JPWO2017200076A1 - Dialogue method, dialogue system, dialogue apparatus, and program - Google Patents

Abstract

  Improve responsiveness and keep the dialogue going for a long time. The dialogue system 10 includes at least an input unit 1 that receives a user's utterance and a presentation unit 5 that presents the utterance. The presentation unit 5-1 or the presentation unit 5-2 presents an open utterance. The input unit 1 accepts a user utterance for an open utterance. The presentation unit 5-1 presents the utterance determined based on the user's utterance by the chat dialogue system 6 that determines at least the utterance based on the utterance content of the user. The presentation unit 5-2 presents the utterance determined by the scenario dialog system 7 that determines the utterance based on the scenario stored in advance after the utterance determined by the chat conversation system 6.

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 communication between humans and robots are assumed, and it is expected to realize a robot that can more naturally interact with users. 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 above-described 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 1 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 speaks a question prepared in a scenario to a user, and an utterance of a user's answer to the question corresponds to an option prepared in the scenario, the agent functions to perform an utterance corresponding to the option. . 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 2, the words included in the utterance of the user or the agent while giving more importance to the context-specific conversation in the multi-turn dialogue performed 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 2 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 3 is a system that reduces the cost of rule generation by fusing rules described manually and rules described by 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.

Arimoto Yasuhiro, Yoshikawa Yuichiro, Ishiguro Hiroshi, “Impression Evaluation of Speechless Speech Recognition by Multiple Robots”, Annual Conference of the Robotics Society of Japan, 2016 Sugiyama Hiroaki, Meguro Toyomi, Higashinaka Ryuichiro, Minami Yasuhiro, “Generation of Responses Using Dependence 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

  In the conventional chat dialogue system, since it is basically one question one answer, it is not possible to continue the dialogue with a consistent story. In a conventional scenario dialogue system, there are cases where a response cannot be made if the user's utterance is outside the range assumed in the scenario. In the dialogue system of Non-Patent Document 1, the user may feel that his / her utterance is being played.

  In view of the above points, an object of the present invention is to provide a dialogue technique that can improve a user's responsiveness to an utterance and can continue the dialogue for a long time.

  In order to solve the above-described problem, in the dialogue method according to the first aspect of the present invention, the dialogue system that determines an utterance based on a scenario stored in advance is called a first dialogue system, and at least the utterance content of the user is included. When the dialogue system for determining the utterance based on the second dialogue system is called a second dialogue system, the presentation unit presents a first presentation step for presenting an open utterance, and the input unit accepts a user utterance for the open utterance. A second presentation step in which the presentation unit presents the utterance determined by the second dialogue system based on the user's utterance; and the presentation unit displays the utterance decided by the first dialogue system after the second presentation step. A third presentation step to present.

  In the dialog system according to the second aspect of the present invention, the dialog system that determines the utterance based on the scenario stored in advance is called the first dialog system, and the dialog system that determines the utterance based on at least the user's utterance content. When the second dialogue system is called, an open utterance is decided, and after the open utterance, the utterance is decided by the second dialogue system based on the user's utterance to the open utterance, and after the utterance decided by the second dialogue system An utterance determination unit that determines an utterance by the first dialog system, an input unit that accepts a user's utterance for an open utterance, and an open utterance determined by the utterance determination unit are presented, and the utterance determination unit is determined by the second dialog system Utterance is presented after the user's utterance, and the utterance determined by the utterance determination unit by the first dialog system is displayed in the second dialog. Including a presentation unit that presents after the utterance stem is determined, the.

  A dialog device according to a third aspect of the present invention is a dialog device that determines an utterance to be presented by a dialog system including at least an input unit that receives a user's utterance and a presentation unit that presents the utterance, and is stored in advance. The dialog system that determines the utterance based on the scenario is called the first dialog system, and at least the dialog system that determines the utterance based on the utterance content of the user is called the second dialog system, the open utterance is determined. And an utterance determination unit that determines an utterance by the second dialog system based on the user's utterance for the open utterance after the open utterance, and determines an utterance by the first dialog system after the utterance determined by the second dialog system.

  According to the present invention, an appropriate response can be made to a user's utterance by an interactive system that utters based on at least the content of the user's utterance, and thereafter an interactive system that performs utterance based on a pre-stored scenario is determined. It is possible to realize a dialogue system and a dialogue device that can give a high responsiveness while making the user feel a story and can continue a dialogue with the user for a long time. It becomes.

