Conversation controller
First Claim
1. A conversation controller configured to output a reply sentence according to a user utterance, comprising:
- a plan memory unit comprising a memory, the plan memory unit storing a plurality of plans, wherein each plan has a reply sentence and one or more pieces of next candidate designation information for designating a next candidate reply sentence to be output following the reply sentence;
a plan conversation unit comprising a processor, the plan conversation unit;
selecting one of the plurality of plans stored in the plan memory unit according to a first user utterance and which outputs a reply sentence which the selected plan includes, andselecting one piece of the next candidate designation information which the plan includes according to a second user utterance and which outputs a next candidate reply sentence corresponding to the selected piece of the next candidate designation information, wherein the plan conversation unit withholds an output of the next candidate reply sentence when the second user utterance is received, which is not related to the next candidate reply sentence or it is unclear whether or not there is a relation between the second user utterance and the next candidate reply sentence;
a discourse space conversation unit that outputs a reply sentence about a topic which is not related to the withheld next candidate reply sentence according to the second user utterance;
a morpheme extracting unit that extracts, based on a character string corresponding to the second user utterance, at least one morpheme comprising a minimum unit of the character string, as first morpheme information; and
a conversation database that stores;
a plurality of pieces of topic identification information for identifying a topic,a plurality of pieces of second morpheme information, each of which includes a morpheme including a character, a string of characters or a combination of the character and the string of characters, anda plurality of reply sentences,wherein one piece of topic identification information is associated with one or more pieces of second morpheme information and at least one piece of second morpheme information is associated with one reply sentence,wherein the discourse space conversation unit comprises;
a topic identification information retrieval portion that compares, based on first morpheme information extracted by the morpheme extracting unit, the first morpheme information with the plurality of pieces of topic identification information, and when the topic identification information retrieval portion retrieves topic identification information including a part of the first morpheme information from among the plurality of pieces of topic identification information and the retrieved topic identification information includes second morpheme information including a part of the first morpheme information, the topic identification information retrieval portion outputs the second morpheme information to the reply retrieval portion;
an elliptical sentence supplementation portion that, when the retrieved topic identification information does not include second morpheme information including a part of the first morpheme information, adds topic identification information previously retrieved by the topic identification information retrieval portion to the first morpheme information to provide supplemented first morpheme information;
a topic retrieval portion that, compares the supplemented first morpheme information with one or more pieces of second morpheme information associated with the retrieved topic identification information, retrieves second morpheme information associated with the retrieved topic identification information including a part of the supplemented first morpheme information from among the one or more pieces of second morpheme information and out puts the retrieved second morpheme information to the reply retrieval portion; and
a reply retrieval portion that retrieves, based on the second morpheme information from one of the topic identification information retrieval portion and the topic retrieval portion, a reply sentence associated with the retrieved second morpheme information,wherein some plans have a plurality of reply sentences into which one explanatory sentence is divided.
4 Assignments
0 Petitions
Accused Products
Abstract
A conversation controller outputs a reply sentence according to a user utterance. The conversation controller comprises a conversation database and a conversation control unit. The conversation database stores a plurality of plans. Each plan has a reply sentence and one or more pieces of next plan designation information for designating a next candidate reply sentence to be output following the reply sentence. The conversation control unit selects one of the plans stored in the conversation database according to a user utterance and outputs a reply sentence which the selected plan has. Then, the conversation control unit selects one piece of the next plan designation information which the plan has according to a next user utterance and outputs a next candidate reply sentence on the basis of the selected piece of the next plan designation information. Some plans have a plurality of reply sentences into which one explanatory sentence is divided.
