JP2017064853A - Robot, content deciding device, content deciding method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a robot, a content deciding device, a content deciding method and a program in which a program of high interest degree of a viewer can be selected according to the situation and can be screened.SOLUTION: A robot 2 implementing content to a viewer acquires an attribute of the viewer from an image of the surrounding (the viewer) and a vocalization situation. The robot 2 acquires the response of the viewer against the content, and seeks the interest degree of the viewer on the basis of the acquired response of the viewer. The content to be performed is decided based on the attribute and the interest degree of the viewer. It may be good that response on screening the content is collected from information opened on the Internet 6, and the interest degree of the viewer is sought based on the collected response. Furthermore, it may be good that the interest degree of the viewer is determined from the number of visitors.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a robot, a content determination device, a content determination method, and a program, and more particularly to a technique for controlling an interactive robot.

  In recent years, robots capable of interacting with people have been developed. This type of robot has a voice recognition function and a speech function for dialogue, and can read the surrounding situation such as facial expression, voice emotion, brightness and temperature of the other party with voice / image recognition engine and various sensors. it can. There are also robots that change their emotions according to the read content, visualize the emotions on the display, and utter utterances according to the emotions. Also, it is possible to provide a movable part on the robot's head, arm, waist, etc., and to perform an action corresponding to the content of the dialogue or command.

  For example, Patent Document 1 describes a robot control method that causes a robot to perform an action as if viewing a video together with a user. In this robot control method, a comment about the program being viewed is acquired from the social media server, and the utterance content to be uttered by the robot is determined from the comment of the personality matching speaker that matches the personality set in the robot. Based on the dialogue state and the emotional state of the robot, the action content to be executed by the robot is extracted from the action database, and the action corresponding to the content of the program being viewed is executed by the robot.

JP2015-148701A

  However, the robot control method disclosed in Patent Document 1 determines the content of an utterance or action of a robot based on information obtained from a social media server, and is determined from the situation of many viewers in front of the robot. Not what you want. For example, when talking-style content or quiz-style content is held in front of an event attendee, the robot will perform a performance that draws the interest and interest by judging the reaction of the visitor and the surrounding situation in real time. Therefore, there is a demand to improve the ability to attract customers.

  An object of the present invention is to provide a robot, a content determination device, a content determination method, and a program that can select and screen a performance with a high degree of interest of the viewer according to the situation.

  In order to solve the above-described problems, the first invention is a content execution unit that executes content for a viewer, a reaction acquisition unit that acquires the viewer's response to the content to be executed, and the acquired viewer's response An interest level calculating means for determining the degree of interest of the viewer based on the content; and a content determining means for determining content to be executed by the content executing means based on the interest level of the viewer. is there.

According to the first invention, the robot that executes content for the viewer acquires the viewer's response to the content to be executed, obtains the viewer's interest based on the acquired viewer's response, and views Content to be executed is determined based on the interest level of the person.
As a result, the robot can determine and execute the content based on the reaction (degree of interest) of the viewer without human intervention. It is possible to provide a robot that can be selected and screened according to the situation in which the audience is highly interested.

  In the first aspect of the invention, further comprising attribute acquisition means for acquiring the viewer attribute, wherein the content determination means determines content to be executed by the content execution means based on the viewer attribute and the degree of interest. It is desirable to do. This makes it possible to determine the content to be executed in consideration of the viewer's attributes.

  In the first aspect of the invention, the image processing device further includes an image analysis unit that acquires and analyzes an image obtained by capturing the state of the viewer, and an utterance state acquisition unit that acquires an utterance state of the viewer. Preferably, the viewer's reaction is acquired based on the analysis result of the image and the utterance situation. Thereby, it becomes possible to judge the reaction of the audience from the image of the audience and the utterance situation.

  In the first aspect of the invention, the information processing device further comprises an Internet information collecting unit that collects a reaction related to the screening of the content from information published on the Internet, and the reaction acquisition unit further includes the reaction collected by the Internet information collecting unit. It should be included in the viewer's response. Moreover, it is desirable that the response acquisition means further includes the number of visitors in the viewer's response. As a result, it is possible to judge the reaction and topicality related to the screening from information such as comments and reviews published on the Internet and the number of visitors, and reflect it in the determination of the screening performance.

  In the first invention, the recording apparatus may further include a recording unit that records the interest level of each content in association with the viewer attribute, and the content determination unit is configured to record the content based on the viewer attribute and the interest level. It is desirable to select a plurality of optimum contents from the contents recorded in the recording means and generate a series of screening lists. Thereby, a screening list can be created in advance, and a screening list with a high degree of interest can be selected and executed according to the attributes (gender, age group, etc.) of the visitors. In addition, for example, a screening list corresponding to a desired degree of interest such as emphasis on excitement and concentration can be generated and stored in advance.

