JP6583754B2 - Information processing device, mirror device, program - Google Patents

Description

  The present invention relates to an information processing apparatus, a mirror device, and a program.

  In recent years, smart speakers equipped with voice recognition technology have attracted attention. The smart speaker analyzes a voice command uttered by the user and performs an operation (for example, voice output) according to the analysis result.

  For example, the voice recognition terminal device of Patent Literature 1 acquires a voice command spoken by a user, transmits the acquired voice command to a voice recognition server, and requests voice recognition processing. When receiving a response message from the voice recognition server, the voice recognition terminal device executes an output process based on the acquired response message.

  In the smart speaker, interactive interaction occurs between a computer mounted on the smart speaker and a user. As a result, the user can use the smart speaker to talk to the smart speaker.

JP2015-135420A

  In general, a mirror is an indispensable device in human daily life. In particular, women frequently use mirrors for beauty behavior (eg, makeup, clothing confirmation, or facial expression confirmation). However, since the user uses the mirror to observe the mirror image, no interaction occurs between the mirror and the user.

  For example, as a method for generating an interaction between a user and a mirror, a method of mounting the technique of Patent Document 1 on a mirror is conceivable. In this case, when the user speaks a voice command, the mirror performs output processing according to the voice command. As a result, an interaction occurs between the user and the mirror via a voice command issued by the user. This interaction is an interaction by voice input and voice output.

  On the other hand, a person who performs a beauty behavior places importance on visual information. For example, the reason why a woman applies makeup while looking at a mirror image in a mirror is to visually confirm the face on which the makeup is applied.

  However, even if the technique disclosed in Patent Document 1 is implemented in a mirror, visual information that is important for an observer reflected in the mirror does not intervene in the interaction because of the interaction by voice input and voice output. That is, the information that is important for the observer is not reflected in the output voice.

  Thus, with the conventional speech recognition technology, it is not possible to generate an interaction reflecting visual information important for an observer who observes a mirror image.

  An object of the present invention is to provide an interaction in which visual information is reflected to an observer who observes a mirror image.

One embodiment of the present invention provides:
An information processing apparatus connected to a mirror device capable of acquiring an observer image corresponding to a mirror image of the observer and outputting sound,
Means for acquiring the observer image from the mirror device;
Means for determining a state of an observer reflected in the mirror image based on the observer image;
Based on the result of the determination, comprising means for generating an anthropomorphic character voice,
Means for transmitting the generated audio to the mirror device;
Information processing apparatus.

  ADVANTAGE OF THE INVENTION According to this invention, the interaction in which visual information was reflected can be provided with respect to the observer who observes a mirror image.

It is a block diagram which shows the structure of the information processing system of this embodiment. It is a functional block diagram of the information processing system of FIG. It is explanatory drawing of the outline | summary of this embodiment. It is a figure which shows the data structure of the observer information database of this embodiment. It is a figure which shows the data structure of the character information master of this embodiment. It is a figure which shows the data structure of the pattern matching table of this embodiment. It is a figure which shows the data structure of the log information database of this embodiment. It is a sequence diagram of the information processing of this embodiment. It is a figure which shows the example of an output in the information processing of FIG. It is a figure which shows the example of an output in the information processing of FIG. It is a figure which shows the example of an output in the information processing of FIG. FIG. 9 is a detailed flowchart of reaction generation in FIG. 8. FIG. It is a figure which shows the example of an output in the information processing of the modification 1. It is a detailed flowchart of the reaction generation of the modification 2. 12 is an explanatory diagram of an outline of modification example 7. FIG. It is explanatory drawing of the outline | summary of the modification 8.

  Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings. Note that in the drawings for describing the embodiments, the same components are denoted by the same reference symbols in principle, and the repetitive description thereof will be omitted.

(1) Configuration of Information Processing System The configuration of the information processing system will be described. FIG. 1 is a block diagram showing the configuration of the information processing system of this embodiment. FIG. 2 is a functional block diagram of the information processing system of FIG.

As illustrated in FIG. 1, the information processing system 1 includes a mirror device 10 and a server 30.
The mirror device 10 and the server 30 are connected via a network (for example, the Internet or an intranet) NW.

The mirror device 10 is an example of an information processing apparatus that transmits a request to the server 30.
The mirror device 10 includes a display 18 and a half mirror 19. The mirror device 10 presents an image to the observer via the display 18 and presents a mirror image to the observer via the half mirror 19. That is, the mirror device 10 can simultaneously present an image and a mirror image to the observer.

The half mirror 19 is disposed outside the display 18 (FIG. 1).
When the observer turns his / her line of sight toward the half mirror 19, the light of the display 18 that has passed through the half mirror 19 enters the eyes of the observer. Thereby, the observer can observe the image displayed on the display 18 and the mirror image (for example, an image of the observer himself) reflected on the half mirror 19 at the same time.

  The server 30 is an example of an information processing apparatus that provides the mirror device 10 with a response corresponding to the request transmitted from the mirror device 10. The server 30 is, for example, a web server.

(1-1) Configuration of Mirror Device The configuration of the mirror device 10 will be described.

  As shown in FIG. 2, the mirror device 10 includes a storage device 11, a processor 12, an input / output interface 13, a communication interface 14, a camera 15, a speaker 16, a microphone 17, a display 18, and a half mirror. 19.

  The storage device 11 is configured to store a program and data. The storage device 11 is, for example, a combination of ROM (Read Only Memory), RAM (Random Access Memory), and storage (for example, flash memory or hard disk).

The programs include, for example, the following programs.
・ OS (Operating System) programs ・ Application programs that execute information processing

The data includes, for example, the following data.
-Database referenced in information processing-Data obtained by executing information processing (that is, information processing execution result)

  The processor 12 is configured to realize the function of the mirror device 10 by activating a program stored in the storage device 11. The processor 12 is an example of a computer.

The input / output interface 13 is configured to acquire an instruction from an observer from an input device connected to the mirror device 10 and output information to an output device connected to the mirror device 10.
The input device includes a camera 15, a microphone 17, and an input device (for example, a keyboard, a pointing device, a touch panel, or a combination thereof).
The output device includes a speaker 16 and a display 18.

  The communication interface 14 is configured to control communication between the mirror device 10 and the server 30.

The camera 15 is configured to capture an image.
The camera 15 is, for example, a COMMS (Complementary Metal Oxide Semiconductor) camera or an infrared camera.

  The speaker 16 is configured to output audio content created by the processor 12.

  The microphone 17 is configured to input sound (for example, observer's sound). The processor 12 analyzes the observer's voice input by the microphone 17 to identify the context of the observer's voice. The processor 12 treats the specified context as an observer's instruction. Thereby, the mirror device 10 shows the behavior according to the voice of the observer.

  Display 18 is configured to display an image generated by processor 12.

