Information processing apparatus and information processing method

Abstract

An information processing apparatus according to the present invention has a scoring unit and an interaction control unit. A scoring unit calculates a score of a response characteristic of a user based on a response history of the user in interaction with the user. The interaction control unit controls the number of queries made to the user during the interaction based on the score.

Description

Technical Field

[0001] The present invention relates to an information processing apparatus and an information processing method. Background Art

[0002] There is known an agent system that interacts with a user via video or voice. The user responds to an inquiry from the system by voice or the like. Therefore, necessary operations can be easily performed.

[0003] Citation List

[0004] Patent Document 1: JP 2002-099404 A Summary of the Invention

[0005] Technical Problem

[0006] Whether a user likes an inquiry depends on the user's personality, situation, etc., and thus setting rules in a unified manner alone does not necessarily achieve a comfortable interaction desired by the user.

[0007] Therefore, the present disclosure proposes an information processing apparatus and an information processing method that achieve a comfortable interaction.

[0008] Solution to the Problem

[0009] According to the present disclosure, there is provided an information processing apparatus including: a scoring unit that calculates a score of a response characteristic of a user based on a response history of the user in an interaction with the user; and an interaction control unit that controls the number of inquiries to the user in the interaction based on the score. According to the present disclosure, there is provided an information processing apparatus including: a data transmission unit that transmits response data from the user in an interaction with the user; a data reception unit that receives a score of a response characteristic of the user calculated based on the response history of the user; and an interaction control unit that controls the number of inquiries to the user in the interaction based on the score. According to the present disclosure, there is provided an information processing method in which information processing of an information processing apparatus is executed by a computer. Brief Description of the Drawings

[0010] Figure 1 is a diagram showing an overview of an agent system.

[0011] Figure 2 is a block diagram showing a configuration example of an agent system.

[0012] Figure 3 is a diagram showing an example of an interaction.

[0013] Figure 4 is a diagram showing an example of an abnormal event in which a user cannot respond.

[0014] Figure 5 It is a table showing an example of score management.

[0015] Figure 6 It is a table showing an example of score management.

[0016] Figure 7 It is a table showing an example of score management.

[0017] Figure 8 It is a table showing an example of score management.

[0018] Figure 9 It is a table showing an example of the conditions for inquiry and information presentation.

[0019] Figure 10 It is a diagram showing an example of an inquiry to the user.

[0020] Figure 11 It is a diagram showing an example of information presentation to the user.

[0021] Figure 12 It is a flowchart showing an exemplary processing procedure of the score.

[0022] Figure 13 It is a flowchart showing an exemplary processing procedure of the restriction on interaction.

[0023] Figure 14 It is a diagram showing a modification of the control of the interaction.

[0024] Figure 15 It is a diagram showing an exemplary hardware configuration of the information processing device. Detailed Description of the Invention

[0025] Embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. In the following embodiments, the same parts are denoted by the same reference numerals, and their descriptions will not be repeated.

[0026] Please note that the description will be given in the following order.

[0027] [1. Overview of the proxy system]

[0028] [2. Configuration example of the proxy system]

[0029] [2-1. Client terminal]

[0030] [2-2. Server]

[0031] [3. Interaction example]

[0032] [4. Score management]

[0033] [5. Examples of inquiry and information presentation]

[0034] [6. Processing Procedure]

[0035] [7. Modification]

[0036] [7-1. First Modification]

[0037] [7-2. Second Modification]

[0038] [7-3. Third Modification]

[0039] [8. Exemplary Hardware Configuration]

[0040] [9. Effects]

[0041] [1. Overview of the Agent System]

[0042] Figure 1 is a diagram showing an overview of the agent system 1.

[0043] The agent system 1 interacts with the user US by using video or voice. In Figure 1 the example, an agent system 1 called a dialogue agent is disclosed, which receives inquiries and requests from the user US through a spoken dialogue, performs information presentation for the user US, and makes inquiries (including suggestions) to the user US.

[0044] The functions of the dialogue agent are executed by the voice assistant VA. The voice assistant VA is software that combines speech recognition technology and natural language processing to receive input from the user US, present information to the user US, and make inquiries to the user US. The voice assistant VA outputs text, video, and voice to the information presentation unit 16 for performing information presentation and making inquiries, such as a television. For example, the text and video from the voice assistant VA are displayed overlaid on the content CT being viewed by the user US.

[0045] The agent system 1 can be an independent system configured by a single information processing device PD, or can be a client / server system in which the agent functions are distributed to multiple information processing devices PD (client terminal 10, server 20: see Figure 2 ). In the example of the client / server system, the system developer can appropriately set which functions are shared by the client terminal 10 and which functions are shared by the server 20 or not.

[0046] The agent system 1 includes a scoring unit 22 and an interaction control unit 15. The agent system 1 interacts with the user US based on the dialogue via the voice assistant VA. The agent system 1 actively presents information corresponding to the user US and makes inquiries as needed.

[0047] The agent system 1 obtains responses to inquiries or information presentations from the user US by using sensors such as cameras or microphones. In Figure 1 the example, the user US plays a game by using the controller CR. As the game progresses, the agent system 1 provides game-related information to the user US in a timely manner.