In the first invention, the content determination means may preferentially include content that has not been shown for a while in the screening list. In addition, pre-designated recommended content may be preferentially included in the screening list. Also, newly created content may be preferentially included in the screening list.
As a result, not only shows with a high degree of interest but also performances that the director wants to preferentially show or shows with a low screening frequency can be executed, so viewers will not get bored and a new level of interest will be added. It becomes possible to find the performance that becomes.

  According to a second aspect of the present invention, content execution means for executing content for a viewer, reaction acquisition means for acquiring the viewer's response to the content to be executed, and the viewer's response based on the acquired viewer's response An interest level calculating means for obtaining an interest level, and a content determination means for determining content to be executed by the content execution means based on the interest level of the viewer.

According to the second invention, the viewer's response to the content to be executed is acquired, the viewer's interest level is obtained based on the acquired viewer's response, and the content to be executed is determined based on the viewer's interest level And execute the content to the viewer.
Accordingly, it is possible to provide an apparatus that can determine and execute content based on the reaction (degree of interest) of a viewer without human intervention. Performances with high viewer interest can be selected and screened according to the situation.

  In the second invention, it is desirable that the content execution means execute the content using a robot. Thereby, the robot can execute the content determined by the content determination device of the second invention. Using a robot, it is possible to select and screen performances with high viewer interest according to the situation.

  According to a third aspect of the present invention, a computer causes a viewer to execute content by using a robot, a step of acquiring the viewer's response to the content executed by the computer, and a viewer's response acquired by the computer. A content determination method comprising: obtaining an interest level of the viewer based on the content; and determining content to be executed by a computer based on the interest level of the viewer.

According to the third invention, by controlling the computer, the viewer is caused to execute the content using a robot, the viewer's response to the executed content is acquired, and the viewer is based on the acquired viewer's response. The content to be executed is determined based on the interest level of the viewer.
This makes it possible to determine content based on the viewer's reaction (degree of interest) without human intervention and execute the content using a robot. It is possible to provide a robot that can be selected and screened according to the situation in which the audience is highly interested.

  A fourth invention is a program for causing a computer to function as the content determination device of the second invention. According to the fourth invention, it is possible to cause a computer to function as the content determination apparatus of the second invention.

  According to the present invention, it is possible to provide a robot, a content determination device, a content determination method, and a program capable of selecting and showing a performance with a high degree of interest of the viewer according to the situation.

Overall configuration diagram of the content determination system 1 Robot 2 internal configuration diagram Internal configuration diagram of computer 3 (content determination device) Configuration diagram regarding content determination function of robot 2 Data structure diagram of screening list data 82 Data configuration diagram of interest level calculation table 83 Data structure diagram of screening history information 84 Data structure of recommended screening list 85 Data structure diagram of content table 81 The flowchart explaining the flow of the content determination process which the robot 2 performs

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings.
FIG. 1 is a diagram illustrating an example of a system configuration of the content determination system 1. The content determination system 1 includes a robot 2, a computer 3, a distance image sensor 4, and a microphone 5. The robot 2 or the computer 3 is connected to an SNS server 7 (Social Networking Service) on the Internet 6 via a network. Communication connection is possible. The robot 2, the distance image sensor 4, and the microphone 5 are communicatively connected to a computer 3 and controlled by the computer 3 so that the operations of the devices cooperate with each other. The form of communication connection between the robot 2, the distance image sensor 4, and the microphone 5 and the computer 3 may be wired or wireless.

  The distance image sensor 4 is an apparatus that captures an image and measures distance information (depth information). The distance image sensor 4 captures an image of the viewer around the robot 2 and the distance (depth information) from the viewer. Input to the computer 3.

The microphone 5 acquires the utterance status of the viewers around the robot 2 and inputs it to the computer 3.
Note that it is also possible to take a picture of the viewer using the camera 210 built in the robot 2 instead of the distance image sensor 4 and acquire distance information between the viewer and the robot 2. In that case, the distance image sensor 4 may be omitted. Similarly, for the microphone 5, the utterance status of the viewer can be acquired using the microphone 211 built in the robot 2. In that case, the microphone 5 may be omitted.

  In addition to controlling the operation of the robot 2, the computer 3 controls acquisition of images and distance information by the distance image sensor 4 and voice acquisition of the microphone 5.

  The robot 2 is an interactive robot. As shown in FIG. 2, the robot 2 includes a control unit 201, a storage unit 202, an input unit 203, a display unit 204, a media input / output unit 205, a communication I / F 206, a movable unit drive mechanism 207, and a peripheral device I / F unit. 208, a sound processing unit 209, a camera 210, a microphone 211, a sensor 212, and a speaker 213 are connected via a bus 220.