  The half mirror 19 is configured to capture a mirror image (for example, an image of the observer itself).

(1-2) Server Configuration The configuration of the server 30 will be described with reference to FIG.

  As illustrated in FIG. 1, the server 30 includes a storage device 31, a processor 32, an input / output interface 33, and a communication interface 34.

  The storage device 31 is configured to store a program and data. The storage device 31 is, for example, a combination of ROM, RAM, and storage (for example, flash memory or hard disk).

The programs include, for example, the following programs.
-OS program-Application program that executes information processing

The data includes, for example, the following data.
・ Database referenced in information processing ・ Results of information processing

  The processor 32 is configured to realize the function of the server 30 by starting a program stored in the storage device 31. The processor 32 is an example of a computer.

The input / output interface 33 is configured to acquire an instruction from the observer from an input device connected to the server 30 and output information to an output device connected to the server 30.
The input device is, for example, a keyboard, a pointing device, a touch panel, or a combination thereof.
The output device is, for example, a display.

  The communication interface 34 is configured to control communication between the server 30 and the mirror device 10.

(2) Overview of Embodiment An overview of this embodiment will be described. FIG. 3 is an explanatory diagram of the outline of the present embodiment.

  As shown in FIG. 3, this embodiment is applicable when the observer OBJ performs a cosmetic action (for example, makeup) while observing a mirror image reflected on the half mirror 19 of the mirror device 10.

When the observer OBJ puts on makeup and stands in front of the mirror device 10, a mirror image MI0 of the face of the observer OBJ before makeup appears on the half mirror 19.
When the observer OBJ starts makeup while looking at the half mirror 19, the mirror device 10 acquires an image IMG1 of the face of the observer OBJ during makeup. On the half mirror 19, a mirror image MI1 of the face of the observer OBJ during makeup is shown. The mirror image MI1 corresponds to the image IMG1. The mirror device 10 reproduces the sound SC1 (for example, the sound of the anthropomorphic character that “a brighter color is better”) corresponding to the image IMG1 (that is, the face of the observer OBJ being makeup).
When the observer OBJ finishes the makeup, the mirror device 10 acquires an image IMG2 of the face of the observer OBJ after the makeup. On the half mirror 19, a mirror image MI2 of the face of the observer OBJ after makeup is shown. The mirror image MI2 corresponds to the image IMG2. The mirror device 10 reproduces the sound SC2 (for example, the sound of the anthropomorphic character “Looks good”) corresponding to the image IMG2 (that is, the face of the observer OBJ after makeup).

  Thus, the mirror device 10 reproduces the audio content according to the state of the observer OBJ. As a result, the mirror device 10 behaves as an object person (so-called partner) who interacts with the voice for the observer OBJ who is performing the beauty. As a result, the observer OBJ can perform the beauty behavior while listening to the reaction by the voice of the anthropomorphic character. In this way, a user experience of new beauty behavior can be provided to the observer OBJ.

  In particular, the observer OBJ attaches importance to the observer OBJ by reproducing the sounds SC1 to SC2 which are auditory information in accordance with the image IMG1 which is visual information important for the observer OBJ observing the mirror image MI1. Output according to information can be obtained. This is because the observer OBJ can give an output corresponding to the visual information simply by giving the visual information that he / she attaches importance to without unconsciously giving a command to the mirror device 10. means.

(3) Database The database of this embodiment will be described. The following database is stored in the storage device 31.

(3-1) Observer information database The observer information database of this embodiment is demonstrated. FIG. 4 is a diagram illustrating a data structure of the observer information database according to the present embodiment.

As shown in FIG. 4, the observer information database stores observer information related to the observer.
The observer information database includes an “observer ID” field, a “password” field, an “observer name” field, and an “observer attribute” field. Each field is associated with each other.

  The “observer ID” field stores an observer ID for identifying the observer OBJ (an example of “observer identification information”).

  In the “password” field, a password for authentication when the observer OBJ logs in to the server 30 is stored. Information in the “password” field is arbitrarily determined by the observer OBJ when the observer information is registered in the server 30.

  Information related to the name of the observer OBJ is stored in the “observer name” field. Information in the “observer name” field is arbitrarily determined by the observer OBJ when the observer information is registered in the server 30.

  In the “observer attribute” field, observer attribute information regarding the attribute of the observer is stored. Information in the “observer attribute” field is arbitrarily determined by the observer OBJ when the observer information is registered in the server 30. The “observer attribute” field includes a “gender” field, an “age” field, and an “occupation” field.

  In the “gender” field, information related to the gender of the observer is stored.

  Information relating to the age of the observer is stored in the “age” field.

  In the “occupation” field, information on the occupation of the observer is stored.

(3-2) Character Information Master The character information master of this embodiment will be described. FIG. 5 is a diagram showing the data structure of the character information master of this embodiment.

The character information related to the anthropomorphic character is stored in the character information master of FIG. The “personified character” is a person who is personified in the computer space for the user of the mirror device 10 and can communicate with the user by voice.
The character information master includes a “character ID” field, a “character parameter” field, and a “reaction” field.

  The “character ID” field stores a character ID for identifying anthropomorphic characters (an example of “character identification information”).

The “character parameter” field stores character parameters set for the personified character.
The “character parameter” field includes a “sex” field, an “age” field, an “occupation” field, and a “personality” field.

  In the “gender” field, information on the gender of the personified character is stored.

  Information related to the age of the anthropomorphic character is stored in the “age” field.

  In the “occupation” field, information on the occupation of the anthropomorphic character is stored.

  The “character” field stores information related to the character of the anthropomorphic character.

(3-3) Pattern Matching Table The pattern matching table of this embodiment will be described. FIG. 6 is a diagram showing the data structure of the pattern matching table of this embodiment.

As shown in FIG. 6, the pattern matching table stores pattern matching information related to the relationship between the content and the image feature amount pattern (hereinafter referred to as “image pattern”).
The pattern matching table includes a “pattern file” field, a “status” field, and a “reaction file” field. Each field is associated with each other.
The pattern matching table is associated with the character ID.

The “pattern file” field stores an image pattern file in which an image pattern is described. The image pattern is, for example, one of the following.
・ Image pattern of makeup ・ Image pattern of facial expression ・ Image pattern of skin condition ・ Image pattern of face shape ・ Image pattern of personality ・ Image pattern of clothes ・ Image pattern of body shape

In the “status” field, a status code for identifying the status of the observer OBJ determined based on the feature amount of the observer image is stored. The status code indicates, for example, at least one of the following states.
・ Makeup condition (for example, makeup color, frequently used makeup items)
・ Expression status (for example, emotions, surprises, frustration, and memorization)
-Skin condition (for example, skin transparency, skin health, skin brightness, skin fineness, skin firmness, acne condition, wrinkle condition, wrinkle condition, and pore condition)
-Face state (for example, two-dimensional state (specifically, contour, eye shape, nose shape, mouth shape, eyelid shape, and eyebrow shape), and three-dimensional state (specifically The state of the conical line, the state of slack, and the state of the E line))
・ Personality / clothing conditions (As an example, the combination of clothes with the observer OBJ, and the state of wrinkles)
・ Body shape (with body as an example)

  A content file is stored in the “reaction file” field. The “reaction file” field includes a “voice” field, an “AR” field, and a “text” field.