[0048] The agent system 1 accumulates response data from the user US sequentially obtained by the sensors, the scoring unit 22 calculates a score for the response characteristics of the user US based on the response history of the user US in the interaction with the user US, and the interaction control unit 15 controls the number of inquiries to the user US in the interaction based on the calculated score.

[0049] [2. Configuration example of the agent system]

[0050] Figure 2 is a block diagram showing a configuration example of the agent system 1.

[0051] The agent system 1 includes a client terminal 10 and a server 20. The client terminal 10 and the server 20 are connected via a network NW such as the Internet. The agent system 1 is configured as a client / server system in which the agent function is distributed to the client terminal 10 and the server 20 that serve as information processing devices PD.

[0052] [2-1. Client terminal]

[0053] The client terminal 10 interacts with the user US by using the voice assistant VA. The client terminal 10 obtains response data from the user US in the interaction and transmits the response data to the server 20. The server 20 analyzes the obtained response data and scores the response characteristics of the user US. The client terminal 10 controls the interaction based on the result of the score.

[0054] For example, the client 10 includes a sensor unit 11, a context detection unit 12, a user identification unit 13, a response detection unit 14, an interaction control unit 15, an information presentation unit 16, a data transmission unit 17, and a data reception unit 18.

[0055] <2-1-1. Sensor unit>

[0056] The sensor unit 11 includes various sensors such as a camera, a microphone, a clock, and a fingerprint sensor. For example, the sensor unit 11 detects the video and voice of the user US by using the camera and the microphone. The sensor unit 11 detects the time by using the internal clock thereof. The sensor unit 11 detects the fingerprint of the user US by using the fingerprint sensor built in the controller CR. The sensor unit 11 obtains information on the content viewed by the user US based on the operation information input by the user US to the system. The sensor unit 11 outputs the detected various information as sensor information.

[0057] <2-1-2. Context Detection Unit>

[0058] The context detection unit 12 detects the context at the time of response based on the sensor information. The context detection unit 12 outputs the detected context as context information in association with the time.

[0059] The context represents the overall situation and environment of the user US. When the context is different, even if the same information, questions, etc. are presented, the response characteristics of the user US are different. In the present disclosure, various elements that can change the response characteristics are called context. For example, the context includes the type of content CT viewed by the user US, the type of information presented to the user US, or the time slot for interaction.

[0060] <2-1-3. User Identification Unit>

[0061] The user identification unit 13 identifies the user US interacting with the agent system 1 based on the sensor information. The user identification unit 13 outputs the information of the identified user US as user identification information. For example, the user identification unit 13 is operable to identify the user US based on the video of the user US captured by the camera, the voice of the user US collected by the microphone, and the fingerprint of the user US obtained by the fingerprint sensor.

[0062] The agent system 1 stores the personal information (face, voiceprint, fingerprint, etc.) of the registered user. The agent system 1 is configured to check the personal information of the registered user against the user identification information to determine whether the identified user US is a registered user.

[0063] <2-1-4. Response Detection Unit>

[0064] The response detection unit 14 detects a response from the user US based on sensor information. For example, the response detection unit 14 analyzes the speech of the user US collected by the microphone to obtain the speech content of the user US. The response detection unit 14 analyzes the video of the user US captured by the camera to detect the expression and actions (including nodding, shaking the head, eye contact, etc.) of the user US. The response detection unit 14 outputs the speech information extracted from the speech content and the non-verbal information extracted from the expression and actions as response data in association with the response time.

[0065] The response detection unit 14 detects a preset abnormal event based on the video and speech of the user US. An abnormal event means a special situation where the user US cannot respond. For example, when the user US leaves the seat or is on the phone, the user US cannot interact. The response detection unit 14 detects such a situation as an abnormal event to identify the response data to be excluded from the score calculation. The response detection unit 14 outputs a flag related to the presence / absence of the abnormal event in association with the response data.

[0066] <2-1-5. Data sending unit and data receiving unit>

[0067] The data sending unit 17 and the data receiving unit 18 are connected to the network NW, such as the Internet, in a wired or wireless manner. The data sending unit 17 and the data receiving unit 18 transmit data to the server 20 via the network NW and receive data from the server 20.

[0068] For example, the data sending unit 17 selectively extracts the response data from the user US identified as a registered user based on the user identification information. The data sending unit 17 transmits the extracted response data from the user US to the server 20 together with the context information. The data receiving unit 18 receives the score of the response characteristics of the user US calculated based on the response history of the user US from the server 20.

[0069] <2-1-6. Interaction control unit>

[0070] The interaction control unit 15 controls the interaction based on the score of the response characteristics of the user US. The control of the interaction includes the control of the number of questions asked to the user US and the amount of information provided to the user US during the interaction. The interaction control unit 15 is also configured to change the timing of asking questions, the timing of information provision, etc. according to the score.

[0071] The interaction control unit 15 starts restricting the interaction based on the score. The restriction of the interaction is performed by prohibiting inquiries or information provision to the user US under specific conditions based on the score. For example, the interaction control unit 15 divides the score into high-level, medium-level, and low-level categories. The interaction control unit 15 restricts the interaction step by step according to the level of the score. The criteria for classifying into each level can be appropriately set by the system developer.