  The control unit 201 includes a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), and the like. The CPU calls and executes a program stored in the storage unit 202, ROM, recording medium, or the like to a work memory area on the RAM, and drives and controls each unit connected via the bus 220.

  The CPU of the control unit 201 executes content determination processing (see FIG. 10). Details of this processing will be described later.

  The ROM permanently stores programs such as a boot program for the robot 2 and a BIOS, data, and the like. The RAM temporarily stores the loaded program and data, and includes a work area used by the control unit 201 for performing various processes.

  The storage unit 202 stores a program executed by the control unit 201, data necessary for program execution, an operating system, and the like. These program codes are read as necessary by the control unit 201, transferred to the RAM, and read and executed by the CPU.

  The storage unit 202 holds databases such as a content DB (database) 31, an utterance DB (database) 32, and an action DB (database) 33. Further, the storage unit 202 includes a content table 81, a screening list data 82, an interest level calculation table 83, a screening history information 84, a recommended screening list 85, and the like as data referred to in the content determination process executed by the control unit 201 ( (See FIG. 4). Details of the content DB 31, utterance DB 32, action DB 33, content table 81, screening list data 82, interest level calculation table 83, screening history information 84, and recommended screening list 85 will be described later.

The input unit 203 includes, for example, an input device such as a touch panel and various operation buttons, and outputs input data to the control unit 201.
The display unit 204 includes a display device such as a liquid crystal panel and a logic circuit for executing display processing in cooperation with the display device, and displays display information input by the control of the control unit 201 on the display device. Let Note that the touch panel of the input unit 203 is configured integrally with the display of the display unit 204.

  The media input / output unit 205 is, for example, a drive device such as a recording medium (media) such as a magnetic disk, an optical disk, or a semiconductor memory, and inputs / outputs data.

  A communication I / F (interface) 206 is an interface that mediates communication with a network. The network includes a LAN (Local Area Network), a WAN (Wide Area Network) connected to a wider area, a public communication line such as the Internet 6, a base station, and the like. The communication connection with the network may be wired or wireless. The robot 2 can access an SNS (Social Networking Service) server or the like on the Internet 6 via a network, acquire information published on the Internet 6, and transmit / receive various programs and data from the Web server.

  The movable part drive mechanism 207 includes a mechanism and a drive device for driving a movable part provided on the head, neck, shoulders, arms, waist, legs, and the like of the robot 2. The operation of the movable unit driving mechanism 207 is controlled by the control unit 201. The control unit 201 reads the movable part control sequence from the action DB 33 of the storage unit 202, sends a control signal to each movable part drive mechanism 207, and operates each drive part.

  The peripheral device I / F (interface) unit 208 is a port for connecting a peripheral device, and transmits / receives data to / from the peripheral device via the peripheral device I / F unit 208. The connection form with the peripheral device may be wired or wireless. The robot 2 can communicate with the computer 3 or the like via the peripheral device I / F unit 208.

  The voice processing unit 209 includes a voice synthesis unit, a voice recognition unit, and the like. The voice synthesizer generates synthesized voice data corresponding to character information (text data) based on a command from the control unit 201, and outputs the synthesized voice to the speaker 213. The synthesized speech is generated based on the correspondence between the character information stored in advance in the utterance DB 32 of the storage unit 202 and the speech data. The speech synthesizer realizes speech by the robot 2, dialogue with people, and the like.

  The voice recognition unit performs voice recognition processing on voice data (for example, human voice) input from the microphone 211, generates character information (text data) corresponding to the voice data, and inputs the character information to the control unit 201. The control unit 201 recognizes the content of a person's story based on the character information input by the speech recognition unit, determines the surrounding situation, changes the emotion of the robot 2, and determines the content of the action or utterance. decide.

  The camera 210 is, for example, a CCD (Charge-Coupled Device) camera built in the robot 2 and shoots the surroundings. An image captured by the camera 210 is input to the control unit 201. The control unit 201 recognizes an image input from the camera 210 (image analysis processing), acquires a surrounding situation, changes an emotion according to the recognition result, and determines an utterance and an action.

The microphone 211 inputs external sound and sends it to the control unit 201.
The speaker 213 outputs a synthesized voice generated by the voice synthesis unit of the voice processing unit 209 or outputs a predetermined voice such as an alarm sound based on a command from the control unit 201.