FIG. 1 is a diagram illustrating a functional configuration of a dialogue system using a humanoid robot according to the first embodiment. FIG. 2 is a diagram illustrating a processing procedure of the interactive method according to the first embodiment. FIG. 3 is a diagram illustrating a functional configuration of the dialogue system using the humanoid robot according to the second embodiment. FIG. 4 is a diagram illustrating a processing procedure of the interactive method according to the second embodiment. FIG. 5 is a diagram illustrating a functional configuration of the interactive system using group chat according to the first embodiment.

  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.

<First embodiment>
The dialogue system of the first embodiment is a system in which a plurality of humanoid robots cooperate with each other to interact with a user. That is, the dialogue system of the first embodiment is an example in the case where the agent is a humanoid robot. As shown in FIG. 1, the dialogue system 10 includes an input unit 1, a speech recognition unit 2, an utterance determination unit 3, a speech synthesis unit 4, and a presentation unit 5. The interactive method of the first embodiment is realized by the processing of each step described later by the interactive system 10. As shown in FIG. 1, a part formed by the speech recognition unit 2, the utterance determination unit 3, and the speech synthesis unit 4 of the dialogue system 10 is a dialogue device 11 of the first embodiment. The utterance determination unit 3 includes an interface capable of communicating with the chat dialogue system 6 and the scenario dialogue system 7 existing outside. The chat dialogue system 6 and the scenario dialogue system 7 may be configured in the dialogue apparatus 11 as processing units having similar functions. The chat dialogue system 6 is an example of a dialogue system in which an agent utters at least based on the content of the user's utterance, and the scenario dialogue system 7 is an example of a dialogue system in which the agent performs utterance based on a scenario stored in advance. is there.

  In order to improve the sense of response, the dialogue system 10 of the present embodiment presents the utterance determined by the chat dialogue system 6 based on the user's utterance to the open utterance, and subsequently chats the utterance decided by the scenario dialogue system 7. Presented after the utterance determined by the dialog system 6. Since the chat dialogue system 6 determines the utterance based on at least the user's utterance content, it can explicitly reply to the user's utterance. For this reason, it is possible to avoid giving the user a feeling that his / her remarks are being shed, as compared with the case of simply hitting a conflict such as “Soka”. Thereby, it is possible to give the user an impression that the dialogue system 10 is seriously responding, and it is possible to continue the dialogue for a long time in the subsequent scenario dialogue.

  The interactive device 11 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 apparatus 11 executes each process under the control of the central processing unit. The data input to the dialogue device 11 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 11 may be configured by hardware such as an integrated circuit.

  The input unit 1 is an interface for the dialog system 10 to acquire a user's utterance. In other words, the input unit 1 is an interface for the user to input an utterance to the dialogue system 10. For example, the input unit 1 is a microphone that picks up a user's speech and uses it as an audio signal. The input unit 1 inputs the voice signal of the collected user's uttered voice to the voice recognition unit 2.

  The voice recognition unit 2 converts the voice signal of the user's uttered voice picked up by the input unit 1 into text representing the content of the user's utterance. The voice recognition unit 2 inputs text representing the user's utterance content to the utterance determination unit 3. Any existing speech recognition technology may be used as the speech recognition method, and an optimal method may be selected as appropriate in accordance with the usage environment.

  The utterance determination unit 3 communicates with the chat dialog system 6 or the scenario dialog system 7 and determines text representing the utterance content from the dialog system 10 for the user's utterance based on the input text. The utterance determination unit 3 inputs text representing the determined utterance content to the speech synthesis unit 4.

  The speech synthesizer 4 converts the text representing the utterance content determined by the utterance determination unit 3 into a speech signal representing the utterance content. The voice synthesis unit 4 inputs a voice signal representing the utterance content to the presentation unit 5. As a speech synthesis method, any existing speech synthesis technology may be used, and an optimum method may be selected as appropriate in accordance with the usage environment.

