KR102298070B1 - System for providing active artificial intelligence image character based on mobile device - Google Patents

Abstract

The present invention discloses an active artificial intelligence character system. More specifically, the present invention provides emotional exchange through an image-based active artificial intelligence image character with a virtual personality in utilizing various functions based on mobile devices carried by users, such as smartphones and tablet PCs. It relates to a mobile device-based active AI character system that implements
According to an embodiment of the present invention, a virtual active AI video character is installed in a mobile device, and various call functions, text transmission functions, music and video playback functions, etc. provided by the mobile device, stored photos, images, and call records , messages, KakaoTalk contents, friend list, purchase and payment history, vehicle operation or movement records, and web search records are used to automatically execute specific functions determined by the artificial intelligence video character in addition to the functions performed by the user. It can provide the same effect as interacting with a character.

Description

Mobile device-based active AI image character system {SYSTEM FOR PROVIDING ACTIVE ARTIFICIAL INTELLIGENCE IMAGE CHARACTER BASED ON MOBILE DEVICE}

The present invention relates to an artificial intelligence image character system, which implements emotional exchange through an artificial intelligence character with a virtual personality in utilizing various functions based on a mobile device carried by a user such as a smartphone or tablet PC. It is related to a mobile device-based active artificial intelligence image character system.

BACKGROUND With the rapid development of electronic information technology, mobile devices such as smartphones and tablet PCs have become indispensable in daily life for modern people. Users can download applications for phone calls, text transmission, music playback, video playback, Internet search, payment, etc. to the mobile device and use various functions, such as sending text messages, or watching videos or music. It can be said that functions play a significant role in our daily life.

On the other hand, the modern people living in the present are experiencing a rapid increase in the number of young generations who are experiencing a growing sense of loss and middle-aged people who wander without a job. , the number of patients with depression and schizophrenia is rapidly increasing.

It is no exaggeration to say that the person most often encountered in daily life is a cell phone, that is, a mobile device, that modern people who are unable to relieve loneliness in a society where stress is extremely high.

As one of the services using such a mobile device, an expression service through the application of artificial intelligence of a cyber character on the desktop to the occurrence of an event in a mobile communication terminal has been proposed.

Patent Laid-Open Publication No. 10-2005-0019368 creates a cyber character equipped with artificial intelligence on the desktop of a mobile communication terminal and expresses all events occurring in the mobile communication terminal through the movement of the cyber character and sound source, thereby allowing users to communicate through mobile communication. Disclosed is a service that makes it easier to recognize the operation and state of the terminal and at the same time allows the user to use the mobile communication terminal more interestingly.

However, the above-mentioned prior art is at a level of simply adding a passive expression of a character to the output of a function provided by a mobile device. There is a limit in that it cannot provide a realistic two-way service that provides

Laid-open Patent Publication No. 10-2005-0019368 (published on March 3, 2005)

The present invention has been devised to overcome the above-described limitations, and the present invention is an application program in the form of an application program that operates based on a mobile device and enables interaction with a user through an artificial intelligence character with a virtual personality. There is a challenge in providing a video character system.

In addition, according to an embodiment of the present invention, the existing form of providing desired information by reacting passively according to a user's command is boldly escaped, and based on artificial intelligence, it determines by itself based on the knowledge and records built into the character and falls within a certain range. Another objective is to provide a new concept image character system that attempts an active action without a separate command.

In order to solve the above problems, the mobile device-based active artificial intelligence character system according to a preferred embodiment of the present invention is a service system executed in a mobile device, according to a plurality of driving modes, a sound in response to an input user action and a character implementation unit that provides a response by a virtual image character implemented in a graphic form in a communication form, derives an autonomous response by a video character interacting with a user action by an AI engine, and responds to the derived response of the character It may include a control unit for controlling an action and an application interworking unit for interworking with one or more other applications installed in the mobile device, and participating in the execution of the other application, the video character.

The plurality of driving modes include a dialogue mode providing a dialogue action between the user and a video character, a game mode in which a user and a video character or a character and a third-party game are performed, and a video character at a point in time that meets the conditions set by the user. may include at least one of the assistant modes for performing a specific action.

The game mode sets the face of each image character in the game environment as a combination of the user's own photo stored in advance in the mobile device and the other's photo designated by the user on the face part of each character, and the name entered by the user The method of displaying on the character may include a function of proceeding with the game.