[0072] For example, when the score is at a high level, the interaction control unit 15 does not restrict the interaction. When the score is at a medium level, the interaction control unit 15 prohibits inquiries to the user US as a restriction on the interaction. When the score is at a low level, the interaction control unit 15 prohibits inquiries and information provision to the user US as a restriction on the interaction.

[0073] For example, the response characteristics of the user US are scored based on the response history at the start of the session. During the same session, the interaction control unit 15 does not change the restriction on the interaction set at the start of the session. A session refers to the period of a series of processes repeated interactively using requests and responses. Processes such as the notification and browsing of updated information at system startup, and the notification and browsing of the latest game title information at game startup each establish a session. For example, a session is the continuous period from the start to the termination of the voice assistant VA.

[0074] <2-1-7. Information presentation unit>

[0075] The information presentation unit 16 presents various types of information to the user US through video and voice. The information presentation unit 16 is configured to control a control device for video / audio devices such as a television. The information presentation unit 16 may include the controlled video / audio device. The information presentation unit 16 interacts with the user US using a tone adjusted by the interaction control unit 15 to match the response preference of the user US.

[0076] [2-2. Server]

[0077] The server 20 analyzes the response data and context information obtained from the client terminal 10. The server 20 calculates a score reflecting the response preference of the user US based on the analysis result. For example, the server 20 includes a data storage unit 21, a scoring unit 22, a data receiving unit 23, and a data sending unit 24.

[0078] <2-2-1. Data storage unit>

[0079] The data storage unit 21 accumulates the response data and context information sequentially obtained from the client terminal 10. The data storage unit 21 stores the accumulated time-series data as the response history of the user US.

[0080] <2-2-2. Scoring Unit>

[0081] The scoring unit 22 calculates a score for the response characteristics of the user US based on the response history of the user US. For example, the scoring unit 22 can be operated to calculate a score based on the presence / absence of a response from the user US to an inquiry. In addition, the scoring unit 22 can be operated to calculate a score based on the time from the inquiry to the response.

[0082] The response status of the user US changes according to the context at the time of response (e.g., the situation of the user US). Even when the same inquiry is made, etc., the response becomes difficult when the user US is immersed in a game, chat, etc. When determining the response characteristics only based on the response data, accurate determination results are not necessarily obtained. Therefore, the scoring unit 22 makes the calculation criteria for the score different for the response characteristics according to the context at the time of response.

[0083] When an abnormal event such as the absence of the user occurs, appropriate response data matching the response characteristics of the user US cannot be obtained. Therefore, the scoring unit 22 performs a calculation that excludes such inappropriate response data. For example, the scoring unit 22 identifies the period during which the abnormal event occurs as an abnormal period, and calculates the score by selectively using the response data from the user US during the period other than the abnormal period.

[0084] The response data from the user US identified as a registered user based on the user identification information is transmitted to the server 20. The scoring unit 22 selectively uses the response data from the user US identified as a registered user to calculate the score. The response data includes verbal information extracted from the speech content of the user US and non-verbal information extracted from the expressions and actions of the user US. The scoring unit 22 calculates the score based on the verbal information and non-verbal information obtained as the response from the user US.

[0085] <2-2-3. Data Sending Unit and Data Receiving Unit>

[0086] The data sending unit 24 and the data receiving unit 23 are connected to the client terminal 10 via the network NW. The data receiving unit 23 receives the response data and context information transmitted from the client terminal 10, and outputs the response data and context information to the data storage unit 21. The data sending unit 24 transmits the score of the response characteristics calculated by the scoring unit 22 to the client terminal 10.

[0087] [3. Interaction Example]

[0088] Figure 3 is a diagram showing an example of the interaction.

[0089] In Figure 3In the example, the process of selecting and browsing update information is performed interactively when the game is activated. First, user US activates the voice assistant VA by using an activation phrase (e.g., "Hey, game console"). The text and video from the voice assistant VA are overlaid and displayed on the game screen SCR. The activation of the voice assistant VA can be performed by using an activation button.

[0090] User US asks the voice assistant VA about the existence / non-existence of update information (e.g., "Is there an update?" or "What's new?"). The voice assistant VA responds to the query from user US, searches the network NW for update information, and overlays the search results on the game screen SCR.

[0091] The voice assistant VA notifies user US of the most relevant update information among the several pieces of update information displayed, and asks user US if they want to browse (e.g., "Item 1 seems relevant to you. View?"). User US responds to the query from the voice assistant VA by answering whether to browse (e.g., "Yes"). The voice assistant VA, in response to the request from user US, displays the target update information on the game screen SCR.

[0092] The voice assistant VA asks user US if they want to browse other update information (e.g., "Want to view the next item?"). User US ignores the query from the voice assistant VA and does not respond.

[0093] When there is no response for a certain period of time, the voice assistant VA determines that there is no response from user US. The voice assistant VA performs a process of changing the tone in the interaction and notifies user US of the tone change (e.g., "I'll turn it down a bit while you're playing the game"). The agent system 1 determines that user US does not intend to continue the browsing process and ends the voice assistant VA.