The sensor 212 includes various sensors such as a 3D sensor, a touch sensor, a gyro sensor, a sonar sensor, a laser sensor, a bumper sensor, and an infrared sensor. Each sensor inputs the detected signal to the control unit 201. The control unit 201 changes emotions according to input signals from each sensor, and determines the content and action of the utterance of the robot 2.
The bus 220 is a path that mediates transmission / reception of control signals, data signals, and the like.

  The computer 3 is a control device that controls the operation of the entire content determination system 1, and controls the operation of the robot 2 and also controls the acquisition of images and distance information by the distance image sensor 4 and the acquisition of sound by the microphone 5.

  As shown in FIG. 3, the computer 3 includes a control unit 301, a storage unit 302, a media input / output unit 303, a peripheral device I / F unit 304, an input unit 306, a display unit 307, and a communication I / F 308 via a bus 309. Connected and configured.

  The control unit 301 includes a CPU, ROM, RAM, and the like. The CPU calls a program stored in the storage unit 302, ROM, recording medium, or the like to a work memory area on the RAM and executes it, and drives and controls each unit connected via the bus 309.

  The CPU of the control unit 301 performs shooting control by the distance image sensor 4 and voice acquisition control of the microphone 5 in cooperation with content determination processing (see FIG. 10) executed by the robot 2.

  The ROM permanently stores a boot program for the computer 3, a program such as BIOS, data, and the like. The RAM temporarily holds the loaded program and data, and includes a work area used by the control unit 301 to perform various processes.

  The storage unit 302 stores a program executed by the control unit 301, data necessary for program execution, an operating system, and the like. These program codes are read as necessary by the control unit 301, transferred to the RAM, and read and executed by the CPU.

  The media input / output unit 303 is a drive device such as a recording medium (media) such as a magnetic disk, an optical disk, or a semiconductor memory, and inputs / outputs data.

  A peripheral device I / F (interface) unit 304 is a port for connecting a peripheral device, and transmits and receives data to and from the peripheral device via the peripheral device I / F unit 304. The connection form with the peripheral device may be wired or wireless. The robot 2, the distance image sensor 4, the microphone 5, and the like can be connected to the computer 3 via the peripheral device I / F unit 304.

  The input unit 306 is an input device such as a keyboard, a mouse, or a touch panel, and outputs input data to the control unit 301.

  The display unit 307 includes a display device such as a liquid crystal panel and a logic circuit for executing display processing in cooperation with the display device, and displays display information input by the control of the control unit 301 on the display device. Let The touch panel of the input unit 306 is configured integrally with the display of the display unit 307.

  A communication I / F (interface) 308 is an interface that mediates communication with a network such as the Internet 6. The computer 3 can access the SNS server 7 and the like on the Internet 6 via the network, obtain information disclosed on the Internet 6, and transmit / receive various programs and data from the Web server.

  The bus 309 is a path that mediates transmission / reception of control signals, data signals, and the like.

Next, the functional configuration of the robot 2 will be described with reference to FIG.
As shown in FIG. 4, the robot 2 includes a content execution unit 21, an attribute acquisition unit 22, a reaction acquisition unit 23, an interest level calculation unit 24, a recording unit 25, a content determination unit 26, a content DB 31, an utterance DB 32, an action DB 33, A content table 81, a screening list data 82, an interest level calculation table 83, a screening history information 84, and a recommended screening list 85 are provided.

  The content execution unit 21 controls the execution of content by the robot 2. The content execution unit 21 refers to the screening list data 82 stored in the storage unit 202, and receives content data of a plurality of contents included in a specified screening list or a screening list that meets a predetermined condition from the content DB 31. Read and execute in order. The content to be screened includes, for example, quiz questions, progress, stories, and other various performances by the robot 2.

  FIG. 5 shows an example of the screening list data 82 stored in the storage unit 202. The screening list data 82 stores a plurality of screening lists. The screening list is a set of which content is executed in which order for each performance unit. “Screening No.” is screening list identification information. For example, in the daytime performance, screening No. No. 1 screening list (contents B, A, D) is executed. No. 3 screening list (contents E, A, F, D) is executed, and so on. You can specify a series of content to be shown.

  The content DB 31 stores a content table 81 and a plurality of content data. The content table 81 is a table that stores the title and type (type of quiz, story, etc.) of the content. The content data includes sequence data of utterances and actions executed by the robot 2 in the content, and related video and music data.

  When executing the content, the content execution unit 21 generates a synthesized voice corresponding to the utterance content defined in the content data by the voice processing unit 209 of the robot 2 and causes the robot 2 to read it out (voice output). The voice processing unit 209 acquires voice data corresponding to the utterance content (text data) defined in the content data with reference to the utterance DB 32, and generates synthesized voice.

