WO2017163515A1 - 情報処理システム、情報処理装置、情報処理方法、および記録媒体 - Google Patents
情報処理システム、情報処理装置、情報処理方法、および記録媒体 Download PDFInfo
- Publication number
- WO2017163515A1 WO2017163515A1 PCT/JP2016/088253 JP2016088253W WO2017163515A1 WO 2017163515 A1 WO2017163515 A1 WO 2017163515A1 JP 2016088253 W JP2016088253 W JP 2016088253W WO 2017163515 A1 WO2017163515 A1 WO 2017163515A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- user
- agent
- information
- attribute
- information processing
- Prior art date
Links
- 230000010365 information processing Effects 0.000 title claims abstract description 55
- 238000003672 processing method Methods 0.000 title claims abstract description 7
- 238000004891 communication Methods 0.000 claims abstract description 72
- 230000004044 response Effects 0.000 claims abstract description 53
- 238000003860 storage Methods 0.000 claims abstract description 19
- 230000003993 interaction Effects 0.000 claims description 42
- 239000003795 chemical substances by application Substances 0.000 description 487
- 238000012545 processing Methods 0.000 description 173
- 238000000034 method Methods 0.000 description 70
- 230000008569 process Effects 0.000 description 67
- 230000006870 function Effects 0.000 description 22
- 238000003780 insertion Methods 0.000 description 21
- 230000037431 insertion Effects 0.000 description 21
- 238000010586 diagram Methods 0.000 description 20
- 238000002360 preparation method Methods 0.000 description 14
- 230000009471 action Effects 0.000 description 13
- 238000004364 calculation method Methods 0.000 description 9
- 241000282414 Homo sapiens Species 0.000 description 8
- 230000006399 behavior Effects 0.000 description 8
- 238000003745 diagnosis Methods 0.000 description 8
- 230000000694 effects Effects 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 5
- 239000000284 extract Substances 0.000 description 5
- 238000003384 imaging method Methods 0.000 description 5
- 238000013508 migration Methods 0.000 description 5
- 230000005012 migration Effects 0.000 description 5
- 238000010187 selection method Methods 0.000 description 5
- 241001122315 Polites Species 0.000 description 4
- 239000008280 blood Substances 0.000 description 4
- 210000004369 blood Anatomy 0.000 description 4
- 230000008859 change Effects 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000012546 transfer Methods 0.000 description 4
- 239000013598 vector Substances 0.000 description 4
- 238000013459 approach Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 206010041349 Somnolence Diseases 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 235000019219 chocolate Nutrition 0.000 description 2
- 238000004590 computer program Methods 0.000 description 2
- 230000008451 emotion Effects 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000001151 other effect Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 125000002066 L-histidyl group Chemical group [H]N1C([H])=NC(C([H])([H])[C@](C(=O)[*])([H])N([H])[H])=C1[H] 0.000 description 1
- 241001465754 Metazoa Species 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000036772 blood pressure Effects 0.000 description 1
- 230000036760 body temperature Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000009223 counseling Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 229910044991 metal oxide Inorganic materials 0.000 description 1
- 150000004706 metal oxides Chemical class 0.000 description 1
- 239000008267 milk Substances 0.000 description 1
- 210000004080 milk Anatomy 0.000 description 1
- 235000013336 milk Nutrition 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000005236 sound signal Effects 0.000 description 1
- 235000013555 soy sauce Nutrition 0.000 description 1
- 210000004243 sweat Anatomy 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/50—Business processes related to the communications industry
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L51/00—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
- H04L51/02—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages
Definitions
- the present disclosure relates to an information processing system, an information processing apparatus, an information processing method, and a recording medium.
- a user can check a message transmitted from another terminal or transmit a message using an information processing terminal such as a smartphone, a mobile phone terminal, or a tablet terminal.
- Patent Document 2 based on past conversations between the user and the assistant (a user interface based on character animation) and other communication history, personality / feelings based on the history, learning data, etc., assistant image generation and behavior An interactive operation support system in which a pattern is generated is described.
- Patent Document 3 a system for selecting and displaying an initial impression attribute of an online search result such as an online date search is described.
- a system is described in which a potential candidate having a taste desired by an individual is searched for in a date service (lover introduction office).
- Patent Document 4 describes a matching system that stores photo data of male members and female members, each member browses photo data of opposite sex members, and introduces ideal images based on brain waves for the opposite sex photos. Yes.
- the conventional agent system is merely an automatic response by a machine imitating a human being, and even if it communicates with an agent, it does not connect with a real human being.
- the present disclosure proposes an information processing system, an information processing apparatus, an information processing method, and a recording medium that can seamlessly connect an interaction with an agent to communication with a person in the real world.
- a storage unit that stores information on a plurality of agents that have different attributes and can interact with a user, receives a message from the user from the client terminal, and sends a response message to the client terminal.
- a communication unit that sends a reply, a specific agent selected from the plurality of agents according to an instruction from the user, and an attribute of the specific agent updated according to the interaction between the specific agent and the user Is identified as the attribute of the user agent, and by comparing the attribute of the user agent with the attributes of a plurality of existing counterpart users, the partner user most similar to the attribute of the user agent is identified,
- To notify the user of the presence of the other user at the timing of It includes a control unit, and to propose an information processing system.
- a message from a user is transmitted to a server device that stores information on a plurality of agents that can interact with the user, each having a different attribute, and a response message to the message is received.
- a communication unit that selects a specific agent from the plurality of agents according to an instruction from the user, and for the user whose attribute of the specific agent is updated according to an interaction between the specific agent and the user
- a control unit that controls to receive, from the server device via the communication unit, a notification indicating that there is a real partner user who is most similar to the attribute of the agent. Propose the device.
- the processor stores information on a plurality of agents having different attributes and capable of interacting with the user in the storage unit, receives a message from the user from the client terminal, and sends a response message.
- Replying to the client terminal by the communication unit selecting a specific agent from the plurality of agents according to an instruction from the user, and selecting the specific agent according to an interaction between the specific agent and the user
- the attribute of the user agent is recorded as an attribute of the user agent, and the attribute of the user agent is compared with the attributes of a plurality of existing users, so that the partner most similar to the attribute of the user agent is recorded.
- the user is identified, and the user of the other user is sent to the user at a predetermined timing. It includes controlling to notify the standing, and proposes an information processing method.
- a computer transmits a message from a user to a server device that stores information on a plurality of agents each having different attributes and capable of interacting with the user, and responds to the message.
- a specific agent is selected from the plurality of agents according to a communication unit that receives a message and an instruction from a user, and an attribute of the specific agent is updated according to an interaction between the specific agent and the user.
- Function as a control unit that controls to receive a notification indicating that there is a real user who is most similar to the attribute of the user agent from the server device via the communication unit at a predetermined timing.
- a communication control system (agent system) according to an embodiment of the present disclosure makes it possible to seamlessly connect an interaction with an agent to communication with a person in the real world.
- the outline of the communication control system according to the present embodiment will be described below with reference to FIG.
- FIG. 1 is a diagram illustrating an overview of a communication control system according to an embodiment of the present disclosure.
- Users interact with agents with personality on a daily basis, depending on the situation, recommending content on the real world and the Internet, providing information such as news and weather forecasts, providing games, providing directions, etc.
- Agent services The interaction with the agent is performed, for example, via the display unit 106, and an image or conversation of the agent can be displayed on the display unit 106. Further, the agent voice is reproduced from a speaker (not shown). The user's utterance is picked up by a microphone (not shown), and language analysis is performed on the agent system side.
- a plurality of agent characters having different personalities are prepared as agents, and the user selects and purchases an arbitrary agent character and uses the agent service by the agent character.
- the conventional agent system is merely an automatic response by a machine imitating a human being, and even if it communicates with an agent, it does not connect with a real human being.
- a real-world-oriented agent system that matches a real person with the same personality and preference as the agent according to the interaction between the user and the agent, and seamlessly communicates with and encounters the real person. suggest. By connecting with not only a virtual person (agent) but also a real person, it is possible to give the user a fulfilling life and a satisfying heart.
- the communication control system is not limited to the voice agent that makes a response by voice, but may be a text-compatible agent that makes a text-based response in a client terminal such as a smartphone.
- the communication control system according to the present embodiment may be installed in an information processing apparatus such as a smartphone, a tablet terminal, or a PC, or may be incorporated in a client server system including a home system, an in-vehicle system, and a client terminal and a server. Good. Further, the communication control system according to the present embodiment may be mounted on an anthropomorphic device such as a robot. In the case of a robot, expression control and action control can be performed in addition to voice dialogue.
- FIG. 2 is a diagram showing the overall configuration of the communication control system according to the present embodiment.
- the communication control system includes a client terminal 1 and an agent server 2.
- the agent server 2 is connected to the client terminal 1 via the network 3 and transmits / receives data. Specifically, the agent server 2 generates a response voice for the uttered voice collected and transmitted by the client terminal 1 and transmits the response voice to the client terminal 1.
- the agent server 2 has a phoneme DB (database) corresponding to one or more agents, and can generate a response voice with a voice of a specific agent.
- the agent may be a character such as a cartoon, an animation, a game, a drama, a movie, a celebrity, a celebrity, a historical person, or the like. It may be an average person.
- the agent may be an animal or a personified character.
- the agent may be a person reflecting the personality of the user, or a person reflecting the personality of the user's friend, family, acquaintance, or the like.
- agent server 2 can generate response contents reflecting the characteristics of each agent.
- the agent server 2 can provide various services such as user schedule management, message transmission / reception, information provision, and the like through interaction with the user via the agent.
- the client terminal 1 is not limited to the smart phone as shown in FIG. 2, for example, a mobile phone terminal, a tablet terminal, a PC (personal computer), a game machine, a wearable terminal (smart eyeglass, smart band, smart watch, smart neck). Etc.).
- the client terminal 1 may be a robot.
- FIG. 3 is a block diagram illustrating an example of the configuration of the agent server 2 according to the present embodiment.
- the agent server 2 includes a voice agent I / F (interface) 20, a dialogue processing unit 30, a phoneme storage unit 40, a conversation DB generation unit 50, a phoneme DB generation unit 60, an advertisement insertion processing unit 70, An advertisement DB 72 and a feedback acquisition processing unit 80 are included.
- the voice agent I / F 20 functions as a voice data input / output unit, a voice recognition unit, and a voice generation unit.
- As the input / output unit a communication unit that performs transmission and reception with the client terminal 1 via the network 3 is assumed.
- the voice agent I / F 20 can receive the user's uttered voice from the client terminal 1 and convert it into text by voice recognition. Also, the voice agent I / F 20 converts the agent answer text data (text) output from the dialogue processing unit 30 into voice using the phoneme data corresponding to the agent, and generates the generated response voice of the agent on the client terminal 1. Send to.
- the dialogue processing unit 30 functions as an arithmetic processing unit and a control unit, and controls the overall operation in the agent server 2 according to various programs.
- the dialogue processing unit 30 is realized by an electronic circuit such as a CPU (Central Processing Unit) or a microprocessor, for example. Further, the dialogue processing unit 30 according to the present embodiment functions as a basic dialogue processing unit 31, a character A dialogue processing unit 32, a person B dialogue processing unit 33, and a person C dialogue processing unit 34.
- the character A dialogue processing unit 32, the person B dialogue processing unit 33, and the person C dialogue processing unit 34 realize a dialogue specialized for each agent.
- “Character A”, “Person B”, and “Person C” are given as examples of the agent.
- the present embodiment is not limited to this, and each dialogue that realizes a dialogue specialized for a large number of agents. You may have a process part.
- the basic dialogue processing unit 31 realizes a general-purpose dialogue that is not specialized for each agent.
- FIG. 4 is a diagram illustrating a configuration example of the dialogue processing unit 300 according to the present embodiment.
- the dialogue processing unit 300 includes a question sentence search unit 310, an answer sentence generation unit 320, a phoneme data acquisition unit 340, and a conversation DB 330.
- the conversation DB 330 stores conversation data in which question sentence data and answer sentence data are paired.
- conversation data specialized for the agent is stored in the conversation DB 330
- general-purpose dialogue processing unit general-purpose conversation data (that is, basic conversation that is not specialized for the agent) is stored in the conversation DB 330. Data) is stored.
- the question sentence search unit 310 searches the conversation DB 330 for question sentence data that matches the question sentence that is output from the voice agent I / F 20 and recognized as a text by recognizing the user's question voice (an example of uttered voice).
- the answer sentence generation unit 320 extracts answer sentence data stored in association with the question sentence data searched by the question sentence search unit 310 from the conversation DB 330, and generates answer sentence data.
- the phoneme data acquisition unit 340 acquires phoneme data for converting the answer sentence generated by the answer sentence generation unit 320 from the phoneme storage unit 40 of the corresponding agent. For example, in the case of the character A dialogue processing unit 32, phoneme data for reproducing the answer sentence data with the voice of the character A is acquired from the character A phoneme DB 42. Then, the dialogue processing unit 300 outputs the generated answer sentence data and the acquired phoneme data to the voice agent I / F 20.
- the phoneme storage unit 40 stores a phoneme database for generating speech for each agent.
- the phoneme storage unit 40 can be realized by a ROM (Read Only Memory) and a RAM (Random Access Memory).
- a basic phoneme DB 41, a character A phoneme DB 42, a person B phoneme DB 43, and a person C phoneme DB 44 are stored.
- Each phoneme DB stores, for example, a phoneme piece and a prosodic model that is control information thereof as phoneme data.
- the conversation DB generation unit 50 has a function of generating the conversation DB 330 of the conversation processing unit 300. For example, the conversation DB generation unit 50 collects assumed question sentence data, collects answer sentence data corresponding to each question, and then saves the question sentence data and the answer sentence data in pairs. When a predetermined number of conversation data (a set of question sentence data and answer sentence data, for example, 100 sets) is collected, the conversation DB generation unit 50 registers the conversation data set in the conversation DB 330 as an agent conversation data set.