[0094] The voice assistant VA monitors the response from user US and scores the response characteristics. The change in tone in the interaction is implemented as a result of changing the score and controlling the interaction based on the changed score (e.g., controlling the number of queries to user US). When there is no response, the score is changed to reduce the number of queries, but when an abnormal event occurs where user US cannot respond, the score is not changed.

[0095] Figure 4 is a diagram showing an example of an abnormal event where user US cannot respond.

[0096] In Figure 4Four situations are shown as examples of abnormal events. The first situation is when user US temporarily leaves the seat. The second situation is when there is no response due to a phone call or the like. The third situation is when an unauthorized third party TP (a user other than the logged-in registered user) is making a response. The fourth situation is when user US is paying attention to another thing.

[0097] The response detection unit 14 detects abnormal events based on sensor information obtained from cameras, microphones, fingerprint sensors, etc. For example, the first, second, and fourth situations can be detected by analyzing camera images through known image analysis techniques, and the third situation can be detected by analyzing the voiceprint of the voice collected by the microphone or the fingerprint detected by the fingerprint sensor through known voice authentication techniques or fingerprint authentication techniques. Note that the abnormal events are not limited to Figure 4 the examples shown in. System developers can appropriately set whether to detect a certain situation as an abnormal event.

[0098] [4. Score Management]

[0099] Figures 5 to 8 are diagrams each showing an example of score management.

[0100] Scores are calculated based on preset calculation criteria. For example, when user US responds to an inquiry from the voice assistant VA, the score increases, and when user US does not respond, the score decreases. In the calculation criteria, it is defined under what circumstances or to what extent the score increases or decreases. General calculation criteria can be used for all cases, but different calculation criteria can be used for different contexts. The agent system 1 is configured to associate multiple scores with different contexts.

[0101] Figure 5 shows an example of using general calculation criteria for all cases. The score S_all is determined only based on the response data. Even if the context or the situation of user US is different, the same score is calculated when the same response data is provided.

[0102] Figure 6 shows an example of managing scores for each piece of viewing content. Score calculation criteria are set for each piece of viewing content. Even when the same response data is used, different scores are calculated when different viewing content is provided. In Figure 6In the example, as scores managed according to different calculation criteria, the response score S_sys_ui on the system UI screen, the score S_game during the game, and the score S_video during media viewing are shown. In addition, another response score on the voice chat screen can be prepared, or different response scores can be prepared according to the immersion level of the game or chat. The immersion level can be determined based on controller operations, the distance between the user US and the game screen SCR, the content of the voice chat, etc.

[0103] Figure 7 An example of managing scores for each type of presentation information is shown. A score calculation criterion is set for each type of presentation information. Even when the same response data is used, different scores are calculated when different presentation information is provided. In Figure 7 the example, as examples of scores managed according to different calculation criteria, the response score S_info_game for game title information, the response score S_info_sys for system information, the response score S_info_notif for notification information, the response score S_info_friend for friend information, the response score S_info_voice for voice command information, and the response score S_info_dlc for download content (DLC) information are shown.

[0104] Figure 8 An example of managing scores for each time slot of an execution interaction is shown. A score calculation criterion is set for each time slot. Even when the same response data is used, different scores are calculated when the time slots are different. In Figure 8 the example, as examples of scores managed according to different calculation criteria, the response score S_time_am in the "morning" time slot, the response score S_time_pm in the "afternoon" time slot, the response score S_time_weekday for "weekday", the response score S_time_weekend for "weekend", the response score S_time_ps_wakeup within a certain period of time since system startup, and the response score S_time_ps_middle after a certain period of time since system startup are shown.

[0105] The score represents the degree of expectation (degree of interest in the interaction) of the user US for an inquiry or information presentation. Therefore, the score can be regarded as a numerical value indicating the probability that the user US allows an inquiry or information presentation. In the agent system 1, the score calculation criterion is set such that the score increases as positive response data is obtained.

[0106] In actual interactions, multiple contexts can be detected in a combined manner. In combined detection, the interaction control unit 15 implements the integration of individual scores that are related to each other, and controls the interaction based on the score obtained through integration (the integrated score). For example, the integrated score can be calculated by obtaining the sum of scores with a predetermined weight added. The weights are configured to be preset for each context.

[0107] [5. Examples of Inquiry and Information Presentation]

[0108] Figure 9 It is a table showing examples of conditions for inquiry and information presentation.

[0109] The voice assistant VA can not only implement information presentation in response to an inquiry or request from the user US, but also initiate an inquiry or information presentation to the user US on its own. When a predetermined condition is met, an inquiry or information presentation is made on its own. Figure 9 Seven types are shown as examples of inquiries and information provided by the voice assistant VA on its own.

[0110] For example, the voice assistant VA determines whether to present system update information based on the user US's past browsing status. When it is determined based on the past browsing status that the user US has not browsed the latest notification about system update, the voice assistant VA asks the user US whether to browse. When a browsing request is received from the user US, the voice assistant VA presents the system update information.

[0111] The voice assistant VA determines whether to present information about new game titles based on information such as game titles owned by the user US or the wish list. When a game of the same type as the game owned by the user US or a game of the same type as the game registered in the wish list, etc. is released, the voice assistant VA asks the user US whether to browse. When a browsing request is received from the user US, the voice assistant VA presents the title information of the target game.