The other applications include a video player that plays a video, a music player that plays a sound source, a game application, a messenger application that transmits and receives messages with other systems, a call processor that connects calls with other systems, and the user's It may include at least one of a scheduler application for recording a schedule and a viewer application for displaying an image or text.

A storage in which character information in which a plurality of attributes for implementing the video character are defined and action information in which a plurality of actions performed by the video character are defined, and when communication between the user and the video character occurs, the content of the communication It may further include a pattern accumulator for recording the user pattern in the storage, and transmitting the user pattern recorded in the storage to the management server at regular intervals.

The character implementation unit may include one or more of the video character's name or nickname, appearance, body shape, color, voice, personality, and behavior according to a user's input within a certain form and attribute range of the basic image character presented by the character service. It may include a customizing function for selecting a property and generating or changing a certain part of the shape and property of the image character.

It may further include a character linking unit that implements communication and game execution between the image character and the character of another user by interworking with another user's mobile device.

The controller may control the video character to implement an action corresponding to an emotion in a person through an expression and an action according to a current situation.

One or more mobile devices are connected through an information communication network to manage member information of the user, and share the user's basic information, files and memory stored in the mobile device with the artificial intelligence of the image character, the image It may be connected to a central management server that provides action information in which a plurality of actions performed by the character are defined to the mobile device.

The central management server includes a communication unit that performs data communication with the mobile device connected through the information communication network, a member management unit that generates member information of the user, and performs a member registration procedure and a login procedure for a mobile device attempting to access , a learning engine for deriving an appropriate response according to the user's behavior by applying learning data to one or more predefined learning models based on deep learning technology;

When the database in which the learning data and learning results are stored and the previously distributed learning results are updated, an update unit for distributing the updated learning results to the mobile device to update the AI engine may be included.

The communication unit receives, from the mobile device, a user pattern including the contents of the communication between the user and the image character according to a predetermined period, and the learning engine can apply the user pattern to a learning model by using the user pattern as learning data. have.

According to an embodiment of the present invention, a virtual artificial intelligence video character is installed in a mobile device, and various call functions, text transmission functions and music playback functions provided by the mobile device, such as stored photos, images, call logs, messages, In addition to functions directly executed by the user using KakaoTalk contents, friend list, purchase and payment history, vehicle operation or movement records, and web search records, by automatically executing specific functions determined by the AI video character, virtual people and images are created. It can provide the same effect as actually interacting with a virtual person in the real world as it is a function in which a virtual person takes over a part of the user's role as well as having a conversation, playing a game, or exchanging a message through it.

1 is a diagram showing the overall structure for operating a mobile device-based active artificial intelligence image character system according to an embodiment of the present invention.
2 is a diagram showing the structure of a mobile device-based active AI image character system and a mobile device in which the system is installed according to an embodiment of the present invention.
3 is a diagram showing the structure of a central management server interworking with a mobile device-based active artificial intelligence image character system according to an embodiment of the present invention.
4 is a diagram illustrating a service providing method by a mobile device-based active artificial intelligence image character system according to an embodiment of the present invention.
5A to 5E are diagrams illustrating screens provided by a mobile device-based active AI image character system according to an embodiment of the present invention.

When a part "includes" or "includes" a certain element throughout the specification, this means that other elements may be further included, rather than excluding other elements, unless otherwise stated. In addition, terms such as "...unit" and "...system" described in the specification mean a unit that processes one or more functions or operations, which is hardware, software, or hardware and a combination of software.

Also, in this specification, the term "embodiment" is meant to serve as an illustration, example, or illustration, but the subject matter of the invention is not limited by such examples. Also, "comprising", "comprising", "having" and other similar terms are used, but when used in the claims, "comprising" is an open transition word that does not exclude any additional or other elements. may be used generically in a manner similar to the term "Comprising".

The various techniques described herein may be implemented with hardware or software, or a combination of both, where appropriate. As used herein, terms such as "...Unit" and "...System" likewise refer to computer-related entities, i.e. hardware, a combination of hardware and software, software or execution. It can be treated as equivalent to the city's software. In addition, each function executed in the program of the present invention may be configured in units of modules, and may be recorded in one physical memory or may be recorded while being dispersed between two or more memories and recording media.

1 is a diagram showing the overall structure for operating a mobile device-based active artificial intelligence image character system according to an embodiment of the present invention.

