterance type of a user voice, the method comprising:

receiving, at the interactive server, data corresponding to a user voice from a user terminal;

calculating a probability of the utterance type of the user voice being a search utterance type using the received data and a search language model established based on search utterances;

calculating a probability of the utterance type of the user voice being a chatting utterance type using the received data and a chatting language model established based on chatting utterances;

calculating a disparity value between the probability of the utterance type of the user voice being the search utterance type and the probability of the utterance type of the user voice being the chatting utterance type;

in response to the disparity value being less than a predetermined value, determining that the utterance type of the user voice is an integrated utterance type;

in response to the disparity value exceeding the predetermined value and the probability of the utterance type of the user voice being the search utterance type being greater than the probability of the utterance type of the user voice being the chatting utterance type, determining that the user voice is the search utterance type; and

in response to the disparity value exceeding the predetermined value and the probability of the utterance type of the user voice being the search utterance type being less than the probability of the utterance type of the user voice being the chatting utterance type, determining that the utterance type of the user voice is the chatting utterance type;

in response to determining that the utterance type of the user voice is the integrated utterance type, generating a search response and a chatting response in response to the user voice and generating an integrated response by modifying at least one from among the search response and the chatting response and combining the search response and the chatting response; and

transmitting the integrated response to the user terminal.

2. The method as claimed in claim 1, further comprising:

in response to determining that the utterance type of the user voice is the search utterance type, generating a search response to the user voice, and transmitting the search response to the user terminal; and

in response to determining that the utterance type of the user voice is the chatting utterance type, generating a chatting response to the user voice, and transmitting the chatting response to the user terminal.

3. The method as claimed in claim 1, wherein the generating the integrated response comprises:

determining whether it is possible to combine the search response and the chatting response by determining a sentence type of the search response and the chatting response and whether common information is included in the search response and the chatting response;

in response to determining that it is possible to combine the search response and the chatting response, determining a combination order of the search response and the chatting response,

wherein the modifying the at least one from among the search response and the chatting response comprises modifying the at least one from among the search response and the chatting response according to the combination order that is determined.

4. The method as claimed in claim 3, further comprising, in response to determining that it is impossible to combine the search response with the chatting response, transmitting one of the search response and the chatting response to the user terminal.

5. The method as claimed in claim 3, wherein the determining the combination order comprises determining the combination order of the search response and the chatting response by comparing a probability of the utterance type of the user voice being the chatting utterance type with a probability of the utterance type of the user voice being the search utterance type.

6. The method as claimed in claim 3, wherein the modifying the at least one from among the search response and the chatting response comprises modifying an ending of the at least one from among the search response and the chatting response.

7. The method as claimed in claim 1, wherein the data corresponding to the user voice is text data.

8. The method as claimed in claim 1, wherein the data corresponding to the user voice is voice data,

the method further comprising performing voice recognition with respect to the voice data to obtain corresponding text data of the user voice.

9. An interactive server for providing a response according to an utterance type of a user voice, the interactive server comprising:

at least one memory configured to store instructions;

at least one processor configured to execute the stored instructions to implement;

an input interface configured to receive data corresponding to a user voice from a user terminal;

a determiner comprising;

a probability calculator configured to calculate a probability of the utterance type of the user voice being a search utterance type using the received data and a search language model established based on search utterances, and calculate a probability of the utterance type of the user voice being a chatting utterance type using the received data and a chatting language model established based on chatting utterances;

a disparity calculator configured to calculate a disparity value between the probability of the utterance type of the user voice being the search utterance type and the probability of the utterance type of the user voice being the chatting utterance type; and

an utterance type determiner configured to;

in response to the disparity value being less than a predetermined value, determine that the utterance type of the user voice is an integrated utterance type,

in response to the disparity value exceeding the predetermined value and the probability of the utterance type of the user voice being the search utterance type being greater than the probability of the utterance type of the user voice being the chatting utterance type, determine that the utterance type of the user voice is the search utterance type, and

in response to the disparity value exceeding the predetermined value and the probability of the utterance type of the user voice being the search utterance type being less than the probability of the utterance type of the user voice being the chatting utterance type, determine that the utterance type of the user voice is the chatting utterance type;

a response generator configured to, in response to determining that the utterance type of the user voice is the integrated utterance type, generate a search response and a chatting response in response to the user voice, and generate an integrated response by modifying at least one from among the search response and the chatting response and combining the search response and the chatting response; and

a transmitter configured to transmit the integrated response to the user terminal.