[0112] The voice assistant VA determines whether to present update information about the owned game based on the user US's latest game status. When there is update information about the game that the user US has recently played (for example, a game played within a week), the voice assistant VA asks the user US whether to browse. When a browsing request is received from the user US, the voice assistant VA presents the update information about the target game.

[0113] The voice assistant VA determines whether to present DLC information about the owned game based on the user US's latest game status. When there is DLC information about the game that the user US has recently played (for example, a game played within a week), the voice assistant VA asks the user US whether to browse. When a browsing request is received from the user US, the voice assistant VA presents the DLC information about the target game.

[0114] The voice assistant VA determines whether to present notification information based on the presence / absence of unread items. When there are unread items, the voice assistant VA asks the user US if they want to view them. When a view request is received from the user US, the voice assistant VA presents the unread items.

[0115] The voice assistant VA determines whether to present friend information based on the presence / absence of new friend requests or activities of friends. Friend information includes information about users who have sent friend requests and information about the activities of users who have been registered as friends. When there is a new friend request or activity, the voice assistant VA asks the user US if they want to view the friend information. When a view request is received from the user US, the voice assistant VA presents the friend information.

[0116] The voice assistant VA determines whether to present game subscription information based on information such as free games. A free game means a special benefit of a paid subscription service that allows some released games to be played without additional payment. When a game of the same type as the game owned by the user US is added to free games, etc., the voice assistant VA asks the user US if they want to view it. When a view request is received from the user US, the voice assistant VA presents the game subscription information.

[0117] Note that when the presentation conditions for multiple pieces of information are detected simultaneously, all the multiple pieces of information can be presented to the user US, or one piece of information can be selected to be presented to the user US. When the number of presented information is limited to one, the information with a higher score can be preferentially selected.

[0118] Figure 10 It is a diagram showing an example of asking the user US. Figure 11 It is a diagram showing an example of presenting information to the user US.

[0119] The method for asking or presenting information can be appropriately set by the system developer. Asking and presenting information can be done only through text or voice or through non-verbal information such as eye contact or gestures of an avatar.

[0120] Various methods can also be adopted for the selection of presented information. For example, in Figure 10 's example, the information to be viewed is determined by the user US through spoken dialogue. In Figure 11 's example, the browsing priority is automatically determined by the system and presented in descending order of priority while obtaining the permission of the user US. Note that the methods for selecting presented information and prompting the system voice for selection are not limited to the Figure 10 and Figure 11 methods shown.

[0121] [6. Processing Procedure]

[0122] Figure 12It is a flowchart showing an exemplary process for processing scores.

[0123] The voice assistant VA asks the user US (step ST1). The response detection unit 14 determines whether there is an input from the user US (step ST2). When there is no input from the user US (step ST2: No), the scoring unit 22 selects the score to be changed based on the context information (step ST3). The scoring unit 22 changes the selected score based on the calculation criterion represented by the following formula (1) (step ST4), and the process is completed.

[0124] S = S -– C × step (1)

[0125] In formula (1), "S" represents the score. "C" represents a constant. "step" represents the coefficient set for each context.

[0126] When there is an input from the user US (step ST2: Yes), the response detection unit 14 determines whether the input is a response to the question (step ST5). When the input is a response to the question (step ST5: Yes), the scoring unit 22 obtains the time from the question to the response (step ST6). The scoring unit 22 selects the score to be changed based on the context information (step ST7). The scoring unit 22 changes the selected score based on the calculation criterion represented by the following formula (2) (step ST8), and the process is completed.

[0127] S = S + C × step × 1 / Tr (2)

[0128] In formula (2), "Tr" represents the time required for the response obtained in step ST6.

[0129] When the input is not a response to the question (step ST5: No), the scoring unit 22 completes the process without changing the score.

[0130] Figure 13 It is a flowchart showing an exemplary process for restricting interactions.

[0131] The scoring unit 22 obtains context information from the client terminal 10 (step ST11). The scoring unit 22 selects a specific score associated with the context (step ST12). The scoring unit 22 calculates the score based on the calculation criterion according to the context and transmits the score to the client terminal 10.

[0132] The interaction control unit 15 determines whether the score is at a high level (step ST13). When the score is at a high level (step ST13: Yes), the interaction control unit 15 makes an inquiry to the user US (step ST14). The scoring unit 22 updates the score based on the response status of the user US to the inquiry (step ST15).

[0133] When the score is not at a high level (step ST13: No), the interaction control unit 15 determines whether the score is at a medium level (step ST16). When the score is at a medium level (step ST16: Yes), the interaction control unit 15 performs information presentation to the user US (step ST17). When the score is not at a medium level (step ST16: No), the interaction control unit 15 prohibits making an inquiry and information presentation to the user US.

[0134] [7. Modification]

[0135] Hereinafter, modifications to the above-described embodiments will be described.

[0136] [7-1. First Modification]

[0137] Figure 14 FIG. is a diagram showing a modification of the interaction control. In the above-described embodiment, the interaction control unit 15 changes the number of inquiries based on the corresponding score. However, the interaction control is not limited to the control of the number of inquiries. The interaction control unit 15 may change the timing of the inquiry according to the score.