1, the mobile device-based active artificial intelligence image character system according to an embodiment of the present invention is installed and executed in a portable terminal possessed by a user, such as a smartphone or a tablet PC, and may be implemented in the form of an application, A service in the form of a character created by a user interacting with the user to perform an action derived based on artificial intelligence and displaying it on a mobile device may be provided.

The mobile device 100 in which the active AI image character system of the present invention is installed may include at least one processing unit and a memory. Here, the processing unit may include, for example, a central processing unit (CPU), a graphic processing unit (GPU), a microprocessor, an Application Specific Integrated Circuit (ASIC), Field Programmable Gate Arrays (FPGA), etc., It may have a plurality of cores. The memory 1120 may be a volatile memory (eg, RAM, etc.), a non-volatile memory (eg, ROM, flash memory, etc.), or a combination thereof. Also, the mobile device 100 may include additional storage. Storage includes, but is not limited to, magnetic storage, optical storage, and the like.

In addition, the mobile device 100 of the present invention may interwork with the central management server 200 through an information communication network for user authentication and updating of the behavior information of the AI character.

The central management server 200 may be connected to one or more mobile devices 100 through an information communication network, and may be linked to the mobile device 100 for user authentication for service provision and character creation and interaction of the artificial intelligence image character system. You can provide and update behavior information for

If it is assumed that the AI engine of the active AI image character system of the present invention is implemented in a stand-alone manner, the central management server 200 has a limited behavior pattern, and provides a large amount of learning data to the learning model. As there is a limit to its application, it plays a role of continuously updating the colorful behavior of the AI character.

The above-described mobile device 100 and the central management server 200 may perform data communication using a known communication technology. For example, long-term evolution (LTE), LTE Advance (LTE-A), code division multiple access (CDMA), wideband CDMA (WCDMA), universal mobile telecommunications system (UMTS), Wireless Broadband (WiBro), or GSM (Global System for Mobile Communications) and the like may be used.

The mobile device-based active AI image character system according to the above-described structure may provide the following functions.

First, the mobile device-based active AI image character system of the present invention is based on the fact that a mobile device owned by a user is always deeply involved in his decision-making and connection with a third party in the same living space as the user. , characterized by providing a virtual person - hereinafter referred to as "video character" - given a unique personality and a unique name and shape by personifying the mobile device 100 .

The present invention provides a video character with artificial intelligence to the user's basic personal information and photos, videos, call records, messages, Kakao Talk, friends list, purchase and payment history, vehicle operation or movement records stored in the mobile device 100, Interaction is realized by storing not only web browsing history, but also conversations with the character himself and activities with the character by date and time, and continuously updating them.

Here, the image character may be blocked from accessing or recognizing important items related to security, such as public certificates, biometric security data, financial account numbers and passwords stored separately in the mobile device 100 . However, the video character is a virtual person based on virtual artificial intelligence, and a unique name can be set, and most information related to the user can be shared. That is, the user's basic information, files, and memory stored in the mobile device may be shared with the artificial intelligence of the image character.

The name or nickname of the video character can be freely set according to the user's intention, but when interworking with the AI video character system of the present invention, it should be displayed as a video character of a similar shape with a common name to anyone who uses it. . This is because a uniform name and shape is essential for everyone in all countries or regions for the convenience of marketing or related businesses using characters created by individuals.

Accordingly, assuming that the name of the video character is set as "Keumdori" by the user, it can be set to have a similar meaning or pronunciation, such as "Dream", "Dreamer" or "ゆめ丸", depending on the country. The shape of the video character may also be expressed as a single, unified shape.

In addition, the image character may change clothes according to the user's selection, and may wear various types of accessories and items.

In addition, the video character implements memory ability, learning ability, limited cognitive ability, and autonomous behavior and language expression within a certain range through the AI engine, but includes an automatic upgrade function that gradually changes according to the user's taste or personality. can Accordingly, they speak and move autonomously within a certain range, changing their facial expressions according to the situation, exclamation, screaming, laughing, crying, grimacing, groaning, sighing, yawning, drowsiness, surprise, anger, joy, sadness, frustration, coughing, sneezing And even moans can be expressed with voice, facial expressions, and body movements, and in the initial setting where expression is possible up to a certain level, the ability is increased according to the frequency of use of the user's video character, and gradually complex expressions and movements can be realistically developed. It can be implemented as much as possible.