10. The interactive server as claimed in claim 9, wherein the response generator is configured to, in response to determining that the utterance type of the user voice is the search utterance type, generate a search response to the user voice, and in response to determining that the utterance type of the user voice is the chatting utterance type, generate a chatting response to the user voice,

wherein the transmitter is configured to transmit one of the search response and the chatting response to the user terminal.

11. The interactive server as claimed in claim 9, wherein the response generator comprises:

a search response generator configured to generate a search response to the user voice;

a chatting response generator configured to generate a chatting response to the user voice;

a combination determiner configured to determine whether it is possible to combine the search response with the chatting response by determining a sentence type of the search response and the chatting response and whether common information is included in the search response and the chatting response;

a combination order determiner configured to, in response to determining that it is possible to combine the search response and the chatting response, determine a combination order of the search response and the chatting response,

wherein the modifying the at least one from among the search response and the chatting response comprises modifying the at least one from among the search response and the chatting response according to the combination order that is determined.

12. The interactive server as claimed in claim 11, wherein the transmitter is configured to, in response to determining that it is impossible to combine the search response with the chatting response, transmit one of the search response and the chatting response to the user terminal.

13. The interactive server as claimed in claim 11, wherein the combination order determiner is configured to determine the combination order of the search response and the chatting response by comparing a probability of the utterance type of the user voice being the chatting utterance type with a probability of the utterance type of the user voice being the search utterance type.

14. The interactive server as claimed in claim 11, wherein the modifying the at least one from among the search response and the chatting response comprises modifying an ending of at least one from among the search response and the chatting response.

15. A method of providing a response according to an utterance type of a user voice, the method comprising:

receiving data corresponding to a user voice from a user terminal;

generating a search response and a chatting response in response to the user voice;

calculating a probability of the utterance type of the user voice being a search utterance type using the received data and a search language model established based on search utterances;

calculating a probability of the utterance type of the user voice being a chatting utterance type using the received data and a chatting language model established based on chatting utterances;

calculating, a disparity value between the calculated probability of the utterance type of the user voice being the search utterance type and the calculated probability of the utterance type of the user voice being the chatting utterance type;

in response to the disparity value being less than a predetermined value, determining that the utterance type of the user voice is an integrated utterance type;

in response to the disparity value exceeding the predetermined value and the probability of the utterance type of the user voice being the search utterance type being greater than the probability of the utterance type of the user voice being the chatting utterance type, determining that the user voice is the search utterance type; and

in response to the disparity value exceeding the predetermined value and the probability of the utterance type of the user voice being the search utterance type being less than the probability of the utterance type of the user voice being the chatting utterance type, determining that the utterance type of the user voice is the chatting utterance type;

in response to determining the utterance type of the user voice is the integrated utterance type, generating an integrated response by modifying at least one from among the search response and the chatting response and combining the search response and the chatting response; and

transmitting the integrated response to the user terminal.

16. An interactive server for providing a response according to an utterance type of a user voice, the interactive server comprising:

at least one memory configured to store instructions;

at least one processor configured to execute the stored instructions to implement;

an input interface configured to receive data corresponding to a user voice from a user terminal;

a response generator configured to generate a search response and a chatting response in response to the user voice;

a determiner comprising;

a probability calculator configured to calculate a probability of the utterance type of the user voice being a search utterance type using the received data and a search language model established based on search utterances, and calculate a probability of the utterance type of the user voice being a chatting utterance type using the received data and a chatting language model established based on chatting utterances;

a disparity calculator configured to calculate a disparity value between the probability of the utterance type of the user voice being the search utterance type and the probability of the utterance type of the user voice being the chatting utterance type; and

an utterance type determiner configured to;

in response to the disparity value being less than a predetermined value, determine that the utterance type of the user voice is an integrated utterance type,

in response to the disparity value exceeding the predetermined value and the probability of the utterance type of the user voice being the search utterance type being greater than the probability of the utterance type of the user voice being the chatting utterance type, determine that the utterance type of the user voice is the search utterance type, and

in response to the disparity value exceeding the predetermined value and the probability of the utterance type of the user voice being the search utterance type being less than the probability of the utterance type of the user voice being the chatting utterance type, determine that the utterance type of the user voice is the chatting utterance type;

an integrated response generator configured to, in response to determining that the utterance type of the user voice is the integrated utterance type, generate an integrated response by modifying at least one from among the search response and the chatting response and combining the search response with the chatting response; and

a transmitter configured to transmit the integrated response to the user terminal.