  In the “voice” field, a voice reaction file is stored. In the voice reaction file, a voice reaction file in which voice information related to the voice reaction of the anthropomorphic character is described.

  In the “AR” field, an AR reaction file is stored. In the AR reaction file, AR information related to the virtual reality (Augmented Reality) reaction of the anthropomorphic character is described.

  A text reaction file is stored in the “text” field. In the text reaction file, a text reaction file is described in which information that is the basis of the reaction by the text of the personified character is described.

(3-4) Log Information Database The log information database of this embodiment will be described. FIG. 7 is a diagram illustrating a data structure of the log information database according to the present embodiment.

As shown in FIG. 7, log information collected by the mirror device 10 every time the observer OBJ uses the mirror device 10 is stored in the log information database.
The log information database includes a “log ID” field, a “date / time” field, an “input” field, a “status” field, and an “output” field. Each field is associated with each other.
The log information database is associated with the observer ID.

  The “log ID” field stores a log ID for identifying log information (an example of “log identification information”).

  Information relating to the date and time of the log information is stored in the “date and time” field.

  Information acquired by the mirror device 10 is stored in the “log” field. The “log” field includes an “input image” field and an “analysis result” field.

  In the “input image” field, an observer image of an image acquired by the mirror device 10 is stored.

  The “analysis result” field stores a file (hereinafter referred to as “analysis result file”) indicating the result of image analysis on the observer image. The feature value of the observer image is described in the analysis result file.

  In the “status” field, a status code for identifying the status of the observer OBJ corresponding to the analysis result file is stored.

  In the “output” field, reaction information relating to the reaction corresponding to the information in the “input” field is stored. The “output” field includes a “character ID” field and an “output content” field.

  In the “character ID” field, the character ID of the anthropomorphic character showing the reaction corresponding to the information in the “input” field is stored.

  The “output content” field stores a reaction file in which reaction information related to the reaction corresponding to the information in the “input” field is described.

(4) Information processing The information processing of this embodiment is demonstrated. FIG. 8 is a sequence diagram of information processing according to the present embodiment. 9 to 11 are diagrams showing output examples in the information processing of FIG. FIG. 12 is a detailed flowchart of the reaction generation shown in FIG.

As shown in FIG. 8, the mirror device 10 receives a character designation (S100).
Specifically, the processor 12 displays a screen P100 (FIG. 9) on the display 18. The screen P100 is displayed superimposed on a mirror image MI0 (for example, an image of the face of the observer OBJ before starting makeup) reflected on the half mirror 19.

The screen P100 includes a display object A100, a field object F100, and an operation object B100.
The display object A100 displays information related to the current date and time.
The field object F100 is an object for receiving designation of at least one personified character from among a plurality of personified characters. In the field object F100, character information (for example, character name, sex, age, and occupation) of a plurality of anthropomorphic characters is displayed. Each character information is assigned a character ID.
The operation object B100 is an object for receiving an instruction for transmitting the character ID assigned to the character information designated in the field object F100 to the server 30.

  When the observer OBJ designates at least one of the character information displayed on the field object F100 and operates the operation object B100, the processor 12 assigns the character ID assigned to the character information designated by the observer OBJ. And the observer ID of the observer OBJ are transmitted to the server 30.

After step S100, the server 30 executes character determination (S300).
Specifically, the processor 32 determines the character ID transmitted from the mirror device 10 in step S100 as the character ID of the character that should indicate a reaction to the observer OBJ.
The processor 32 transmits a notification indicating that the character ID has been determined to the mirror device 10.

After step S300, the mirror device 10 executes image acquisition (S101).
Specifically, the processor 12 activates the camera 15.
When the camera 15 captures an image of the face of the observer OBJ (hereinafter referred to as “observer image”), the processor 12 stores the observer image captured by the camera 15 in the storage device 11. The observer image corresponds to the mirror image MI0 observed by the observer OBJ. For example, the camera 15 is disposed at a position where an image corresponding to the mirror image MI0 can be captured.

After step S101, the mirror device 10 executes a reaction request (S102).
Specifically, the processor 12 transmits reaction request data to the server 30. The reaction request data includes the following information.
The observer image stored in the storage device 11 in step S101 is included.
-Information about the execution date and time of step S102

  After step S102, the server 30 executes reaction generation (S301).

As shown in FIG. 12, the server 30 performs image analysis (S3010).
Specifically, the processor 32 specifies the image pattern of the observer image based on the feature amount of the observer image included in the reaction request data.

After step S3010, the server 30 executes pattern matching (S3011).
Specifically, the processor 32 refers to the pattern matching table (FIG. 6) associated with the character ID determined in step S300 and describes an image pattern that is the same as or similar to the image pattern of the observer image. Specify the image pattern file. The “similar image pattern” is, for example, an image pattern in which the difference in feature amount from the image pattern of the observer image is a predetermined value or less.

After step S3012, the server 30 executes state determination (S3012).
Specifically, the storage device 31 stores a state determination model. The input and output of the state determination model are as follows.
・ Input: Image pattern ・ Output: Observer state

  The processor 32 inputs the image pattern file specified in step S3011 to the state determination model stored in the storage device 31, thereby determining a state code for identifying the state of the observer OBJ corresponding to the image pattern file. .

After step S3012, the server 30 executes voice reaction generation (S3013).
Specifically, the processor 32 refers to the voice reaction file in the “voice” field associated with the status code determined in step S3012 and generates voice reaction data to be output as voice reaction.

After step S3013, the server 30 executes generation of an AR reaction (S3014).
Specifically, the processor 32 refers to the AR reaction file in the “AR” field associated with the status code determined in step S3012 and generates AR reaction data to be output as the AR reaction.

After step S3014, the server 30 executes text reaction generation (S3015).
Specifically, the processor 32 refers to the text reaction file in the “text” field associated with the status code determined in step S3012 and generates text reaction data to be output as the text reaction.