In addition, the video character autonomously determines according to the active function, outputs a notification of the mobile device 100 to the user, calls a virtual call, and calls the user with appropriate contents, for example, congratulations, consolation, encouragement, etc. Communication can be implemented and conversations using multi-party chat rooms that are actively provided through messages or SNS such as KakaoTalk, for example, communication such as appointments, schedules, guidance, alarms, congratulations, consolation, encouragement, health or mood inquiries can be done with the user.

The active autonomous communication function described above is a virtual friend who can talk with me anytime, anywhere and talk about anything to a lonely modern person who has many single generations and is accustomed to living alone. It can be said that it is a new concept active image expression artificial intelligence service implemented based on a mobile device.

Here, the above-described virtual call function is a method in which a second virtual phone number, that is, another phone number dependent on the terminal number assigned to the mobile device, is given to the video character, or the user is issued and applied two phone numbers. can be implemented as

Hereinafter, the structure of a mobile device-based active artificial intelligence image character system according to an embodiment of the present invention will be described in detail with reference to the drawings.

2 is a diagram showing the structure of a mobile device-based active artificial intelligence character system and a mobile device in which the system is installed according to an embodiment of the present invention.

Referring to FIG. 2 , a mobile device-based active artificial intelligence image character system 300 according to an embodiment of the present invention is a service system executed in the mobile device 100, and user actions are input according to a plurality of driving modes. In response to the character implementation unit 310 that provides a response by a virtual image character implemented in sound and graphics in the form of communication, the AI engine derives an autonomous response by the image character interacting with the user's action, The control unit 320 for controlling the behavior of the video character in response to the derived response, the application linkage unit 330 for interworking with one or more other applications installed in the mobile device 100, and participating in the execution of other applications, the video character When communication between the user and the image character occurs, the storage 340 in which the image character information in which a plurality of properties are defined for implementing the image character information and the behavior information in which a plurality of actions performed by the image character are defined is stored It may include a pattern accumulation unit 350 that records the user pattern in the storage and transmits the user pattern recorded in the storage to the management server at a predetermined period.

The character implementation unit 310 may be implemented in a form that can be displayed and output in the mobile device 100 . The user can create or convert the basic shape of the video character presented by the artificial intelligence character system 300 into a video character of a desired shape within a certain range according to his or her tendency, and then store it. When executed, the image character stored by the user is called and can communicate with the user.

The communication performed by the video character may be made in a plurality of driving modes, and this driving mode is a conversation mode that performs functions related to conversation with a user, a video character or a third person, or uses the video character as a second player. It may include a game mode that provides functions related to a game to be played together, and a secretary mode that provides functions related to schedule management, message transmission, and notification display based on user information.

In addition, the character implementation unit 310, according to the user's input within the range of a certain form and attributes of the basic character presented by the character service, one or more attributes of the name or nickname, appearance, body type, color and voice of the video character. can be selected to provide a customizing function for generating or changing a certain portion of the shape and properties of the image character.

When creating a video character, the user can simply select and set a common name and shape predefined in the system, and can select masculinity and femininity according to preferences. In addition, the user may designate the character's disposition in the basic setting according to the intention.

For example, the tendency of the video character can be set as a fighter type, a talkative type, a conversation type, a sympathetic reaction type, a play type, etc., and can be selected by applying a plurality of forms or personalities. In addition, it is possible to change and wear the clothes to the fashion desired by the user.

At this time, the image character changed and stored by the user is easily recognized as the same image character as the basic shape of the male character or the female character presented by the active artificial intelligence image character system 300 when a third party visually recognizes it. should be in the shape of Therefore, it is recommended to limit the changes in the video character's clothing, changes in height or weight, changes in hair shape or skin color, wear of accessories or equipment, etc. within a certain range that does not damage the basic shape of the video character and can be stored. desirable.

The controller 320 may control the image character by reflecting the plurality of image character actions defined based on artificial intelligence to the current situation or behavior of the character being executed, and may execute other related applications or drive the mobile device. can be controlled

To this end, the controller 320 may include an AI engine 325 . The AI engine 325 defines various actions that a virtual image character can take on the mobile device 100 , and the AI engine 325 responds or autonomously according to a user action input to the mobile device 100 . As an action, an action to be performed by the virtual video character is determined according to the action information set in the video character, and the controller 320 controls the video character displayed on the mobile device 100 according to the determination to display facial expressions and gestures. Change or output voice to communicate.