  The utterance DB 32 stores character information (text data) and voice data in association with each other. The speech processing unit 209 realizes speech by the robot 2 by generating synthesized speech based on the correspondence between the character information stored in the speech DB 32 and the speech data, and outputting it from the speaker 213. The utterance DB 32 may store standard sentence voice data, utterance conversion data for changing the utterance inflection or ending in accordance with the emotional state of the robot 2, and the like.

  Further, the content execution unit 21 controls the movable unit driving mechanism 207 of the robot 2 to cause the robot 2 to execute an action (action) defined in the content data.

  The action DB 33 stores a control sequence table describing the control sequence of the robot 2. In the control sequence table, data in which the action name of the action is associated with the movable part control sequence of the robot 2 is stored. The action name corresponds to the action name of the action in the content data. For example, the robot action when the robot 2 executes a series of moving part control sequences such as “nodding”, “raising one hand”, and “raising both hands”. It shows a state. The movable part control sequence is a list of driving locations and values of each movable part to be controlled (for example, head tilt angle 0 degrees, right arm tilt angle 0 degrees, etc.).

  The attribute acquisition unit 22 acquires the attributes of the viewer who is viewing the content executed by the robot 2. The viewer attribute is, for example, age or sex. The attribute acquisition unit 22 analyzes an image captured by the camera 210 of the robot 2 or an image input from the distance image sensor 4 and determines the attributes (age and gender) of each viewer by, for example, face recognition processing. As a method for recognizing a person's attribute from a face image, a known method (Japanese Patent Application Laid-Open No. 2012-053813) may be used. Since there are a plurality of viewers, the attribute of the person near the robot 2 among the attributes of each viewer may be used as a representative attribute of the entire viewer, or the ratio of age group and gender is obtained and the ratio is large. Age group and gender may be representative attributes of the entire viewer. Further, when the viewer attributes can be collected as data by, for example, a questionnaire survey for the viewers, the viewer attribute value is input to the robot 2 (computer 3) via the input unit 203 or the computer 3 or the like. May be.

  The reaction acquisition unit 23 acquires a viewer's reaction to the content executed by the content execution unit 21. The viewer's reaction is input from, for example, (1) image information input from the camera 210 or the distance image sensor 4 built in the robot 2, and (2) input from the microphone 211 or the external microphone 5 built in the robot 2. Voice information, (3) information published on the Internet 6, and the like.

  Therefore, the reaction acquisition unit 23 includes at least one of the image analysis unit 23a, the utterance situation acquisition unit 23b, and the Internet information collection unit 23c.

  The image analysis unit 23 a reads the viewer's facial expression from image information input from the camera 210 or the distance image sensor 4 built in the robot 2. Facial expressions are classified into several emotion patterns such as joy, surprise, anger, sadness, and expressionlessness. Each emotion pattern can be read from the rising angle of the mouth corner, the positional relationship of the cheek, eyes, eyebrows, and the like. As a method for recognizing emotions and states from facial expressions, a known method (Japanese Patent Application Laid-Open No. 2014-206903) may be used. The image analysis unit 23a obtains several degrees of emotion such as joy level, surprise level, anger level, sadness level, and indifference level from the facial expression of the viewer read from the image information, and notifies the interest level calculation unit 24 of the level.

  The utterance situation acquisition unit 23 b reads the utterance situation of the viewer from the voice information input from the microphone 211 built in the robot 2 or the external microphone 5. The utterance situation is, for example, the volume of utterance, the utterance density, and the like. The utterance density is a numerical value indicating how dense the utterances are (whether the speakers are actively participating), and is obtained from the frequency and timing of the peak at which the volume increases. The utterance status acquisition unit 23b notifies the interest level calculation unit 24 of each numerical value of the obtained utterance status.

  The Internet information collection unit 23c collects reactions related to the content screening from information published on the Internet 6 by a social networking service (SNS) or the like. The Internet information collection unit 23c collects information published on the Internet 6 using, for example, a content title or a phrase included in the content as a search keyword. The content and magnitude of the response to the content are quantified based on the type and number of search hits and words (“interesting”, “not boring”, etc.) corresponding to the impression. The internet information collection unit 23 c notifies the interest level calculation unit 24 of the obtained numerical value.

  The interest level calculation unit 24 obtains the interest level of the viewer based on the reaction of the viewer acquired from the reaction acquisition unit 23 (the image analysis unit 23a, the utterance situation acquisition unit 23b, and the Internet information collection unit 23c). When calculating the interest level, the interest level calculation unit 24 may refer to an interest level calculation table 83 as shown in FIG. The interest level calculation table 83 is a table that defines the relationship between the values of a plurality of items indicating reactions and the level of interest (such as “rising degree”).