After step S3015, the server 30 executes database update (S302).
Specifically, the processor 32 specifies the log information database (FIG. 7) associated with the observer ID transmitted from the mirror device 10 in step S100.
The processor 32 adds a new record to the specified log information database. The following information is stored in each field of the new record.
In the “log ID” field, a new log ID arbitrarily determined by the processor 32 is stored.
In the “date and time” field, information related to the execution date and time of step S102 included in the reaction request data is stored.
In the “input image” field, an observer image included in the reaction request data is stored.
In the “analysis result” field, the image pattern specified in step S3010 is stored.
In the “status” field, the status code determined in step S3012 is stored.
In the “character ID” field, the character ID determined in S300 is stored.
The “reaction” field stores the reaction data (voice reaction data, AR reaction data, and text reaction data) generated in steps S3013 to S3015.

After step S302, the server 30 executes a reaction response (S303).
Specifically, the processor 32 transmits reaction response data to the mirror device 10. The reaction response data includes the reaction data generated in steps S3013 to S3015.

After step S303, the mirror device 10 executes a reaction (S103).
Specifically, the processor 12 displays a screen P101 (FIG. 10) on the display 18. The screen P101 is displayed so as to be superimposed on a mirror image MI1 (for example, an image of the face of the observer OBJ being makeup) reflected on the half mirror 19.

The screen P101 includes a display object A100.
The display object A100 displays information related to the current date and time.

  The processor 12 outputs from the speaker 16 a sound based on the sound reaction data included in the reaction response data (for example, a sound “brighter color is better”). This sound is generated based on the image acquired in step S101. Thereby, the observer OBJ can obtain advice and guidance according to the state of his / her face during makeup through the voice of the anthropomorphic character.

Steps S101 to S103 and S301 to S303 are continuously executed until the observer OBJ gives an instruction to end the information processing (for example, a voice utterance meaning to end the information processing).
Therefore, whenever the makeup of the observer OBJ proceeds, in step S103, the reaction of the anthropomorphic character is presented to the observer OBJ.

  For example, the processor 12 displays the screen P102 (FIG. 11) on the display 18. The screen P102 is displayed superimposed on a mirror image MI2 (for example, an image of the face of the observer OBJ after makeup has advanced with respect to FIG. 10) reflected on the half mirror 19.

  The screen P102 includes display objects A100 and A102, and an image object IMG102.

In the display object A102, text corresponding to the text reaction data generated in step S3015 is displayed.
The image object IMG102 is an image (for example, a makeup navigation image that proposes a makeup method) based on the AR reaction data assigned in step S3014.

  The processor 12 outputs from the speaker 16 a sound based on the sound reaction data included in the reaction response data (for example, a sound “brighter color is better”). This sound is generated based on the image acquired in step S101.

Thereby, the observer OBJ can experience the reaction of the anthropomorphic character with respect to his / her face after the makeup progresses via the voice, the AR image, and the text.
When the image pattern is an image pattern of clothes, the reaction of the anthropomorphic character to the clothes worn by the observer OBJ can be experienced through voice, AR images, and text.
When the image pattern is a body type image pattern, the reaction of the anthropomorphic character to the body shape of the observer OBJ can be experienced through voice, AR image, and text.
When the image pattern is a facial expression image pattern, the anthropomorphic character's reaction to the facial expression of the observer OBJ can be experienced via voice, AR images, and text.

  According to the present embodiment, the server 30 controls the mirror device 10 to behave as anthropomorphic characters corresponding to partners who provide opinions and advice to the observer OBJ. For example, the server 30 presents a reaction including advice and guidance for making the observer OBJ more beautiful to the observer OBJ. That is, the anthropomorphic character is a virtual partner who thinks that the observer OBJ is more attractive to the observer OBJ than anyone else, or a virtual person who can draw out the charm of the observer OBj more than anyone else. Act as a good partner. Thereby, the observer OBJ can experience at least one of beauty behaviors (for example, makeup, clothing coordination, body shape management, and body composition management) while experiencing an objective reaction regarding her beauty from such a partner. Two actions).

  In particular, in this embodiment, it is preferable not to display an image of anthropomorphic characters. By realizing communication between the anthropomorphic character and the observer OBJ by voice or text, the reaction of the anthropomorphic character is strongly impressed by the observer OBJ. As a result, the observer OBJ feels familiar with the personified character.

(5) Modification A modification of the present embodiment will be described.

(5-1) Modification 1
Modification 1 will be described. Modification 1 is an example in which audio content corresponding to n (n ≧ 2) characters selected from a plurality of characters is output. FIG. 13 is a diagram illustrating an output example in the information processing according to the first modification.

  In the first modification, when the observer OBJ designates a plurality of pieces of character information from the character information displayed on the field object F100 in step S100 (FIG. 8), the processor 12 determines the character designated by the observer OBJ. The plurality of character IDs assigned to the information and the observer ID of the observer OBJ are transmitted to the server 30.

  In step S300 (FIG. 8), the processor 32 determines the plurality of character IDs transmitted from the mirror device 10 in step S100 as character IDs of characters that should indicate a reaction to the observer OBJ.

  In step S3011 (FIG. 12), the processor 32 refers to the pattern matching table (FIG. 6) associated with the plurality of character IDs determined in step S300, and describes an image pattern similar to the image pattern of the observer image. The obtained image pattern file is specified for each character ID.

  In steps S3013 to S3015 (FIG. 12), the processor 32 generates voice reaction data, AR reaction data, and text reaction data for each character ID specified in step S3011.

  In step S103 (FIG. 8), the processor 12 displays a screen P110 (FIG. 13) on the display 18. The screen P110 is displayed so as to be superimposed on a mirror image MI1 (for example, an image of the face of the observer OBJ being makeup) reflected on the half mirror 19.

  The processor 12 performs a voice based on a plurality of voice reaction data included in the reaction response data (for example, a voice of character 1 (character ID: CHA001) “a brighter color is better” and a character 2 (character ID: CHA002). From the speaker 16 is output from the speaker 16.

  According to the first modification, the observer OBJ can experience a reaction in which a plurality of anthropomorphic characters are having a conversation with respect to the mirror image MI1 of his / her face during makeup.

(5-2) Modification 2
Modification 2 will be described. Modification 2 is an example in which a reaction to an image of the first observer OBJ1 is presented with reference to information of a second observer OBJ2 similar to the first observer OBJ1. FIG. 14 is a detailed flowchart of reaction generation according to the second modification.

As shown in FIG. 14, after steps S3010 to S3012 (FIG. 12), the server 30 executes reference observer information reference (S3220).
Specifically, the processor 32 refers to the observer information database (FIG. 4) and is associated with the observer ID (hereinafter referred to as “first observer ID”) transmitted from the mirror device 10 in step S100. Specify observer attribute information.
The processor 32 refers to the “observer attribute” field of the observer information database, and at least one observer ID (hereinafter referred to as “second”) associated with the same or similar observer attribute information as the first observer attribute information. "Observer ID"). “Similar observer attribute information” is, for example, observer attribute information whose difference from the observer attribute information associated with the first observer ID is equal to or less than a predetermined value.
The processor 32 refers to the log information database (FIG. 7) associated with the specified second observer ID, and records corresponding to the predetermined condition (for example, information in the “date / time” field from the execution date / time of step S3220). Identify records included within a certain period of time.
The processor 32 specifies the information of the record in which the character ID determined in step S300 is stored in the “character ID” field among the specified records as the reference observer information.

