Device for providing toolkit for agent developer

Patent metadata

Field	Value
Publication number	US-2019156827-A1
Application number	US-201816197171-A
Country	US
Kind code	A1
Filing date	Nov 20, 2018
Priority date	Nov 20, 2017
Publication date	May 23, 2019
Grant date	—

Abstract

Disclosed is a system for providing a toolkit for an agent developer. A system for providing a toolkit for an agent developer according to an embodiment of the present invention includes: an interface unit that obtains an utterance input by a user and outputs the utterance; and a support unit that determines intent of the utterance input by the user when the utterance is received through the interface unit, and provides another utterance or response corresponding to the determined intent through the interface unit.

First claim

What is claimed is:

1 . A device for providing a toolkit for an agent developer, the system comprising:
an interface unit to obtain an utterance input by a user and output the utterance; and a support unit, when the utterance input by the user is received through the interface unit, to determine intent of the received utterance, and to provide an another utterance or a response corresponding to the determined intent through the interface unit.

2 . The device of claim 1 , wherein the support unit obtains similarities between the received utterance and utterances pertaining to various intents and determines intent the received utterance on the basis of the similarities.

3 . The device of claim 1 , wherein the support unit provides the another utterance pertaining to the determined intent or the response pertaining to an answer corresponding to the determined intent.

4 . The device of claim 1 , wherein the support unit obtains a pattern of the received utterance using at least one of an entity name, an entity name word, a parameter, and a common word, obtained from the received utterance.

5 . The device of claim 4 , wherein the support unit provides the another utterance corresponding to the determined intent and having a pattern different from the pattern of the received utterance.

6 . The device of claim 5 , wherein the support unit provides the another utterance having the different pattern by performing at least one of adding, changing, and deleting on at least one of the entity name, the entity name word, the parameter, and the common word, obtained from the received utterance.

7 . The device of claim 4 , wherein, when the pattern of the received utterance is a new pattern, the support unit provides the another utterance having a pattern different from the pattern of the received utterance by reflecting the new pattern.

8 . The device of claim 4 , wherein when the pattern of the received utterance is a registered pattern, the support unit provides a notification indicating that the pattern of the received utterance is the registered pattern.

9 . The device of claim 1 , wherein, when a plurality of utterances defined as one intent by a user are input and intents of some of the plurality of utterances are different from an intent of other utterances, the support unit recommends deleting or movement some of the utterances.

10 . The device of claim 1 , when a plurality of utterances defined as one intent by a user are input and the plurality of utterances are classified in accordance with a plurality of detailed intents, the support unit recommends the plurality of detailed intents.

11 . The device of claim 1 , wherein the support unit classifies and keeps utterances registered by a user in a plurality of groups in accordance with intents, compares similarities between the received utterance and the utterances classified into a plurality of groups, and adds the received utterance to one of the plurality of groups or a new group on the basis of the similarities.

Assignees

• Lg Electronics Inc

Inventors

• Jung Seulki
• Choi Bongjun

Classifications

• G06F40/35

  Discourse or dialogue representation · CPC title

• G06F40/295

  Named entity recognition · CPC title

• G06F3/167 Primary

  Audio in a user interface, e.g. using voice commands for navigating, audio feedback · CPC title

• G10L15/22 Primary

  Procedures used during a speech recognition process, e.g. man-machine dialogue · CPC title

• G10L2015/225

  Feedback of the input speech · CPC title