- a predetermined number of conversation data a set of question sentence data and answer sentence data, for example, 100 sets
- the phoneme DB generation unit 60 has a function of generating a phoneme DB stored in the phoneme storage unit 40.
- the phoneme DB generation unit 60 analyzes speech information read out from a predetermined text, decomposes it into phoneme segments and prosodic models that are control information thereof, and collects a predetermined number or more of speech information as phoneme DB as phoneme data. Process to register with.
- the advertisement insertion processing unit 70 has a function of inserting advertisement information into the agent dialogue.
- the advertisement information to be inserted can be extracted from the advertisement DB 72.
- advertisement information requested by a provider (vendor, supplier) of a company or the like for example, advertisement contents such as text, image, and sound, information on an advertiser, an advertisement period, an advertisement target person, etc. is registered. Yes.
- the feedback acquisition processing unit 80 has a function for inserting a question for acquiring feedback into the agent's dialogue and obtaining feedback from the user.
- the configuration of the agent server 2 according to this embodiment has been specifically described above. Note that the configuration of the agent server 2 according to the present embodiment is not limited to the example shown in FIG. For example, each configuration of the agent server 2 may be configured by other servers on the network.
- FIG. 5 is a flowchart showing a process for generating the conversation DB 330 according to this embodiment. As shown in FIG. 5, first, the conversation DB generation unit 50 stores an assumed question sentence (step S103).
- the conversation DB generating unit 50 stores a (paired) answer sentence corresponding to the question sentence (step S106).
- the conversation DB generation unit 50 determines whether or not a predetermined number of pairs of question sentences and answer sentences (also referred to as conversation data) have been collected (step S109).
- the conversation DB generation unit 50 registers a data set including a large number of pairs of question sentences and answer sentences in the conversation DB 330 (step S109). S112).
- a pair of a question sentence and an answer sentence for example, the following is assumed.
- Pair of question and answer sentences Pair 1 Question: Good morning. Answer: How are you feeling today? Pair 2 Question: What is the weather today? Answer text: Today's weather is ⁇ .
- Such a pair can be registered in the conversation DB 330 as conversation data.
- FIG. 6 is a flowchart showing a phoneme DB generation process according to this embodiment.
- the phoneme DB generation unit 60 displays an example sentence (step S113). For example, an example sentence necessary for generating phoneme data is displayed on a display of an information processing terminal (not shown).
- the phoneme DB generation unit 60 records the voice that reads the example sentence (step S116) and analyzes the recorded voice (step S119). For example, the voice information read out by the person in charge of the agent's voice is collected by the microphone of the information processing terminal, and the phoneme DB generation unit 60 receives and stores it, and further performs voice analysis.
- the phoneme DB generation unit 60 generates a prosody model based on the speech information (step S122).
- the prosody model is used to extract prosodic parameters indicating prosodic features of speech (for example, pitch of a sound, strength of a sound, speech rate, etc.), and differs for each individual.
- the phoneme DB generation unit 60 generates phoneme pieces (phoneme data) based on the voice information (step S125).
- the phoneme DB generation unit 60 stores the prosody model and phoneme pieces (step S128).
- the phoneme DB generation unit 60 determines whether or not a predetermined number of prosodic models and phonemes have been collected (step S131).
- the phoneme DB generation unit 60 registers the prosodic models and phonemes in the phoneme storage unit 40 as a phoneme database for a predetermined agent (Ste S134).
- FIG. 7 is a flowchart showing the dialogue control process according to the present embodiment.
- the voice agent I / F 20 checks whether or not the user's question voice and the agent ID have been acquired (step S143).
- the agent ID is identification information indicating specific agents such as character A, person B, and person C.
- the user can purchase phoneme data for each agent. For example, the ID of the agent purchased during the purchase process is stored in the client terminal 1.
- the voice agent I / F 20 recognizes the question voice and converts it into text (step S149).
- the voice agent I / F 20 outputs the question text converted to text to the dialog processing unit of the specific agent specified by the agent ID. For example, in the case of “agent ID: character A”, the voice agent I / F 20 outputs the question text converted to text to the character A dialogue processing unit 32.
- the dialogue processing unit 30 searches the question DB that matches the question text converted to text from the conversation DB of the specific agent specified by the agent ID (step S152).
- step S155 when there is a matching question (step S155 / Yes), the character A dialogue processing unit 32 obtains answer sentence data corresponding to the question (stored in pairs) from the conversation DB of the specific agent. (Step S158).
- step S155 when there is no matching question (step S155 / No), a question sentence that matches the textualized question sentence is searched from the conversation DB of the basic dialogue processing unit 31 (step S161).
- step S161 / Yes the basic dialogue processing unit 31 obtains answer sentence data corresponding to the question (stored as a pair) from the conversation DB of the basic dialogue processing unit 31. (Step S167).
- step S164 when there is no matching question sentence (step S164 / No), the basic dialogue processing unit 31 returns answer sentence data (for example, an answer sentence such as “I do not understand the question”) when there is no matching question sentence.
- answer sentence data for example, an answer sentence such as “I do not understand the question”
- the character A dialogue processing unit 32 refers to the phoneme DB of the specific agent designated by the agent ID (here, the character A phoneme DB 42), and the phoneme data of the character A for generating the voice of the answer sentence data is obtained. Obtained (step S173).
- the acquired phoneme data and answer sentence data are output to the voice agent I / F 20 (step S176).
- the voice agent I / F 20 converts the response sentence data (text) into speech using the phoneme data (speech synthesis) and transmits it to the client terminal 1 (step S179).
- the answer sentence is reproduced with the voice of the character A.
- Conversation DB update processing> update processing of the conversation DB 330 of each dialogue processing unit 300 will be described.
- the conversation DB 330 can be grown by conversation with the user.
- FIG. 8 is a diagram for explaining a data configuration example of the conversation DB 330 according to the present embodiment.
- each conversation DB 330 has two layers, a personalization layer 331 and a common layer 332.
- the common layer 332A holds conversation data reflecting the character and character of the character A.
- the personalization layer 331A holds conversation data customized for the user by the conversation with the user.
- the conversation data can be customized for the user. That is, for example, when “person B” is “person in 20s”, the common layer 332B holds average conversation data of 20s, and customized conversation data is maintained for each user by continuing the conversation with the user. Of personalization layer 331B.
- the user can also select and purchase favorite phoneme data such as “male”, “female”, “high voice”, and “low voice” from the person B phoneme DB 43 as the voice of the person B.
- FIG. 9 is a flowchart showing the update processing of the conversation DB 330 according to the present embodiment.
- the voice agent I / F 20 acquires (receives) the user's question voice from the client terminal 1, and converts it into text by voice recognition (step S183).
- the text data (question sentence data) is output to the dialogue processing unit (here, for example, the character A dialogue processing unit 32) of the specific agent designated by the agent ID.
- the character A dialogue processing unit 32 determines whether or not the question sentence data is a predetermined command (step S186).
- the character A dialogue processing unit 32 registers the answer text data specified by the user in a pair with the question text data in the personalization layer 331A of the conversation DB 330A (step S189).
- the predetermined command may be words such as “NG” and “setting”, for example.
- the conversation DB of character A can be customized by the following conversation flow.
- NG is a predetermined command
- the character A dialogue processing unit 32 has issued “NG” from the user, the user-specified answer text data “Perform with good spirit”
- the question sentence data “Good morning” is registered in the personalization layer 331A of the conversation DB 330A.
- the character A dialogue processing unit 32 searches the character A conversation DB 330A for answer sentence data held in a pair with the question sentence data.
- the answer sentence data held in a pair with the question sentence data is not held in the character A conversation DB 330A, that is, when the user's question is a question without an answer sentence (step S192 / Yes)
- the character A dialogue processing unit 32 registers the answer sentence data specified by the user in the personalization layer 331A as a pair with the question sentence (step S195).
- the conversation DB of character A can be customized by the following conversation flow.
- Character A “I don't know the question” (An example of answer data when there is no applicable answer) User: “If you ask,“ How are you? ”, Say,“ I ’m fine today. ” Character A: “I'm fine today”
- step S192 the character A dialogue processing unit 32 acquires the answer sentence data and outputs it to the voice agent I / F 20 together with the corresponding phoneme data of the character A.
- the answer sentence is reproduced with the voice of the character A at the client terminal 1 (step S198).
- FIG. 10 is a flowchart showing conversation data migration processing from the personalization layer to the common layer according to the present embodiment.
- the conversation data migration processing from the personalization layer 331A to the common layer 332A of the character A dialogue processing unit 32 will be described.
- the character A dialogue processing unit 32 periodically searches for a personalization layer 331A for each user (step S203), and a conversation pair (question sentence data and answer sentence having substantially the same contents).
- a data pair) is extracted (step S206).
- a conversation pair with substantially the same content is, for example, a pair of a question sentence “How are you?” And an answer sentence “I'm fine today!” And a question sentence “How are you?” And an answer sentence “I'm fine today.
- the “!” Pair is only a difference in whether the question sentence is a polite word or not, and can be determined as a conversation pair having substantially the same content.
- step S209 / Yes when a predetermined number or more of conversation pairs are extracted from the personalization layer 331A for each user (step S209 / Yes), the character A dialogue processing unit 32 registers the conversation pair in the common layer 332A (for each user). (Step S212).
- FIG. 11 is a diagram for explaining the transfer of conversation data to the basic conversation conversation DB 330F according to the present embodiment.
- the conversation processing unit 30 may include an A conversation DB 330A-X, a user Y character A conversation DB 330A-Y, and a user Z person B conversation DB 330B-Z.
- each personalization layer 331A-X, 331A-Y, 331B-Z is registered with its own (customized) conversation pair according to the dialogue with each user X, user Y, and user Z. (See FIG. 9).
- the personalization layers 331A-X and 331A-Y of the same agent they are registered in the common layers 332A-X and 332A-Y for each user (see FIG. 10).
- the conversation processing unit 30 extracts a predetermined number or more of substantially the same conversation pairs from the common layers 332A-X, 332A-Y, and 332B-Z of a plurality of agents (which may include different agents), the conversation processing unit 30 The conversation pair is transferred to the conversation conversation DB 330F.
- the basic conversation conversation DB 330 ⁇ / b> F is a conversation DB included in the basic conversation processing unit 31. This makes it possible to grow the basic conversation conversation DB 330F (expand conversation pairs).
- FIG. 12 is a flowchart showing the conversation data migration processing to the basic dialogue DB 330F according to the present embodiment.
- the dialogue processing unit 30 periodically searches a plurality of common layers 332 in the conversation DB 330 (step S223), and extracts substantially the same conversation pairs (step S226).
- the conversation processing unit 30 registers the conversation pairs in the basic conversation conversation DB 330F (step S232). .
- advertisement information insertion processing by the advertisement insertion processing unit 70 will be described with reference to FIGS.
- the advertisement insertion processing unit 70 can insert the advertisement information stored in the advertisement DB 72 into the utterance of the agent. Advertisement information can be registered in the advertisement DB 72 in advance.
- FIG. 13 is a diagram illustrating an example of advertisement information registered in the advertisement DB 72 according to the present embodiment.
- the advertisement information 621 includes, for example, an agent ID, a question sentence, advertisement contents, conditions, and a probability.
- the agent ID designates an agent that speaks the advertisement contents
- the question sentence designates a question sentence of a user that triggers insertion of the advertisement contents
- the advertisement contents are advertisement sentences to be inserted into the agent's dialogue.
- the condition is a condition for inserting the advertisement content
- the probability indicates the probability of inserting the advertisement content. For example, in the example shown in the first row of FIG.
- the probability of inserting an advertisement may be set in this embodiment. Such a probability may be determined according to the advertisement fee. For example, the higher the advertising fee, the higher the probability.
- FIG. 14 is a flowchart showing the insertion processing of advertisement content according to this embodiment.
- the advertisement insertion processing unit 70 monitors the dialogue between the user and the agent (specifically, dialogue processing by the dialogue processing unit 30) (step S243).
- the advertisement insertion processing unit 70 determines whether or not a question sentence having the same content as the question sentence registered in the advertisement DB 72 has appeared in the dialogue between the user and the agent (step S246).
- the advertisement insertion processing unit 70 checks the advertisement insertion condition and probability associated with the corresponding question sentence (step S249).
- the advertisement insertion processing unit 70 determines whether or not it is currently possible to place an advertisement based on the condition and the probability (step S252).
- the advertisement insertion processing unit 70 temporarily stops the dialogue processing by the dialogue processing unit 30 (step S255), and inserts the advertisement content into the dialogue (step S258). Specifically, for example, the advertisement content is inserted into the agent's answer to the user's question.
- the dialogue (conversation text data) including the advertisement content is output from the dialogue processing unit 30 to the voice agent I / F 20, transmitted from the voice agent I / F 20 to the client terminal 1, and reproduced by the voice of the agent (step S261). ).
- the content of the advertisement can be presented to the user as an utterance of the character A by the following conversation.
- the conversation data registration process As described above, the conversation data registration process, the phoneme DB generation process, the conversation control process, the conversation DB update process, and the advertisement insertion process have been described as basic operation processes of the communication control system according to the present embodiment.
- the dialogue control process according to the present embodiment is not limited to the above-described example.
- the dialogue processing unit 30 according to the present embodiment can seamlessly connect a dialogue with an agent to communication with a person in the real world.
- FIGS. 1-10 a specific description will be given with reference to FIGS.
- FIG. 15 is a diagram illustrating a system configuration example of the communication control system according to the present embodiment.
