JP2018151471A - Dialogue method, dialogue system, dialogue apparatus, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To continue a dialogue in a small number of scenarios regardless of a user's response to a question. A dialogue system 100 presents a first utterance, which is a question, and a second utterance, which is a utterance related to the first utterance, to a user 101. The humanoid robot 50-1 presents the first utterance. The microphone 11-1 receives the user utterance uttered by the user 101 after the first utterance. The humanoid robot 50-1 presents a second utterance in which at least a part of the content of the user's utterance is omitted from the utterance related to the first utterance. [Selection diagram] Fig. 1

Description

  The present invention relates to a technology in which a computer interacts with a human using a natural language, which can be applied to a robot that communicates with a human.

  In recent years, research and development of robots that communicate with people have progressed and have been put to practical use in various fields. For example, in the field of communication therapy, there is a usage form in which a robot is a conversation partner of a person who is lonely. Specifically, in a nursing home for the elderly, the robot can play a role of listening to the resident, so he can play a role in healing the loneliness of the resident and show a conversation with the robot. You can create conversation opportunities with the family and caregivers. Further, for example, there is a usage form in which a robot is a practice partner in a communication training field. Specifically, the foreign language learning can be efficiently advanced by having the robot become a practice partner of the foreign language learner at the foreign language learning facility. Also, for example, in application as an information presentation system, it is basic to let robots talk to each other, but by talking to people from time to time, people can participate in the conversation without being bored, and information that is easy for people to accept Can be presented. Specifically, news, product introductions, accumulation / knowledge when people have time in meeting places in the city, bus stops, station platforms, etc. or when there is room to participate in dialogues at home or in classrooms. Efficient information presentation such as introduction and education (for example, childcare / education for children, general education professor for adults, moral education, etc.) can be expected. Furthermore, for example, in application as an information collection system, there is a utilization form in which a robot collects information while talking to a person. Since communication can be maintained through communication with the robot, information can be collected without giving a sense of pressure that people are listening. Specifically, it is assumed to be applied to personal information surveys, market surveys, product evaluations, preference surveys for recommended products, and the like. As described above, various applications of human-robot communication are expected, and realization of a robot that can more naturally interact with users is expected. In addition, with the spread of smartphones, services such as LINE (registered trademark) that allow users to enjoy conversations with people by chatting in almost real time are also being implemented. If the technology of conversation with the robot is applied to this chat service, it becomes possible to realize a chat service for more natural dialogue with the user even when there is no chat partner.

  In this specification, the hardware used as a conversation partner with a user such as a robot and a chat partner used in these services, and the computer software for causing the computer to function as the hardware as a conversation partner with the user are collectively referred to. It will be called an agent. Since the agent is a conversation partner with the user, the agent may be anthropomorphic, personalized, or have personality or individuality, such as a robot or a chat partner.

  The key to the realization of these services is a technology that enables agents realized by hardware and computer software to naturally interact with humans.

  As an example of the agent described above, there is a voice dialogue system that recognizes a user's utterance, understands / infers the intention of the utterance, and responds appropriately as described in Non-Patent Document 1, for example. Research on speech dialogue systems has been actively promoted with the progress of speech recognition technology, and has been put to practical use in, for example, automatic speech response systems.

  Further, as an example of the agent, there is a scenario dialogue system that performs dialogue with a user on a specific topic according to a predetermined scenario. In the scenario dialogue system, the dialogue can be continued as long as the dialogue develops along the scenario. For example, the dialogue system described in Non-Patent Document 2 is a system that performs a dialogue between a user and a plurality of agents, including an interruption by an agent and an exchange between agents. For example, when an agent utters a question prepared for a scenario to a user, and the utterance of a user's answer to the question corresponds to an option prepared for the scenario, the agent functions to utter corresponding to the option. To do. That is, the scenario dialogue system is a dialogue system in which an agent makes an utterance based on a scenario stored in advance in the system. In this interactive system, when the agent asks the user and receives a response from the user, the conversation is swayed regardless of the content of the user's utterance, or the topic is changed by interrupting the agent. Even when the user's utterance deviates from the original topic, it is possible to respond so as not to make the user feel the story is broken.

  Further, as an example of the above-described agent, there is a chat dialogue system in which a user and an agent have a natural dialogue when the agent utters according to the content of the user's utterance. For example, in the dialogue system described in Non-Patent Document 3, words included in the utterance of the user or the agent while giving more importance to the context-specific ones in a plurality of dialogues between the user and the agent. Is a system that realizes a chat conversation between the user and the system by the system speaking according to the rules described in advance. The rules used by the chat dialogue system are not limited to those described in advance, but may be automatically generated based on the user's utterance content, or uttered in the immediate utterance by the user or agent or in the vicinity thereof. It may be automatically generated based on the utterance, or may be automatically generated based on the utterance including at least the utterance immediately before or near the utterance by the user or the agent. . Non-Patent Document 3 describes a technique for automatically generating a rule based on words that have a co-occurrence relationship or a dependency relationship with words included in a user's utterance. Further, for example, the dialogue system described in Non-Patent Document 4 is a system that reduces the cost of rule generation by fusing rules described manually and rules described using a statistical utterance generation method. The chat dialogue system is different from the scenario dialogue system because the agent does not utter the utterance according to the prepared scenario. Therefore, depending on the user's utterance, the agent's utterance does not correspond to the user's utterance. At least the content of the user's utterance, the utterance spoken immediately before or near the user or agent, or the utterance spoken immediately before or near the user or agent An agent can make an utterance based on the utterance. That is, the chat dialogue system includes at least the utterance content of the user, the utterance spoken immediately before or by the user or agent, or the utterance uttered immediately before or by the user or agent. It is a dialogue system in which an agent utters speech based on. In these chat dialogue systems, it is possible to explicitly respond to the user's utterance.