After step S3220, the server 30 executes voice reaction generation (S3221).
Specifically, the processor 32 includes an audio reaction file in the “audio” field associated with the status code determined in step S3012 and information in the “reaction” field of the reference observer information specified in step S3220. Referring to, voice reaction data to be output as voice reaction is generated.

After step S3221, the server 30 executes generation of an AR reaction (S3222).
Specifically, the processor 32 receives the AR reaction file of the “AR” field associated with the status code determined in step S3012 and the information of the “reaction” field of the reference observer information specified in step S3220. Referring to this, AR reaction data to be output as an AR reaction is generated.

After step S3222, the server 30 executes text reaction generation (S3223).
Specifically, the processor 32 obtains the text reaction file of the “text” field associated with the status code determined in step S3012 and the information of the “reaction” field of the reference observer information specified in step S3220. Reference is made to generate text reaction data to be output as a text reaction.

After step S3223, the server 30 executes steps S302 to S303 (FIG. 8).
After step S303, the mirror device 10 executes step S103.

  According to the second modification, the response data is generated with reference to the log information of the second observer that is the same as or similar to the attribute of the first observer. Thereby, the first observer can experience various reactions of the personified character.

(5-3) Modification 3
Modification 3 will be described. Modification 3 is an example in which a reaction corresponding to the body composition of the observer OBJ is presented.

  In the “pattern file” field of the pattern matching table (FIG. 6) of the third modification, a body composition pattern file is stored in addition to the image pattern file. The body composition pattern file describes the body composition pattern related to the body composition.

In step S102 (FIG. 8), the processor 12 acquires body composition information related to the body composition (for example, at least one of body weight and body fat) of the observer OBJ from a body composition measuring apparatus (not shown).
The processor 12 further includes body composition information in the reaction request data.

  In step S3011 (FIG. 12), the processor 32 specifies a body composition pattern file in which body composition information that is the same as or similar to the body composition information included in the reaction request data is described. The “similar body composition information” is, for example, body composition information whose difference from the body composition information included in the reaction request data is equal to or less than a predetermined value.

The inputs and outputs of the state determination model stored in the storage device 31 are as follows.
・ Input: Image pattern and body composition ・ Output: Observer status

  The processor 32 inputs the image pattern file and the body composition pattern file specified in step S3011 to the state determination model stored in the storage device 31, thereby identifying the state of the observer OBJ corresponding to the pattern file Determine the code.

  In step S3013 (FIG. 12), the processor 32 refers to the audio reaction file in the “audio” field associated with the status code determined in step S3012 and generates audio reaction data to be output as an audio reaction.

  In step S3014 (FIG. 12), the processor 32 refers to the AR reaction file in the “AR” field associated with the status code determined in step S3012 and generates AR reaction data to be output as an AR reaction.

  In step S3015 (FIG. 12), the processor 32 refers to the text reaction file in the “text” field associated with the status code determined in step S3012 and generates text reaction data to be output as a text reaction.

  According to the modification 3, the reaction according to both the image and body composition of the observer OBJ is presented. Thereby, more various reactions can be presented to the observer OBJ.

(5-4) Modification 4
Modification 4 will be described. Modification 4 is an example in which a reaction corresponding to the change of the observer OBJ is presented.

In step S3013 (FIG. 12), the processor 32 of the modified example 4 compares the information in the “analysis result” field (that is, the analysis result file related to the past analysis result) with the analysis result in step S3010. A past analysis result (for example, evaluation of past makeup) and a current analysis result (for example, evaluation of current makeup) are compared.
The processor 32 generates voice reaction data, AR reaction data, and text reaction data in a context (eg, “I got better than yesterday”) based on the result of the comparison.

  According to the modification 4, the reaction according to the change of the observer OBJ is presented. As a result, the observer OBJ can objectively know his / her changes.

(5-5) Modification 5
Modification 5 will be described. Modification 5 is an example in which a reaction corresponding to the comparison result between the observer OBJ and the statistical value is presented.

Statistical information is stored in the storage device 31 of the fifth modification. The statistical information includes at least one of the following.
・ Statistical information about the beauty of makeup ・ Statistical information about the beauty of clothes ・ Statistical information about the beauty of body shape ・ Statistical information about the beauty of facial expressions

In step S3013 (FIG. 12), the processor 32 compares the analysis result of step S3010 with the statistical information stored in the storage device 31, thereby analyzing the analysis result (for example, the current makeup evaluation) and the statistical information. The difference between the information (for example, statistical information on the beauty of makeup) is calculated.
The processor 32 generates voice reaction data, AR reaction data, and text reaction data in a context (for example, “I feel refreshed more than others”) based on the result of the calculation.

  According to the modification 5, the reaction according to the change of the observer OBJ is presented. Thus, the observer OBJ can objectively know the difference between the state of the observer and the statistical result. As a result, the willingness of the observer OBJ to use the mirror device 10 is promoted.

(5-6) Modification 6
Modification 6 will be described. Modification 6 is an example in which a reaction corresponding to the sound of the observer OBJ is presented.

  A sound pattern file is stored in the storage device 31 of the sixth modification. The voice pattern file describes a voice pattern related to the relationship between the voice context and the frequency.

In step S <b> 102 (FIG. 8), the microphone 17 obtains sound information related to the sound of the observer OBJ (for example, sound “What should I do?”).
The processor 12 further includes audio information in the reaction request data.

  In step S <b> 3011 (FIG. 12), the processor 32 refers to the voice pattern file stored in the storage device 31 and interprets the meaning of the voice information included in the reaction request data.

  In step S3013 (FIG. 12), the processor 32 has a context corresponding to the analysis result in step S3010 and the determination result in step S3012 (for example, “It is better to use the cosmetics used yesterday”). Voice reaction data, AR reaction data, and text reaction data are generated.

  According to the modification 6, the reaction according to both the image and sound of the observer OBJ is presented. Thereby, the observer OBJ can use the mirror device 10 as if he / she is talking to the personified character.

(5-7) Modification 7
Modification 7 will be described. The modified example 7 is an example in which the voice of the anthropomorphic character is provided based on the tendency of the state of the observer OBJ specified from the image. FIG. 15 is an explanatory diagram of an outline of the modified example 7.

  As shown in FIG. 15, when the observer OBJ applies bright makeup while looking at the mirror image MI0, the mirror device 10 acquires a face image IMG3 on which bright makeup has been applied. On the half mirror 19, a mirror image MI3 of the face of the observer OBJ is shown. The mirror image MI3 corresponds to the image IMG3.