FIG. 6 is a diagram illustrating an example of the interest level calculation table 83.
For example, when “degree of excitement” is calculated as the degree of interest, each of volume, utterance density, facial expression, looking away (number of viewers or density), distance from the screen, SNS (Internet information), number of visitors, etc. For items, the degree of excitement is classified into several types based on the sum of frequencies and the value of which item is remarkably high.

  For example, the criterion of “satisfying” for the degree of excitement “9” is that the volume is “3” or less, the speech density is “5” or less, the looking away is “3” or less, the distance from the screen is “3” or less, and the SNS is “7” or more and the number of visitors is “7” or more. When viewers are focused and staring, there are few viewers who talk and look away, and they are often approaching the screen (stage). The expression is not questioned.

  On the other hand, the criteria for “big excitement” with the degree of excitement “10” are: volume level “8” or higher, speech density “9” or higher, “joy” facial expression “8” or higher, looking away “3” or lower The distance from the screen is “3” or less, the SNS is “8” or more, and the number of visitors is “8” or more.

  In addition, the criteria for determining the level of excitement as “1” is “being bored and blurred” is that the volume is “3” or less, the speech density is “2” or less, and the expression indicating “no expression” is “8” or more. Is “5” or more, the distance to the screen is “5” or more, the SNS is “5” or less or “8” or more (there are many topics that are not topical or boring), and the number of visitors is “4” or less. And

In addition, the criteria of “noisy and noisy” with a degree of excitement of “1” is that the volume is “8” or higher, the speech density is “9” or higher, the expression indicating “no expression” is “7” or higher, “5” or more, the distance to the screen is “5” or more, SNS is “5” or less or “8” or more (there are many topics that are not topical or boring), and the number of visitors is “5” or less To do.
The interest level calculation unit 24 notifies the content determination unit 26 of the interest level obtained based on the plurality of indexes as described above.

  Note that the degree of interest is not limited to “the degree of excitement”, and for example, “concentration” indicating whether the audience is concentrated, “impression” indicating the degree of impression, and the like, different determination criteria may be provided.

  The recording unit 25 records the interest level calculated by the interest level calculation unit 24 and the viewer attribute acquired by the attribute acquisition unit 22 in association with each content as a history of content screening (execution).

FIG. 7 is an example of the screening history information 84 recorded by the recording unit 25.
As shown in FIG. 7, the screening history information 84 includes a content screening date and a content number of the executed content. Are sequentially recorded. Each content No. The attribute information such as the gender and age group of the viewer and the interest level calculated by the interest level calculation unit 24 are recorded in association with each other. The gender and age group of the attribute information records the gender and age group with a large proportion of the entire viewer. Alternatively, the ratio of men and women and the ratio of each age group may be recorded.

  The content determination unit 26 determines the content to be executed by the robot 2 (content execution unit 21) based on the viewer attributes acquired by the attribute acquisition unit 22 and the interest level calculated by the interest level calculation unit 24. Specifically, the content determination unit 26 selects a plurality of optimal contents from the contents recorded in the content table 81 based on the attributes and interests of the viewers, and recommends suitable for the attributes and interests of the viewers. A screening list 85 is generated. The generated recommended screening list 85 may be included in the screening list data 82 of the storage unit 202 or may be stored in another storage area of the storage unit 202.

FIG. 8 is an example of the recommended screening list 85 generated by the content determination unit 26.
As shown in FIG. 8, a list (screening list) in which some contents are selected based on gender, age group, and degree of interest is generated. For example, the content determination unit 26 may preferentially include content having a high interest level in the corresponding gender and age group in the screening list.

  Specifically, content A, C, D, ... for boys 0-5 years old, content B, A, C, ..., 6-12 years old for girls 0-5 years old A screening list is generated for contents B, C, E,... For boys, and contents A, D, E,.

  In addition, the content determination unit 26 may refer to the screening history information 84 to preferentially include content that has not been screened for a predetermined period in the recommended screening list 85 described above. Moreover, priority content designated in advance may be preferentially included in the recommended screening list 85. Further, the newly created new content may be preferentially included in the recommended screening list 85. Therefore, it is desirable for the user to set a “priority” flag and a “new work” flag for each content stored in the content DB 31 as shown in FIG. It is assumed that the content determination unit 26 preferentially incorporates content having these flags set to “1” into the recommended screening list 85. Further, the content determination unit 26 may select and determine the content to be screened so that the screening frequency is not biased.

  The content table 81 shown in FIG. 9 is a table in which the title, type, priority, etc. of each content stored in the content DB 31 are set. The content table 81 is stored in the content DB 31 together with the content data. “Content No.” in the content table 81 is content identification information, “Title” is the title of each content, “Type” is the content type of quiz, story, etc., and “Priority” is the screening It is a flag that specifies whether or not to be preferentially included in the list, and “new work” is a flag that indicates whether or not it is a new work.