- the present embodiment as an example, a case will be described in which matching with a marriage hunting member is performed using member information of a marriage consulting office in accordance with a dialogue between a user and an agent.
- the communication control system includes a client terminal 1, an agent server 2, and a management server 4.
- the management server 4 has a function of managing member information (marital affiliation member information 41) of the marriage consulting office, and provides member information in response to a request from the agent server 2.
- the dialogue processing unit 30a realizes a real-world working type agent system that seamlessly links a dialogue with an agent to communication with a person in the real world.
- FIG. 16 is a diagram illustrating a configuration example of the dialogue processing unit 30a according to the present embodiment.
- the dialogue processing unit 30a includes a basic dialogue processing unit 31, a character A dialogue processing unit 32, a person B dialogue processing unit 33, a person C dialogue processing unit 34, a matching unit 35, and a communication unit 36. .
- the basic dialogue processing unit 31, the character A dialogue processing unit 32, the person B dialogue processing unit 33, and the person C dialogue processing unit 34 are as described above with reference to FIG. Character A, person B, and person C are all examples of agent characters.
- the communication unit 36 can transmit / receive data to / from an external device via a network.
- the communication unit 36 receives marriage hunting member information from the management server 4 or transmits a consent notice to the matched partner.
- the matching unit 35 has a function of matching an actual person who has the same personality and preference as the agent according to the dialogue between the user and the agent. The detailed configuration of the matching unit 35 will be described next with reference to FIG.
- FIG. 17 is a diagram illustrating a configuration example of the matching unit 35 according to the present embodiment.
- the matching unit 35 includes a user agent dialogue processing unit 350, a user / agent information management unit 351, a user information DB 352, an agent information DB 360, an agent learning unit 353, a user agent DB 354, an introduction user selection unit. 355, a closeness calculation unit 356, an introduction processing unit 357, a scenario management unit 358, and a scenario DB 359.
- the user / agent information management unit 351 has a function of registering, changing, updating, deleting, etc. user information for the user information DB 352 or agent information for the agent information DB 360.
- the user information is input by the user at the client terminal 1 and transmitted to the agent server 2.
- the user information is extracted from the married member information 41 of the management server 4.
- the user information includes basic information such as user ID, age, occupation, family structure, family living, annual income, residence, blood type and the like. Each item has a public attribute relating to whether or not it can be disclosed to a final introduction partner (matching partner). Further, the user information includes basic condition information for specifying a favorite partner.
- Basic condition information indicates the user's desired conditions in information that can be items when searching for a matching partner at a marriage counselor, such as age, occupation, family structure, family living, annual income, residence, blood type. is there.
- a priority can be set for each item, and it is possible to first select an agent having the attribute with emphasis on an item with a high priority.
- the user information includes user preference information.
- preference information may be set in the range of -1.0 (dislike) to 1.0 (like) for each item, or entered in a questionnaire format such as "I love, like, neither, hate, hate"
- the registered information may be registered.
- the user's preference information may be input in advance by the user, or the preference information may be generated or edited in accordance with the interaction between the user and the agent. For example, if the user says “I have no eyes on ramen” in the conversation with the agent, the item “ramen” may be automatically generated and “1.0” may be set.
- agent information DB 360 The same information regarding the agent is stored in the agent information DB 360. Default values are set for the basic information, basic condition information, and preference information of the agent.
- FIGS. 1-10 Specific examples of basic information of users and agents are shown in FIGS.
- the agent learning unit 353 has a function of changing the personality (attribute) of the user agent in the user agent dialogue processing unit 350 to the user preference by learning the dialogue with the user.
- the initial value of the user agent dialogue processing unit 350 is, for example, an agent selected by the user from a plurality of agent characters prepared in advance.
- the agent learning unit 353 gradually changes the agent to a user preference as the dialogue is performed. To do.
- the user agent DB 354 stores user agent attributes (for example, basic information, basic condition information, and preference information items) that have been changed to user preferences by the agent learning unit 353 and are updated as appropriate.
- the user agent attributes may also include the appearance of the agent (face type, hairstyle, clothing type, etc.).
- the user agent dialogue processing unit 350 has a function of realizing an automatic dialogue with the user by the user agent that is appropriately changed by the agent learning unit 353. Specifically, the user agent interaction processing unit 350 analyzes the speech or text transmitted from the client terminal 1 and outputs a corresponding response sentence.
- the introduction user selection unit 355 has a function of searching for a real person having an attribute similar to the attribute of the agent (user agent) changed to the user preference and matching the user as an introduction user.
- the attribute of the agent changed to the user preference is grasped with reference to the user agent DB 354.
- the introduction user selection unit 355 searches the marriage member information 41 for a real person having the same attribute as the agent attribute changed to the user preference, and extracts it as an introduction user.
- the introduction user attribute may be output to the agent learning unit 353 so that the user agent is brought closer to the introduction user attribute extracted by the introduction user selection unit 355.
- the familiarity calculating unit 356 calculates the familiarity between the user agent and the user reflecting the attributes of the introducing user. For example, the familiarity calculation unit 356 calculates the familiarity according to the content of the interaction between the user agent and the user.
- the introduction processing unit 357 performs various processes for introducing the introducing user (real person) to the user when the familiarity calculated by the familiarity calculating unit 356 exceeds the threshold. For example, the introduction processing unit 357 transmits an approval notification asking the introduction user whether introduction is possible or not, or displays an introduction screen on both client terminals 1.
- the scenario management unit 358 performs various processes for supporting the encounter between the user and the introduction user according to the scenario registered in the scenario DB 359. For example, the scenario management unit 358 controls to notify both of the date and place and the plan content to a predetermined supplier according to a scenario arbitrarily selected by the user from the scenario DB 359.
- the configuration of the matching unit 35 according to the present embodiment has been described above.
- FIG. 18 is a block diagram illustrating an example of the configuration of the client terminal 1 according to the present embodiment.
- the client terminal 1 includes a control unit 100, a communication unit 101, an operation input unit 102, a sensor 103, a camera 104, a microphone (abbreviation of microphone) 105, a display unit 106, a speaker 107, and a storage unit 108.
- Control unit 100 The control unit 100 is realized by a processor such as a CPU (Central Processing Unit) included in the client terminal 1, for example.
- the control unit 100 controls the agent response sound transmitted from the agent server 2 via the communication unit 101 to be reproduced from the speaker 107, or displays the agent image on the display unit 106, for example. To do.
- a processor such as a CPU (Central Processing Unit) included in the client terminal 1, for example.
- the control unit 100 controls the agent response sound transmitted from the agent server 2 via the communication unit 101 to be reproduced from the speaker 107, or displays the agent image on the display unit 106, for example. To do.
- control unit 100 transmits selection information (for example, selection of an agent, etc.) by the user input from the operation input unit 102 to the agent server 2 via the communication unit 101.
- selection information for example, selection of an agent, etc.
- control unit 100 controls to provide an agent service such as an automatic dialogue by the agent selected by the user according to the control of the agent server 2.
- the communication unit 101 is a communication interface configured with, for example, a communication device for connecting to the network 3.
- the communication unit 101 can be, for example, a communication card for LAN (Local Area Network), Bluetooth (registered trademark), Wi-Fi, or WUSB (Wireless USB).
- the communication unit 101 may be a router for optical communication, a router for ADSL (Asymmetric Digital Subscriber Line), a modem for various communication, or the like.
- the communication unit 101 transmits and receives signals and the like with a predetermined protocol such as TCP / IP, for example, with the Internet and other communication devices.
- the network 3 connected to the communication unit 101 is a network connected by wire or wireless, and may include, for example, the Internet, a home LAN, infrared communication, radio wave communication, satellite communication, or the like.
- the operation input unit 102 has a function of receiving an input of a user operation and outputting it to the control unit 100.
- the operation input unit 102 is realized by, for example, a mouse, a keyboard, a touch panel, a button, a switch, or a lever.
- the sensor 103 has a function of detecting a user or a surrounding situation.
- the sensor 103 is a biological sensor (pulse meter, heart rate monitor, sweat sensor, body temperature sensor, blood pressure sensor, electroencephalograph, etc.), environmental sensor (temperature sensor, illuminance sensor, pressure gauge, etc.), acceleration sensor, gyro sensor, direction sensor. , A vibration sensor, a position measurement sensor, or the like.
- the camera 104 photoelectrically converts imaging light obtained by a lens system including an imaging lens, a diaphragm, a zoom lens, and a focus lens, a drive system that causes the lens system to perform a focus operation and a zoom operation, and the lens system.
- a lens system including an imaging lens, a diaphragm, a zoom lens, and a focus lens, a drive system that causes the lens system to perform a focus operation and a zoom operation, and the lens system.
- the solid-state imaging device array may be realized by, for example, a CCD (Charge Coupled Device) sensor array or a CMOS (Complementary Metal Oxide Semiconductor) sensor array.
- the microphone 105 picks up the user's voice and surrounding environmental sound and outputs it to the control unit 100 as voice data.
- the display unit 106 has a function of displaying characters, diagrams, images, videos, and the like.
- the display unit 106 is realized by, for example, a liquid crystal display (LCD) device, an OLED (Organic Light Emitting Diode) device, or the like.
- the speaker 107 has a function of reproducing an audio signal.
- the storage unit 108 stores programs and parameters for the control unit 100 to execute each function.
- the storage unit 108 may store user information such as a user ID, name, age, gender, occupation, family structure, family living, annual income, residence, and preference information.
- the operation process according to the present embodiment mainly includes four phases of search, fixing, introduction, and support.
- FIG. 19 is a flowchart showing the search phase operation processing according to this embodiment. As shown in FIG. 19, first, when one agent is selected by the user (step S270), the user agent interaction processing unit 350 starts an interaction with the user by the selected agent (step S273).
- the agent learning unit 353 learns the interaction between the agent and the user by the user agent interaction processing unit 350, and changes the agent's preference and personality attributes so as to approach the user's preference (step S276).
- the introduction user selection unit 355 matches the agent according to the user's preference with the attributes of each real person (the person to be introduced) and selects the introduction user (step S279).
- the agent attributes are changed, and the agent and the real person are matched behind the scenes without being known by the user.
- step S282 The above process is repeated until the number of times the same person is selected exceeds a predetermined number.
- step S282 / Yes when the number of times the same person has been selected exceeds the predetermined number (step S282 / Yes), the fixed phase process described in FIG. 20 is started.
- the number of times the same person is selected is used as a reference here, the present embodiment is not limited to this, and the case where the same person continues to be selected for a certain period may be used as a reference for starting the fixed phase.
- FIG. 20 is a flowchart showing the operation process of the fixed phase according to the present embodiment.
- a matched person selected introducer, also referred to as “real person”
- intimacy whether the person and the user are really compatible (in this case, determined by “intimacy”) is checked. It is a phase for. Therefore, here, when the attribute of the real person changes, the attribute of the corresponding agent is changed according to the real person.
- the user agent interaction processing unit 350 continues the user interaction with the user agent that matches the real person (step S303).
- the agent learning unit 353 continuously checks the information of the real person stored in the user information DB 352, and if the attribute of the matched real person has changed, reflects this in the attribute of the user agent (step). S306). Specifically, the agent attribute stored in the user agent DB 354 is updated so as to approach the attribute of the real person.
- the familiarity calculation unit 356 calculates the familiarity between the agent and the user (step S309).
- the agent attributes are brought to the real person who matched the attributes, there is a possibility that the personality and preference of the user and the agent may shift as the dialogue proceeds. Accordingly, the closeness between the agent similar to the real person and the user is calculated, and it is determined whether or not the matched real person is really appropriate for the user.
- the familiarity can be calculated based on, for example, the degree of matching between the agent attribute item and the user attribute item, the conversation amount with the user, the smile amount of the user during the conversation, and the like.
- the familiarity may be calculated from wording in a user's dialogue, word positive / negative, or the like.
- step S312 / Yes the introduction processing unit 357 notifies the matching actual person whether or not the introduction is approved (step S315).
- step S318 / Yes when the introduction is approved by the real person (step S318 / Yes), the processing of the introduction phase described in FIG. 21 is started.
- step S318 / No when the introduction approval is not recognized by the real person (introduction is rejected) (step S318 / No), the process returns to the search phase shown in FIG. In the matching process in step S279 of the search phase, persons who have already refused introduction approval are excluded.
- FIG. 21 is a flowchart showing the operation processing of the introduction phase according to this embodiment.
- the introduction phase the user is notified for the first time that there is a real person that can be introduced.
- the introduction processing unit 357 notifies the user that there is a real person that can be introduced to the user (step S323).
- the introduction processing unit 357 displays on the display unit 106 of the user's client terminal 1 the profile of the real person to be introduced (in a range that the other party has made public) (step S326), and further, the real person ( In the following, “Would you like to meet / do not want to meet a button” for selecting whether or not to actually meet (referred to as “introducing partner”) is displayed (step S329). Thereby, the user can confirm the profile of the introduction partner.
- step S332 the dialogue between the user and the real person through the agent by voice or text is continued (step S332).
- the chat of the introduction partner is played with the voice of the agent or the face image of the agent is displayed.
- the user continues the dialogue with the introduction partner through the agent, determines whether or not to actually meet the introduction partner, and taps the “I want to meet / I do not want to meet button” button.
- Such buttons can also be displayed on the other party's screen.
- step S335 / Yes when both the user and the introduction partner have expressed their intention to meet (the “I want to meet button” is selected) (step S335 / Yes), the support phase shown in FIG. 22 is started.
- step S338 / Yes the matching with the actual person is canceled, and the search phase returns to FIG.
- the user can select whether or not to reselect the agent (step S341).