The server 30 stores a preference model related to the preference of the observer OBJ. The preference of the observer OBJ means the tendency of the observer OBJ specified from the face image (that is, the visual tendency of the face of the observer OBJ).
The server 30 updates the preference model based on the information in the “output” field of the log information database (FIG. 7). As a result, the preference model describes the visual preference of the face of the observer OBJ.
The tendency described in the preference model includes, for example, the following.
-Frequent use of light-colored cheeks.
・ On Monday, there is a high frequency of dark expressions.

The input and output of the preference model are, for example, as follows.
・ Input: Status code ・ Output: Voice reaction data

  The server 30 inputs the status code determined in step S3012 in the preference model instead of the voice reaction file (FIG. 6) in the creation of the voice reaction (step S3013 in FIG. 12 or step S3221 in FIG. 14). Generate voice reaction data.

  Thus, the mirror device 10 reproduces the audio content corresponding to the tendency of the state of the observer OBJ. Thereby, the observer OBJ can feel the anthropomorphic character as a partner who can view himself more objectively than himself. In this way, a user experience of new beauty behavior can be provided to the observer OBJ.

The preference model of the modified example 7 can be applied not only to the generation of the voice reaction but also to the following processing.
Generation of AR reaction (step S3014 in FIG. 12 and step S3222 in FIG. 14)
Generation of text reaction (step S3015 in FIG. 12 and step S3223 in FIG. 14)

(5-8) Modification 8
Modification 8 will be described. The modification 8 is an example in which the voice of the anthropomorphic character is provided based on the reaction tendency of the observer OBJ according to the voice of the anthropomorphic character. FIG. 16 is an explanatory diagram of an outline of Modification 8.

As shown in FIG. 16, when the observer OBJ views the half mirror 19, the mirror device 10 reads the image IMGm of the face of the observer OBJ being makeup (m is at least one of integers 1 to n−1). To get. On the half mirror 19, a mirror image MIm of the face of the observer OBJ is shown. The mirror image MIm corresponds to the image IMGm. The mirror device 10 reproduces a sound SCm (for example, a sound of anthropomorphic character “It's a good smile”) corresponding to the image IMGm.
When the observer OBJ does not change the facial expression even when the voice SCm is reproduced (that is, when the observer OBJ shows a reaction contrary to the voice of the anthropomorphic character), the mirror device 10 reacts against the voice of the anthropomorphic character. The voice SCn (for example, “more laughing”) corresponding to the image IMGn (n = m + 1) of the face of the observer OBJ that indicates The voice of the anthropomorphic character “Mite” is reproduced.

Specifically, the server 30 stores a preference model related to the preference of the observer OBJ. The preference of the observer OBJ means a tendency of reaction indicated by the observer OBJ according to the voice of the personified character.
The server 30 updates the preference model based on the relationship between the information in the “input” field of the mth record in the log information database (FIG. 7) and the information in the “output” field of the nth record. As a result, the preference of the observer OBJ determined by whether or not the observer OBJ has accepted the context indicated by the voice of the anthropomorphic character is described in the preference model.
The tendency described in the preference model includes, for example, the following.
・ Even if it is recommended to use a light-colored cheek, a dark-colored cheek is often used.
・ Even on Mondays, even if the voice is lightly praised, there is little change in the facial expression of the observer OBJ.

The input and output of the preference model are, for example, as follows.
・ Input: Status code ・ Output: Voice reaction data

  The server 30 inputs the status code determined in step S3012 in the preference model instead of the voice reaction file (FIG. 6) in the creation of the voice reaction (step S3013 in FIG. 12 or step S3221 in FIG. 14). Generate voice reaction data.

  As described above, the mirror device 10 reproduces the audio content based on the reaction of the observer OBJ according to the sound of the personified character. Thereby, the observer OBJ can feel the anthropomorphic character as a partner who can view himself more objectively than himself. In this way, a user experience of new beauty behavior can be provided to the observer OBJ.

The preference model of the modified example 8 can be applied not only to the generation of voice reaction but also to the following processing.
Generation of AR reaction (step S3014 in FIG. 12 and step S3222 in FIG. 14)
Generation of text reaction (step S3015 in FIG. 12 and step S3223 in FIG. 14)

(5-9) Modification 9
Modification 9 will be described. The modification 9 is an example in which the voice of the personified character corresponding to the combination of the personified character and the observer OBJ is generated.

  The pattern matching table (FIG. 6) of Modification 9 is associated with a combination of character parameters (FIG. 5) and observer attribute information (FIG. 4).

In step S3013 (FIG. 12) or S3221 (FIG. 14), the processor 32 associates the character parameter associated with the character ID of the anthropomorphic character determined in step S300 and the observer ID transmitted in step S100. A pattern matching table associated with the combination of the observed observer attribute information is specified.
The processor 32 refers to the identified pattern matching table and identifies the voice reaction file in the “voice” field associated with the status code determined in step S3012.
The processor 32 refers to the specified audio reaction file and generates audio reaction data to be output as an audio reaction.

  According to the modification 9, it is possible to provide the voice of the anthropomorphic character according to the relationship (for example, compatibility) between the parameter of the anthropomorphic character and the observer attribute of the observer OBJ. As a result, the observer OBJ can experience more diverse sounds of the anthropomorphic character.

(5-10) Modification 10
Modification 10 will be described. The modification 10 is an example in which the voice of the anthropomorphic character corresponding to the character of the anthropomorphic character and the personality of the observer OBJ is generated.

  The pattern matching table (FIG. 6) of the modified example 9 is associated with the combination of information in the “personality” field (FIG. 5) and the personality of the observer.

The storage device 31 stores a personality determination model. The input / output of the personality determination model is as follows.
・ Input: Image features (for example, the arrangement of each part of the face)
・ Output: person's personality

  After determining the state of the observer OBJ in S3012, the processor 32 inputs the image pattern file specified in step S3011 to the personality determination model stored in the storage device 31, thereby corresponding to the image pattern file. The character of the observer OBJ is determined.

In step S3013 (FIG. 12) or S3221 (FIG. 14), the processor 32 specifies a pattern matching table associated with the combination of the information related to the determined personality and the personalized character.
The processor 32 refers to the identified pattern matching table and identifies the voice reaction file in the “voice” field associated with the status code determined in step S3012.
The processor 32 refers to the specified audio reaction file and generates audio reaction data to be output as an audio reaction.

  According to the tenth modification, the voice of the anthropomorphic character according to the combination (for example, compatibility) of the character of the anthropomorphic character and the personality of the observer OBJ can be provided. As a result, the observer OBJ can experience more diverse sounds of the anthropomorphic character.

(6) Summary of this embodiment This embodiment is summarized.