Tatsuya Kawahara, “Spoken Dialogue System by Spoken Language”, Information Processing, vol. 45, no. 10, pp. 1027-1031, October 2004 Arimoto Yasuhiro, Yoshikawa Yuichiro, Ishiguro Hiroshi, “Impression Evaluation of Speechless Speech Recognition by Multiple Robots”, Annual Conference of the Robotics Society of Japan, 2016 Hiroaki Sugiyama, Toyomi Meguro, Ryuichiro Higashinaka, Yasuhiro Minami, “Generation of response sentences using dependency and examples for user utterances with arbitrary topics”, Transactions of the Japanese Society for Artificial Intelligence, vol. 30 (1), pp. 183-194, 2015 Toyomi Meguro, Hiroaki Sugiyama, Ryuichiro Higashinaka, Yasuhiro Minami, “Construction of a dialogue system based on the fusion of rule-based utterance generation and statistical utterance generation”, Proceedings of National Conference of the Japanese Society for Artificial Intelligence, vol. 28, pp. 1-4 ,2014

  It takes a huge amount of money to build a dialogue scenario manually. Although it is possible to collect existing resources such as articles on the Web and generate utterance sentences on many topics, it is not usually a dialogue-like sentence, and since the amount of information is large, it is understood in the conversation. Often difficult.

  In order to encourage the user's understanding, the sentence generated by the method described above can be divided into multiple sentences, and multiple agents can share and speak, but if the user only listens passively, the feeling of dialogue Is reduced. Therefore, it is conceivable to improve the sense of dialogue by causing the user to speak by inserting a question to the user. However, since it is difficult to completely control the user's utterance, it is necessary to prepare a scenario for each assumed user's utterance, and it is difficult to greatly reduce the cost.

  In view of the above-described points, an object of the present invention is to realize a dialogue system and a dialogue device capable of continuing a dialogue in a small number of scenarios prepared in advance regardless of a user's response to a question from the dialogue system. That is.

  In order to solve the above-described problem, the dialogue method according to the first aspect of the present invention is a dialogue system that presents a user with a first utterance that is a question and a second utterance that is an utterance related to the first utterance. In the interactive method to be executed, a presentation unit presents a first utterance, a first presentation step, an input unit accepts a user utterance uttered by a user after the first utterance, and a presentation unit A second presentation step of presenting a second utterance in which at least a part of the content of the user utterance is omitted from the utterance related to the one utterance.

  In the dialogue method according to the second aspect of the present invention, there is provided a dialogue system for presenting a user with a first utterance that is a question for uttering a word belonging to a specific word class and a second utterance that is an utterance related to the first utterance. A dialogue method to be executed, wherein a presentation unit presents a first utterance, a first presentation step, an input unit accepts a user utterance uttered by a user after the first utterance, and a presentation unit A second presentation step for presenting an additional utterance selected from a predetermined utterance sentence according to whether or not a word belonging to the word class is included in the utterance, and a third presentation step for the presentation unit to present the second utterance And including.

  The dialogue method according to the third aspect of the present invention is a dialogue method executed by a dialogue system that presents to a user a first utterance that is a question asking a certain knowledge and a second utterance that is an utterance related to the first utterance. The presentation unit presents the first utterance, the input unit accepts the user utterance uttered by the user after the first utterance, and the presentation unit represents the knowledge in the user utterance. Depending on whether or not is included, a second presentation step of presenting an additional utterance selected from a predetermined utterance sentence and a third presentation step of presenting a second utterance by the presenting unit are included.

  According to the present invention, it is possible to return to a scenario prepared in advance regardless of the content of the user's utterance to the question presented from the dialogue system, so that the dialogue can be continued only by preparing a small number of scenarios. An interactive system and an interactive device can be realized.

FIG. 1 is a diagram illustrating a functional configuration of the interactive system according to the embodiment. FIG. 2 is a diagram illustrating a processing procedure of the interactive method according to the embodiment. FIG. 3 is a diagram illustrating a functional configuration of a dialog system according to a modification.

  In the present invention, when a dialogue is performed in accordance with a scenario stored in advance by the dialogue system, a question is asked to the user in order to improve the feeling of dialogue, and the subsequent utterance is determined according to the content of the user's utterance with respect to the question. Change the contents. If the content spoken by a user in response to a certain question is included in the content spoken immediately after by the dialog system, the same content is repeated if the speech is spoken as it is, giving a redundant impression. For this reason, the utterance content of the user is omitted from the content that the dialog system utters immediately, and the utterance is presented. In addition, when a word class spoken by a user can be assumed for a certain question, the utterance when a word in the word class is included in the user's utterance and the word in the word class are not included in the user's utterance Utterance at the time is prepared in advance, and the content to be uttered immediately after the user's utterance is switched according to whether or not the content uttered by the user is included in the word class. Furthermore, when asking a question that asks for certain knowledge, an utterance when the user's utterance includes the correct answer and an utterance when the user's utterance does not include the correct answer are prepared in advance, and the content uttered by the user is correct. The content to be uttered immediately after the user's utterance is switched depending on whether or not it is included. Conventionally, it has been necessary to prepare a large number of scenarios covering the user's response to the question from the interactive system, and the cost for that has been enormous, but by configuring as above, the content of the user's response Regardless of this, it is possible to cope with a small number of scenarios, and the cost for preparing the scenarios can be reduced.

  Hereinafter, embodiments of the present invention will be described in detail. In addition, the same number is attached | subjected to the component which has the same function in drawing, and duplication description is abbreviate | omitted.

  The dialogue system of the embodiment is a system in which at least one humanoid robot interacts with a user. That is, the dialogue system of the embodiment is an example in the case where the agent is a humanoid robot. As shown in FIG. 1, the dialogue system 100 includes, for example, a dialogue device 1, an input unit 10 including a microphone 11, and a presentation unit 50 including at least a speaker 51. The dialogue apparatus 1 includes, for example, a voice recognition unit 20, an utterance determination unit 30, and a voice synthesis unit 40. The interactive method of the embodiment is realized by the processing of each step described later by the interactive system 100.

  The interactive device 1 is a special device configured by reading a special program into a known or dedicated computer having a central processing unit (CPU), a main storage device (RAM), and the like. Device. For example, the interactive device 1 executes each process under the control of the central processing unit. The data input to the interactive device 1 and the data obtained in each process are stored in, for example, the main storage device, and the data stored in the main storage device is read out as necessary and used for other processing. The Further, at least a part of each processing unit of the interactive apparatus 1 may be configured by hardware such as an integrated circuit.