  The content determination unit 26 refers to the priority flag, new work flag, or screening history information 84 in the content table 81 and selects content to be included in the recommended screening list 85 from the content table 81. Further, the order of the contents may be determined under conditions specified in advance by the director, such as arranging the contents of the same type so as not to be continuous, or intentionally continuing the same type of content.

  Next, the flow of content determination processing executed by the robot 2 will be described with reference to FIG.

  As shown in the flowchart of FIG. 10, the control unit 201 of the robot 2 first acquires the viewer attributes of the viewers around (step S101). As described above, the viewer attributes are, for example, age, sex, etc., and an image taken by the camera 210 of the robot 2 or the external distance image sensor 4 is analyzed, and the attributes (age, sex) of each viewer are determined. to decide. Further, attribute data obtained by a questionnaire survey or the like may be input to the robot 2.

  When the performance starts, the robot 2 executes the content (step S102). The control unit 201 controls content execution by the robot 2. That is, the control unit 201 refers to the screening list data 82 stored in the storage unit 202, and sequentially reads content data of content set in advance or set in the specified screening list from the content DB 31. And execute. The content to be screened includes, for example, quiz questions, progress, stories, and other various performances by the robot 2. The control unit 201 controls the utterance and action of the robot 2 according to the content data.

  During the execution of the content, the control unit 201 of the robot 2 acquires the viewer's reaction (step S103). As described above, the viewer's reaction is, for example, (1) image information input from the camera 210 or the distance image sensor 4 built in the robot 2, and (2) the microphone 211 built in the robot 2 or an externally attached device. It can be acquired from voice information input from the microphone 5, (3) information published on the Internet 6, and the like. The image information (1) and the audio information (2) are acquired in real time while the robot 2 is executing the content. The information disclosed on the Internet 6 in (3) can usually be acquired after the end of the screening because the viewer uses SNS and the like to write impressions after the end of viewing the content.

  The control unit 201 acquires the viewer's facial expression (joy, surprise, anger, sadness, no expression, etc.) based on image information input from the camera 210 or the distance image sensor 4 built in the robot 2. Further, the control unit 201 acquires the venue (viewer) volume, speech density, and the like based on audio information input from the microphone 211 built in the robot 2 or the external microphone 5.

  The control unit 201 calculates the degree of interest in the content based on the viewer's response acquired in step S103 (step S104). The control unit 201 refers to the interest level calculation table 83 (see FIG. 6) stored in the storage unit 202, and considers several indexes such as volume, speech density, facial expression, looking away, distance to the screen, etc. Calculate the interest level of each content being shown. The control unit 201 may further calculate the interest level in consideration of input data such as the number of visitors. For example, the degree of interest may be high when the number of visitors is large, and the degree of interest may be low when the number of visitors is small.

  The control unit 201 records the interest level of the viewer calculated in step S104 and the viewer attribute acquired in step S101 in the screening history information 84 (see FIG. 7) in association with the content (step S105).

  During the execution of the content, the processing from step S103 to step S105 is performed, and the attribute (viewer's gender and age group) is recorded in the screening history information 84 for each executed content, and the viewer's reaction (degree of interest) is shown. Recorded in the history information 84.

  After the screening, the control unit 201 searches for the content shown from the information published on the Internet 6, acquires the number of search hits, impressions, etc., recalculates the interest level in consideration of the Internet information, and re-records it. May be.

  When the screening is completed, the control unit 201 refers to the screening history information 84 and generates a recommended screening list 85 for each gender and age of the viewer (step S106). When generating the recommended screening list 85, the control unit 201 considers the interest level of each content, and also considers the number of visitors and the priority (priority flag, new work flag) set for the content. Create

The control unit 201 adds the created recommended screening list 85 to the screening list data 82 of the storage unit 202, and ends a series of content determination processing.
When the control unit 201 of the robot 2 determines content to be screened, an optimal screening list may be selected from the screening list data 82 (including the recommended screening list 85) according to the viewer's attributes and age. . For example, it is possible to select a screening list corresponding to the desired degree of interest, such as emphasis on excitement and importance on the degree of concentration, according to the viewer's attributes and age group.

  As described above, when the robot 2 executes the content for the viewer, the robot 2 acquires the viewer attribute, acquires the viewer's response to the content to be executed, and based on the acquired viewer's response. The interest level of the viewer is obtained, and the content to be executed by the robot 2 is determined based on the attribute and the interest level of the viewer. The degree of interest of the viewer is obtained from an image of the viewer watching the content, audio information of the venue (viewer), Internet information such as SNS, the number of visitors, and the like. As a result, it is possible to determine content with a high degree of interest of the viewer without causing human intervention to be executed by the robot 2.