  The presentation unit 5 is an interface for presenting the utterance content determined by the utterance determination unit 3 to the user. For example, the presentation unit 5 is a humanoid robot imitating a human shape. This humanoid robot, for example, produces a speech signal representing the utterance content obtained by converting the text representing the utterance content determined by the utterance determination unit 3 into an audio signal by the speech synthesis unit 4 from, for example, a speaker mounted on the head. Present the utterance. When the presenting unit 5 is a humanoid robot, one humanoid robot is prepared for each personality participating in the dialogue. In the following, it is assumed that there are two humanoid robots 5-1 and 5-2 as an example in which two personalities participate in the dialogue.

  The input unit 1 may be configured integrally with the presentation unit 5. For example, when the presentation unit 5 is a humanoid robot, a microphone can be mounted on the head of the humanoid robot and used as the input unit 1.

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

  In step S11, the humanoid robot 5-1 or 5-2 outputs a voice representing the content of the open utterance selected by the scenario dialogue system 7 or the chat dialogue system 6 from the speaker. An open utterance is an utterance that a partner can freely respond to, and is an utterance including at least one of an open question and an open comment. The open question is the so-called 5W1H (when, where, who, who, what, why, how) This is a question that can be answered. In contrast to the open question, a question whose answer range is limited such as “Yes / No” or “A or B or C” is called a closed question. An open comment is an utterance that does not particularly require an answer, such as an unconcerned impression of a topic, that is, an utterance that can be freely spoken by the other party. The text representing the content of the open utterance is determined according to the scenario selected by the scenario dialogue system 7 in response to a request from the utterance determination unit 3. The scenario dialogue system 7 may select a scenario at random from a plurality of scenarios prepared in advance, or may select a scenario according to a rule determined in advance based on the content of the previous dialogue. As a method for selecting a scenario based on the content of the previous conversation, for example, for a conversation including about the last five utterances, the words included in each utterance, the focus words constituting each utterance, the words included in each scenario, There is a method of selecting a scenario in which the distance between words with the focal word is closer than a predetermined distance.

  In step S <b> 12, the microphone 1 accepts a user utterance for an open utterance. The voice recognition unit 2 recognizes the voice signal of the user's utterance voice collected by the microphone 1 and inputs the text obtained as the voice recognition result to the utterance determination unit 3 as text representing the user's utterance content.

  In step S <b> 13, the humanoid robot 5-1 outputs a voice representing the content of the chat utterance determined by the chat dialogue system 6 based on the text representing the content of the user's utterance from the speaker. The text representing the content of the chat utterance is determined by the utterance determination unit 3 using the chat dialogue system 6 based on the text representing the content of the user's utterance as the voice recognition result of the user's utterance. The humanoid robot that outputs the voice representing the contents of the chat utterance may be a humanoid robot that outputs the voice representing the contents of the open utterance, or the humanoid robot that outputs the voice representing the contents of the open utterance. Different humanoid robots may be used.

  In step S <b> 14, the humanoid robot 5-2 outputs a voice representing the content of the scenario utterance determined by the scenario dialogue system 7 from the speaker. Step S14 is performed after step S13. That is, the dialogue system outputs the voice representing the content of the chat utterance determined by the chat dialogue system 6 based on the user's utterance from the speaker, and then the voice representing the content of the scenario utterance decided by the scenario dialogue system 7 from the speaker. Output. The text representing the contents of the scenario utterance is determined by the utterance determination unit 3 in accordance with the scenario selected by the scenario dialogue system 7 in response to a request from the utterance determination unit 3. The humanoid robot that outputs the voice that represents the contents of the scenario utterance may be a humanoid robot that outputs the voice that represents the contents of the open utterance, or the humanoid robot that outputs the voice that represents the contents of the open utterance Different humanoid robots may be used. Thereafter, for example, the dialogue system outputs a voice representing the utterance content of the scenario utterance determined by the scenario dialogue system 7 from the speaker so that a dialogue according to the scenario is executed between the user and the dialogue system. The humanoid robot that performs subsequent speech may be any one humanoid robot or a plurality of humanoid robots.

  Hereafter, the specific example of the content of the conversation of 1st embodiment is shown. Here, R represents a humanoid robot, and H represents a user. The number in parentheses after R is the identifier of the humanoid robot. “R (1 → H)” indicates that the humanoid robot 5-1 is speaking with the intention of speaking to the user, and “R (1 → 2)” indicates that the humanoid robot 5-1 is humanoid robot 5- Indicates that you are speaking with the intention of speaking to 2. It should be noted that the intention of the humanoid robot to speak to may be expressed by, for example, the movement of the head or line of sight of the humanoid robot, but may not be expressed.