[Input unit 10]
The input unit 10 may be configured integrally or partially with the presentation unit 50. In the example of FIG. 1, microphones 11-1 and 11-2 that are part of the input unit 10 are mounted on the heads (ear positions) of the humanoid robots 50-1 and 50-2 that are the presentation unit 50. ing. In the example of FIG. 1, the presentation unit 50 is configured by two humanoid robots 50-1 and 50-2, but may be configured by a single humanoid robot, and may include three or more humanoids. You may be comprised from the robot.

  The input unit 10 is an interface for the dialog system 100 to acquire a user's utterance. In other words, the input unit 10 is an interface for inputting a user's utterance to the dialogue system 100. For example, the input unit 10 is a microphone 11 that picks up a user's voice and converts it into a voice signal. The microphone 11 only needs to be able to collect the uttered voice uttered by the user 101. That is, FIG. 1 is an example, and one of the microphones 11-1 and 11-2 may not be provided. Further, one or more microphones installed in a place different from the humanoid robots 50-1 and 50-2 such as the vicinity of the user 101 or a microphone array including a plurality of microphones is used as an input unit, and the microphone 11- It is good also as a structure which is not provided with both of 1 and 11-2. The microphone 11 outputs a voice signal of the user's uttered voice obtained by the conversion. The voice signal output from the microphone 11 is input to the voice recognition unit 20.

[Voice recognition unit 20]
The speech recognition unit 20 recognizes the speech signal of the user's speech input from the microphone 11 and converts the speech signal into text representing the user's speech content, and outputs the text to the speech determination unit 30. The speech recognition method performed by the speech recognition unit 20 may be any existing speech recognition technology, and an optimal one may be selected as appropriate in accordance with the usage environment.

[Speech determination unit 30]
The utterance determination unit 30 determines text representing the utterance content from the dialogue system 100 and outputs the text to the speech synthesis unit 40. When text representing the utterance content of the user is input from the speech recognition unit 20, the text representing the utterance content from the dialogue system 100 is determined based on the input text representing the utterance content of the user, and speech synthesis is performed. Output to the unit 40. When the presentation unit 50 of the interactive system 100 is configured with a plurality of humanoid robots, the utterance determination unit 30 may determine which humanoid robot presents the utterance. In this case, information representing the humanoid robot that presents the utterance is also output to the speech synthesizer 40. In this case, the utterance determination unit 30 determines whether to present the utterance, that is, whether to present the utterance to the user or any humanoid robot. Also good. In this case, information representing the partner who presents the utterance is also output to the speech synthesizer 40.

[Speech synthesizer 40]
The voice synthesis unit 40 converts the text representing the utterance content input from the utterance determination unit 30 into a voice signal representing the utterance content, and outputs the voice signal to the presentation unit 50. The speech synthesis method performed by the speech synthesizer 40 may be any existing speech synthesis technology, and an optimal method may be selected as appropriate according to the usage environment. In addition, when the presentation unit 50 of the interactive system 100 is configured by a plurality of humanoid robots, when information representing a humanoid robot that presents the utterance is input from the utterance determination unit 30 together with text representing the utterance content. The voice synthesizer 40 outputs a voice signal representing the utterance content to the humanoid robot corresponding to the information. In addition, when the text representing the utterance content and the information representing the humanoid robot that presents the utterance are also input from the utterance determining unit 30 together with the information representing the partner to present the utterance, the speech synthesizer 40 The voice signal representing the utterance content and the information representing the partner who presents the utterance are output to the humanoid robot corresponding to the information.

[Presentation section 50]
The presentation unit 50 is an interface for presenting the utterance content determined by the utterance determination unit 30 to the user. For example, the presentation unit 50 is a humanoid robot imitating a human shape. This humanoid robot generates a voice corresponding to the voice signal representing the utterance content input from the voice synthesizer 40, for example, from the speaker 51 mounted on the head, that is, presents the utterance. The speaker 51 only needs to be able to produce sound corresponding to the sound signal representing the utterance content input from the sound synthesizer 40. That is, FIG. 1 is an example, and one of the speakers 51-1 and 51-2 may not be provided. Further, a speaker array including one or more speakers or a plurality of speakers is installed in a place different from the humanoid robots 50-1 and 50-2 such as the vicinity of the user 101, and the speakers 51-1 and 51- are installed. 2 may be provided. In addition, the humanoid robot may present the utterance content determined by the utterance determination unit 30 based on non-verbal behavior such as facial expressions and body movements to the user. For example, when presenting consent to the previous utterance, present a non-verbal action such as swinging the head vertically and presenting the disagreement by shaking the head sideways. Is mentioned. In addition, when presenting an utterance, a humanoid robot directs the face or body toward the user or another humanoid robot, thereby allowing the user or other humanoid robot facing the face or body. In contrast, it is possible to express that an utterance is being presented. When the presentation unit 50 is a humanoid robot, for example, one humanoid robot is prepared for each personality (agent) participating in the dialogue. In the following, it is assumed that there are two humanoid robots 50-1 and 50-2 as an example in which two personalities participate in the dialogue. When the presentation unit 50 of the interactive system 100 is configured by a plurality of humanoid robots, if the utterance determination unit 30 determines from which humanoid robot the utterance is to be presented, speech synthesis is performed. The humanoid robot 50-1 or 50-2 that has received the audio signal representing the utterance content output by the unit 40 presents the utterance. Further, when information representing a partner who presents the utterance determined by the utterance determination unit 30 is also input, the humanoid robot 50-1 or 50-2 selects a humanoid corresponding to the information representing the partner presenting the utterance. The utterance is presented with the face or line of sight directed at the utterance of the robot or user.

  Hereinafter, with reference to FIG. 2, a processing procedure of the interactive method of the embodiment will be described.

  In step S1, the dialogue system 100 outputs a voice representing the content of the first utterance from the speaker 51-1 included in the humanoid robot 50-1, that is, presents the first utterance. The voice representing the content of the first utterance is obtained by converting the text representing the content of the first utterance determined by the utterance determining unit 30 into a voice signal by the voice synthesizing unit 40.