The first aspect of this embodiment is
An information processing apparatus (for example, server 30) connected to a mirror device 10 that acquires an observer image corresponding to a mirror image of the observer and that can output sound,
Means for acquiring an observer image from the mirror device 10 (for example, the processor 32 for executing the process of step S301);
Means for determining the state of the observer reflected in the mirror image based on the observer image (for example, the processor 32 for executing the process of step S3012);
Based on the result of the determination, a means for generating an anthropomorphic character voice (for example, the processor 32 that executes the process of step S3013) is provided.
Means for transmitting the generated sound to the mirror device 10 (for example, the processor 32 that executes the process of step S303);
Information processing device.

According to the first aspect, the voice of the anthropomorphic character corresponding to the state of the observer determined based on the observer image is provided to the observer via the mirror device 10. Thereby, the interaction in which visual information is reflected can be provided to the observer.
In particular, the observer can have an experience that the mirror device 10 speaks according to his / her state. As a result, the observer feels familiarity with the mirror device 10. This gives the observer a motivation to use the mirror device 10, so that the observer frequently uses the mirror device.

The second aspect of this embodiment is
By means of analyzing the feature amount of the observer image, a means for specifying the image pattern of the observer image (for example, the processor 32 that executes the process of step S3010) is provided.
The means for determining determines the state of the observer based on the specified image pattern.
Information processing apparatus.

  According to the second aspect, the voice of the anthropomorphic character corresponding to the state of the observer determined based on the image pattern of the observer image is provided to the observer via the mirror device 10. Thereby, the interaction in which visual information is reflected can be provided to the observer.

The third aspect of this embodiment is
Means (for example, the observer information database in FIG. 4 and the log in FIG. 7) for associating and storing the observer identification information for identifying the observer, the observer attribute information regarding the attributes of the observer, and the specified image pattern Information database)
The means for specifying includes means for specifying second observer identification information associated with observer attribute information that is the same as or similar to the observer attribute information of the first observer,
The means for determining is based on the first image pattern specified from the observer image of the first observer and the second image pattern associated with the second observer identification information. Determine the state of the person,
Information processing apparatus.

  According to the third aspect, when determining the state of the first observer who should provide the voice of the anthropomorphic character, not only the first image pattern of the first observer but also the attribute is similar to the first observer. Reference is made to the second image pattern of the second observer. Thereby, the accuracy of the determination of the state of the first observer can be improved. In this way, it is possible to provide an interaction that is more suitable for the actual state of the first observer.

The fourth aspect of this embodiment is
The means for determining determines a change in the state of the observer.
Information processing apparatus.

  According to the fourth aspect, the voice of the anthropomorphic character corresponding to the change in the state of the observer is provided to the observer. As a result, the observer OBJ can objectively know his / her changes.

The fifth aspect of this embodiment is
The means for determining determines the evaluation of makeup applied to the face of the observer.
Information processing apparatus.

  According to the fifth aspect, the voice of the anthropomorphic character corresponding to the state of the face of the observer who applied makeup is provided to the observer. Thereby, the observer can perform makeup while interacting with the voice of the anthropomorphic character. In this way, a completely new user experience is born in the cosmetic behavior of makeup. As a result, it is expected that the beauty awareness of the observer will be improved.

The sixth aspect of this embodiment is
The means for determining determines the evaluation of the clothes worn by the observer.
Information processing apparatus.

According to the sixth aspect,
The voice of the anthropomorphic character corresponding to the state of the observer wearing the clothes is provided to the observer. Thus, the observer can select clothes while interacting with the voice of the anthropomorphic character. In this way, a completely new user experience is born in the behavior of clothing selection. As a result, it is expected that the viewer's awareness of clothes will also be improved.

The seventh aspect of this embodiment is
The means for determining determines the evaluation of the observer's body shape,
Information processing apparatus.

  According to the seventh aspect, the voice of the anthropomorphic character corresponding to the body shape of the observer is provided to the observer. Thereby, the observer can confirm and select the figure while interacting with the voice of the anthropomorphic character. In this way, a completely new user experience is born in the behavior of body shape management. As a result, it is expected that the observer's awareness of the experience will be improved.

The eighth aspect is
The means for determining determines the evaluation of the observer's facial expression.
Information processing apparatus.

  According to the eighth aspect, the voice of the anthropomorphic character corresponding to the facial expression of the observer is provided to the observer. Thus, the observer can observe the facial expression while interacting with the voice of the anthropomorphic character. In this way, a completely new user experience is born in the action of confirming facial expressions. As a result, it is expected that the consciousness of the observer with respect to facial expressions will be improved.

The ninth aspect of this embodiment is
Means for receiving an instruction to specify at least one anthropomorphic character from among a plurality of anthropomorphic characters (for example, the processor 32 that executes the process of step S300);
The generating means outputs the voice of the anthropomorphic character designated by the received instruction.
Information processing apparatus.

  According to the ninth aspect, the sound of any anthropomorphic character of the observer is provided to the observer. As a result, the observer learns a greater sense of affinity in the mirror device 10.

The tenth aspect of this embodiment is
When the receiving means receives an instruction to specify a plurality of anthropomorphic characters, the generating means generates the sound of the plurality of anthropomorphic characters.
Information processing apparatus.

  According to the tenth aspect, sounds of a plurality of anthropomorphic characters are provided to the observer. Thus, the observer can observe the mirror image with a sense of communicating with a plurality of anthropomorphic characters.

The eleventh aspect of this embodiment is
Means for obtaining the voice of the observer from the mirror device,
The means for generating outputs the voice of the anthropomorphic character according to the result of the determination and the voice of the observer.
Information processing apparatus.

  According to the eleventh aspect, the voice of the anthropomorphic character corresponding to both the observer image and the voice of the observer is provided to the observer. As a result, the observer can observe the mirror image with a sense closer to real communication.

The twelfth aspect of this embodiment is
The means for generating generates an anthropomorphic character voice for presenting advice or guidance according to the result of the determination.
Information processing apparatus.

  According to the twelfth aspect, it is possible to prompt the observer OBJ who uses the mirror device 10 to take out an attraction.

The thirteenth aspect of this embodiment is
The means for generating refers to the preference model describing the tendency of the state of the observer, and generates the voice of the anthropomorphic character.
Information processing apparatus.

  According to the 13th aspect, the audio | voice content according to the tendency of the state of the observer OBJ is provided. Thereby, the observer OBJ can feel the anthropomorphic character as a partner who can view himself more objectively than himself. In this way, a user experience of new beauty behavior can be provided to the observer OBJ.

The means for generating refers to the preference model in which the tendency of the reaction shown by the observer to the voice of the personified character is described, and generates the voice of the personified character.
Information processing apparatus.