R (1 → H): “What do you like if it was food?” (* 2)
H: “I like ramen!”
R (1 → H): “I like pork bones” (* 1)
R (2 → 1): “Because we are robots, we can't eat anything.” (* 2)
R (1 → 2): “It can't be helped” (* 2)
R (2 → 1): “So soft.” (* 2)

  Here, * 1 is the utterance content determined by the chat dialogue system 6. * 2 is the utterance content determined by the scenario dialogue system 7. Note that the utterances of the humanoid robots after * 1 may be the opposite of the humanoid robot that speaks, or the person to talk to may be a human or another humanoid robot, or both humans and other humanoid robots.

  With the configuration as described above, according to the dialogue technique of this embodiment, even if the user's utterance with respect to the open utterance is outside the scenario, the chat dialogue system can respond to the user's utterance. The responsiveness felt by the user is improved. Thereby, the user is motivated to continue the dialogue with the system, and the dialogue can be continued for a long time.

<Variation 1 of the first embodiment>
In the first embodiment, the example in which the open utterance selected by the scenario dialogue system 7 in step S11 is output from the speaker has been described. However, the open utterance output from the speaker in step S11 may be generated in any way.

  For example, the user's utterance performed before step S11 is picked up by the input unit 1 as a voice signal, the voice recognition unit 2 obtains text representing the user's utterance content, and the utterance determination unit 3 receives the user's utterance. It is also possible to determine the text representing the utterance content of the open utterance by the chat dialogue system 6 based on at least the text representing the content, and to output the audio signal representing the utterance content of the open utterance determined by the chat dialogue system 6 from the speaker. Good.

<Modification 2 of the first embodiment>
In the first embodiment, an example of outputting voice representing the contents of the chat utterance determined by the chat dialog system 6 based on the user's utterance in step S13 from the speaker, that is, one determined by the chat dialog system 6 in step S13. Although the example which outputs the audio | voice showing the content of chat utterance from a speaker was demonstrated, you may output the audio | voice showing the content of the several chat utterance which the chat dialog system 6 determined in step S13 from a speaker.

  For example, in step S13, the humanoid robot 5-1 first outputs a voice representing the contents of the chat utterance determined by the chat dialogue system 6 based on the user's utterance from the speaker, and then the humanoid robot 5- 2 may output from the speaker a voice representing the content of the chat utterance determined by the chat dialogue system 6 based on the utterance of the humanoid robot 5-1 output from the speaker.

<Modification 3 of the first embodiment>
The dialogue system 10 of the first embodiment is a system in which a plurality of humanoid robots cooperate with each other to interact with a user. However, a single humanoid robot performs all or part of the dialogue with the user. It may be a system. For example, the same one humanoid robot performs the open utterance in step S11, the chat utterance determined by the chat dialog system 6 based on the user's utterance in step S13, and the scenario utterance determined by the scenario dialog system 7 in step S14. You may do it. In this case, for example, the dialogue with the user after step S14 may be performed in cooperation by a plurality of humanoid robots, or may be performed by the same one humanoid robot as in step S14.

<Second embodiment>
In the second embodiment, in order to improve the user's sense of participation in the topic, the scenario dialogue is changed to the chat dialogue, and then the chat dialogue is changed to the scenario dialogue again. By inserting a chat dialogue between scenario dialogues, the flow of the dialogue becomes natural, and it is possible to give the user a sense that he / she is also participating in the topic determination. As a result, the user can continue the dialogue for a long time in the subsequent scenario dialogue.

  As shown in FIG. 3, the dialogue system 12 of the second embodiment includes an input unit 1, a speech recognition unit 2, an utterance determination unit 3, a speech synthesis unit 4, and a presentation unit 5, as in the first embodiment. Further, a dialog control unit 8 is provided. The dialog system according to the second embodiment is realized by the processing of each step described later by the dialog system 12. In addition, as shown in FIG. 3, let the part by the speech recognition part 2, the speech determination part 3, the speech synthesizing part 4, and the dialogue control part 8 of the dialogue system 12 be the dialogue apparatus 13 of the second embodiment. The dialogue control unit 8 controls another processing unit included in the dialogue system 12 to execute a dialogue flow that is a dialogue in which the reception of the user's utterance and the presentation of the utterance to the user are executed at least once.