[0138] For example, the time from when the voice assistant VA is activated (or after the user US inputs a command to the voice assistant VA) until the voice assistant VA makes an inquiry to the user US is defined as the proximity time. In Figure 14 the example, when the score is at a high level, the proximity time is set to be shorter, and when the score is at a low level, the proximity time is set to be longer.

[0139] The score reflects the degree of interest of the user US in the interaction. The higher the score, the higher the interest of the user US in the interaction. Therefore, when the score is high, the proximity time is set to be shorter. Thereby, the inquiry and response proceed at a good rhythm, and a comfortable interaction matching the response preference of the user US is achieved.

[0140] [7-2. Second Modification]

[0141] In Figure 13 the example, the inquiry and information provision of the voice assistant VA are restricted periodically. However, the restriction on the interaction is not limited to this. For example, the method of the present disclosure can also be applied to the process of determining whether to set the voice of which character as text-to-speech (TTS).

[0142] For example, when the score is at a high level, the interaction control unit 15 enables the user US to determine the role for voice through a dialogue. When the score is at a low level, the interaction control unit 15 automatically assigns the voice of the role frequently used by the user US without having a dialogue with the user US.

[0143] The method of the present disclosure can also be applied to the process of determining whether to reproduce video content. For example, when the score is at a high level, the interaction control unit 15 enables the user US to determine whether to reproduce video content through a dialogue. When the score is at a low level, the interaction control unit 15 automatically determines whether to reproduce video content based on the behavior tendency of the user US obtained in advance without having a dialogue with the user US. For example, for a user US with a high possibility of watching video content, the video content is configured to be automatically reproduced without relying on a dialogue.

[0144] Similarly, the method of the present disclosure can also be applied to the process of determining whether to switch to the "next content". For example, when the user US tends to frequently watch the "next content" and the score is at a low level, the interaction control unit 15 is configured to automatically switch to the "next content" without relying on a dialogue.

[0145] [7-3. Third Modification]

[0146] In Figure 3 the example of, the agent system 1 of the present disclosure has been applied to a game console. However, the agent system 1 of the present disclosure can be applied to various devices other than game consoles. For example, the agent system 1 is configured to be installed on a vehicle system to select and browse restaurant information, café information, etc. through a spoken dialogue.

[0147] The agent system 1 is also configured to be installed on a television or a video content device for a streaming service, etc. to determine the type of video content to be watched, etc. through a spoken dialogue. In addition, the agent system 1 is configured to be installed on a home robot (e.g., a cleaning robot) to select and execute various processes through a spoken dialogue. In addition, the agent system 1 can also be applied to communication using an avatar (such as a virtual store or a virtual chat, e.g., a massively multiplayer online (MMO)).

[0148] [8. Exemplary Hardware Configuration]

[0149] Figure 15 is a diagram showing an exemplary hardware configuration of the information processing device PD.

[0150] For example, the information processing in the information processing device PD is implemented by the computer 1000. The computer 1000 includes a central processing unit (CPU) 1100, a random access memory (RAM) 1200, a read-only memory (ROM) 1300, a hard disk drive (HDD) 1400, a communication interface 1500, and an input / output interface 1600. Each unit of the computer 1000 is connected via a bus 1050.

[0151] The CPU 1100 operates based on programs (program data 1450) stored in the ROM 1300 or the HDD 1400, and controls each unit. For example, the CPU 1100 loads the programs stored in the ROM 1300 or the HDD 1400 into the RAM 1200, and executes the processing corresponding to each of the respective programs.

[0152] The ROM 1300 stores a startup program (such as the basic input / output system (BIOS)) to be executed by the CPU 1100 when the computer 1000 starts up, and the program depends on the hardware of the computer 1000, etc.

[0153] The HDD 1400 is a computer-readable non-transitory recording medium that non-transitorily records programs executed by the CPU 1100, data used by the programs, etc. Specifically, the HDD 1400 is a recording medium that records an information processing program according to an embodiment as an example of the program data 1450.

[0154] The communication interface 1500 is an interface for connecting the computer 1000 to an external network 1550 (such as the Internet). For example, the CPU 1100 receives data from another device via the communication interface 1500 or transmits data generated by the CPU 1100 to another device.

[0155] The input / output interface 1600 is an interface for connecting an input / output device 1650 and the computer 1000. For example, the CPU 1100 receives data from an input device such as a keyboard or a mouse via the input / output interface 1600. In addition, the CPU 1100 transmits data to an output device such as a display device, a speaker, or a printer via the input / output interface 1600. In addition, the input / output interface 1600 can be used as a media interface for reading programs, etc. recorded on a predetermined recording medium. For example, the medium includes optical recording media such as digital versatile discs (DVDs) or phase change rewritable discs (PDs), magneto-optical recording media such as magneto-optical discs (MOs), tape media, magnetic recording media, semiconductor memories, etc.

[0156] For example, when the computer 1000 is used as the information processing device PD according to the embodiment, the CPU 1100 of the computer 1000 executes the information processing program loaded into the RAM 1200 to implement the functions of each of the above units. In addition, the HDD 1400 stores the information processing program, various models, and various data according to the present disclosure. Note that the CPU 1100 executes the program data 1450 read from the HDD 1400, but in another example, the CPU 1100 may obtain these programs from another device via the external network 1550.