The application interworking unit 330 may link the image character with another application 400 installed in the mobile device 100, rather than the artificial intelligence character system 300 of the present invention, and execute it as communication.

Other applications 400 installed in the mobile device 100 include a video player 410 that plays a video, a music player 420 that plays a sound source, a game 430, and a messenger application that transmits and receives messages with other systems ( 440), a call processor 450 that connects a call with another system, a scheduler application 460 that records a user's schedule on a specific date or period, and a viewer application 470 that displays an image or text.

Communication that can be implemented through the linkage between the image character and other applications by the application linkage unit 330 is as follows.

First, as an interworking function with the video player 410 , the shape of a video character can be displayed on the screen of the mobile device 100 , and a simple conversation can be exchanged with the user, and a reaction according to the situation can be expressed through voice and motion. In addition, a predetermined motion ordered by the user may be displayed for a specified period of time such as singing, dancing, and motion.

In order to implement such a function, the application linkage unit 330 may convert the rendering of a video character into a standard codec, convert it to a video format, and play it through the video player 410 .

As a function of interworking with the music player 420 , when a user requests a specific song, a service in which the video character sings the corresponding song may be provided. Such a function may be implemented in the form of searching for a corresponding song source online and playing it through a music player in the form of download or streaming.

As an interlocking function with the game 430, using a specific shape called on the screen by displaying the face picture or name of a third party designated by the user, in various genres of games such as fighting games, punch balls, corporal punishment games, and role games. The game 430 may be executed in the form of directly participating as a gamer on behalf of the user or participating as a supporter of the game.

In addition, as an interworking function with the game 430 , a method of playing a game between a user's image character and a guest character in a game environment may be applied. It may be set as a combination of a picture of a user and a user-specified counterpart on the face, respectively, and the name entered by the user may be displayed on the video character to proceed with the game.

In addition, the game method may be a form of relieving the user's stress according to the intention thereof, and may be set so that the guest character takes a separate action corresponding to the user's action during the course of the game.

As the type of the game 430, a fighting game between a third party and a video character designated by the user, a punch ball game in which the video character touches a punch ball, a judo and wrestling game, a water balloon throwing or using a hammer It may be a hitting game that strikes a hitting target in which the face or name of the designated opponent is expressed, and a role reversal game in which the user takes the desired action on the opponent designated by the user.

As an interworking function with the messenger application 440, the video character conducts a virtual one-to-one conversation with the user through a message or Kakao Talk connection, listens to complaints, and provides consoling comments and reactions to reduce stress. can provide In addition, an emoticon provided by the messenger application 440 may be transmitted or a special character, a picture, or the like may be expressed during a conversation.

As a function of interworking with the call processor 450 , a video character may transmit a video or audio (recording) designated by the user to another user designated by the user at the reserved date and time through a call method.

For example, a user records a certain amount of video for a predetermined purpose and stores it in the mobile device 100, and a video file stored with a specific title is set to another user, 'HongXX', as a recipient on March D, 2019 If you command the character to transmit at H, the video character makes a call to another user at the time and when the call is connected, the user's image character image is automatically loaded on the screen of the other user's mobile device and the image file has arrived. A guide message is displayed, and the video can be played depending on whether you agree to receive the video.

Thereafter, when the video playback is finished, the video character inquires to the video sender if there is something to reply to the video sender, and if there is a reply, the video character can be temporarily stored or transmitted to the user's mobile device for storage. Here, the temporarily stored reply may be delivered to the user by asking the video character in a conversational manner at a required time.

In addition, if the content that the user wants to convey to other users in the form of a reservation is a voice recording, a video character is loaded on the receiver's screen to express the voice recording in a speaking form instead, or to play the voice recording content depending on whether the user agrees to the reception. can Here, the processing of the reply may be processed in a similar manner to the reply of the above-described video call.

In addition, as a function of interworking with the call processor 450 and the scheduler application 460, the video character includes important appointments, anniversaries, important events, happy days, sad days, lucky events and unlucky events included in the user's schedule information. may provide a function of making a ringtone by extracting the , requesting the call processor 450 according to autonomous judgment and calling a call whose caller is a video character. Such a function may be implemented in the form of attempting a congratulatory or consoling conversation by making a call autonomously without a user's command at an arbitrary time determined by a random function, or attempting a message or SNS conversation.