  Hereinafter, with reference to FIG. 4, the processing procedure of the dialogue method of the second embodiment will be described.

  In step S <b> 21, the humanoid robot 5-1 or 5-2 outputs a voice representing the content of the first scenario utterance included in the first scenario arbitrarily selected by the scenario dialogue system 7 from the speaker. The selection of the first scenario is performed in response to a request from the utterance determination unit 3. The scenario dialogue system 7 may select a first scenario randomly from a plurality of scenarios prepared in advance, or may select a first scenario according to a rule determined in advance based on previous dialogue contents. The method by which the scenario dialogue system 7 selects the first scenario is the same as the method described in step S11 of the first embodiment.

  In step S22, the microphone 1 accepts a first user utterance uttered by the user in response to the first scenario utterance. The voice recognition unit 2 recognizes the voice signal of the user's utterance collected by the microphone 1 and inputs the text obtained as the voice recognition result to the utterance determination unit 3 as text representing the content of the first user utterance.

  In step S23, the dialogue system 12 determines whether or not a condition for switching from the scenario dialogue to the chat dialogue is satisfied. If it is determined that the condition is satisfied, the process proceeds to step S24. If it is determined that the condition is not satisfied, the process returns to step S21, and the processes of steps S21 to S22 are executed again.

  The conditions for switching from the scenario dialogue to the chat dialogue are, for example, A1. If all dialogues according to the first scenario are completed, A2. For example, when the progress of the first scenario fails. For example, A2-1. When the user's utterance after speaking to the user is not included in the assumed range of the scenario dialogue system, A2-2. For example, when the user's behavior with respect to an ongoing scenario is recognized that the user does not want to talk about the scenario or that the user's willingness to interact is reduced. In addition, the case of A2-1 and the case of A2-2 are combined, and the case where the reaction of the user after speaking to the user is not good is also included. When the user's utterance is not included in the assumed range of the scenario dialogue system, for example, the user's utterance matches the utterance determined for the user's utterance based on the scenario stored in advance by the scenario dialogue system This is the case. User behavior includes non-verbal signals, paralinguistic signals (including information between them), and the like. The case where the user's reaction is not good includes a case where an intention not to speak is recognized from a user's behavior and a case where a decrease in the user's desire to interact is recognized. Recognizing intentions that you do not want to talk about or diminishing willingness to talk, for example, if you utter a specific phrase (phrase) that the user has set in advance (for example, “This story, I did it before”), talk about that topic. It may be judged that it expresses the intention that you do not want.

  In step S <b> 24, the humanoid robot 5-1 or 5-2 outputs a voice representing the content of the chat utterance determined by the chat dialogue system 6 based on the text representing the content of the first user utterance from the speaker. The humanoid robot that outputs the voice representing the content of the chat utterance may be a humanoid robot that outputs the voice representing the content of the last utterance out of one or more utterances based on the first scenario. The humanoid robot may be different from the humanoid robot that outputs the voice representing the content of the utterance.

  The text representing the content of the chat utterance is based on the text representing the content of the utterance sequence, which is the speech recognition result of the sequence of utterances between the humanoid robot and the user who are talking in the first scenario. This is determined using the system 6. The range of the utterance sequence input to the chat dialogue system 6 may be limited to the immediately preceding utterance, may be part or all of the first scenario, or the chat dialogue or scenario dialogue performed before the first scenario. All may be included. The first utterance when switching to chat conversation may be set in advance for each scenario. For example, it may be possible to prepare a question such as “What do you like except eating?” After a scenario of talking about food.

  In step S <b> 25, the microphone 1 accepts a second user utterance uttered by the user in response to the chat utterance. The voice recognition unit 2 recognizes the voice signal of the second user utterance picked up by the microphone 1 and inputs the text obtained as the voice recognition result to the utterance determination unit 3 as text representing the content of the second user utterance. .

  In step S26, the dialogue system 12 determines whether or not a condition for switching from the chat dialogue to the scenario dialogue is satisfied. If it is determined that the condition is satisfied, the process proceeds to step S27. If it is determined that the condition is not satisfied, the process returns to step S24, and the processes of steps S24-S25 are executed again.