  The text representing the content of the first utterance is determined by the utterance determination unit 30 according to, for example, the content of the utterance until immediately before. As a technique for determining the utterance content according to the utterance content up to immediately before, a technique used in a conventional dialogue system may be used. For example, a scenario dialogue system described in Non-Patent Document 2 may be used. it can. When the utterance determination unit 30 uses the technology used in the scenario dialogue system, for example, the utterance determination unit 30 has a word included in each utterance and a focal point constituting each utterance with respect to a conversation including about the last five utterances. Select a scenario in which the distance between the words and the words included in each scenario stored in the storage unit (not shown) in the utterance determination unit 30 or the focal word is closer than a predetermined distance, and select the text included in the selected scenario By doing so, the text representing the content of the first utterance is determined.

  The scenario stored in the storage unit (not shown) in the utterance determination unit 30 is obtained, for example, by acquiring an existing resource such as an article on the Web, dividing it into a plurality of sentences, and converting each divided sentence into a dialogue style. It consists of a plurality of utterance sentences generated in this way. The scenario may be a scenario in which simple utterances such as a match are inserted into a plurality of generated utterance sentences in order to perform a dialogue between a plurality of agents. The scenario includes an utterance sentence that asks the user a question about the topic of an article or a question about a keyword in the article in order to improve the feeling of interaction. Hereinafter, the utterance sentence up to the question in the scenario is called the first utterance, and the utterance sentence after the question is called the post-question utterance.

  A scenario generation method stored in advance in a storage unit (not shown) in the utterance determination unit 30 will be described. For example, suppose that an article about current events with the following contents has been acquired.

A. "The transport aircraft being trained in XX city in XX prefecture crashed, wrecked, and injured."
By dividing this sentence A, converting it into a dialogue style, and including a question, for example, a scenario composed of the following utterance sentences A-1 to A-4 is generated. Here, an utterance sentence (A-2) for asking the user questions regarding knowledge about the incident is generated and included in the scenario. In this example, utterance sentences A-1 to A-2 included in the scenario are first utterances, and utterance sentences A-3 to A-4 are utterances after a question.
A-1. "Hey, in XX prefecture"
A-2. "Are you talking about transport planes?"
A-3. “△△ It ’s a city, but it crashed and was wrecked during training.
A-4. "I'm scared"

Also, for example, suppose that an article about current events with the following contents has been acquired.
B. “A long-established movie magazine“ XX ”announced the best ten of the year, and the animated film“ XXX ”was decided to be the first Japanese film to control strong players such as“ △△△ ”and“ □□□ ”.
By dividing this sentence B, converting it into a dialogue style, and including a question, for example, a scenario composed of the following utterance sentences B-1-1 to B-1-3 is generated. Here, an utterance sentence (B-1-1) for asking the user a question for answering an arbitrary movie name is generated and included in the scenario. In this example, the utterance sentence B-1-1 included in the scenario is the first utterance, and the utterance sentences B-1-2 to B-1-3 are utterances after the question.
B-1-1. "Did you watch a movie recently?"
B-1-2. “XXX was the first place.”
B-1-3. "Yes, that's amazing"

Also, for example, by dividing the above sentence B, converting it into a dialogue style and including a question, a scenario composed of the following utterance sentences B-2-1 to B-2-3 is generated. To do. Here, an utterance sentence (B-2-1) for asking the user a question for answering a specific movie name is generated and included in the scenario. In this example, the utterance sentence B-2-1 included in the scenario is the first utterance, and the utterance sentences B-2-2 to B-2-3 are utterances after the question.
B-2-1. "Recently, this year's Japanese movie Best Ten was announced, do you know what?"
B-2-2. "The correct answer was XXX! I decided on the first place."
B-2-3. "Hey, amazing"

  In step S2, the microphone 11 accepts an utterance uttered by the user 101 after presenting the first utterance. Hereinafter, this utterance is referred to as user utterance. A voice signal representing the user's utterance content acquired by the microphone 11 is input to the voice recognition unit 20. The speech recognition unit 20 recognizes a speech signal representing the user's speech content acquired by the microphone 11 and outputs the text obtained as a recognition result to the speech determination unit 30 as text representing the user's speech content.

  In step S <b> 3, the utterance determination unit 30 receives the text representing the utterance content of the user output from the speech recognition unit 20, and based on the text representing the content of the first utterance and the text representing the utterance content of the user, the second utterance The text representing the content of is determined and output to the speech synthesizer 40. The second utterance may be one utterance or a plurality of utterances. When the presentation unit 50 of the dialogue system 100 includes a plurality of humanoid robots, the utterance determination unit 30 may determine a humanoid robot that presents the second utterance, and in this case, the content of the second utterance is determined. Information representing the humanoid robot that presents the second utterance together with the text to be output is output. Moreover, the utterance determination unit 30 may determine a partner who presents the second utterance. In this case, the utterance determination unit 30 outputs information representing the partner who presents the second utterance together with text representing the content of the second utterance.

  The utterance determination unit 30 includes a text representing the content of the second utterance, among utterances included in the same scenario as the first utterance among scenarios stored in a storage unit (not shown) in the utterance determination unit 30. It is generated from text representing the content of the utterance after the question, that is, the text of the utterance sentence after the question. At the time of generation, the utterance determination unit 30 selects a method for generating the text representing the content of the second utterance according to the text representing the content of the first utterance and the text representing the content of the user's utterance. Specifically, the utterance determination unit 30 selects a method for generating text representing the content of the second utterance from the following methods.

  1. In the utterance determination unit 30, the text representing the content of the first utterance is a question asking any information about a certain topic, and the text representing the utterance content of the user uses at least a part of the text representing the content of the utterance after the question. If included, a text obtained by omitting all or part of the text representing the content of the post-question utterance is generated as text representing the content of the second utterance.