[0157] [9. Effects]

[0158] The information processing device PD includes a scoring unit 22 and an interaction control unit 15. The scoring unit 22 calculates a score for the response characteristics of the user US based on the response history of the user US in the interaction with the user US. The interaction control unit 15 controls the number of inquiries to the user US in the interaction based on the score. In the information processing method of the present disclosure, the processing in the information processing device PD is executed by a computer.

[0159] According to this configuration, the interaction suitable for the response preference of the user US is controlled. Thus, a comfortable interaction desired by the user US is achieved.

[0160] The scoring unit 22 calculates a score based on the presence / absence of the response of the user US to the inquiry.

[0161] According to this configuration, an appropriate score reflecting the response state of the user US is calculated.

[0162] The scoring unit 22 calculates a score based on the time from the inquiry to the response.

[0163] According to this configuration, an appropriate score reflecting the response state of the user US is calculated.

[0164] The information processing device PD includes a response detection unit 14. The response detection unit 14 detects an abnormal event in which the user US cannot respond. The scoring unit 22 calculates a score by selectively using the response data from the user US during the period other than the period when the abnormal event occurs.

[0165] According to this configuration, an appropriate score reflecting the response state of the user US is calculated.

[0166] The scoring unit 22 makes the calculation criteria of the score different according to the context at the time of response.

[0167] According to this configuration, the score is calculated considering the context at the time of response. The response characteristics are analyzed considering the context, and an appropriate score is calculated.

[0168] The context includes the type of content viewed by the user US, the type of information presented to the user US, or the time slot for interaction.

[0169] The content being viewed at the time of response, etc. can greatly affect the response characteristics of the user US. Therefore, appropriate interaction control is achieved by considering this information.

[0170] The interaction control unit 15 restricts interaction based on a score.

[0171] With this configuration, the burden of the response is appropriately reduced according to the response preference of the user US.

[0172] The interaction control unit 15 classifies the score. The interaction control unit 15 restricts interaction step by step according to the level of the score.

[0173] This configuration enables a fine response to the response preference of the user US.

[0174] When the score is at a high level, the interaction control unit 15 does not restrict interaction. When the score is at a medium level, as a restriction on interaction, the interaction control unit 15 prohibits asking questions to the user US. When the score is at a low level, as a restriction on interaction, the interaction control unit 15 prohibits asking questions to the user US and providing information.

[0175] With this configuration, the burden of the response is appropriately reduced according to the score level.

[0176] The interaction control unit 15 changes the timing of asking questions according to the score.

[0177] This configuration enables asking questions at an appropriate pace according to the response preference of the user US.

[0178] In the same session, the interaction control unit 15 does not change the restriction on interaction set at the start of the session.

[0179] This configuration enables the tone in the interaction to be stable.

[0180] The scoring unit 22 calculates a score based on the verbal information and non-verbal information obtained as the response from the user US.

[0181] With this configuration, the response preference of the user US is accurately identified based on both verbal information and non-verbal information.

[0182] The information processing device PD includes a user identification unit 13. The user identification unit 13 identifies the user US. The scoring unit 22 selectively uses the response data from the user US identified as a registered user to calculate the score.

[0183] This configuration enables an accurate calculation of the score based on the response data from only the registered users.

[0184] An information processing device PD includes a data transmission unit 17, a data reception unit 18, and an interaction control unit 15. The data transmission unit 17 transmits response data from a user US obtained in interaction with the user US. The data reception unit 18 receives a score of the response characteristics of the user US calculated based on the response history of the user US. The interaction control unit 15 controls the number of inquiries to the user US in the interaction based on the score. In the information processing method of the present disclosure, the processing in the information processing device PD is executed by a computer.

[0185] According to this configuration, the interaction suitable for the response preference of the user US is controlled. Therefore, a comfortable interaction desired by the user US is achieved.

[0186] Note that the effects described herein are merely examples and are not limited to the description, and other effects may be provided.

[0187] [Supplementary Explanation]

[0188] Note that the present technology may also have the following configurations.

[0189] (1) An information processing device, comprising:

[0190] a scoring unit that calculates a score of the response characteristics of a user based on the response history of the user in interaction with the user; and

[0191] an interaction control unit that controls the number of inquiries to the user in the interaction based on the score.

[0192] (2) The information processing device according to (1), wherein

[0193] the scoring unit calculates the score based on the presence / absence of the user's response to an inquiry.

[0194] (3) The information processing device according to (2), wherein

[0195] the scoring unit calculates the score based on the time from the inquiry to the response.

[0196] (4) The information processing device according to any one of (1) to (3), further comprising

[0197] a response detection unit that detects an abnormal event in which the user cannot respond, wherein

[0198] the scoring unit selectively uses the response data from the user in a period other than the period in which the abnormal event occurs to calculate the score.

[0199] (5) The information processing apparatus according to any one of (1) to (4), wherein

[0200] the scoring unit calculates the score according to the context at the time of response with different calculation criteria.