- a plurality of agents are prepared, and each agent has a different personality, that is, an attribute of an initial value is different. Therefore, when starting again from a different type of agent, the user reselects the agent.
- the agent is not selected again, the attribute of the current agent is changed again to the user-preferred attribute in accordance with the dialogue with the user.
- matching is not performed with a person who has been resolved once by matching for a certain period of time.
- FIG. 22 is a flowchart showing the operation process of the support phase according to this embodiment. Since the connection with the real person has been achieved in the introduction phase, the support phase does not necessarily need to be executed, but can be performed according to the user's request. Here, it is possible to select the situation and direction when two people meet in the form of a “scenario”.
- an encounter plan is selected by the user from a plurality of encounter plans (scenarios) stored in the scenario DB 359 (step S353).
- the scenario management unit 358 may automatically recommend an encounter plan based on the contents of the conversation between the user and the introduction partner so far.
- the encounter plan includes location (restaurant, park, amusement park, aquarium, etc.), contents (surprise production, with gifts, special treatment, discounts), date (weekdays / holidays, morning / noon / evening), cost, etc.
- a content such as “a surprise restaurant gives her a surprise present ...” is set.
- the encounter plan can be viewed only by the user depending on the content.
- the scenario management unit 358 notifies the date and place of the encounter to the user and the real person according to the activated encounter plan (step S356).
- the scenario management unit 358 confirms whether or not the schedules match each other and are OK (step S359). Note that when selecting an encounter plan, the user may be allowed to specify a date and time when the schedule is vacant.
- step S362 the director who executes the encounter plan is notified (step S362).
- the director may be a facility such as a restaurant or an amusement park.
- the facility has the advantage of being an advertisement because it can be used by two people through the encounter plan.
- step S368 an effect by the supplier is executed (step S368).
- an effect by the supplier is executed (step S368).
- the referral partner or a seat with the best view in a place with a good view.
- performers may be arranged around, so that the user can talk about the topic of a tourist spot that leads to the next date, or how wonderful the marriage is.
- the user inputs user basic information from the client terminal 1 (step S393).
- Information input at the client terminal 1 is transmitted to the agent server 2 and stored in the user information DB 352 by the user / agent information management unit 351.
- FIG. 24A shows an example of user basic information.
- the basic user information includes information on age, occupation, family structure, living with family, annual income, residence, and blood type.
- each item of the user basic information has a public attribute indicating whether or not it may be viewed by another person.
- the user inputs user basic condition information from the client terminal 1 (step S396).
- the user basic condition information is information related to the attributes of an ideal introduction partner.
- FIG. 24B shows an example of user basic condition information.
- the user basic condition information includes information on age, occupation, family structure, family living, annual income, residence, and blood type.
- the user can set priorities for each item, and can select the first agent with emphasis on the items with high priorities.
- the items of the user basic condition information are not limited to the example illustrated in FIG. 24B, and may include items such as appearance, personality, speech, and voice tone.
- the user preference information is information related to user preferences.
- FIG. 25 shows an example of user preference information.
- the user preference information is set, for example, in the range of -1.0 (dislike) to 1.0 (like) for each item.
- the present invention is not limited to numerical value input, and when user preference information is input, “love, like, neither, dislike, dislike” or the like may be input for each item in a questionnaire format.
- the user / agent information management unit 351 performs user personality diagnosis (step S402).
- the personality diagnosis of a user can be classified into several types of personality based on answers to questions presented to the user in a questionnaire format, or a radar chart or There is a method of diagnosing with a line graph.
- egogram test for example, by giving users questions such as “I am not good at refusing?” Or “Is it difficult to keep time?” Scores are calculated for each element (called CP (Controlling Parent), NP (Nurturing Parent), A (Adult ego state), FC (Free Child), AC (adapted ⁇ child)), and finally a polygonal line called an egogram Personality diagnosis results can be expressed in a graph. When presenting to the user, it is expressed using a line graph or a radar chart.
- the score of each element is stored as user personality information, and the score of each user's element is compared to compare the personalities. It can be determined whether the personality is a degree or user preference.
- the user / agent information management unit 351 performs a personality diagnosis of the ideal person of the user (step S405).
- the personality of the user's ideal person can be set, for example, by the user inputting the egogram test as an ideal partner.
- the score of each element is stored in the user information DB 352 as ideal personality information.
- the basic information, basic condition information, preference information, personality information, and ideal personality information described above are also possessed by the agent and stored in the agent information DB 360.
- An example of the basic information of the agent is shown in FIG. 26A
- An example of the basic condition information is shown in FIG. 26B
- an example of the preference information is shown in FIG.
- Basic information, basic condition information, preference information, personality information, and ideal personality information can be set as initial values for defining the personality of the agent.
- the basic information, basic condition information, preference information, personality information, and ideal personality information prepared in the preparation phase have the same values for the actual person (marriage hunting member) on the marriage counselor side managed by the management server 4. It is stored in the marriage hunting member information 41.
- FIG. 28 is a diagram illustrating an example of an agent selection screen according to the present embodiment.
- the illustrated screen 200 displays an agent image (hand-drawn picture, CG, photo, etc.), the basic profile of the agent, and the personality and preferences of the agent.
- the personality of the agent may be displayed as a personality chart as shown in FIG. 28, or may be displayed as a line graph based on the egogram described above.
- the agent selection candidates may be generated reflecting user preferences. Agent selection candidates are selected from the agents set in the agent information DB 360 in a profile that matches the user's preference, and can be an index such as popularity or evaluation order regardless of the user's preference. It may be selected based on this.
- FIG. 29A and FIG. 29B are flowcharts showing the search phase operation process according to the present embodiment.
- an agent is selected by the user (steps S413 to S437).
- three agent selection methods are prepared.
- the user can select an agent by a favorite selection method.
- step S413 / Yes images of a plurality of female agents registered in the agent information DB 360 are displayed, and the user selects a female agent who looks like (step S416).
- the agent learning unit 353 extracts the ideal basic condition and ideal personality of the user input in the preparation phase from the user information DB 352, and sets the selected female agent as it is (registered in the user agent DB 354) ( Step S419, Step S422).
- the agent learning unit 353 has a female agent having an attribute close to the agent information DB 360 based on the basic condition information of the user and the ideal personality information input in the preparation phase.
- the search result is presented to the user, and agent selection is accepted (step S434).
- the user's preference information may be taken into consideration, and a female agent having preference information similar to the user may be searched. Further, among the items of the basic condition information, priority may be given to the female agent having the highest number of attributes with priority given to the item with the highest priority.
- step S425 / No the user arbitrarily selects from all agents while confirming the appearance and profile of the agent (step S437).
- the user agent interaction processing unit 350 acquires the user's behavior history and biometric information (steps S440 and S443).
- the user's action history is a history of an activity place, an activity content history, a purchase history, an SNS posting history, etc., and is acquired from the client terminal 1 or a predetermined server.
- the user's behavior can be grasped in detail by behavior recognition based on position information, Internet history, acceleration sensor data, gyro sensor data, and the like.
- the user's biometric information is acquired from the client terminal 1 in real time, and the current state of the user (tensed, sleepy, laughing, etc.) is grasped.
- the user agent dialogue processing unit 350 performs dialogue processing with the user by the agent (step S446).
- the user agent interaction processing unit 350 may generate a response sentence with reference to the acquired user behavior history and biometric information and output it as an agent utterance.
- FIG. 30 to FIG. 32 show an example of the interaction by the agent using the user action history and the biometric information.
- the agent and the user interact with each other via the chat screen of the display unit 106.
- FIG. 30 is a diagram illustrating an example of an agent utterance using a user action history according to the present embodiment.
- the dialogue between the user M and the agent “Saki” is displayed in a chat format.
- a text input field and a send button are displayed at the bottom of the screen 202.
- the user inputs a message in the text input field and taps the send button, the input message is transmitted to the agent server 2 as an utterance text. .
- the user agent interaction processing unit 350 generates a response of the agent “Saki” based on, for example, a preset question sentence and response sentence data set. Specifically, for example, "Do you like XX?" For questions such as “I'm nailed to XX", “I'm crazy about XX", “I don't have eyes on XX”. To respond.
- the user agent dialogue processing unit 350 refers to, for example, an action history that the user M got off at Y station today, and as shown in FIG. There are many ramen shops! It is also possible to output the utterance. As a result, the user M can be expected to feel a sense of familiarity with the agent or to develop a consciousness of being a partner of his own.
- FIG. 31 is a diagram for explaining an example of an agent utterance using the user's biometric information according to the present embodiment.
- the biometric information is obtained from a biometric sensor provided in a wearable device worn by the user or a camera that captures the user's face image, and is transmitted from the client terminal 1 to the agent server 2 in real time.
- the user agent dialogue processing unit 350 estimates a user's state (psychological state, emotion) based on the biological information, and generates an utterance reflecting the estimation result as an agent's utterance. For example, when the heart rate of the user M is faster than usual, as shown on the screen 203 in FIG. Is it something different from usual? The utterance that synchronizes with the other person's psychological state is output, such as “I ’ve been formed”. As a result, the user M can expect a closer psychological distance from the agent and feel closer.
- FIG. 32 is a diagram illustrating an example of an agent utterance using SNS information according to the present embodiment.
- the user agent dialogue processing unit 350 searches a topic that matches the user's preference information from an SNS or a website and uses it for the utterance of the agent. For example, as displayed on the screen 204 in FIG. 32, an utterance such as “I don't know well, but it seems that the SNS was excited by the draft topic of the AA team a while ago” is output. Thus, by providing a topic that matches the preference of the user M, the story is excited, and the user M can be expected to feel the conversation with the agent more enjoyable.
- FIGS. 33A and 33B Details of the dialogue processing by the user agent dialogue processing unit 350 described above are shown in FIGS. 33A and 33B.
- FIG. 33A and 33B are flowcharts of the agent dialogue process according to the present embodiment.
- the user agent dialogue processing unit 350 makes a greeting when a certain time has elapsed since the previous conversation with the user (step S473 / Yes) (step S476).
- the user agent dialogue processing unit 350 When the introduction user is not selected for a predetermined period (step S479 / Yes), the user agent dialogue processing unit 350 generates a personality improvement (or ideal improvement) message for the user and speaks ( Step S482). The selection of introduction users will be described in detail later.
- the agent server 2 performs voice recognition on the uttered voice (step S488) and performs text analysis on the uttered text (step S488). Step S491).
- the user agent dialogue processing unit 350 of the agent server 2 generates utterance candidates according to the user attributes (basic information, basic condition information, and preference information items) (step S494).
- step S497 when the user's action history is available (step S497 / Yes), the user agent interaction processing unit 350 modifies the utterance candidate according to the action history (step S498).
- the user agent interaction processing unit 350 modifies the utterance candidate according to the biometric information (step S503). Note that the user agent dialogue processing unit 350 may modify the utterance candidate using both the action history and the biological information.
- step S506 / Yes when there is an utterance candidate (step S506 / Yes), the user agent interaction processing unit 350 outputs the utterance candidate as an agent utterance (step S509). If there is no utterance candidate (step S506 / No), the agent does not speak.
- the response when there is a new input (utterance) from the user has been described.
- the dialogue according to the present embodiment is not limited to this, and the agent side may make a statement.
- step S485 when there is no new input (utterance) from the user (step S485 / Yes), as shown in FIG. 33B, the user agent interaction processing unit 350 adds a topic to the user's action history. It is determined whether or not there is (step S512). For example, the user agent dialogue processing unit 350 determines from the behavior history whether or not there has been a noticeable behavior such as going to a sightseeing spot, getting off at a different station, or doing expensive shopping.
- step S512 when there is a topic (step S512 / Yes), the user agent dialogue processing unit 350 generates an utterance candidate using the action history (step S515).
- step S528 it is determined whether or not there is a topic in the user's biometric information.
- the user agent dialogue processing unit 350 determines a notable state or emotion different from usual, such as smile, fun, sleepy, tired, excited, angry, based on the biological information.
- step S518 / Yes when there is a topic (step S518 / Yes), the user agent dialogue processing unit 350 generates an utterance candidate using the biological information (step S521).
- the user's favorite topic is searched from the most recent SNS, Web site, etc., and when there is a user's favorite topic (step S524 / Yes), the user agent interaction processing unit 350 uses the topic such as the SNS. To generate utterance candidates (step S527).
- FIG. 33B shows a flow of generating an utterance candidate using any one of action history, biometric information, SNS information, and the like, but this embodiment is not limited to this, and the action history, It is also possible to generate an utterance candidate by combining any one or more of biometric information and SNS information.
- the agent learning unit 353 appropriately learns the agent according to the interaction between the agent and the user (step S449). Specifically, the agent learning unit 353 updates the agent attributes so that they become user-preferred. Hereinafter, this will be specifically described with reference to FIGS.
- FIG. 34 is a flowchart showing an agent learning operation process according to this embodiment. As shown in FIG. 34, first, the agent learning unit 353 performs text analysis on the user's input sentence (utterance text) (step S530).
- the agent learning unit 353 performs a positive / negative determination of the user with respect to the object (the event to be a preference object) (step S533).
- the agent learning unit 353 confirms whether or not the object is an item included in the agent's preference information (step S536). If not, the agent learning unit 353 includes the object in the agent's preference information. Is added (step S539).
- the agent learning unit 353 adjusts the preference degree (preference value) of the corresponding item in the agent preference information (step S541). That is, the agent learning unit 353 can determine the positive / negative of the user for a certain event based on the conversation content, and reflect the determination result in the preference information of the agent, thereby bringing the preference of the agent closer to the preference of the user. it can. Note that the user's preference may be reflected not only in the agent's preference information but also in the basic information and basic condition information of the agent.