  2. When the text representing the content of the first utterance is a question asking any information belonging to a specific word class regarding a certain topic, the utterance determination unit 30 is information that the text representing the content of the user's utterance belongs to the word class. Utterance sentence stored in a storage unit (not shown) in the utterance determination unit 30, that is, a text representing the content of an additional utterance selected from a predetermined utterance sentence, What is added to the text representing the content is generated as text representing the content of the second utterance. The text representing the content of the additional utterance represents the content for transitioning the topic from the text representing the content of the user's utterance to the text representing the content of the post-question utterance.

  For example, if the text representing the utterance content of the user includes information belonging to the word class, the text representing the content of the additional utterance synchronized with the user, such as “That ’s good” or “That ’s right” The text representing the content of the utterance after the question is determined. Also, for example, when the text representing the user's utterance content does not include information belonging to the word class, the word class for presentation by a humanoid robot different from the humanoid robot that presented the first utterance The text representing the content of the additional utterance including the belonging information and the text representing the content of the post-question utterance are determined.

  3. When the text representing the content of the first utterance is a question asking a certain knowledge (that is, there is a correct answer), the utterance determination unit 30 determines whether the text representing the content of the user's utterance includes the correct answer. If it does not include the correct answer, it is determined whether or not it includes information belonging to the same word class as the correct answer, and an utterance sentence stored in a storage unit (not shown) in the utterance determination unit 30, that is, a predetermined sentence A text representing the content of the second utterance is generated by adding the text representing the content of the additional utterance selected from the utterance to the text representing the content of the utterance after the question. The text representing the content of the additional utterance represents the content for transitioning the topic from the text representing the content of the user's utterance to the text representing the content of the post-question utterance.

  For example, if the text describing the user's utterance content contains the correct answer to the question, the correct answer such as “Yes!” Or “I understand well!” For the humanoid robot that presented the first utterance. And a text representing the content of the additional utterance representing the text and a text representing the content of the post-question utterance. In addition, when the text representing the user's utterance content includes information that is not the correct answer but belongs to the same word class as the correct answer, “Yes!” “No” for the humanoid robot that presented the first utterance to present. The text indicating the content of the additional utterance indicating the incorrect answer such as “Yo!” And the text indicating the content of the utterance after the question are determined.

  Furthermore, for example, when the text representing the user's utterance content includes information belonging to the same word class as the correct answer, the incorrect answer information to be presented by a humanoid robot different from the humanoid robot that presented the first utterance The text representing the content of the additional utterance including the text and the text representing the content of the post-question utterance are determined. In addition, when the text representing the user's utterance content does not include information belonging to the same word class as the correct answer, it includes correct information for presentation by a humanoid robot different from the humanoid robot that presented the first utterance. A text representing the content of the additional utterance and a text representing the content of the post-question utterance are determined.

  In step S4, the speech synthesizer 40 synthesizes the text representing the content of the second utterance output from the utterance determination unit 30 to generate speech representing the content of the second utterance, and the humanoid robot 50-1 includes. Output from the speaker 51-2 or the speaker 51-2 included in the humanoid robot 50-2, that is, the second utterance is presented. When information representing a humanoid robot that presents the second utterance together with text representing the content of the second utterance is input from the utterance determination unit 30, the speech synthesizer 40 includes a speaker included in the humanoid robot 50 corresponding to the information. From 51, a voice representing the content of the second utterance is output. When information representing a partner who presents the second utterance is input from the utterance determining unit 30 together with text representing the content of the second utterance, the presenting unit 50 moves the humanoid robot 50 in the direction of the partner corresponding to the information. A voice representing the content of the second utterance is output with the face or the whole body facing. In FIG. 2, it is described that the second utterance is presented in step S <b> 4, but when the second utterance includes a plurality of utterances (for example, when an additional utterance and an after-question utterance are included). The content of step S4 is executed by a plurality of substeps.

  Thereafter, the dialogue system continues the dialogue with the user by performing the utterance with the content of the second utterance as a topic. For example, when the post-question utterance is an utterance in the middle of the scenario selected by the technology used in the scenario dialogue system, the dialogue according to the scenario is executed between the user and the dialogue system. The dialogue system outputs a voice representing the utterance content of the scenario utterance determined by the technology used in the scenario dialogue system from the speaker. Also, for example, when the post-question utterance is the last utterance of the scenario selected by the technique used in the scenario dialogue system, the chat decided by the technique used in the chat dialogue system based on the user's utterance A voice representing the utterance content of the utterance is output from the speaker. The humanoid robot that presents the subsequent utterances may be any one humanoid robot or a plurality of humanoid robots.

[Concrete example]
Hereinafter, specific examples of dialogue contents by the dialogue system of the embodiment will be shown. In the following description of specific examples, R represents a robot and H represents a user. The number after R is the identifier of the humanoid robot. t (i) (i = 0, 1, 2,...) represents an utterance during conversation, and i is a number representing the order of utterances. The part enclosed in parentheses during utterance represents omitting the utterance.

(Specific Example 1-1)
Specific example 1-1 is an example in which after presenting a first utterance including a question asking about arbitrary information regarding a certain topic, a second utterance in which a part of information included in the user utterance is omitted from the post-question utterance is presented. is there.
t (1) R1: Hey, in XX prefecture
t (2) H: Yeah?
t (3) R1: Are you talking about transport aircraft?
t (4) H: Ah, △△ of the city
t (5) R2: (Yes, it ’s a city), but crashed and wrecked during training and two people were injured
t (6) R1: scary

  In this example, first, the dialogue system 100 presents the first utterances t (1) and t (3) related to the topic of the sentence A. At this time, when the user utters between the first utterance t (1) and the first utterance t (3), the user utterance t (1) and the first utterance t (3) 2) may be included. Next, the dialogue system 100 determines a user utterance t (4) that is a user's utterance with respect to the first utterance t (3) that is a question asking any information about the topic of the sentence A that is the last utterance of the first utterance. get. Next, since the dialogue system 100 includes the information of “△△ city” in the user utterance t (4), the second utterance t in which the part of “△△ city is what it is” is omitted from the utterance after the question. Present (5). Thereby, it is possible to avoid giving redundant impressions with respect to “ΔΔ city”, which is the premise knowledge spoken by the user, and avoiding redundant impressions.

