KR20230071053A - Method for analyzing personality or aptitude based on metaverse - Google Patents

Abstract

Provided is a method for analyzing a personality based on metaverse and artificial intelligence comprising: a step of obtaining at least one among speech information and visual information of a user; a step of analyzing a personality of the user using at least one among the speech information, the visual information, and the bio information; a step of generating data per user comprising a personality analysis result of the user; and a step of generating an avatar of the user using the data per user.

Description

Personality analysis method based on metaverse and artificial intelligence {METHOD FOR ANALYZING PERSONALITY OR APTITUDE BASED ON METAVERSE}

The present invention relates to a personality analysis method, and more particularly, to a personality analysis method based on metaverse and artificial intelligence.

Metaverse is a compound word of 'meta', meaning virtual and transcendent, and 'universe', meaning world and universe. It means a virtual world where learning, learning, etc. are connected. In other words, metaverse is a concept that has evolved one step further than virtual reality (VR), and is characterized by using avatars to not only enjoy games or virtual reality, but also to engage in social and cultural activities like real life.

The Acceleration Studies Foundation (ASF), a non-profit technology research organization, categorizes metaverses into four types: augmented reality, life logging, mirror worlds, and virtual worlds.

Augmented reality refers to an environment that interacts through virtual overlapping objects expressed in 2D or 3D in real space. It has the characteristics of reducing people's resistance to the virtual world and increasing their immersion. An example of augmented reality is a scene in which a digitally constructed building in the past is superimposed on a user terminal when a user photographs a trace of the current ruins with a terminal camera.

Lifelogging is a technology that captures, stores, and describes everyday experiences and information about objects and people. Users can capture every moment in their daily life as text, video, sound, etc., store the contents on the server, organize them, and share them with other users. An example of a daily record is an act of storing and sharing information such as distance run, calories consumed, music selection, etc. by interlocking sensor-attached sportswear with an MP3 player that can be connected to a network.

The mirror world refers to a virtual world that reflects the real world as realistically and as it is, but is expanded informationally. A typical example is Google Earth. Google Earth collects satellite images from all over the world and updates them at regular intervals, reflecting the ever-changing real world as it is. As technology continues to develop, the mirror world that reflects reality will gradually approach the real world, which will become a great immersive factor in virtual reality in the future. The user of the mirror world obtains information about the real world by viewing the virtual world.

The virtual world is an alternative world that is similar to or completely different from the real world built with digital data. In the virtual world, users are characterized by performing activities similar to economic and social activities in the real world through avatars. The virtual world is the most familiar type of metaverse, from online role-playing games to living virtual worlds such as SANDBOX, ZEPETO, and Roblox, a community implemented in a 3D computer graphic environment. is a concept that encompasses

Korean Patent Publication No. 10-2019-0108523 (2019.09.24.)

Since the personality test is conducted on people in real life, it is difficult to easily conduct the test unless the participant conducts it face-to-face. Therefore, there is a need for a relatively simple way to conduct a personality test based on the metaverse.

An object of the present invention is to acquire user voice information and/or visual information, analyze the user's personality using artificial intelligence, generate user-specific data, and use the user-specific data to create avatars of users active in the metaverse. It is to provide a personality analysis method based on metaverse and artificial intelligence that can generate

Another object of the present invention is to provide a personality analysis method based on a metaverse and artificial intelligence capable of updating a personality in real time according to words or actions of an avatar created using initial user-specific data in the metaverse.

Another object of the present invention is to provide a personality analysis method based on metaverse and artificial intelligence that can be used in various fields such as career path, learning, communication, corporate personnel/recruitment, travel recommendation, and product recommendation through an avatar reflecting the user's personality. is to provide

However, the problem to be solved by the present invention is not limited thereto, and may be expanded in various ways without departing from the spirit and scope of the present invention.

The personality analysis method based on the metaverse and artificial intelligence performed in the personality analysis system according to an embodiment of the present invention includes the steps of acquiring at least one of the user's voice information, visual information and bio information such as brain waves and DNA; Analyzing the personality of the user using at least one of the voice information and visual information and bio information such as brain waves and DNA; generating data for each user including the result of analyzing the user's personality; and generating an avatar of the user using data for each user.

Analyzing the personality of the user includes analyzing the voice information, predicting the personality of the user based on the analysis result of the voice information, and recommending the personality of the user based on the prediction. steps may be included.

Analyzing the voice information may include removing noise from the voice information and normalizing the voice information; generating voice data that can be marked on a timeline based on the voice information; The method may include generating a voice style including at least one of a tempo, an idle time, a tone, and a volume based on the voice information.