  According to the 14th aspect, the audio | voice content based on the reaction of the observer OBJ according to the audio | voice of anthropomorphic character is provided. Thereby, the observer OBJ can feel the anthropomorphic character as a partner who can view himself more objectively than himself. In this way, a user experience of new beauty behavior can be provided to the observer OBJ.

The fifteenth aspect of this embodiment is
The generating means generates the voice of the anthropomorphic character based on the parameter of the anthropomorphic character and the combination of the observer attribute information regarding the attribute of the observer and the result of the determination.
Information processing apparatus.

  According to the fifteenth aspect, it is possible to provide the voice of the anthropomorphic character according to the relationship (for example, compatibility) between the parameter of the anthropomorphic character and the observer attribute of the observer OBJ. As a result, the observer OBJ can experience more diverse sounds of the anthropomorphic character.

The sixteenth aspect of this embodiment is
Means for determining the personality of the observer according to the observer image with reference to the personality determination model having the feature quantity of the image as an input and the personality as an output;
The means for generating generates the voice of the anthropomorphic character based on the information on the character of the anthropomorphic character, the combination of the determined personality of the observer, and the state of the determined observer.
Information processing apparatus.

  According to the sixteenth aspect, it is possible to provide the voice of the anthropomorphic character according to the combination (for example, compatibility) of the character of the anthropomorphic character and the personality of the observer OBJ. As a result, the observer OBJ can experience more diverse sounds of the anthropomorphic character.

The seventeenth aspect of this embodiment is
A mirror (for example, a half mirror 19) that mirrors the observer's mirror image;
A camera 15 for capturing an observer image corresponding to the mirror image reflected on the mirror;
A speaker 16 that outputs the sound of the anthropomorphic character corresponding to the state of the observer determined based on the observer image captured by the camera 15;
This is a mirror device 10.

  According to the seventeenth aspect, the sound of the anthropomorphic character corresponding to the state of the observer is transmitted to the mirror device 10. Thereby, the interaction in which visual information is reflected can be provided to the observer.

The eighteenth aspect of this embodiment is
This is a program for causing a computer (for example, the processor 12 or 32) to function as each of the above means.

(7) Other Modifications Other modifications will be described.

  The storage device 11 may be connected to the mirror device 10 via the network NW. The storage device 31 may be connected to the server 30 via the network NW.

Each step of the above information processing can be executed by either the mirror device 10 or the server 30.
For example, the information processing performed by the server 30 may be performed by the mirror device 10. In this case, the storage device 11 and the processor 12 of the mirror device 10 replace the functions of the storage device 31 and the processor 32 of the server 30.

In the configuration of the mirror device 10 (FIG. 2), the microphone 17 and the display 18 can be omitted. In this case, the following points are different from the present embodiment.
The half mirror 19 is replaced by a mirror that totally reflects light.
Steps S3014 to S3015 (FIG. 12) and S3222 to S3223 (FIG. 14) are omitted.

  As mentioned above, although embodiment of this invention was described in detail, the scope of the present invention is not limited to said embodiment. The above-described embodiment can be variously improved and changed without departing from the gist of the present invention. Moreover, said embodiment and modification can be combined.

1: Information processing system 1: Character 2: Character 10: Mirror device 11: Storage device 12: Processor 13: Input / output interface 14: Communication interface 15: Camera 16: Speaker 17: Microphone 18: Display 19: Half mirror 30: Server 31: Storage device 32: Processor 33: Input / output interface 34: Communication interface

Claims (14)

An information processing apparatus connected to a mirror device comprising a mirror, a camera that acquires an observer image corresponding to a mirror image of the observer reflected in the mirror, and a speaker that outputs sound,
Means for acquiring the observer image from the mirror device;
By analyzing the feature quantity of the observer image, comprising means for specifying an image pattern of the observer image,
Means for determining a state of an observer reflected in the mirror image based on the identified image pattern;
The state of the observer is the state of makeup applied to the face of the observer, the state of the facial expression of the observer, the state of the skin of the observer, the state of the clothes of the observer, and the state of the observer It represents at least one of the body conditions,
Based on the result of the determination, comprising means for generating an anthropomorphic character voice,
By transmitting voice said generated in the mirror device, Ru comprising means for outputting the voice from the speaker,
Information processing device.   Means for associating and storing observer identification information for identifying an observer, observer attribute information relating to the attributes of the observer, and the identified image pattern;
  The means for specifying the second observer identification information of the second observer associated with the same or similar observer attribute information as the observer attribute information associated with the first observer identification information of the first observer. With means to identify,
    The second observer is an observer different from the first observer,
  The means for determining is based on the first image pattern specified from the observer image of the first observer and the second image pattern associated with the second observer identification information. Determine the state of the person,
The information processing apparatus according to claim 1.   The observer attribute information is information indicating at least one of the gender, age, and occupation of the observer.
The information processing apparatus according to claim 2. The generating means generates a voice of anthropomorphic characters according to the tendency of the observer state with reference to a preference model in which the tendency of the observer state is described .
The information processing apparatus according to claim 1 .   The tendency of the state of the observer is an occurrence frequency for each state of the observer.
The information processing apparatus according to claim 4. The means for generating refers to the preference model tendency of change in facial expression in response to the sound of the anthropomorphic character is described, to produce the sound of the anthropomorphic character according to the trend of change of the facial expression,
The information processing apparatus according to claim 1 . The generating means generates the voice of the anthropomorphic character based on the parameter of the anthropomorphic character, the combination of the observer attribute information regarding the attribute of the observer, and the result of the determination .
The information processing apparatus according to claim 1 . The determining means determines a change in the state of the observer.
The information processing apparatus according to any one of claims 1 to 7 . Means for receiving an instruction to specify at least one personified character from among a plurality of personified characters;
Said means for generating generates the sound of the specified personified character by the accepted instruction,
The information processing apparatus according to any one of claims 1-8. When the receiving unit receives an instruction to specify a plurality of anthropomorphic characters, the generating unit generates voices of a plurality of anthropomorphic characters, thereby generating the plurality of anthropomorphic characters according to the state of the observer Make characters talk to each other ,
The information processing apparatus according to claim 9 . Means for obtaining the voice of the observer from the mirror device;
The means for generating generates the voice of the anthropomorphic character according to the result and the observer's voice of the determination,
The information processing apparatus according to any one of claims 1-10. The generating means generates voice of the anthropomorphic character for presenting advice or guidance according to the result of the determination.
The information processing apparatus according to any one of claims 1 to 11. A mirror device connectable to an information processing apparatus according to any one of claims 1 to 12,
With a mirror ,
A camera for capturing an observer image corresponding to a mirror image of the observer reflected in the mirror;
A speaker that outputs the sound of the anthropomorphic character generated based on the observer image captured by the camera;
Mirror device. Program for causing a computer to function as each means according to any one of claims 1 to 12.

Images