(Specific example 1-2)
Specific example 1-2 is an example in which after presenting a first utterance including a question asking about arbitrary information related to a certain topic, a second utterance in which a part of information included in the user utterance is omitted from the post-question utterance is presented. is there.
t (1) R1: Hey, in XX prefecture
t (2) H: Yeah?
t (3) R1: Are you talking about transport aircraft?
t (4) H: Oh, you've crashed
t (5) R2: Yes, it ’s a △ city, but during the training, it crashed and two people were injured.
t (6) R1: scary

  In this example, first, the dialogue system 100 presents first utterances t (1) and t (3) asking for arbitrary information related to the topic of the sentence A. At this time, when the user utters between the first utterance t (1) and the first utterance t (3), the user utterance t (1) and the first utterance t (3) 2) may be included. Next, the dialogue system 100 determines a user utterance t (4) that is a user's utterance with respect to the first utterance t (3) that is a question asking any information about the topic of the sentence A that is the last utterance of the first utterance. get. Next, the dialogue system 100 presents the second utterance t (5) in which the “falling” part is omitted from the post-question utterance because the user utterance t (4) includes the information “falling”. .

(Specific example 2-1)
Specific example 2-1 determines that a user utterance is an utterance including information belonging to the word class after presenting a first utterance that is a question asking about arbitrary information belonging to a specific word class regarding a certain topic. This is an example of presenting a second utterance including an additional utterance to synchronize with the user and a post-question utterance after the additional utterance.
t (1) R1: How many movies have you seen recently?
t (2) H: I saw □□□
t (3) R2: That's popular
t (4) R1: Speaking of which, XXX has been decided as the first place
t (5) R2: Yeah, awesome

  In this example, first, the dialogue system 100 presents the first utterance t (1), which is a question that can predict that the user utters a word belonging to the movie name class regarding the topic of the sentence B. Next, the dialogue system 100 acquires a user utterance t (2) that is a user's utterance for the first utterance t (1). Next, since the movie name “□□□” is included in the user utterance t (2), the dialogue system 100 adds an additional utterance t (3) indicating that the user tunes to the user like “That ’s popular”. ). Furthermore, the dialogue system 100 presents the post-question utterance t (4) to which the content of transition of the topic is added as “speaking so”. This makes it possible to return to the same topic regardless of the movie name spoken by the user, and to prevent the user from feeling that his / her utterance has been ignored.

(Specific example 2-2)
Specific example 2-2 determines that the user utterance is an utterance that does not include information belonging to the word class after presenting the first utterance that is a question asking about arbitrary information belonging to a specific word class related to a certain topic. This is an example in which a second utterance including an additional utterance including information belonging to the word class and a post-question utterance after the additional utterance is presented.
t (1) R1: How many movies have you seen recently?
t (2) H: I haven't seen a movie recently
t (3) R2: I saw □□□
t (4) R1: That's popular. Speaking of which, XXX was the first place
t (5) R2: Yeah, awesome

  In this example, first, the dialogue system 100 presents the first utterance t (1), which is a question that can predict that the user utters a word belonging to the movie name class regarding the topic of the sentence B. Next, the dialogue system 100 acquires a user utterance t (2) that is a user's utterance for the first utterance t (1). Next, since the dialogue system 100 does not include the movie name in the user utterance t (2), the humanoid robot R2 other than the humanoid robot R1 that has presented the question t (1) has a movie “□□□”. An additional utterance t (3) including the name is presented. Furthermore, the dialogue system 100 returns the original topic by presenting post-question utterances t (4) and t (5) regarding the additional utterance t (3) between the humanoid robots.

(Specific example 3-1)
In specific example 3-1, after presenting the first utterance which is a question with a correct answer, it is determined that the user utterance is an utterance including the correct answer, an additional utterance including an incorrect answer, and an additional utterance after the additional utterance. It is an example which shows the 2nd utterance containing the utterance after a question.
t (1) R1: Recently, this year's Japanese movie top ten was announced, do you know what?
t (2) H: XXX
t (3) R2: Is it □□□?
t (4) R1: The correct answer was XXX! I decided on the first place
t (5) R2: Hey, amazing

  In this example, first, the dialogue system 100 presents a first utterance t (1), which is a question asking a movie name determined as a Japanese movie best ten on the topic of the sentence B. Next, the dialogue system 100 acquires a user utterance t (2) that is a user's utterance for the first utterance t (1). Next, since the correct utterance name “xxx” is included in the user utterance t (2), the dialogue system 100 has a human type other than the humanoid robot R1 that presented the first utterance t (1). An additional utterance t (3) including an incorrect movie name “□□□” is presented from the robot R2. Furthermore, the dialogue system 100 returns the original topic by presenting post-question utterances t (4) and t (5) regarding the additional utterance t (3) between the humanoid robots. If you present a post-question utterance t (4) that reveals the correct answer immediately after the user's utterance t (2), including the correct answer, the user will get the impression that his utterance t (2) was ignored, so avoid this Therefore, the additional utterance t (3) of the incorrect answer is presented by the humanoid robot R2.

(Specific Example 3-2)
Specific Example 3-2 shows that after presenting the first utterance that is a question with a correct answer, it is determined that the user utterance is an utterance that does not include the correct answer, and an additional utterance including the same word class as the correct answer is added. It is an example which presents the 2nd utterance including the utterance after the question after the utterance.
t (1) R1: Recently this year's Japanese movie Best Ten was announced, do you know what?
t (2) H: □□□
t (3) R2: Is it □□□?
t (4) R1: The correct answer was XXX! I decided on the first place
t (5) R2: Hey, amazing

  In this example, first, the dialogue system 100 presents a first utterance t (1), which is a question asking a movie name determined as a Japanese movie best ten on the topic of the sentence B. Next, the dialogue system 100 acquires a user utterance t (2) that is a user's utterance for the first utterance t (1). Next, since the user's utterance t (2) includes an incorrect movie name “□□□” in the dialogue system 100, a person other than the humanoid robot R1 who presented the first utterance t (1). The additional utterance t (3) including the incorrect movie name “□□□” is presented from the robot R2. Furthermore, the dialogue system 100 returns the original topic by presenting post-question utterances t (4) and t (5) regarding the additional utterance t (3) between the humanoid robots. In this case, the additional utterance t (3) of the humanoid robot R2 may be correct or incorrect as long as it belongs to the word class of the movie name.