In addition, when an acquaintance of the user uses the active AI image character system of the present invention, a function to talk with the image character of the user's acquaintance with reference to the aforementioned schedule information, or to inform the user of an important schedule, event, or accident can provide a function.

As an interworking function with the viewer application 470 , a picture stored in the mobile device is displayed, and an effect as if the video character appears in the picture can be provided by superimposing and displaying the image character on the picture.

When communication between the user and the image character occurs, the pattern accumulator 350 may store the details as user pattern information in the storage 340 . In an embodiment of the present invention, the user's behavior is an index indicating the user's pattern and can be used for deep learning learning, and the pattern accumulator 350 records the user's behavior according to the execution of the active AI image character service in the pattern information. and to be reflected as learning data by providing it to the management server 200 according to a certain period to be utilized to update the AI engine 325 later.

In addition, the active artificial intelligence image character system according to an embodiment of the present invention may further include a character linkage unit 360 that interworks with another user's mobile device to implement communication and game execution between the image character and the character of another user. have.

As a function provided by this character linkage unit 360, when a user invites another user's character through the active artificial intelligence image character system 300, he meets his image character in a virtual space to have a conversation or play a game, or It is possible to provide a function for performing a limited action within a certain range and allowing the user to observe it.

Here, on the screen of the mobile device 100 , the two characters of the user and the other user may be simultaneously displayed on the two screens of the two mobile devices 100 of the user and the other user, respectively. In this case, an autonomous action of preventing one of the two characters from viewing the corresponding screen by turning off the screen may be applied.

The storage 340 may store character information created by a user, behavior information defining an action performed by an image character, and user information for implementing other services.

Hereinafter, a central management server interworking with a mobile device-based active artificial intelligence image character system according to an embodiment of the present invention will be described with reference to the drawings.

3 is a diagram showing the structure of a central management server interworking with a mobile device-based active artificial intelligence image character system according to an embodiment of the present invention.

Referring to FIG. 3 , the central management server 200 according to an embodiment of the present invention generates and accesses a communication unit 210 that performs data communication with the mobile device connected through the information communication network, and user's member information. Member management unit 220 that performs the membership registration procedure and login procedure for the attempted mobile device, applying the learning data to one or more predefined learning models based on deep learning technology to derive an appropriate response according to the user's behavior When the learning engine 230, the database 240 in which the learning data and the learning results are stored, and the previously distributed learning results are updated, an update unit that distributes the updated learning results to the mobile device to update the AI engine ( 250) may be included.

The communication unit 210 may be connected to an information communication network, and may be connected to one or more mobile devices 100 in which the active AI image character system of the present invention is stored and executed to perform data communication with each other.

The member management unit 220 may perform a membership registration procedure, user authentication procedure, etc. of a user who wants to use the active AI image character system of the present invention. A user who wants to use the artificial intelligence image character service of the present invention must perform a membership registration procedure by inputting an account and password, a phone number of the mobile device, or the user's biometric information, and the member management unit 220 includes the mobile device ( 100) may generate member information for the user according to the request and store it in the database 240 .

In addition, as the AI image character system is executed in the mobile device 100 , the member management unit 220 is the AI image of the present invention only for the mobile device of the user who is normally authenticated through the user authentication procedure through the registered member information. Make the character service available.

The learning engine 230 defines one or more learning models for implementing the AI engine mounted on the active artificial intelligence image character system, and applies one or more learning data stored in the database 240 to interact with user actions. The behavior of the video character can be defined and the behavior information can be reflected in the AI engine.

In addition, when user pattern information is provided from each mobile device 100 later, the learning engine 230 may use it as learning data to reinforce the learning result, and to implement a more personalized AI engine.

The database 240 may store the aforementioned member information, behavior information, and learning data for deep learning. In addition, the database 240 may store various setting values for implementing the active AI image character system, for example, attribute values and items used when generating image characters, which are distributed to the mobile device 100 . can be

When the version of the AI engine installed in the mobile device 100 is updated, the update unit 250 may reflect this and distribute the update file to the mobile device 100 . In addition, the update unit 250 may also update the behavior information of the video character personalized for each user, which specifies the user and the mobile device 100 with reference to member information, and updates only the mobile device 100 . information can be provided.

Hereinafter, a service providing method by a mobile device-based active artificial intelligence image character system according to an embodiment of the present invention will be described with reference to the drawings.