In the step of predicting the personality of the user based on the analysis result of the voice information, predicting the personality of the user based on the voice data, taking the voice data as an input, and using the voice data for learning classified by personality type The personality of the user may be predicted by the first engine learned by

In the step of predicting the personality of the user based on the analysis result of the voice information, the personality of the user is predicted based on the voice style, but the average value of the voice style for a specific part that is easy to read is input. , Personality of the user may be predicted by the second engine learned by learning voice styles classified by personality types.

In the step of predicting the personality of the user based on the analysis result of the voice information, predicting the personality of the user based on the voice style, using the voice style marked at regular time intervals as an input, data for each marking Alternatively, the user's personality may be predicted by a third engine learned from continuous pattern data of the marking section.

In the step of predicting the personality of the user based on the analysis result of the voice information, the personality of the user is predicted based on the voice data and the voice style, and the predicted result based on the voice data and the voice style It is possible to predict the personality of the user by statistically processing the predicted result based on .

Analyzing the personality of the user includes analyzing the visual information, predicting the personality of the user based on the analysis result of the visual information, and recommending the personality of the user based on the prediction. steps may be included.

Analyzing the visual information may include recognizing the user's face from the visual information and extracting the user's face, and based on the visual information, one scene of a specific image or the entire specific image as an axis of time. (timeline) generating face data that can be marked, and based on the visual information, the shape and size of the face, the shape and position of the eyes, the shape and position of the nose, the shape and position of the mouth, the shape and position of the ears, and the eyebrows It may include generating a face style including at least one of a shape and position, a facial expression, and a hairstyle of the face.

In the step of predicting the personality of the user based on the analysis result of the visual information, predicting the personality of the user based on the face data, taking the face data as an input, and using the face data for learning classified by personality type The personality of the user may be predicted by the fourth engine learned by

The step of predicting the personality of the user based on the analysis result of the visual information may include predicting the personality of the user based on the face style, based on the face style of one scene of the specific image that is easy to read With the generated 2D or 3D face pattern as an input, the user's personality can be predicted by the fifth engine learned by the learning face style classified by personality type.

In the step of predicting the personality of the user based on the analysis result of the visual information, predicting the personality of the user based on the face style, based on the face style marked for each time with respect to the entirety of the specific image With the generated 2D or 3D face pattern as an input, the personality of the user can be predicted by the sixth engine learned by the face style for learning classified by personality type.

Predicting the personality of the user based on the analysis result of the visual information may include predicting the personality of the user based on the face data and the face style, and the predicted result based on the face data and the face style. It is possible to predict the personality of the user by statistically processing the predicted result based on .

Analyzing the personality of the user includes analyzing the bio information, predicting the personality of the user based on the analysis result of the bio information, and recommending the personality of the user based on the prediction. steps may be included.

The step of analyzing the bio-information may include an EEG analysis step of acquiring and analyzing brain waves responding to at least one of various stimuli among visual, auditory, taste, touch, and smell for a specific period of time, and DNA element-specific structural characteristics analyzed data It may include bioinformation analysis including a DNA information analysis step of analyzing unique DNA information by comparing it with an information map.

The step of predicting the personality of the user based on the analysis result of the bio-information. It may include a step of predicting the predicted results through the accumulated bio-analysis and the analyzed personality.

The step of recommending the personality of the user based on the personality prediction may include guiding the personality result predicted through one or more analyzes to the final recommended personality through a single or integrated analysis and asking whether to apply it when creating an avatar. there is. The personality analysis method based on the metaverse and artificial intelligence updates the user-specific data including the user's personality analysis result in real time or periodically by reflecting the generated avatar's actions, conversations, words or writings in the metaverse. It may include more steps.

The personality analysis method based on the metaverse and artificial intelligence, from the viewpoint of multimodal behavioral patterns, words, conversations or texts used by the created avatar in the metaverse and the behavior of the avatar-the behavior is the avatar's hands, head, The method may further include analyzing at least one of a field of interest and a tendency of a user using the avatar in consideration of including eye movement.

The created avatar can have artificial intelligence properties, can be active without user access, and can be operated by interacting with other avatars. Also, depending on the conditions, it can be moved to another place, copied, and used for various experiments or metaverse (virtualization, digital twinning) configuration.

The disclosed technology may have the following effects. However, it does not mean that a specific embodiment must include all of the following effects or only the following effects, so it should not be understood that the scope of rights of the disclosed technology is limited thereby.