  The condition for switching from the chat dialogue to the scenario dialogue is, for example, B1. If it becomes difficult to continue the chat conversation, B2. For example, it is possible to select the next scenario with sufficient confidence through chat conversation. The case where it is difficult to continue the chat conversation is, for example, B1-1. When the user's utterance after speaking to the user is not included in the assumed range of the chat dialogue system, B1-2. When it is recognized from the user's behavior with respect to the ongoing chat conversation that the user does not want to talk about the scenario or the user's willingness to interact is reduced. There are cases where the next scenario is not selected with sufficient reliability and the chat utterance is repeated a predetermined number of times. In addition, the case of B1-1 and the case of B1-2 are combined, and the case where the reaction of the user after speaking to the user is not good is also included. The case of B1-3 can be combined with the case of B1-1 and / or B1-2. For example, when the user's utterance is not included in the assumed range of the chat dialogue system and / or from the user's behavior The configuration may be such that it is determined that it is difficult to continue the chat conversation when the intention that the user does not want to speak is recognized a predetermined number of times. The case where the user's utterance is not included in the assumed range of the chat dialogue system is, for example, a case where the user's utterance and the utterance determined by the chat dialogue system based on at least the user's utterance do not match. User behavior includes non-verbal signals, paralinguistic signals (including information between them), and the like. The case where the user's reaction is not good includes a case where an intention not to speak is recognized from a user's behavior and a case where a decrease in the user's desire to interact is recognized. The recognition of the intention that the user does not want to speak and the decrease in the willingness to talk include, for example, a case where a user has issued a specific phrase (phrase) (for example, “This story, I've done it before”).

  When switching to the scenario dialogue, the previous user's utterance is received with a so-called “Soka”, and an interval is inserted immediately before the first utterance to change the topic, such as “By the way”, “Yes”, “Hey”, etc. It is good to make the utterance showing that it is going to speak to a humanoid robot different from the humanoid robot uttered at the end of the chat conversation. Thereby, the uncomfortable feeling caused by the discontinuity of the topic can be reduced. At this time, the content to be uttered may be changed according to the reliability when the next scenario is selected. Here, the reliability of selection is an index representing the degree of similarity between the contents of the chat conversation and the selected scenario. For example, if the selection reliability is relatively high, a short utterance such as “Yes, ...” is inserted, and if the selection reliability is relatively low, “By the way, it may not be related at all. It is conceivable to utter content that expressly expresses that the topic will change, such as "No ...". As a specific example when the reliability of the selection is low, for example, after speaking that “I ate ramen” in the chat dialogue, I talked about “Restaurant management” in the scenario dialogue. One example is when the similarity between terms is low. Also, for example, after speaking that “I do not like sports” in the chat dialogue, the topic similarity is high, but the user denies the topic. Case. In addition, for example, in a chat conversation, there is an utterance about “drag”, and the topic word is an ambiguous word, such as when it is not possible to determine whether it is a topic of mouse drag or drug drag, and spoken in any sense There are cases where it is impossible to determine whether or not.

  In step S <b> 27, the humanoid robot 5-1 or 5-2 outputs a voice representing the content of the second scenario utterance included in the second scenario selected by the scenario dialogue system 7 from the speaker. The selection of the second scenario is performed in response to a request from the utterance determination unit 3. If switching from the scenario dialogue to the chat dialogue has been performed before the completion of the first scenario in step S23, the remaining portion of the first scenario may be set as the second scenario. Thereafter, a dialogue according to the second scenario is executed between the user and the dialogue device.

  The scenario dialogue system 7 selects the second scenario according to a rule determined in advance based on the contents of the chat dialogue performed in steps S24 and S25. Since it is considered that the utterance during the chat conversation includes a phrase representing a topic that the user may be interested in, a scenario related to an appropriate topic can be selected as a second scenario by using these as clues. For example, a keyword representing a topic is set in each scenario, and the scenario is selected according to the similarity between the keyword and one or both of the utterance of the humanoid robot and the user's utterance during the chat conversation. Further, for example, when the sentence generated by the chat dialogue system 6 with respect to the user's utterance is similar to the head sentence of any scenario, the scenario is selected.