4 is a diagram illustrating a service providing method by a mobile device-based active artificial intelligence image character system according to an embodiment of the present invention.

Referring to FIG. 4 , according to the service providing method by the mobile device-based active artificial intelligence image character system according to the embodiment of the present invention, first, the user uses his/her mobile device to create the active artificial intelligence image character system of the present invention. Download and install, and connect to the central management server by executing the active AI image character system (S100).

Next, the user who first accesses the system enters personal information such as his or her name and age, and user information such as account, password, and phone number of the mobile device to proceed with the membership registration procedure (S110).

According to step S110, the user who is normally registered as a member creates his/her own character by setting attributes such as appearance, voice, and color of the image character using his/her mobile device (S120).

After that, the active AI image character system maintains the input standby state until the user's action occurs (S130), determines whether the user's input or an action such as voice occurs (S140), and responds to the user's action in the driving mode is set (S150), and a response is derived by calculating the interaction by the AI engine based on artificial intelligence (S160).

Here, although not shown, the video character may perform an autonomous action according to a specific condition even if the user's action does not occur in step S140.

After step S160, the active AI image character system controls the character to perform an action corresponding to the response derived by the AI engine (S170).

On the other hand, after step S170, the active AI image character system creates a user pattern by recording information about communication between the user and the image character according to a predetermined period, and is provided to the central management server to be utilized for user pattern analysis (S180) )

Hereinafter, the technical idea of the present invention will be described in detail through an example of a character screen provided by an active AI image character system according to an embodiment of the present invention.

5A to 5E are diagrams illustrating screens provided by a mobile device-based active AI image character system according to an embodiment of the present invention.

First, referring to FIG. 5A , the mobile device-based active AI image character system of the present invention may provide a function for a user to directly create his/her own character. The character creation screen W100 provided by the active AI image character system of the present invention includes the character window W11 in which the entire character is displayed, and specific attributes of the character's appearance, for example, face shape, hair shape, and head. A first template window (W120) for selecting color, eye shape, eye color, height, body type, ear shape, etc., and a second template for selecting clothes and accessories such as glasses, hats, tops, bottoms, watches and shoes, etc. It may include a window W125 and a name input window W130 into which a character name or nickname can be input.

A user may create a video character having a desired name or nickname and an external appearance through a user interface (UI) provided by the system.

Referring to FIG. 5B , the generated image character interacts with the user's action to perform a specific action, and the image may be displayed on the screen W200 of the mobile device. As an example, when the user calls the name or nickname of the video character, the video character may output a sound or display text in response to the voice, and may take an appropriate posture.

Referring to FIG. 5C , as a function of interworking with other applications, a function in which a user plays a video through interaction with a video character is exemplified.

A user may execute a service interworking with various applications installed on a mobile device, and the active AI image character system of the present invention may provide useful services through interworking with various other applications according to a user's command.

As illustrated in the figure, when a user executes an active AI video character system and commands his/her video character to play a specific video in the form of voice or text input, the video character recognizes it and stores it in the mobile device By extracting the file from the video file, the video can be played and played on the screen W300 of the mobile display in connection with the mounted video player.

In this case, the user's image character may be displayed at the bottom of the screen of the mobile device, and a response to the user's command may be output in the form of sound or text. Also, at the top of the screen of the mobile device, the playback screen W310 of the corresponding video file may be displayed by the video player.

Referring to FIG. 5D , as an interworking function with other applications, a call connection function is exemplified by a user interacting with a video character.

In the active AI character system according to an embodiment of the present invention, when a user commands a call with another user through voice or input, as another application installed on the mobile device, it interworks with a call processor to establish a call connection through the phone number of the other party. you can try

At this time, the video character extracts the phone number for another user input by the user on the call connection screen W400 from the storage of the mobile device, attempts a call connection to the corresponding phone number through the call processor, and at the same time, the current call The content being connected can be output in the form of voice or text.

Referring to FIG. 5E , as an interworking function with a game through an active AI image character system according to an embodiment of the present invention, the user can run the game together with his or her image character, and relieve the user's stress during the game. As a result, the punch ball game screen W500 that touches other characters is exemplified.

In such a punch ball game, the user plays a hitting game by touching another character displayed on the screen of the mobile device. can In this case, the other character may output a response corresponding to the touched area (T1, T2) as voice or text, and the user may use the game using the other character as well as his/her own image character.