33 Citations
9 Claims
-
1. A conversation controller configured to output a reply sentence according to a user utterance, comprising:
-
a plan memory unit comprising a memory, the plan memory unit storing a plurality of plans, wherein each plan has a reply sentence and one or more pieces of next candidate designation information for designating a next candidate reply sentence to be output following the reply sentence; a plan conversation unit comprising a processor, the plan conversation unit; selecting one of the plurality of plans stored in the plan memory unit according to a first user utterance and which outputs a reply sentence which the selected plan includes, and selecting one piece of the next candidate designation information which the plan includes according to a second user utterance and which outputs a next candidate reply sentence corresponding to the selected piece of the next candidate designation information, wherein the plan conversation unit withholds an output of the next candidate reply sentence when the second user utterance is received, which is not related to the next candidate reply sentence or it is unclear whether or not there is a relation between the second user utterance and the next candidate reply sentence; a discourse space conversation unit that outputs a reply sentence about a topic which is not related to the withheld next candidate reply sentence according to the second user utterance; a morpheme extracting unit that extracts, based on a character string corresponding to the second user utterance, at least one morpheme comprising a minimum unit of the character string, as first morpheme information; and a conversation database that stores; a plurality of pieces of topic identification information for identifying a topic, a plurality of pieces of second morpheme information, each of which includes a morpheme including a character, a string of characters or a combination of the character and the string of characters, and a plurality of reply sentences, wherein one piece of topic identification information is associated with one or more pieces of second morpheme information and at least one piece of second morpheme information is associated with one reply sentence, wherein the discourse space conversation unit comprises; a topic identification information retrieval portion that compares, based on first morpheme information extracted by the morpheme extracting unit, the first morpheme information with the plurality of pieces of topic identification information, and when the topic identification information retrieval portion retrieves topic identification information including a part of the first morpheme information from among the plurality of pieces of topic identification information and the retrieved topic identification information includes second morpheme information including a part of the first morpheme information, the topic identification information retrieval portion outputs the second morpheme information to the reply retrieval portion; an elliptical sentence supplementation portion that, when the retrieved topic identification information does not include second morpheme information including a part of the first morpheme information, adds topic identification information previously retrieved by the topic identification information retrieval portion to the first morpheme information to provide supplemented first morpheme information; a topic retrieval portion that, compares the supplemented first morpheme information with one or more pieces of second morpheme information associated with the retrieved topic identification information, retrieves second morpheme information associated with the retrieved topic identification information including a part of the supplemented first morpheme information from among the one or more pieces of second morpheme information and out puts the retrieved second morpheme information to the reply retrieval portion; and a reply retrieval portion that retrieves, based on the second morpheme information from one of the topic identification information retrieval portion and the topic retrieval portion, a reply sentence associated with the retrieved second morpheme information, wherein some plans have a plurality of reply sentences into which one explanatory sentence is divided. - View Dependent Claims (2, 3)
-
-
4. A non-transitory computer-readable medium encoded with computer-executable instructions which, when executed by a processor, cause the processor to:
-
designate a next candidate reply sentence to be output following a reply sentence according to a plan of a plurality of plans, wherein each plan has a reply sentence and one or more pieces of next candidate designation information; select one of the plurality of plans according to a first user utterance and output a reply sentence which the selected plan has, and select one piece of the next candidate designation information which the plan has according to a second user utterance and output a next candidate reply sentence corresponding to the selected piece of the next candidate designation information, wherein some plans have a plurality of reply sentences into which one explanatory sentence is divided; withhold an output of the next candidate reply sentence when receiving the second user utterance which is not related to the next candidate reply sentence or it is unclear whether or not there is a relation between the second user utterance and the next candidate reply sentence; output a reply sentence about a topic which is not related to the withheld next candidate reply sentence according to the second user utterance; extract, based on a character string corresponding to the second user utterance, at least one morpheme comprising a minimum unit of the character string, as first morpheme information; store a plurality of pieces of topic identification information for identifying a topic, a plurality of pieces of second morpheme information, each of which includes a morpheme including a character, a string of characters or a combination of the character and the string of characters, and a plurality of reply sentences, wherein one piece of topic identification information is associated with one or more pieces of the second morpheme information, and at least one piece of the second morpheme information is associated with one reply sentence; compare, based on first morpheme information extracted the first morpheme information with the plurality of pieces of topic identification information when topic identification information including a part of the first morpheme information is retrieved from among the plurality of pieces of topic identification and the retrieved topic identification information includes second morpheme information including a part of the first morpheme information, output the second morpheme information; when the retrieved topic identification information does not include second morpheme information including a part of the first morpheme information, add topic identification information previously retrieved to the first morpheme information to provide supplemented first morpheme information; compare the supplemented first morpheme information with one or more pieces of second morpheme information associated with the retrieved topic identification information, retrieve second morpheme information associated with the retrieved topic identification information including a part of the supplemented first morpheme information from among the one or more pieces of second morpheme information, and output the retrieved second morpheme information; retrieve, based on the output second morpheme information, a reply sentence associated with the retrieved second morpheme information; and output the retrieved reply sentence according to a user utterance. - View Dependent Claims (5, 6, 7)