  In FIG. 4, the control unit 201 and the storage unit 202 in the robot 2 include the function units included in the content determination system 1 (content execution unit 21, attribute acquisition unit 22, reaction acquisition unit 23, image analysis unit 23 a, utterance status). An acquisition unit 23b, an Internet information collection unit 23c, an interest level calculation unit 24, a recording unit 25, a content determination unit 26, and various data, etc.), but these are provided in a computer 3 provided separately from the robot 2. Each function unit may be provided, and the computer 3 may execute the content determination process illustrated in FIG. 10 or control the robot 2. The processing of each functional unit may be executed by a program. Programs related to these processes may be stored in the storage unit 202 of the robot 2 or the storage unit 302 of the computer 3, or may be installed in the robot 2 or the computer 3 via a network such as the Internet 6.

  The preferred embodiments of the robot, the content determination device, the content determination method, and the like according to the present invention have been described above with reference to the accompanying drawings, but the present invention is not limited to such examples. It will be apparent to those skilled in the art that various changes or modifications can be conceived within the scope of the technical idea disclosed in the present application, and these are naturally within the technical scope of the present invention. Understood.

1 …………………… Content Determination System 2 …………………… Robot 201 ……………… Control Unit 21 …………………… Content Execution Unit 22 ………………… Attribute acquisition unit 23 …………………… Reaction acquisition unit 23a ……………… Image analysis unit 23b ……………… Speech status acquisition unit 23c ……………… Internet information collection unit 24 ………… ………… Interest Calculation Unit 25 ………………… Recording Unit 26 …………………… Content Determination Unit 202 ……………… Storage Unit 31 …………………… Content DB
32 ………………… Speech DB
33 …………………… Action DB
81 …………………… Contents table 82 ………………… Show list data 83 ………………… Interest calculation table 84 ………………… Show history information 85 …………… …… Recommended screening list 3 …………………… Computer 301 ……………… Control unit 302 ……………… Storage unit 4 …………………… Distance image sensor 5 ………… ………… Microphone 6 …………………… Internet 7 …………………… SNS Server

Claims (13)

Content execution means for executing content for viewers;
Reaction acquisition means for acquiring a response of the viewer to the content to be executed;
An interest level calculating means for calculating the viewer's interest level based on the acquired viewer's reaction;
Content determination means for determining content to be executed by the content execution means based on the degree of interest of the viewer;
A robot characterized by comprising: Attribute acquisition means for acquiring attributes of the viewer,
The robot according to claim 1, wherein the content determination unit determines the content to be executed by the content execution unit based on the attribute of the viewer and the degree of interest. Image analysis means for acquiring and analyzing an image of the viewer's appearance;
Utterance status acquisition means for acquiring the utterance status of the viewer,
The robot according to claim 1, wherein the response acquisition unit acquires the viewer's response based on the analysis result of the image and the utterance state. Internet information collecting means for collecting a reaction related to the screening of the content from information published on the Internet,
The robot according to any one of claims 1 to 3, wherein the response acquisition unit further includes the response collected by the Internet information collection unit in the viewer's response.   The robot according to claim 1, wherein the response acquisition unit further includes the number of visitors in the viewer's response. Recording means for recording the degree of interest for each content in association with the attribute of the viewer;
The content determination means selects a plurality of optimum contents from contents recorded in the recording means based on viewer attributes and the degree of interest, and generates a series of screening lists. 2. The robot according to 2.   The robot according to claim 6, wherein the content determination unit preferentially includes content that has not been screened for a while in the screening list.   The robot according to claim 6 or 7, wherein the content determination means preferentially includes recommended content designated in advance in the screening list.   The robot according to claim 6, wherein the content determination unit preferentially includes newly created content in the screening list. Content execution means for executing content for viewers;
Reaction acquisition means for acquiring a response of the viewer to the content to be executed;
An interest level calculating means for calculating the viewer's interest level based on the acquired viewer's reaction;
Content determination means for determining content to be executed by the content execution means based on the degree of interest of the viewer;
A content determination apparatus comprising:   The content determination apparatus according to claim 10, wherein the content execution unit executes the content using a robot. A computer causing a viewer to execute content using a robot;
Obtaining a response of the viewer to computer-executed content;
Determining the viewer's interest based on a viewer response obtained by a computer;
Determining content to be executed by a computer based on the viewer's interest;
The content determination method characterized by including.   The program which functions a computer as a content determination apparatus of Claim 10.

Images