- the agent learning unit 353 since the agent learning unit 353 becomes unnatural when the preference and personality of the agent change suddenly, the numerical value may be updated within a certain range. For example, as shown in FIG. 35, the user said, “Today, I was stuck in the ramen ranking on TV and I wanted to eat tonkotsu ramen. However, “TV”, “Ramen ranking”, “Tonkotsu ramen”, and “soy sauce ramen” are extracted as objects (preference target words) by text analysis. The agent learning unit 353 performs positive / negative determination for each object as shown in the upper right table of FIG. 35 according to each corresponding word indicating the user's identification for the object. In the example shown in FIG.
- the agent learning unit 353 may normalize and set the preference degree so that 1.0 is maximum and -1.0 is minimum.
- the agent learning unit 353 determines the user's word usage category (step S544).
- the word usage category is divided into, for example, polite, brilliant (unfriendly), gentle, violent, fast-paced, and slowly.
- the agent learning unit 353 adjusts the agent's word usage category parameter (step S547).
- the agent word usage category parameters may be stored in the user agent DB 354.
- the word usage category determination the most common category among the word usage categories collected from the user's dialogue during a certain period may be used as the word usage category representing the user. For example, in the example shown in FIG. 35, as shown in the lower right of FIG. 35, the “Polite” category is extracted three times, and the “Blitter” category is extracted once. It may be determined as “polite”.
- the agent learning unit 353 may use the appearance probability of each word usage category extracted from the user's dialogue as the determination result. Then, the agent learning unit 353 reflects the word usage category representing the user or the appearance probability of each word usage category in the word usage category parameter of the agent.
- the agent learning unit 353 confirms whether or not the change in the preference information or the word usage parameter described above affects the personality of the agent (step S550), and adjusts the parameter of the personality information of the agent if affected. (Step S553). Since personality appears in preferences and word usage, the personality parameters are adjusted according to the changed preferences and language usage.
- the agent gradually grows into a user-preferred agent while interacting with the user.
- the introduction user selection unit 355 determines whether or not the update interval of the matching described below has exceeded a threshold value (step S452).
- the matching update interval is set as appropriate, for example, one day or one week.
- the introduction user selection unit 355 performs matching between the real person and the agent (step S455).
- matching between an actual person and an agent is performed periodically while the dialogue between the agent and the user proceeds.
- the introduction user selection unit 355 refers to the marriage hunting member information 41 and searches for an actual person similar to the agent.
- the agent for the user since the agent for the user has changed to the user's preference, it is possible to more effectively search for a partner who is compatible with the user by matching a real person similar to the agent.
- FIG. 36 is a flowchart showing a matching operation process according to this embodiment.
- the introduction user selection unit 355 searches the marriage hunting member information for a person who matches all the basic condition information of the agent (step S563).
- FIG. 37 is a diagram for explaining search using basic condition information of an agent.
- the marriage hunting member information is screened based on whether or not the item of the basic condition satisfies the condition.
- the introduction user selection unit 355 sends the basic condition information to the management server 4 and acquires one or more member IDs that satisfy the condition from the married member information 41.
- step S566 it is determined whether or not the number of candidates searched (number of active members) exceeds a threshold value (a certain number) (step S566). Thereby, it is possible to secure the number of candidates for selecting matching partners, for example, at least 10 people.
- the introduction user selection unit 355 searches for candidates that satisfy the priority “high / medium” item of the basic condition information (step S569).
- the introduction user selection unit 355 further searches for candidates that satisfy the priority “high” item of the basic condition information (step S575).
- the introduction user selection unit 355 obtains a correlation between the agent and the personality diagnosis result of each candidate (step S581).
- the personality diagnosis explained in the preparation phase is also performed in advance by each mariting member.
- the introduction user selection unit 355 adds the squares of the differences between the values of the agent and each candidate egogram, and obtains the cross-correlation.
- the introduction user selection unit 355 excludes candidates whose correlation value is lower than a predetermined threshold (step S584).
- the introduction user selection unit 355 calculates the similarity between the remaining candidate and the preference of the agent (step S587), and selects the candidate (person P) having the highest similarity (step S590).
- the preference similarity is calculated using preference information.
- the introduction user selection unit 355 generates an n-dimensional vector in which numerical values of matching items among the items of preference information of an agent and a candidate are arranged, and the agent's n-dimensional vector and the candidate Finds the inner product of cos ⁇ of n-dimensional vectors. If the numerical value is 1.0, the similarity is completely coincident, and if it is 0.0, it is completely inconsistent.
- n is smaller than a predetermined threshold, the candidate is excluded from the candidates without calculating the similarity. This is because the reliability is low even if the number of items matches.
- the introduction user selection unit 355 generates a five-dimensional vector in which the numerical values of these five items are arranged.
- step S578 / No if the number of candidates does not exceed the threshold value (step S578 / No), since there are few candidates, the matching is canceled and the process returns to step S440 shown in FIG. 29B.
- the introduction user selection unit 355 checks whether or not the selected person P is the same as the previously selected person (step S461). As described above, the matching of the real person can be performed periodically (for example, once a day) while the interaction between the agent and the user is performed.
- step S461 when the selected person P is the same as the previously selected person (step S461 / Yes), the introduction user selection unit 355 increments the same person count Cs (step S464).
- step S470 when the selected person P is different from the previously selected person (step S461 / No), the introduction user selection unit 355 resets the same person count Cs (step S470).
- the introduction user selection unit 355 periodically performs matching with the actual person until the same person count Cs exceeds the threshold (step S467).
- the same person count Cs exceeds the threshold value (step S467 / Yes)
- the fixed phase shown in FIG. 39 is started.
- FIG. 39 4-3-3. Fixed phase
- FIG. 39 In the fixed phase, while the user and the agent continue to interact as in the search phase, the agent's preference and personality attributes are gradually brought closer to the real person matched in the search phase, and the real existence is made without the user's awareness. Affinity (intimacy) with a person is determined.
- FIG. 39 is a diagram showing an operation process of a fixed phase according to the present embodiment. As shown in FIG. 39, first, the approval waiting flag is set to False (step S603).
- the familiarity calculation unit 356 sets the familiarity to 0 (step S606).
- the user agent interaction processing unit 350 continues the interaction processing with the user using the user's behavior history and biometric information in the same manner as the above steps S440 to S446 (steps S609 to S615).
- the agent learning unit 353 confirms whether or not a change has occurred in the preference, basic information, and personality information attributes of the selected person P (step S618). And reflected in attributes such as basic information and personality information (step S621). At this time, the changed value may be copied as it is, but the user may feel unnaturalness when the preference of the agent changes abruptly. Therefore, the amount to be changed at a time may be limited. The agent's preference is matched with the preference of the person P over a certain period of time.
- FIG. 40 shows an example of dialogue when the preference of the agent changes.
- the agent “Saki” shown in FIG. 40 is an attribute that does not like the AA team at the time of the search phase in which the user M was interacting (in the search phase, the agent is gradually adjusted to the user's preference). I should have become an AA team fan over time.
- the preference of the agent “Saki” approaches the preference of the real person this time. If the real person happens to be an AA team fan by accident, the preference of the agent “Saki” is adjusted to like the AA team. For example, as shown on the screen 206 in FIG. 40, “I also like the AA team” I came. Yesterday's match was amazing! "A more positive remarks than before are output on the topic of the AA team.
- the closeness calculation unit 356 calculates the closeness between the user and the agent (step S636).
- the calculation method of the familiarity may be the same as the matching method between the agent and the real person performed by the introduction user selection unit 355 using the basic condition information, the preference information, and the like. That is, the degree of matching of items such as basic condition information and preference information is calculated as the familiarity. Further, a numerical value based on the conversation amount between the user and the agent or the smile amount during the conversation may be added to the intimacy.
- FIG. 41 shows an example of an introduction approval approval / disapproval notification screen.
- the profile of the user M is displayed on the screen 207 (in a public range), and an approval button 207a and a rejection button 207b are further displayed.
- the person P who has received the introduction approval approval / disapproval confirmation confirms the profile of the user M, taps the approval button 207a when approving the introduction of himself / herself, and taps the reject button 207b when not approving.
- step S648 the user M side waits for approval (step S648), and dialogue with the agent and learning of the agent are continued (steps S609 to S624).
- step S627 / Yes when there is no response from the person P for a certain period of time (step S627 / Yes), the introduction processing unit 357 sets the approval waiting flag to false (step S630), and the familiarity calculation unit 356 determines the familiarity. It is set to 0 (step S633). In this case, the matching with the person P is canceled and the process returns to the search phase.
- step S651 / No when the approval from the person P is not recognized (step S651 / No), the closeness is reset to 0 (step S633), the matching with the person P is released, and the process returns to the search phase.
- step S651 / Yes when the approval from the person P is accepted (step S651 / Yes), the introduction phase shown in FIG. 42 is started.
- the introduction phase of this embodiment will be described with reference to FIGS.
- the user is notified that there is a person who can be introduced, triggered by the approval of a real person. From this point, the user knows the existence of a real person and can view the profile. In addition, the profile of the real person presented to the user side is only that whose public attribute of the basic information is “Yes”.
- FIG. 42 is a flowchart showing the operation process of the introduction phase according to the present embodiment.
- the introduction user selection unit 355 notifies the user that there is a person who can be introduced (step S663).
- FIG. 43 shows an example of an introduction notification.
- an agent “Saki” and a chat of the user M are displayed on the screen 208. If the referral approval is permitted from the person P linked to the agent “Saki”, the referral user selection unit 355 “suddenly informs the user from the operation! A woman similar to Saki was found. Want to get in touch with the heart button at the bottom of the screen? ”Is interrupted. Since the message is an interrupt notification, the dialogue with the agent “Saki” continues.
- FIG. 44 shows an example of a detailed information display screen of the introduction partner.
- a profile of the introduction partner person P
- a button 209a that the user wants to meet and a button 209b to quit are displayed on the screen 209 (step S669).
- the conversation between the user and the person P is performed by chat or voice through the agent (step S672).
- the dialogue here is a dialogue through the agent, and the user's voice is output as the agent's voice or displayed as the agent's voice without directly listening to the voice of the other party.
- the user knows the partner's profile for the first time in the introduction phase, but since the matching between the user and the partner is sufficiently performed in the process up to the fixed phase, the profile information of the person P displayed from the user side is It is the same as the preference of the agent who has been in contact with it so far, and the personality is expected to be similar, so that it is possible to shift to a conversation with the person P with a good impression.
- step S675 / Yes when one of the user and the person P presses the “stop” button (step S675 / Yes), the matching with the person P is canceled.
- the agent learning unit 353 deletes the agent (step S681), and starts again from the search phase agent selection (steps S413 to S437).
- step S678 / No the current agent remains unchanged (matching with the person P is canceled), and the process returns to the search phase dialog loop (steps S440 to S446) and again. Matching is performed (at this time, the person P whose matching is canceled may be excluded for a certain period).
- step S684 when both sides press the “I want to meet” button (step S684 / Yes) and the user desires support (step S685), the support phase shown in FIG. 45 is started.
- the agent learning unit 353 performs an agent erasure process (an agent program activated on the client terminal 1) (step S688).
- the user agent dialogue processing unit 350 may perform dialogue control such that the agent leaves the user by giving a greeting as will be described later with reference to FIG. Further, when the support phase is unnecessary, the user decides a meeting date and place, exchanges contact information, etc. in a dialogue with the person P via the agent.
- FIG. 45 is a flowchart showing the operation process of the support phase according to this embodiment.
- the user selects an arbitrary plan from a plurality of encounter plans presented by the scenario management unit 358 (step S693).
- FIG. 46 shows an example of an encounter plan selection screen.
- detailed information such as the title, content, and cost of the encounter plan, a selection button 210a, and a return button 210b are displayed on the screen 210.
- the encounter plan has a title such as “A nice Italian restaurant makes a happy surprise to her”, and the contents of the surprise menu include preparation of a special menu, preparation of a special seat, preparation of a present, and the like.
- Restaurants and facilities appearing in the encounter plan are tie-ups as advertisements, for example, and even with a free encounter plan, it can be expected to attract customers by guiding the user to a specific restaurant or facility.
- a paid dating plan may be prepared to guide the user to a place unrelated to the sponsor.
- FIG. 47 shows an example of an encounter plan notification screen.
- 47 shows a screen 211 displayed on the display unit 106A of the client terminal 1 of the person P.
- a message of the user M such as “I made a reservation here. I am looking forward to seeing you!”
- an accept button 211a and a decline button 211b are displayed.
- the person P confirms the date and time location and the contents of the plan, and if there is no problem, taps the accept button 211a, and taps the reject button 211b to refuse the invitation.
- a screen 212 displayed on the display unit 106B of the user M's client terminal 1 is shown.
- the screen 212 displays that the person P has been invited, detailed information on the encounter plan, an OK button 212a, and a redo button 212b.
- the user M confirms the date / time location and the plan contents, and taps the OK button 212a if there is no problem, and taps the redo button 212b to change the date / time and the plan.
- the scenario management unit 358 notifies the encounter plan (restaurant, facility, etc.) of the encounter plan (step S699).
- step S702 it waits until the designated date and time (step S702), and the scenario management unit 358 confirms whether the two people meet at the designated date and time when the designated date and time comes (step S705). Whether or not the two people have actually met may be automatically determined from the position information of the two people detected by GPS or the like, or the user M may be notified that the two have met manually.
- step S705 when the two people meet at the designated date and time place (step S705 / Yes), the scenario management unit 358 notifies the related companies that the two people have met (step S708).
- the scenario management unit 358 In the case of a restaurant, it can be clearly understood that two people who have made a reservation have arrived. Then, related companies provide the contents of the encounter plan to two people.