-
-
8. A conversation controller configured to output a reply sentence according to a user utterance, comprising:
-
a plan memory unit comprising a memory, the plan memory unit storing a plurality of plans, wherein each plan has a reply sentence and one or more pieces of next candidate designation information for designating a next candidate reply sentence to be output following the reply sentence; a plan conversation unit comprising a processor, the plan conversation unit; selecting one of the plurality of plans stored in the plan memory unit according to a first user utterance and which outputs a reply sentence which the selected plan includes, and selecting one piece of the next candidate designation information which the plan includes according to a second user utterance and which outputs a next candidate reply sentence corresponding to the selected piece of the next candidate designation information, wherein the plan conversation unit withholds an output of the next candidate reply sentence when the second user utterance is received, which is not related to the next candidate reply sentence or it is unclear whether or not there is a relation between the second user utterance and the next candidate reply sentence; a discourse space conversation unit that outputs a reply sentence about a topic which is not related to the withheld next candidate reply sentence according to the second user utterance; a morpheme extracting unit that extracts, based on a character string corresponding to the second user utterance, at least one morpheme comprising a minimum unit of the character string, as first morpheme information; and a conversation database that stores; a plurality of pieces of topic identification information for identifying a topic, a plurality of pieces of second morpheme information, each of which includes a morpheme including a character, a string of characters or a combination of the character and the string of characters, and a plurality of reply sentences, wherein one piece of topic identification information is associated with one or more pieces of second morpheme information and at least one piece of second morpheme information is associated with one reply sentence, wherein the discourse space conversation unit comprises; a topic identification information retrieval portion that compares, based on first morpheme information extracted by the morpheme extracting unit, the first morpheme information with the plurality of pieces of topic identification information, and when the topic identification information retrieval portion retrieves topic identification information including a part of the first morpheme information from among the plurality of pieces of topic identification information and the retrieved topic identification information includes second morpheme information including a part of the first morpheme information, the topic identification information retrieval portion outputs the second morpheme information to the reply retrieval portion; an elliptical sentence supplementation portion that, when the retrieved topic identification information does not include second morpheme information including a part of the first morpheme information, adds topic identification information previously retrieved by the topic identification information retrieval portion to the first morpheme information to generate supplemented first morpheme information, retrieves second morpheme information including a part of the supplemented first morpheme information, and outputs the retrieved second morpheme information to the reply retrieval portion; and a reply retrieval portion that retrieves, based on the second morpheme information from one of the topic identification information retrieval portion and the elliptical sentence supplementation portion, a reply sentence associated with the retrieved second morpheme information, wherein some plans have a plurality of reply sentences into which one explanatory sentence is divided.
-
-
9. A non-transitory computer-readable medium encoded with computer-executable instructions which, when executed by a processor, cause the processor to:
-
designate a next candidate replay sentence to be output following a reply sentence according to a plan of a plurality of plans, wherein each plan has a reply sentence and one or more pieces of next candidate designation information; select one of the plurality of plans according to a first user utterance and output a reply sentence which the selected plan has, and select one piece of the next candidate designation information which the plan has according to a second user utterance and output a next candidate reply sentence corresponding to the selected piece of the next candidate designation information, wherein some plans have a plurality of reply sentences into which one explanatory sentence is divided; withhold an output of the next candidate reply sentence when receiving the second user utterance which is not related to the next candidate reply sentence or it is unclear whether or not there is a relation between the second user utterance and the next candidate reply sentence; output a reply sentence about a topic which is not related to the withheld next candidate reply sentence according to the second user utterance; extract, based on a character string corresponding to the second user utterance, at least one morpheme comprising a minimum unit of the character string, as first morpheme information; store a plurality of pieces of topic identification information for identifying a topic, a plurality of pieces of second morpheme information, each of which includes a morpheme including a character, a string of characters or a combination of the character and the string of characters, and a plurality of reply sentences, wherein one piece of topic identification information is associated with one or more pieces of the second morpheme information, and at least one piece of the second morpheme information is associated with one reply sentence; compare, based on first morpheme information extracted, the first morpheme information with the plurality of pieces of topic identification information; when topic identification information including a part of the first morpheme information is retrieved from among the plurality of pieces of topic identification information and the retrieved topic identification information includes second morpheme information including a part of the first morpheme information, output the second morpheme information; when the retrieved topic identification information does not include second morpheme information including a part of the first morpheme information, add topic identification information previously retrieved to the first morpheme information to generate supplemented first morpheme information; retrieve second morpheme information including a part of the supplemented first morpheme information and output the retrieved second morpheme information; retrieve, based on the output second morpheme information, a reply sentence associated with the retrieved second morpheme information; and output the retrieved reply sentence according to a user utterance.
-
Specification