(Specific Example 3-3)
Specific Example 3-3 determines that the user utterance is an utterance that does not include the correct answer after presenting the first utterance that is a question for which there is a correct answer, but that includes information belonging to the same word class as the correct answer. In this example, the second utterance including the additional utterance indicating the incorrect answer and the post-question utterance after the additional utterance is presented.
t (1) R1: Recently this year's Japanese movie Best Ten was announced, do you know what?
t (2) H: □□□
t (3) R1: Yes, the correct answer was XXX! I decided on the first place
t (4) R2: Hey, amazing

  In this example, first, the dialogue system 100 presents a first utterance t (1), which is a question asking a movie name determined as a Japanese movie best ten on the topic of the sentence B. Next, the dialogue system 100 acquires a user utterance t (2) that is a user's utterance for the first utterance t (1). Next, since the incorrect utterance movie name “□□□” is included in the user utterance t (2), the dialogue system 100 receives “Yeah” from the humanoid robot R1 that presented the first utterance t (1). After adding an additional utterance indicating that it is an incorrect answer such as “”, a post-question utterance t (3) that reveals the correct answer is presented. If the correct movie name is included in the user utterance t (2), an utterance after the question that reveals the correct answer is added by adding an additional utterance indicating that the answer is correct, such as "I understand well!" Even if t (3) is presented, a redundant impression can be avoided.

(Specific Example 3-4)
Specific example 3-4 determines that the user utterance is an utterance that does not include the correct answer after presentation of the first utterance that is a question with a correct answer, and does not include information that belongs to the same word class as the correct answer. In this example, the second utterance including the additional utterance including the same word class as the correct answer and the post-question utterance after the additional utterance is presented.
t (1) R1: Recently this year's Japanese movie Best Ten was announced, do you know what?
t (2) H: I don't know because I haven't seen a movie recently.
t (3) R2: Is it □□□?
t (4) R1: The correct answer was XXX! I decided on the first place
t (5) R2: Hey, amazing

  In this example, first, the dialogue system 100 presents a first utterance t (1), which is a question asking a movie name determined as a Japanese movie best ten on the topic of the sentence B. Next, the dialogue system 100 acquires a user utterance t (2) that is a user's utterance for the first utterance t (1). Next, since the user utterance t (2) does not include a word belonging to the word class of the movie name, the dialogue system 100 has a humanoid robot R2 other than the humanoid robot R1 that presented the first utterance t (1). Presents an additional utterance t (3) that includes the incorrect movie name “□□□”. Furthermore, the dialogue system 100 returns the original topic by presenting post-question utterances t (4) and t (5) regarding the additional utterance t (3) between the humanoid robots. The additional utterance t (3) presented by the humanoid robot R2 may include the correct movie name, and in that case, the utterance of the humanoid robot R1 is correct as "Yes!" An additional utterance to represent may be added.

[Modification]
In the above-described embodiment, an example in which a voice conversation is performed using a humanoid robot as an agent has been described. However, even if the presentation unit of the above-described embodiment is a humanoid robot having a body or the like, it has a body or the like. There may be no robot. In addition, the dialogue technique of the present invention is not limited to these, and it is also possible to adopt a form in which a dialogue is performed using an agent that does not have an entity such as a human body and does not have an utterance mechanism like a humanoid robot. As such a form, for example, a form in which dialogue is performed using an agent displayed on a computer screen can be cited. More specifically, in a group chat in which multiple accounts interact by text messages, such as “LINE” (registered trademark) and “2 channel” (registered trademark), the user's account and the dialog device account interact. It is also possible to apply to the form which performs. In this form, the computer having the screen for displaying the agent needs to be in the vicinity of the person, but the computer and the interactive device may be connected via a network such as the Internet. That is, this dialogue system can be applied not only to a dialogue in which speakers such as a person and a robot actually talk each other but also to a conversation in which the speakers communicate via a network.

  As shown in FIG. 3, the interactive system 200 according to the modification includes, for example, a single interactive device 2. The interactive apparatus 2 according to the modification includes, for example, an input unit 10, a voice recognition unit 20, an utterance determination unit 30, and a presentation unit 50. The interactive device 2 may include a microphone 11 and a speaker 51, for example.

  The interactive apparatus 2 according to the modified example is an information processing apparatus such as a mobile terminal such as a smartphone or a tablet, or a desktop or laptop personal computer. In the following description, it is assumed that the interactive device 2 is a smartphone. The presentation unit 50 is a liquid crystal display included in the smartphone. A chat application window is displayed on the liquid crystal display, and conversation contents of the group chat are displayed in time series in the window. The group chat is a function in which a plurality of accounts post a text message to each other and develop a conversation in the chat. It is assumed that a plurality of virtual accounts corresponding to a virtual personality controlled by the dialogue apparatus 2 and a user account participate in this group chat. That is, this modification is an example in which the agent is a virtual account displayed on a liquid crystal display of a smartphone that is an interactive device. The user can input the utterance content to the input unit 10 which is an input area provided in the group chat window using the software keyboard, and can post to the group chat through his / her account. The utterance determination unit 30 determines the utterance content from the dialogue apparatus 2 based on the posting from the user's account, and posts it to the group chat through each virtual account. In addition, it is good also as a structure which uses the microphone 11 mounted in the smart phone and a speech recognition function, and a user inputs speech content to the input part 10 by speech. Moreover, it is good also as a structure which outputs the utterance content obtained from each dialog system from the speaker 51 by the audio | voice corresponding to each virtual account, using the speaker 51 mounted in the smart phone and a voice synthesizing function.

  As described above, the embodiments of the present invention have been described, but the specific configuration is not limited to these embodiments, and even if there is a design change or the like as appropriate without departing from the spirit of the present invention, Needless to say, it is included in this invention. The various processes described in the embodiments are not only executed in time series according to the order described, except for the utterance order presented by the presentation unit, but also in parallel according to the processing capability of the apparatus that executes the process or as necessary. Or may be performed individually.