- the scenario management unit 358 waits for a completion notification from the affiliated company (step S711), and when the completion notification is received, the agent asks the user M whether or not the two have come together (step S714).
- the timing for inquiring may be after a certain period has elapsed.
- FIG. 48 shows an example of the last congratulatory message screen by the agent.
- a message asking “Did you get along with her?” From the agent “Saki” to the user M on the screen 213, a “Dating” button 213a, and a “No” button 213b is displayed.
- the user M taps the “associate” button 213a.
- agent "Saki” says "Yes ... irritation! I'm glad, because I'm a part of her. My role ends today. I ’m happy because I ’m here. I ’ll say goodbye. Goodbye. ”
- the agent learning unit 353 performs an agent erasure process (an agent application activated on the client terminal 1) (step S720).
- step S723 when the two cannot meet at the designated date and time (step S723 / No), the user agent interaction processing unit 350 presents a message to the user M that the user is disappointed (step S273).
- the user agent dialogue processing unit 350 presents a comfort or encouragement message from the agent to the user M (step S726).
- the agent learning unit 353 deletes the agent, and starts again from the search phase agent selection (steps S413 to S437).
- step S729 / No if the user does not reselect the agent (step S729 / No), the current agent remains unchanged (matching with the person P is canceled), and the process returns to the search phase dialog loop (steps S440 to S446) and again. Matching is performed (at this time, the person P whose matching is canceled may be excluded for a certain period).
- a computer-readable storage medium storing the computer program is also provided.
- the configuration in which various functions are realized by the agent server 2 on the Internet is shown.
- the present embodiment is not limited to this, and at least a part of the configuration of the agent server 2 is a user.
- Client terminal 1 smart phone, wearable terminal, etc.
- the entire configuration of the agent server 2 may be provided in the client terminal 1 so that all processing can be performed by the client terminal 1.
- the present embodiment is widely applied to person matching in various scenes.
- a storage unit for storing information of a plurality of agents each having different attributes and capable of interacting with a user;
- a communication unit that receives a message from the user from the client terminal and returns a response message to the client terminal; Selecting a specific agent from the plurality of agents in response to an instruction from a user; Records the updated attribute of the specific agent in response to the interaction between the specific agent and the user as the attribute of the user agent; Comparing the attribute of the user agent with the attributes of a plurality of existing counterpart users to identify the counterpart user most similar to the attribute of the user agent;
- a control unit that controls to notify the user of the presence of the counterpart user at a predetermined timing;
- An information processing system comprising: (2) The controller is Record the updated attribute of the specific agent so as to be close to the attribute of the user as the attribute of the user agent;
- the information processing system according to (1), wherein a partner user who is most similar to the attribute of the user agent is selected every certain period, and is specified as the partner user when the same
- the control unit when receiving a notification from the user to start dating with the other user, sends a congratulatory message to the user and controls the agent application in the client terminal to be deleted.
- a communication unit that transmits a message from the user and receives a response message to the message to a server device that stores information on a plurality of agents each having a different attribute and capable of interacting with the user; Selecting a specific agent from the plurality of agents in response to a user instruction; In response to the interaction between the specific agent and the user, a notification indicating that there is a real partner user who is most similar to the attribute of the user agent whose attribute of the specific agent has been updated is sent from the server device.
- a control unit that controls reception at a predetermined timing via the communication unit;
- An information processing apparatus comprising: (12) The information according to (11), wherein the control unit receives, from the server device, a notification of the presence of the counterpart user at a timing when it is determined that a closeness between the user and the user agent exceeds a threshold value. Processing equipment. (13) The information processing apparatus according to (12), wherein the control unit receives a scenario of encounter with the partner user by the server apparatus.
- the controller is In response to the user's instruction, a notification to start dating with the other user is sent to the server device; In response to the notification to start the association, the server device receives a congratulatory message and a control signal instructing to delete the agent application installed in the information processing device, and deletes the agent application in response to the control signal.
- the information processing apparatus according to (13), wherein the information processing apparatus is controlled to perform.
- the information processing apparatus according to any one of (11) to (14), wherein the attribute is at least one of profile information, preference information, personality information, ideal profile condition information, and ideal personality information. .
- Computer A communication unit that transmits a message from the user and receives a response message to the message to a server device that stores information on a plurality of agents each having a different attribute and capable of interacting with the user; Selecting a specific agent from the plurality of agents in response to a user instruction; In response to the interaction between the specific agent and the user, a notification indicating that there is a real partner user who is most similar to the attribute of the user agent whose attribute of the specific agent has been updated is sent from the server device.
- a control unit that controls reception at a predetermined timing via the communication unit; Program to function as.
- Agent server 30 Dialog processing part 300 Dialog processing part 310 Question sentence search part 320 Answer sentence generation part 330 Conversation DB 340 Phoneme Data Acquisition Unit 30a Dialogue Processing Unit 31 Basic Dialogue Processing Unit 32 Character A Dialogue Processing Unit 33 Person B Dialogue Processing Unit 34 Person C Dialogue Processing Unit 35 Matching Unit 350 User Agent Dialogue Processing Unit 351 User / Agent Information Management Unit 351 352 User information DB 353 Agent Learning Unit 354 User Agent DB 355 Introduction user selection unit 356 Intimacy calculation unit 357 Introduction processing unit 358 Scenario management unit 359 Scenario DB 360 Agent information DB 36 Communication unit 40 Phoneme storage unit 41 Basic phoneme DB 42 Character A Phoneme DB 43 Person B Phoneme DB 44 Person C Phoneme DB 50 Conversation DB Generation Unit 60 Phoneme DB Generation Unit 70 Advertisement Insertion Processing Unit 72 Advertising DB 80 Feedback acquisition processing unit 3 Network 10 Agent
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Biomedical Technology (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Transfer Between Computers (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Operations Research (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
Description
1.本開示の一実施形態による通信制御システムの概要
2.構成
2-1.システム構成
2-2.サーバの構成
3.システム動作処理
3-1.会話データ登録処理
3-2.音素DB生成処理
3-3.対話制御処理
3-4.会話DB更新処理
3-5.広告挿入処理
4.対話制御処理
4-1.構成
4-2.動作処理
4-3.実施例
(4-3-1.準備フェーズ)
(4-3-2.探索フェーズ)
(4-3-3.固定フェーズ)
(4-3-4.紹介フェーズ)
(4-3-5.支援フェーズ)
5.まとめ
本開示の一実施形態による通信制御システム(エージェントシステム)は、エージェントとの対話をシームレスに実世界の人物とのコミュニケーションに繋げることを可能とする。以下、図1を参照して本実施形態による通信制御システムの概要について説明する。
ここで、従来のエージェントシステムでは、人間の代わりをするエージェントがユーザの会話相手になったり、ユーザのスケジュール管理や情報の整理の手伝いをする等、エンターテイメントまたはエンターテイメント性を持った実用ツールとして提案されてきた。また、エージェントを複数から選ぶことができたり、会話内容に応じてエージェントを学習、成長させることが可能であった。
<2-1.システム構成>
続いて、上述した本実施形態による通信制御システムの全体構成について図2を参照して説明する。図2は、本実施形態による通信制御システムの全体構成を示す図である。
図3は、本実施形態によるエージェントサーバ2の構成の一例を示すブロック図である。図3に示すように、エージェントサーバ2は、音声エージェントI/F(インタフェース)20、対話処理部30、音素記憶部40、会話DB生成部50、音素DB生成部60、広告挿入処理部70、広告DB72、およびフィードバック取得処理部80を有する。
<3-1.会話データ登録処理>
図5は、本実施形態による会話DB330の生成処理を示すフローチャートである。図5に示すように、まず、会話DB生成部50は、想定される質問文を保存する(ステップS103)。
ペア1
質問文:おはよう。
回答文:今日の調子はどうですか?
ペア2
質問文:今日の天気は?
回答文:今日の天気は○○です。
図6は、本実施形態による音素DBの生成処理を示すフローチャートである。図6に示すように、まず、音素DB生成部60は、例文の表示を行う(ステップS113)。例文の表示は、例えば図示しない情報処理端末のディスプレイに、音素データ生成のために必要な例文を表示する。