[0201] (6) The information processing apparatus according to (5), wherein

[0202] the context includes the type of content viewed by the user, the type of information presented to the user, or the time slot for interaction.

[0203] (7) The information processing apparatus according to any one of (1) to (6), wherein

[0204] the interaction control unit restricts the interaction based on the score.

[0205] (8) The information processing apparatus according to (7), wherein

[0206] the interaction control unit divides the score into levels and restricts the interaction step by step according to the level of the score.

[0207] (9) The information processing apparatus according to (8), wherein

[0208] the interaction control unit

[0209] does not restrict the interaction when the score is at a high level,

[0210] prohibits asking questions to the user as a restriction on the interaction when the score is at a medium level, and

[0211] prohibits asking questions to the user and providing information as a restriction on the interaction when the score is at a low level.

[0212] (10) The information processing apparatus according to any one of (7) to (9), wherein

[0213] the interaction control unit changes the timing of asking questions according to the score.

[0214] (11) The information processing apparatus according to any one of (7) to (10), wherein

[0215] the interaction control unit does not change the restriction on the interaction set at the start of the session during the same session.

[0216] (12) The information processing apparatus according to any one of (1) to (11), wherein

[0217] the scoring unit calculates the score based on the verbal information and non-verbal information obtained as a response from the user.

[0218] (13) The information processing device according to any one of (1) to (12) further includes a user identification unit that identifies the user, wherein

[0219] The scoring unit selectively uses the response data from the user identified as a registered user to calculate the score.

[0220] (14) An information processing device includes:

[0221] A data sending unit that transmits response data from the user in an interaction with the user;

[0222] A data receiving unit that receives a score of the response characteristics of the user calculated based on the response history of the user; and

[0223] An interaction control unit that controls the number of inquiries to the user in the interaction based on the score.

[0224] (15) An information processing method executed by a computer, the method includes:

[0225] Calculating a score of the response characteristics of the user based on the response history of the user in an interaction with the user; and

[0226] Controlling the number of inquiries to the user in the interaction based on the score.

[0227] (16) An information processing method executed by a computer, the method includes:

[0228] Transmitting response data from the user in an interaction with the user;

[0229] Receiving a score of the response characteristics of the user calculated based on the response history of the user, and

[0230] Controlling the number of inquiries to the user in the interaction based on the score.

[0231] List of Reference Signs

[0232] 10 Client terminal (information processing device)

[0233] 13 User identification unit

[0234] 14 Response detection unit

[0235] 15 Interaction control unit

[0236] 17 Data sending unit

[0237] 18 Data receiving unit

[0238] PD information processing device

[0239] US user

Claims

1. An information processing device, comprising: a scoring unit that calculates a score of a user's response characteristic based on the user's response history in an interaction with the user; and an interaction control unit that controls the number of inquiries to the user in the interaction based on the score.

2. The information processing device according to claim 1, wherein the scoring unit calculates the score based on the presence / absence of the user's response to an inquiry.

3. The information processing device according to claim 2, wherein the scoring unit calculates the score based on the time from the inquiry to the response.

4. The information processing device according to claim 1, further comprising: a response detection unit that detects an abnormal event in which the user cannot make a response, wherein the scoring unit selectively uses the user's response data during a period other than the period in which the abnormal event occurs to calculate the score.

5. The information processing device according to claim 1, wherein the scoring unit has different calculation criteria for the score according to the context at the time of the response.

6. The information processing device according to claim 5, wherein the context includes the type of content the user is viewing, the type of information presented to the user, or the time period of the interaction.

7. The information processing device according to claim 1, wherein the interaction control unit restricts the interaction based on the score.

8. The information processing device according to claim 7, wherein the interaction control unit divides the score into levels and restricts the interaction step by step according to the level of the score.

9. The information processing device according to claim 8, wherein the interaction control unit: when the score is at a high level, does not restrict the interaction, when the score is at a medium level, as a restriction on the interaction, prohibits inquiries to the user, and when the score is at a low level, as a restriction on the interaction, prohibits inquiries to the user and information provision.

10. The information processing device according to claim 7, wherein the interaction control unit changes the timing of making inquiries according to the score.

11. The information processing device according to claim 7, wherein in the same session, the interaction control unit does not change the restriction on the interaction set at the start of the session.

12. The information processing device according to claim 1, wherein the scoring unit calculates the score based on the verbal information and non-verbal information obtained as the user's response.

13. The information processing device according to claim 1, further comprising: a user identification unit that identifies the user, wherein the scoring unit selectively uses the response data of the user identified as a registered user to calculate the score.

14. An information processing device, comprising: a data sending unit that sends the user's response data in an interaction with the user; a data receiving unit that receives a score of the user's response characteristic calculated based on the user's response history; and an interaction control unit that controls the number of inquiries to the user in the interaction based on the score.

15. An information processing method executed by a computer, the method comprises: calculating a score of a user's response characteristic based on the user's response history in an interaction with the user; and controlling the number of inquiries to the user in the interaction based on the score.

16. An information processing method executed by a computer, the method comprises: sending the user's response data in an interaction with the user; receiving a score of the user's response characteristic calculated based on the user's response history; and controlling the number of inquiries to the user in the interaction based on the score.

Images