  When it is determined that it is difficult to continue the conversation because the utterance during the chat conversation cannot be selected as a clue, the second scenario is selected at random from a plurality of prepared scenarios. In this case, the selection probability may be set for each scenario when preparing the scenario in advance. That is, it is not always necessary to make an equal random selection. Further, the selection probability may be weighted based on the similarity calculated from the observation information so far.

  Each step in the dialogue method of the second embodiment is executed under the control of the dialogue control unit 8. The dialogue control unit 8 presents the content of the first scenario utterance based on the first scenario determined by the scenario dialogue system 7, and accepts the first user utterance uttered by the user in response to the first scenario utterance. And the second user utterance uttered by the user in response to the chat utterance and step S24 in which the chat dialog system 6 presents the contents of the chat utterance determined based on the user utterance. The dialogue method of the second embodiment is realized by performing control to execute the second dialogue flow in which step S25 for accepting each is executed at least once.

  Hereinafter, a specific example of the conversation contents of the second embodiment will be shown. Here, R represents a humanoid robot, and H represents a user. The number in parentheses after R is the identifier of the humanoid robot. As in the first embodiment, the intention that the humanoid robot intends to talk to may be expressed by, for example, the movement of the head or line of sight of the humanoid robot. Good.

R (1 → 2): “What do you like if it was food?” (* 1: The content of the utterance determined based on the first scenario selected by the scenario dialogue system 7)
R (2 → 1): “I like sushi.” (* 1: The content of the utterance determined based on the first scenario selected by the scenario dialogue system 7)
H: “This story, I said a while ago.” (* This is a specific example of “A2-2. When a user's behavior regarding an ongoing scenario is recognized as not to talk about the scenario”)
R (1 → H): “What do you like except eating?” (* 1: Specifics of utterances included in the first scenario selected by the scenario dialogue system 7 based on the previous dialogue “food”) An example.)
H: "Is it reading?"
R (1 → H): “What is your favorite book genre?” (* 2: A specific example of an utterance determined by the chat dialogue system 6 based on “reading” of a user utterance.)
H: “I like sports comics”
R (2 → H): “Do you often do sports?” (* 2: This is a specific example in which the chat dialogue is repeated multiple times. Here, the chat dialogue system 6 determines the utterance content based on the “sport” of the user utterance. doing.)
H: “I do n’t really like sports.”
R (2 → H): “Soka” (* 3: Difficult to continue chat dialogue when “B1-2. User's behavior for ongoing chat dialogue recognizes intention to not talk about the scenario”) This is a specific example when the user denies the topic “I don't like it so much,” so it is judged that it is difficult to continue the chat conversation. .)
R (1 → 2): “Speaking of me, I've been skiing last time” (* 3: Speaking of “speaking of me”, which means that I ’m trying to change the topic first, followed by chat conversation (Speech included in the second scenario selected by the scenario dialogue system 7 based on the content "Sports")
R (2 → 1): “Isn't it difficult to ski with that figure?” (* 3: The utterances included in the second scenario selected by the scenario dialogue system 7 based on the “sports” that was the content of the chat dialogue This is a specific example.)

  By configuring as described above, according to the dialogue technique of this embodiment, after a scenario dialogue is finished, the topic of the next scenario dialogue is determined according to the contents of the chat dialogue, so participation in the dialogue felt by the user A feeling improves. As a result, the user is motivated to continue the dialogue with the dialogue system, and the dialogue can be continued for a long time.

<Modification>
In the embodiment described above, an example in which a robot is used as an agent to perform a voice conversation has been described. However, the robot in the embodiment described above is a humanoid robot having a body or the like, but a robot having no body or the like. There may be. 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. 5, the interactive system 20 according to the modification includes an input unit 1, an utterance determination unit 3, and a presentation unit 5. In the example of FIG. 5, the interactive system 20 according to the modified example includes one interactive device 21, and the interactive device 21 according to the modified example includes an input unit 1, an utterance determining unit 3, and a presenting unit 5. The utterance determination unit 3 includes an interface capable of communicating with the chat dialogue system 6 and the scenario dialogue system 7 existing outside. The chat dialogue system 6 and the scenario dialogue system 7 may be configured in the dialogue apparatus as processing units having similar functions.