図7は、本実施形態による対話制御処理を示すフローチャートである。図7に示すように、まず、音声エージェントI/F20は、ユーザの質問音声およびエージェントIDを取得したか否かを確認する(ステップS143)。エージェントIDは、キャラクターA、人物B、人物Cといった特定のエージェントを示す識別情報である。ユーザは、エージェント毎の音素データを購入することができ、例えば購入処理時に購入したエージェントのIDがクライアント端末1に保存される。
次に、各対話処理部300の会話DB330の更新処理について説明する。本実施形態では、ユーザとの会話によって会話DB330を成長させることが可能である。
キャラクターA:「おはよう」
ユーザ:「NG。元気で頑張ってと答えて」
キャラクターA:「元気で頑張って」
キャラクターA:「質問がわかりません」(該当する回答が無い場合の回答データ例)
ユーザ:「『元気?』と聞いたら、『今日も元気だよ』と答えて」
キャラクターA:「今日も元気だよ」
続いて、広告挿入処理部70による広告情報の挿入処理について図13~図14を参照して説明する。本実施形態では、広告挿入処理部70により、エージェントの発言に広告DB72に格納されている広告情報の挿入を行うことが可能である。広告DB72には、予め広告情報が登録され得る。図13は、本実施形態による広告DB72に登録されている広告情報の一例を示す図である。
キャラクターA:「おはよう!今日の調子はどうですか?」
ユーザ:「元気だよ。何か美味しい物食べたいな」
キャラクターA:「CC店の焼肉が美味しいらしいよ」
<4-1.構成>
(4-1-1.システム構成)
図15は、本実施形態による通信制御システムのシステム構成例を示す図である。本実施形態では、一例として、ユーザとエージェントとの対話に応じて、結婚相談所の会員情報を用いて、婚活会員とのマッチングを行う場合について説明する。
次に、本実施形態によるエージェントサーバ2に含まれる対話処理部30aの構成例について図16を参照して説明する。本実施形態による対話処理部30aは、エージェントとの対話をシームレスに実世界の人物とのコミュニケーションに繋げる実世界実働型のエージェントシステムを実現する。
図17は、本実施形態によるマッチング部35の構成例を示す図である。図17に示すように、マッチング部35は、ユーザ用エージェント対話処理部350、ユーザ/エージェント情報管理部351、ユーザ情報DB352、エージェント情報DB360、エージェント学習部353、ユーザ用エージェントDB354、紹介ユーザ選出部355、親密度算出部356、紹介処理部357、シナリオ管理部358、およびシナリオDB359を有する。
続いて、本実施形態によるクライアント端末1の構成について図18を参照して説明する。図18は、本実施形態によるクライアント端末1の構成の一例を示すブロック図である。
制御部100は、例えばクライアント端末1が有するCPU(Central Processing Unit)のようなプロセッサによって実現される。本実施形態による制御部100は、例えば通信部101を介してエージェントサーバ2から送信されたエージェントの応答音声をスピーカ107から再生するよう制御したり、エージェントの画像を表示部106に表示するよう制御したりする。
通信部101は、例えば、ネットワーク3に接続するための通信デバイスなどで構成された通信インターフェースである。通信部101は、例えば、LAN(Local Area Network)、Bluetooth(登録商標)、Wi-Fi、またはWUSB(Wireless USB)用の通信カードなどでありうる。また、通信部101は、光通信用のルータ、ADSL(Asymmetric Digital Subscriber Line)用のルータ、または、各種通信用のモデムなどであってもよい。通信部101は、例えばインターネットや他の通信機器との間で、TCP/IPなどの所定のプロトコルを用いて信号などを送受信する。また、通信部101に接続されるネットワーク3は、有線または無線によって接続されたネットワークであり、例えば、インターネット、家庭内LAN、赤外線通信、ラジオ波通信または衛星通信などを含みうる。
操作入力部102は、ユーザ操作の入力を受け付け、制御部100に出力する機能を有する。操作入力部102は、例えばマウス、キーボード、タッチパネル、ボタン、スイッチ、またはレバーなどにより実現される。
センサ103は、ユーザまたは周辺状況を検知する機能を有する。例えばセンサ103は、生体センサ(脈拍計、心拍計、発汗センサ、体温センサ、血圧センサ、脳波計等)、環境センサ(温度センサ、照度センサ、圧力計等)、加速度センサ、ジャイロセンサ、方位センサ、振動センサ、または位置測位センサなどにより実現される。
カメラ104は、撮像レンズ、絞り、ズームレンズ、及びフォーカスレンズ等により構成されるレンズ系、レンズ系に対してフォーカス動作やズーム動作を行わせる駆動系、レンズ系で得られる撮像光を光電変換して撮像信号を生成する固体撮像素子アレイ等を各々有する。固体撮像素子アレイは、例えばCCD(Charge Coupled Device)センサアレイや、CMOS(Complementary Metal Oxide Semiconductor)センサアレイにより実現されてもよい。
マイク105は、ユーザの音声や周囲の環境音を収音し、音声データとして制御部100に出力する。
表示部106は、文字、図、画像、映像等を表示する機能を有する。表示部106は、例えば液晶ディスプレイ(LCD)装置、OLED(Organic Light Emitting Diode)装置等により実現される。
スピーカ107は、音声信号を再生する機能を有する。
記憶部108は、制御部100が各機能を実行するためのプログラムやパラメータを格納する。例えば記憶部108は、ユーザID、氏名、年齢、性別、職業、家族構成、家族同居、年収、居住地、嗜好情報等のユーザ情報を記憶していてもよい。
本実施形態による動作処理は、探索、固定、紹介、支援の大きく4つのフェーズから成る。
図19は、本実施形態による探索フェーズの動作処理を示すフローチャートである。図19に示すように、まず、ユーザにより一人のエージェントが選択されると(ステップS270)、ユーザ用エージェント対話処理部350は、選択されたエージェントによるユーザとの対話を開始する(ステップS273)。
図20は、本実施形態による固定フェーズの動作処理を示すフローチャートである。固定フェーズでは、マッチングした人物(選出した紹介人物、以下「実在人物」とも称す)を固定し、本当に当該人物とユーザの相性(ここでは「親密度」により判断される)が良いかを確認するためのフェーズである。したがって、ここでは、実在人物の属性が変化した場合には対応するエージェントの属性が実在の人物に合わせて変化される。
図21は、本実施形態による紹介フェーズの動作処理を示すフローチャートである。紹介フェーズでは、ユーザ側に紹介できる実在の人物がいることが初めて通知される。
図22は、本実施形態による支援フェーズの動作処理を示すフローチャートである。紹介フェーズにおいて実在人物との繋がりは達成しているため、支援フェーズは必ずしも実行する必要はないが、ユーザの要求に応じて実施され得る。ここでは、二人が出会う時のシチュエーションや演出を「シナリオ」という形で選択することが可能である。
続いて、本実施形態について実施例を用いて詳細に説明する。ここでは、結婚相談所(管理サーバ4)と連携した例について説明する。本実施例では、システムを利用しているユーザを仮に成人男性のユーザとし、結婚相談所の会員情報に基づいて紹介相手の女性をマッチングする。また、本実施例では、上述した4つのフェーズの前に、準備を行うための準備フェーズについて説明する。
ユーザは、本システムを利用する準備としてプロフィール(ユーザ情報)を入力する。以下、図23を参照して具体的に説明する。
続いて、探索フェーズの実施例について図28~図38を参照して具体的に説明する。
次に、固定フェーズの実施例について図39~図41を参照して具体的に説明する。固定フェーズでは、ユーザとエージェントが探索フェーズの際と同様に対話を続けながら一方でエージェントの嗜好や性格の属性が探索フェーズでマッチングされた実在人物に徐々に近付けられ、ユーザが気付かないうちに実在人物との相性(親密度)が判断される。
次に、本実施例の紹介フェーズについて図42~図44を参照して説明する。紹介フェーズでは、実在の人物が承認を行ったことをトリガとして、ユーザ側に紹介できる人物がいる旨が通知される。ユーザはこの時点から実在の人物の存在を知り、プロフィールを見ることができるようになる。なおユーザ側に提示される実在人物のプロフィールは、基本情報の公開属性が「はい」になっているものだけである。
次に、支援フェーズの実施例について図45~図48を参照して説明する。支援フェーズでは、シナリオDB359に登録されている出会いを支援するシナリオリストからユーザが任意でシナリオ(以下、「出会いプラン」とも称す)を選択し、シナリオに従った出会い演出が関係業者によって実施される。シナリオは有料制にしてもよい。
上述したように、本開示の実施形態による通信制御システムでは、エージェントとの対話をシームレスに実世界の人物とのコミュニケーションに繋げることが可能である。
(1)
それぞれ異なる属性を有し、ユーザと対話可能な複数のエージェントの情報が記憶される記憶部と、
ユーザからのメッセージをクライアント端末から受信すると共に、応答メッセージを当該クライアント端末に返信する通信部と、
ユーザからの指示に応じて、前記複数のエージェントから特定のエージェントを選択し;
前記特定のエージェントと前記ユーザとの対話に応じて、当該特定のエージェントの属性を更新したものをユーザ用エージェントの属性として記録し;
前記ユーザ用エージェントの属性と、実在する複数の相手ユーザの属性とを比較することにより、当該ユーザ用エージェントの属性に最も類似する相手ユーザを特定し;
所定のタイミングで前記ユーザに前記相手ユーザの存在を通知するように制御する制御部と、
を備える、情報処理システム。
(2)
前記制御部は、
前記特定のエージェントの属性を前記ユーザの属性に近付けるよう更新したものを当該ユーザ用エージェントの属性として記録し;
一定期間毎に前記ユーザ用エージェントの属性に最も類似する相手ユーザを選出し、同一人物が所定回数選出された場合に相手ユーザとして特定する、前記(1)に記載の情報処理システム。
(3)
前記制御部は、前記ユーザ用エージェントの属性に最も類似する相手ユーザの属性に応じて、前記ユーザ用エージェントの属性を更新する、前記(1)または(2)に記載の情報処理システム。
(4)
前記ユーザ用エージェントの属性に最も類似する相手ユーザが一定期間同一であるとき、当該相手ユーザの属性に応じて、前記ユーザ用エージェントの属性を更新する、前記(2)~(3)のいずれか1項に記載の情報処理システム。
(5)
前記実在する複数の相手ユーザの属性は、外部装置から取得される、前記(4)に記載の情報処理システム。
(6)
前記制御部は、前記ユーザと、前記ユーザ用エージェントとの親密度が閾値を超えると、前記ユーザに前記相手ユーザの存在を通知する、前記(4)または(5)に記載の情報処理システム。
(7)
前記制御部は、さらに、前記相手ユーザに前記ユーザの存在を通知する、前記(6)に記載の情報処理システム。
(8)
前記制御部は、前記ユーザと前記相手ユーザの双方から会いたい旨の要求信号を受信すると、出会いのシナリオを前記ユーザに提供可能となる、前記(7)に記載の情報処理システム。
(9)
前記制御部は、前記ユーザから、前記相手ユーザと交際を始める旨の通知を受け取ると、前記ユーザにお祝いのメッセージを送信すると共に、前記クライアント端末におけるエージェントアプリケーションが消去されるように制御する、前記(8)に記載の情報処理システム。
(10)
前記属性は、プロフィール情報、嗜好情報、性格情報、理想のプロフィール条件情報、および理想の性格情報の少なくともいずれかである、前記(1)~(9)のいずれか1項に記載の情報処理システム。
(11)
それぞれが異なる属性を有し、ユーザと対話可能な複数のエージェントの情報が記憶されるサーバ装置に対して、ユーザからのメッセージを送信すると共に、そのメッセージに対する応答メッセージを受信する通信部と、
ユーザによる指示に応じて、前記複数のエージェントから特定のエージェントを選択し;
前記特定のエージェントと前記ユーザとの対話に応じて、当該特定のエージェントの属性が更新されたユーザ用エージェントの属性に最も類似する実在する相手ユーザが存在することを示す通知を、前記サーバ装置から前記通信部を介して所定のタイミングで受信するように制御する、制御部と、
を備える、情報処理装置。
(12)
前記制御部は、前記サーバ装置から、前記ユーザと前記ユーザ用エージェントとの親密度が閾値を超えると判断されたタイミングで前記相手ユーザの存在の通知を受信する、前記(11)に記載の情報処理装置。
(13)
前記制御部は、前記サーバ装置により、前記相手ユーザとの出会いのシナリオを受信する、前記(12)に記載の情報処理装置。
(14)
前記制御部は、
前記ユーザの指示に応じて、前記相手ユーザと交際を始める旨の通知を前記サーバ装置に送信し;
当該交際を始める旨の通知に応じて、サーバ装置から、お祝いのメッセージおよび情報処理装置に実装されるエージェントアプリケーションの消去を指示する制御信号を受信し、当該制御信号に応じて前記エージェントアプリケーションを消去するように制御する、前記(13)に記載の情報処理装置。
(15)
前記属性は、プロフィール情報、嗜好情報、性格情報、理想のプロフィール条件情報、および理想の性格情報の少なくともいずれかである、前記(11)~(14)のいずれか1項に記載の情報処理装置。
(16)
プロセッサが、
それぞれ異なる属性を有し、ユーザと対話可能な複数のエージェントの情報を記憶部に記憶することと、
ユーザからのメッセージをクライアント端末から受信すると共に、応答メッセージを当該クライアント端末に通信部により返信することと、
ユーザからの指示に応じて、前記複数のエージェントから特定のエージェントを選択し;
前記特定のエージェントと前記ユーザとの対話に応じて、当該特定のエージェントの属性を更新したものをユーザ用エージェントの属性として記録し;
前記ユーザ用エージェントの属性と、実在する複数の相手ユーザの属性とを比較することにより、当該ユーザ用エージェントの属性に最も類似する相手ユーザを特定し;
所定のタイミングで前記ユーザに前記相手ユーザの存在を通知するように制御することと、
を含む、情報処理方法。
(17)
コンピュータを、
それぞれが異なる属性を有し、ユーザと対話可能な複数のエージェントの情報が記憶されるサーバ装置に対して、ユーザからのメッセージを送信すると共に、そのメッセージに対する応答メッセージを受信する通信部と、
ユーザによる指示に応じて、前記複数のエージェントから特定のエージェントを選択し;
前記特定のエージェントと前記ユーザとの対話に応じて、当該特定のエージェントの属性が更新されたユーザ用エージェントの属性に最も類似する実在する相手ユーザが存在することを示す通知を、前記サーバ装置から前記通信部を介して所定のタイミングで受信するように制御する、制御部と、
として機能させるためのプログラム。
2 エージェントサーバ
30 対話処理部
300 対話処理部
310 質問文検索部
320 回答文生成部
330 会話DB
340 音素データ取得部
30a 対話処理部
31 基本対話処理部
32 キャラクターA対話処理部
33 人物B対話処理部
34 人物C対話処理部
35 マッチング部
350 ユーザ用エージェント対話処理部
351 ユーザ/エージェント情報管理部351
352 ユーザ情報DB
353 エージェント学習部
354 ユーザ用エージェントDB
355 紹介ユーザ選出部
356 親密度算出部
357 紹介処理部
358 シナリオ管理部
359 シナリオDB
360 エージェント情報DB
36 通信部
40 音素記憶部
41 基本用音素DB
42 キャラクターA音素DB
43 人物B音素DB
44 人物C音素DB
50 会話DB生成部
60 音素DB生成部
70 広告挿入処理部
72 広告DB
80 フィードバック取得処理部
3 ネットワーク
10 エージェント
Claims (17)
- それぞれ異なる属性を有し、ユーザと対話可能な複数のエージェントの情報が記憶される記憶部と、
ユーザからのメッセージをクライアント端末から受信すると共に、応答メッセージを当該クライアント端末に返信する通信部と、
ユーザからの指示に応じて、前記複数のエージェントから特定のエージェントを選択し;
前記特定のエージェントと前記ユーザとの対話に応じて、当該特定のエージェントの属性を更新したものをユーザ用エージェントの属性として記録し;
前記ユーザ用エージェントの属性と、実在する複数の相手ユーザの属性とを比較することにより、当該ユーザ用エージェントの属性に最も類似する相手ユーザを特定し;
所定のタイミングで前記ユーザに前記相手ユーザの存在を通知するように制御する制御部と、
を備える、情報処理システム。 - 前記制御部は、
前記特定のエージェントの属性を前記ユーザの属性に近付けるよう更新したものを当該ユーザ用エージェントの属性として記録し;
一定期間毎に前記ユーザ用エージェントの属性に最も類似する相手ユーザを選出し、同一人物が所定回数選出された場合に相手ユーザとして特定する、請求項1に記載の情報処理システム。 - 前記制御部は、前記ユーザ用エージェントの属性に最も類似する相手ユーザの属性に応じて、前記ユーザ用エージェントの属性を更新する、請求項1に記載の情報処理システム。
- 前記ユーザ用エージェントの属性に最も類似する相手ユーザが一定期間同一であるとき、当該相手ユーザの属性に応じて、前記ユーザ用エージェントの属性を更新する、請求項2に記載の情報処理システム。
- 前記実在する複数の相手ユーザの属性は、外部装置から取得される、請求項4に記載の情報処理システム。
- 前記制御部は、前記ユーザと、前記ユーザ用エージェントとの親密度が閾値を超えると、前記ユーザに前記相手ユーザの存在を通知する、請求項4に記載の情報処理システム。
- 前記制御部は、さらに、前記相手ユーザに前記ユーザの存在を通知する、請求項6に記載の情報処理システム。
- 前記制御部は、前記ユーザと前記相手ユーザの双方から会いたい旨の要求信号を受信すると、出会いのシナリオを前記ユーザに提供可能となる、請求項7に記載の情報処理システム。
- 前記制御部は、前記ユーザから、前記相手ユーザと交際を始める旨の通知を受け取ると、前記ユーザにお祝いのメッセージを送信すると共に、前記クライアント端末におけるエージェントアプリケーションが消去されるように制御する、請求項8に記載の情報処理システム。
- 前記属性は、プロフィール情報、嗜好情報、性格情報、理想のプロフィール条件情報、および理想の性格情報の少なくともいずれかである、請求項1に記載の情報処理システム。
- それぞれが異なる属性を有し、ユーザと対話可能な複数のエージェントの情報が記憶されるサーバ装置に対して、ユーザからのメッセージを送信すると共に、そのメッセージに対する応答メッセージを受信する通信部と、
ユーザによる指示に応じて、前記複数のエージェントから特定のエージェントを選択し;
前記特定のエージェントと前記ユーザとの対話に応じて、当該特定のエージェントの属性が更新されたユーザ用エージェントの属性に最も類似する実在する相手ユーザが存在することを示す通知を、前記サーバ装置から前記通信部を介して所定のタイミングで受信するように制御する、制御部と、
を備える、情報処理装置。 - 前記制御部は、前記サーバ装置から、前記ユーザと前記ユーザ用エージェントとの親密度が閾値を超えると判断されたタイミングで前記相手ユーザの存在の通知を受信する、請求項11に記載の情報処理装置。
- 前記制御部は、前記サーバ装置により、前記相手ユーザとの出会いのシナリオを受信する、請求項12に記載の情報処理装置。
- 前記制御部は、
前記ユーザの指示に応じて、前記相手ユーザと交際を始める旨の通知を前記サーバ装置に送信し;
当該交際を始める旨の通知に応じて、サーバ装置から、お祝いのメッセージおよび情報処理装置に実装されるエージェントアプリケーションの消去を指示する制御信号を受信し、当該制御信号に応じて前記エージェントアプリケーションを消去するように制御する、請求項13に記載の情報処理装置。 - 前記属性は、プロフィール情報、嗜好情報、性格情報、理想のプロフィール条件情報、および理想の性格情報の少なくともいずれかである、請求項11に記載の情報処理装置。
- プロセッサが、
それぞれ異なる属性を有し、ユーザと対話可能な複数のエージェントの情報を記憶部に記憶することと、
ユーザからのメッセージをクライアント端末から受信すると共に、応答メッセージを当該クライアント端末に通信部により返信することと、