[Program, recording medium]
When various processing functions in each device described in the above embodiment are realized by a computer, the processing contents of the functions that each device should have are described by a program. Then, by executing this program on a computer, various processing functions in each of the above devices are realized on the computer.

  The program describing the processing contents can be recorded on a computer-readable recording medium. As the computer-readable recording medium, for example, any recording medium such as a magnetic recording device, an optical disk, a magneto-optical recording medium, and a semiconductor memory may be used.

  This program is distributed by selling, transferring, or lending a portable recording medium such as a DVD or CD-ROM in which the program is recorded. Furthermore, the program may be distributed by storing the program in a storage device of the server computer and transferring the program from the server computer to another computer via a network.

  A computer that executes such a program first stores, for example, a program recorded on a portable recording medium or a program transferred from a server computer in its own storage device. When executing the process, the computer reads a program stored in its own recording medium and executes a process according to the read program. As another execution form of the program, the computer may directly read the program from a portable recording medium and execute processing according to the program, and the program is transferred from the server computer to the computer. Each time, the processing according to the received program may be executed sequentially. A configuration in which the above-described processing is executed by a so-called ASP (Application Service Provider) type service that realizes a processing function only by an execution instruction and result acquisition without transferring a program from the server computer to the computer. It is good. Note that the program in this embodiment includes information that is used for processing by an electronic computer and that conforms to the program (data that is not a direct command to the computer but has a property that defines the processing of the computer).

  In this embodiment, the present apparatus is configured by executing a predetermined program on a computer. However, at least a part of these processing contents may be realized by hardware.

1, 2 Utterance device 10 Input unit 11 Microphone 20 Speech recognition unit 30 Utterance determination unit 40 Speech synthesis unit 50 Presentation unit 51 Speaker 100, 200 Dialogue system 101 User

Claims (12)

A dialogue method executed by a dialogue system for presenting a user with a first utterance that is a certain question and a second utterance that is an utterance related to the first utterance,
A first presentation step in which the presenting unit presents the first utterance;
An utterance receiving step in which the input unit receives a user utterance uttered by the user after the first utterance;
A second presentation step in which the presenting unit presents the second utterance in which at least a part of the content of the user utterance is omitted from the utterance related to the first utterance;
Interactive method including A dialogue method executed by a dialogue system that presents a user with a first utterance that is a question that utters a word belonging to a specific word class and a second utterance that is an utterance related to the first utterance,
A first presentation step in which the presenting unit presents the first utterance;
An utterance receiving step in which the input unit receives a user utterance uttered by the user after the first utterance;
A second presentation step in which the presenting unit presents an additional utterance selected from a predetermined utterance sentence according to whether or not the user utterance includes a word belonging to the word class;
A third presentation step in which the presenting unit presents the second utterance;
Interactive method including A dialogue method executed by a dialogue system that presents to a user a first utterance that is a question asking a certain knowledge and a second utterance that is an utterance related to the first utterance,
A first presentation step in which the presenting unit presents the first utterance;
An utterance receiving step in which the input unit receives a user utterance uttered by the user after the first utterance;
A second presentation step in which the presenting unit presents an additional utterance selected from a predetermined utterance sentence according to whether or not the word representing the knowledge is included in the user utterance;
A third presentation step in which the presenting unit presents the second utterance;
Interactive method including The dialogue method according to claim 3,
In the case where the user utterance does not include a word representing the knowledge, the second presentation step determines whether the user utterance includes a word belonging to the same word class as the word representing the knowledge. Select and present additional utterances,
How to interact. An interactive system for presenting a user with a first utterance that is a certain question and a second utterance that is an utterance related to the first utterance,
An input unit for receiving a user utterance uttered by the user after the first utterance;
An utterance determination unit for determining the second utterance, in which at least a part of the content of the user utterance is omitted from the utterance related to the first utterance and the first utterance;
Presenting the first utterance and presenting the second utterance after accepting the user utterance;
Interactive system including An interactive system for presenting a user with a first utterance that is a question that utters a word belonging to a specific word class and a second utterance that is an utterance related to the first utterance,
An input unit for receiving a user utterance uttered by the user after the first utterance;
An utterance determination unit that determines an additional utterance selected from a predetermined utterance sentence and the second utterance according to whether the first utterance and a word belonging to the word class are included in the user utterance When,
Presenting the first utterance, presenting the additional utterance after accepting the user utterance, and presenting the second utterance;
Interactive system including An interactive system for presenting a user with a first utterance that is a question asking a certain knowledge and a second utterance that is an utterance related to the first utterance,
An input unit for receiving a user utterance uttered by the user after the first utterance;
An utterance determination unit that determines the first utterance, an additional utterance selected from a predetermined utterance sentence, and the second utterance according to whether or not the word representing the knowledge is included in the user utterance; ,
Presenting the first utterance, presenting the additional utterance after accepting the user utterance, and presenting the second utterance;
Including system. An interactive device for determining an utterance to be presented by an interactive system including at least an input unit that receives a user's utterance and a presentation unit that presents the utterance,
A first utterance that is a certain question and an utterance related to the first utterance, and the second utterance in which at least a part of the content of the user utterance uttered by the user after the first utterance is omitted are determined. An interactive device that includes an utterance determination unit. An interactive device for determining an utterance to be presented by an interactive system including at least an input unit that receives a user's utterance and a presentation unit that presents the utterance,
A first utterance which is a question for uttering a word belonging to a specific word class, and an additional utterance selected from a predetermined utterance sentence according to whether or not the word belonging to the word class is included in the user utterance; An interactive device including the second utterance and an utterance determination unit for determining the second utterance. An interactive device for determining an utterance to be presented by an interactive system including at least an input unit that receives a user's utterance and a presentation unit that presents the utterance,
A first utterance which is a question asking a certain knowledge, an additional utterance selected from a predetermined utterance sentence according to whether or not the word representing the knowledge is included in the user utterance, and the second utterance An interactive device including an utterance determining unit for determining.   A program for causing a computer to execute each step of the interactive method according to claim 1.   A program for causing a computer to function as the interactive device according to claim 8.

Images