  The interactive apparatus of 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 is a smartphone. The presentation unit 5 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 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 1 using the software keyboard and post it to the group chat through his / her account. The utterance determination unit 3 inputs a post from the user's account to the chat dialogue system 6 or the scenario dialogue system 7 and posts the utterance content obtained from each dialogue system to the group chat through each virtual account. In addition, it is good also as a structure which inputs the utterance content to the input part 1 by a utterance using the microphone and voice recognition function which were mounted in the smart phone. Moreover, it is good also as a structure which outputs the utterance content obtained from each dialog system from the speaker with the audio | voice corresponding to each virtual account, using the speaker and speech synthesis function which were mounted in the smart phone.

  If the agent's utterance is displayed on the presentation unit 5 (display) at the same time, it may be positively felt that the conversation is active or exciting, but the user cannot read it all at once. If the amount of text is presented, the user may feel it is difficult to continue the conversation, so depending on the situation used, such as when the user is unable to put a cognitive load or where a calm atmosphere is required, The utterances may be displayed sequentially. In addition, when displaying sequentially, you may open and display a predetermined time interval. The time interval may be fixed or variable.

  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 description order, but may also be executed in parallel or individually as required by the processing capability of the apparatus that executes the processes.

[Program, recording medium]
When various processing functions in the interactive device described in the above embodiment are realized by a computer, the processing contents of the functions that the interactive device should have are described by a program. When various processing functions in the interactive system described in the above modification are realized by a computer, the processing contents of the functions that the interactive system should have are described by a program. Then, by executing this program on a computer, various processing functions in the interactive device and the interactive system 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, this computer reads the program stored in its own storage device and executes the 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.

Claims (7)

When the dialog system that determines the utterance based on the scenario stored in advance is called the first dialog system, and the dialog system that determines the utterance based on at least the user's utterance content is called the second dialog system,
A first presentation step in which the presentation unit presents an open utterance;
An utterance accepting step in which the input unit accepts a user's utterance for the open utterance;
A second presentation step in which the presenting unit presents the utterance determined by the second dialogue system based on the utterance of the user;
A third presentation step in which the presentation unit presents the utterance determined by the first dialogue system after the second presentation step;
Interactive method including The dialogue method according to claim 1,
Each of the above presentation steps is performed by a plurality of robots,
In the first presentation step, any one of the plurality of robots speaks.
The utterance receiving step is to recognize and accept the user's utterance,
In the second presentation step, a first robot that is one of the plurality of robots speaks,
In the third presentation step, a second robot that is different from the first robot among the plurality of robots speaks.
How to interact. The dialogue method according to claim 1,
Each of the above presentation steps is performed by multiple accounts in group chat.
In the first presentation step, the group chat is presented by any one of the plurality of accounts.
The utterance receiving step is to receive the user's post in the group chat,
The second presentation step is to present by the first account that is one of the plurality of accounts in the group chat,
The third presentation step is to present a second account different from the first account among the plurality of accounts in the group chat.
How to interact. When the dialog system that determines the utterance based on the scenario stored in advance is called the first dialog system, and the dialog system that determines the utterance based on at least the user's utterance content is called the second dialog system,
An open utterance is determined, an utterance is determined by the second dialog system based on the user's utterance for the open utterance after the open utterance, and an utterance is determined by the first dialog system after the utterance determined by the second dialog system. An utterance determination unit to determine;
An input unit for receiving the user's utterance for the open utterance;
The open utterance determined by the utterance determination unit is presented, the utterance determination unit presents the utterance determined by the second dialog system after the user's utterance, and the utterance determination unit is displayed by the first dialog system. A presentation unit for presenting the determined utterance after the utterance determined by the second dialogue system;
Interactive system including 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,
When the dialog system that determines the utterance based on the scenario stored in advance is called the first dialog system, and the dialog system that determines the utterance based on at least the user's utterance content is called the second dialog system,
An open utterance is determined, an utterance is determined by the second dialog system based on the user's utterance for the open utterance after the open utterance, and an utterance is determined by the first dialog system after the utterance determined by the second dialog system. An interactive device including an utterance determining unit for determining.   The program for making a computer perform each step of the dialogue method in any one of Claim 1 to 3.   A program for causing a computer to function as the interactive device according to claim 5.

Images