ユーザからの指示に応じて、前記複数のエージェントから特定のエージェントを選択し;
前記特定のエージェントと前記ユーザとの対話に応じて、当該特定のエージェントの属性を更新したものをユーザ用エージェントの属性として記録し;
前記ユーザ用エージェントの属性と、実在する複数の相手ユーザの属性とを比較することにより、当該ユーザ用エージェントの属性に最も類似する相手ユーザを特定し;
所定のタイミングで前記ユーザに前記相手ユーザの存在を通知するように制御することと、
を含む、情報処理方法。 - コンピュータを、
それぞれが異なる属性を有し、ユーザと対話可能な複数のエージェントの情報が記憶されるサーバ装置に対して、ユーザからのメッセージを送信すると共に、そのメッセージに対する応答メッセージを受信する通信部と、
ユーザによる指示に応じて、前記複数のエージェントから特定のエージェントを選択し;
前記特定のエージェントと前記ユーザとの対話に応じて、当該特定のエージェントの属性が更新されたユーザ用エージェントの属性に最も類似する実在する相手ユーザが存在することを示す通知を、前記サーバ装置から前記通信部を介して所定のタイミングで受信するように制御する、制御部と、
として機能させるためのプログラム。
Priority Applications (5)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US16/078,437 US11610092B2 (en) | 2016-03-24 | 2016-12-21 | Information processing system, information processing apparatus, information processing method, and recording medium |
CN201680083674.XA CN108885768A (zh) | 2016-03-24 | 2016-12-21 | 信息处理系统、信息处理设备、信息处理方法以及记录介质 |
JP2018506778A JP6822467B2 (ja) | 2016-03-24 | 2016-12-21 | 情報処理システム、情報処理装置、情報処理方法、および記録媒体 |
KR1020187026403A KR20180123037A (ko) | 2016-03-24 | 2016-12-21 | 정보 처리 시스템, 정보 처리 장치, 정보 처리 방법, 및 기록 매체 |
EP16895538.3A EP3435323A4 (en) | 2016-03-24 | 2016-12-21 | INFORMATION PROCESSING SYSTEM, INFORMATION PROCESSING DEVICE, INFORMATION PROCESSING METHOD, AND RECORDING MEDIUM |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2016-059331 | 2016-03-24 | ||
JP2016059331 | 2016-03-24 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2017163515A1 true WO2017163515A1 (ja) | 2017-09-28 |
Family
ID=59899901
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2016/088253 WO2017163515A1 (ja) | 2016-03-24 | 2016-12-21 | 情報処理システム、情報処理装置、情報処理方法、および記録媒体 |
Country Status (6)
Country | Link |
---|---|
US (1) | US11610092B2 (ja) |
EP (1) | EP3435323A4 (ja) |
JP (3) | JP6822467B2 (ja) |
KR (1) | KR20180123037A (ja) |
CN (1) | CN108885768A (ja) |
WO (1) | WO2017163515A1 (ja) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2020213468A1 (ja) * | 2019-04-16 | 2020-10-22 | ソニー株式会社 | 情報処理システム、情報処理方法、及びプログラム |
JP2021033832A (ja) * | 2019-08-28 | 2021-03-01 | Bhi株式会社 | 統合された予約支援システム |
WO2021049594A1 (ja) * | 2019-09-13 | 2021-03-18 | 株式会社Aill | コミュニケーション支援サーバ、コミュニケーション支援方法、及びコミュニケーション支援プログラム |
CN112534449A (zh) * | 2018-07-27 | 2021-03-19 | 索尼公司 | 信息处理系统、信息处理方法和记录介质 |
JP2021512393A (ja) * | 2018-02-01 | 2021-05-13 | インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Machines Corporation | 会話エージェント学習モデルを動的に構築して構成する方法、コンピュータ・プログラム、およびシステム |
WO2021192991A1 (ja) * | 2020-03-23 | 2021-09-30 | ソニーグループ株式会社 | 情報処理装置、情報処理方法、およびプログラム |
WO2021200307A1 (ja) * | 2020-03-30 | 2021-10-07 | ソニーグループ株式会社 | 情報処理装置、対話型ロボット、制御方法 |
JP2023500995A (ja) * | 2020-01-17 | 2023-01-17 | マッチ グループ, エルエルシー | 第三者によってなされる選択に基づいてユーザをマッチングさせるためのシステム及び方法 |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2020119412A (ja) * | 2019-01-28 | 2020-08-06 | ソニー株式会社 | 情報処理装置、情報処理方法、及びプログラム |
WO2024165932A1 (en) | 2023-02-07 | 2024-08-15 | Ricoh Company, Ltd. | Conversation system, conversation control method, and storage medium |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001195430A (ja) * | 1999-11-02 | 2001-07-19 | Atr Media Integration & Communications Res Lab | 知識共有促進システム |
JP2002149882A (ja) * | 2000-11-14 | 2002-05-24 | Tokai Kiyouhan Kk | 顧客どうしの出会いの場の設定方法及び設定システム |
Family Cites Families (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001076002A (ja) | 1999-09-01 | 2001-03-23 | Kazuhiro Shiina | 情報ニーズ推定機能を備えた情報提供システム |
JP2002041276A (ja) | 2000-07-24 | 2002-02-08 | Sony Corp | 対話型操作支援システム及び対話型操作支援方法、並びに記憶媒体 |
JP2002049778A (ja) | 2000-08-03 | 2002-02-15 | Akuaparu:Kk | メール相手紹介装置、メール相手紹介システム、メール相手紹介方法、及び記憶媒体 |
WO2007059241A2 (en) * | 2005-11-15 | 2007-05-24 | Enpresence, Inc. | Proximity-a ware virtual agents for use with wireless mobile devices |
US7958117B2 (en) | 2006-11-17 | 2011-06-07 | Yahoo! Inc. | Initial impression analysis tool for an online dating service |
JP2011195430A (ja) | 2010-03-18 | 2011-10-06 | Tokuyama Corp | 低収縮性セメント組成物 |
KR101070844B1 (ko) | 2010-10-01 | 2011-10-06 | 주식회사 바로연결혼정보 | 이상형을 연결시켜주기 위한 감성 매칭시스템 및 매칭방법 |
US8732102B2 (en) * | 2011-02-17 | 2014-05-20 | Disney Enterprises, Inc. | System and method for using atomic agents to implement modifications |
JP2013054494A (ja) | 2011-09-02 | 2013-03-21 | Sony Corp | 情報処理装置、情報処理方法、プログラム、記録媒体、及び情報処理システム |
US9349131B2 (en) * | 2012-02-02 | 2016-05-24 | Kodak Alaris Inc. | Interactive digital advertising system |
US20170017501A1 (en) * | 2013-12-16 | 2017-01-19 | Nuance Communications, Inc. | Systems and methods for providing a virtual assistant |
US9924033B2 (en) * | 2014-10-23 | 2018-03-20 | Teletech Holdings, Inc. | Method for collecting data using a user interaction event-driven data collection system |
JP6797481B2 (ja) | 2017-03-01 | 2020-12-09 | 株式会社ディスコ | 半導体インゴットの検査方法、検査装置及びレーザー加工装置 |
-
2016
- 2016-12-21 JP JP2018506778A patent/JP6822467B2/ja active Active
- 2016-12-21 US US16/078,437 patent/US11610092B2/en active Active
- 2016-12-21 KR KR1020187026403A patent/KR20180123037A/ko active Search and Examination
- 2016-12-21 EP EP16895538.3A patent/EP3435323A4/en not_active Ceased
- 2016-12-21 WO PCT/JP2016/088253 patent/WO2017163515A1/ja active Application Filing
- 2016-12-21 CN CN201680083674.XA patent/CN108885768A/zh active Pending
-
2020
- 2020-12-18 JP JP2020210753A patent/JP7070652B2/ja active Active
-
2022
- 2022-04-27 JP JP2022073241A patent/JP7396396B2/ja active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001195430A (ja) * | 1999-11-02 | 2001-07-19 | Atr Media Integration & Communications Res Lab | 知識共有促進システム |
JP2002149882A (ja) * | 2000-11-14 | 2002-05-24 | Tokai Kiyouhan Kk | 顧客どうしの出会いの場の設定方法及び設定システム |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2021512393A (ja) * | 2018-02-01 | 2021-05-13 | インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Machines Corporation | 会話エージェント学習モデルを動的に構築して構成する方法、コンピュータ・プログラム、およびシステム |
JP7263370B2 (ja) | 2018-02-01 | 2023-04-24 | インターナショナル・ビジネス・マシーンズ・コーポレーション | 会話エージェント学習モデルを動的に構築して構成する方法、コンピュータ・プログラム、およびシステム |
US11886823B2 (en) | 2018-02-01 | 2024-01-30 | International Business Machines Corporation | Dynamically constructing and configuring a conversational agent learning model |
CN112534449A (zh) * | 2018-07-27 | 2021-03-19 | 索尼公司 | 信息处理系统、信息处理方法和记录介质 |
WO2020213468A1 (ja) * | 2019-04-16 | 2020-10-22 | ソニー株式会社 | 情報処理システム、情報処理方法、及びプログラム |
JP2021033832A (ja) * | 2019-08-28 | 2021-03-01 | Bhi株式会社 | 統合された予約支援システム |
WO2021049594A1 (ja) * | 2019-09-13 | 2021-03-18 | 株式会社Aill | コミュニケーション支援サーバ、コミュニケーション支援方法、及びコミュニケーション支援プログラム |
JP2023500995A (ja) * | 2020-01-17 | 2023-01-17 | マッチ グループ, エルエルシー | 第三者によってなされる選択に基づいてユーザをマッチングさせるためのシステム及び方法 |
JP7335453B2 (ja) | 2020-01-17 | 2023-08-29 | マッチ グループ, エルエルシー | 第三者によってなされる選択に基づいてユーザをマッチングさせるためのシステム及び方法 |
US11775600B2 (en) | 2020-01-17 | 2023-10-03 | Match Group, Llc | System and method for matching users based on selections made by third parties |
WO2021192991A1 (ja) * | 2020-03-23 | 2021-09-30 | ソニーグループ株式会社 | 情報処理装置、情報処理方法、およびプログラム |
WO2021200307A1 (ja) * | 2020-03-30 | 2021-10-07 | ソニーグループ株式会社 | 情報処理装置、対話型ロボット、制御方法 |
Also Published As
Publication number | Publication date |
---|---|
JP6822467B2 (ja) | 2021-01-27 |
CN108885768A (zh) | 2018-11-23 |
KR20180123037A (ko) | 2018-11-14 |
EP3435323A1 (en) | 2019-01-30 |
JP7070652B2 (ja) | 2022-05-18 |
JP7396396B2 (ja) | 2023-12-12 |
JP2022093479A (ja) | 2022-06-23 |
US20190050708A1 (en) | 2019-02-14 |
JP2021073549A (ja) | 2021-05-13 |
JPWO2017163515A1 (ja) | 2019-01-31 |
US11610092B2 (en) | 2023-03-21 |
EP3435323A4 (en) | 2019-04-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP7070652B2 (ja) | 情報処理システム、情報処理方法、およびプログラム | |
US20220284896A1 (en) | Electronic personal interactive device | |
US11327556B2 (en) | Information processing system, client terminal, information processing method, and recording medium | |
US11922934B2 (en) | Generating response in conversation | |
JP7070638B2 (ja) | 情報処理システムおよび情報処理方法 | |
JP7099589B2 (ja) | サーバ、通信制御方法、およびプログラム | |
US11267121B2 (en) | Conversation output system, conversation output method, and non-transitory recording medium | |
JP7207425B2 (ja) | 対話装置、対話システムおよび対話プログラム | |
WO2020027073A1 (ja) | 情報処理装置および情報処理方法 | |
CN114449297A (zh) | 一种多媒体信息的处理方法、计算设备及存储介质 | |
WO2024225222A1 (ja) | 電子機器、エージェントシステム、及びそのプログラム、スマートグラス、並びに注文端末及びそのプログラム | |
JP7438479B1 (ja) | 音声自動応答装置、音声自動応答方法、音声自動応答プログラム及び音声自動応答システム | |
JP2024159682A (ja) | エージェントシステム、プログラム、スマートグラス | |
JP2024159689A (ja) | 注文端末、プログラム | |
JP2024159681A (ja) | エージェントシステム、プログラム | |
JP2024159686A (ja) | 電子機器 | |
JP2024159685A (ja) | 電子機器 | |
JP2024159688A (ja) | エージェントシステム |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 2018506778 Country of ref document: JP |
|
ENP | Entry into the national phase |
Ref document number: 20187026403 Country of ref document: KR Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2016895538 Country of ref document: EP |
|
ENP | Entry into the national phase |
Ref document number: 2016895538 Country of ref document: EP Effective date: 20181024 |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 16895538 Country of ref document: EP Kind code of ref document: A1 |