Although many matters are specifically described in the above description, these should be construed as examples of preferred embodiments rather than limiting the scope of the invention. Accordingly, the invention should not be defined by the described embodiments, but should be defined by the claims and equivalents to the claims.

100: mobile device 200: management server
210: communication unit 220: member management unit
230: learning engine 240: database
250: update unit 300: artificial intelligence character system
310: character implementation unit 320: control unit
325: AI engine 330: application linkage
340: storage 350: pattern accumulation unit
360: character linkage 400: external application

Claims (11)

A service system running on a mobile device, comprising:
a character implementation unit that provides a response by a virtual image character implemented as sound and graphics in a communication form in response to an input user action according to a plurality of driving modes;
a controller for deriving an autonomous response by a video character interacting with a user action by the AI engine, and controlling the action of the video character in response to the derived response; and
an application interworking unit that interworks with one or more other applications installed in the mobile device, and allows a video character to participate in the execution of the other application; and
It includes a character linkage unit that interworks with another user's mobile device to implement communication and game execution between the video character and the video character of another user,
The plurality of driving modes,
Which of a game mode in which a game is played between a user and a video character or between a user's video character and another user's video character, and a secretary mode in which the video character performs a specific action at a point in time that meets the conditions set by the user? including one,
The other application includes a call processor that connects a call with another system, and a scheduler application that records the user's schedule on a specific date or period,
The app linkage unit,
A mobile device that extracts a specific schedule or event from the user's schedule information according to interworking with the call processor and scheduler application, requests the call processor according to autonomous judgment, and makes a call using the caller as a video character without a user's command Based active AI character system. The method of claim 1,
The plurality of driving modes,
The mobile device-based active AI character system further comprising a conversation mode for providing a dialogue action between the user and the image character. 3. The method of claim 2,
The game mode is
In the game environment, the face of each video character is set as a composite of the user's own photo and the other's photo designated by the user, stored in advance in the mobile device, on the face of each character, and the name input by the user is displayed on the character. The ability to progress the game in a way
A mobile device-based active artificial intelligence character system comprising a. The method of claim 1,
The other applications are
A mobile device-based active AI character system comprising at least one of a video player that plays a video, a music player that plays a sound source, a game application, a messenger application that transmits and receives messages with other systems, and a viewer application that displays images or text . The method of claim 1,
a storage for storing character information in which a plurality of attributes for implementing the image character are defined and behavior information in which a plurality of actions performed by the image character are defined; and
When communication occurs between the user and the video character, the pattern accumulation unit records the contents of the communication as a user pattern in the storage, and transmits the user pattern recorded in the storage to the management server at regular intervals.
A mobile device-based active artificial intelligence character system further comprising a. 6. The method of claim 5,
The character implementation unit,
According to a user's input within a certain form and attribute range of the basic image character presented by the character service, one or more attributes of the image character's name or nickname, appearance, body type, color, voice, personality and behavior are selected and said A customizing function to create or change a certain part of the shape and properties of a video character
A mobile device-based active artificial intelligence character system comprising a. delete The method of claim 1,
The control unit is
A mobile device-based active AI character system that controls the video character to implement an action corresponding to an emotion in a person through facial expression and action according to the current situation. The method of claim 1,
The mobile device is
At least one is connected through an information communication network, manages the user's member information, shares the user's basic information, files, and memory stored in the mobile device with the artificial intelligence of the video character, A mobile device-based active AI character system that is connected to a central management server that provides behavior information in which the behavior of the user is defined to the mobile device. 10. The method of claim 9,
The central management server,
a communication unit for performing data communication with the mobile device connected through the information communication network;
a member management unit generating member information of the user and performing a member registration procedure and a login procedure for a mobile device attempting to access;
a learning engine for deriving an appropriate response according to the user's behavior by applying learning data to one or more predefined learning models based on deep learning technology;
a database in which the learning data and learning results are stored; and
When the previously distributed learning result is updated, an update unit that distributes the updated learning result to the mobile device to update the AI engine
A mobile device-based active artificial intelligence character system comprising a. 11. The method of claim 10,
The communication unit receives, from the mobile device, a user pattern including the contents of the communication generated between the user and the image character according to a predetermined period,
The learning engine is
A mobile device-based active AI character system that uses the user pattern as learning data and applies it to a learning model.

Images