According to the above-described personality analysis method based on the metaverse and artificial intelligence according to the embodiments of the present invention, the user's voice information and / or visual information is acquired, and the user's personality is analyzed using artificial intelligence to obtain data for each user. Avatar of a user who is active in the metaverse can be created using user-specific data.

According to the personality analysis method based on the metaverse and artificial intelligence according to the above-described embodiments of the present invention, the personality can be updated in real time according to the words or actions of the avatar created using the initial user-specific data in the metaverse. there is.

According to the personality analysis method based on the metaverse and artificial intelligence according to the above-described embodiments of the present invention, the avatar reflecting the user's personality is used in various fields such as artificial intelligence learning for personality analysis, communication service provision, and corporate recruitment process. can be utilized in

1 is a flowchart of a personality analysis method based on metaverse and artificial intelligence according to an embodiment of the present invention.
2 schematically illustrates an overall process for generating an avatar according to an embodiment of the present invention.
3 illustrates a process of analyzing voice information according to an embodiment of the present invention.
4 illustrates a process of analyzing personality using voice information according to an embodiment of the present invention.
5 illustrates a process of analyzing visual information according to an embodiment of the present invention.
6 illustrates a process of analyzing personality using visual information according to an embodiment of the present invention.

Since the present invention can make various changes and have various embodiments, specific embodiments will be illustrated in the drawings and described in detail.

However, this is not intended to limit the present invention to specific embodiments, and should be understood to include all modifications, equivalents, or substitutes included in the spirit and technical scope of the present invention.

Terms such as first and second may be used to describe various components, but the components should not be limited by the terms. These terms are only used for the purpose of distinguishing one component from another. For example, a first element may be termed a second element, and similarly, a second element may be termed a first element, without departing from the scope of the present invention.

It should be understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, but other elements may exist in the middle. something to do. On the other hand, when a component is referred to as “directly connected” or “directly connected” to another component, it should be understood that no other component exists in the middle.

Terms used in this application are only used to describe specific embodiments, and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this application, the terms "include" or "have" are intended to designate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, but one or more other features It should be understood that the presence or addition of numbers, steps, operations, components, parts, or combinations thereof is not precluded.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related art, and are not interpreted in an ideal or excessively formal meaning unless explicitly defined in this application. .

Hereinafter, with reference to the accompanying drawings, preferred embodiments of the present invention will be described clearly and in detail so that those skilled in the art can easily practice the present invention.

1 is a flowchart of a personality analysis method based on a metaverse and artificial intelligence performed in a personality analysis system according to an embodiment of the present invention.

Referring to FIG. 1 , in the personality analysis system, at least one of voice information and visual information is acquired in step S110. Voice information can be acquired by ontology (natural language machine learning)-based voice recognition technology, and visual information is text information such as the user's handwriting and model name displayed on the object held by the user, the user's appearance, facial expression, and Information that can be judged by sight, such as face information such as appearance change during activity, user gesture information, and object information such as clothes worn by the user, may be included without limitation. In the obtaining step, various types of information including bio information such as brain waves and DNA that can be acquired by using appropriate sensors may be further used in addition to voice information and visual information. Exchanging conversations, recorded information, and patterns can also be used as analysis and learning information.

In the personality analysis system, in step S130, as a simple screening step for the user's personality, the user's personality is analyzed using at least one of the voice information, the visual information, and the bio information. Specifically, at least one of voice information, visual information, and bio information of the user may be analyzed, and the user's personality may be first analyzed using the analyzed data. The analysis of the bio-information may be performed by acquiring brain waves responding to at least one of visual, auditory, taste, touch, and smell stimuli for a specific period of time. In addition, the analysis of the bio-information may analyze unique DNA information by comparing it with a data information map in which characteristics of each DNA element and structure are analyzed. In the case of the bio information, it is possible to predict the user's personality by comparing and analyzing the predicted results through the accumulated bio analysis.

In the personality analysis system, the personality of the user primarily analyzed in step S130 is primary personality data, and based on this, the personality of the user can be recommended. For example, in the case of generating primary personality data, it is possible to analyze the user's job or personality from an astronomical perspective through artificial intelligence learning. That is, the user's voice or face information can be analyzed as the user's job is president or athlete using artificial intelligence learned for the user's face or voice, and the primary personality data for personality can be analyzed. there is. Such primary personality data may be updated according to actions or words of the user's avatar later to generate secondary personality data.

In the personality analysis system, in step S150, user data including the personality analysis result of the user in the simple screening step is generated. The user-specific data may include, for example, the user's ID, password, and personality analysis result data. The user-specific data may include data obtained by analyzing the user's voice or face image rather than the user's voice data or face image data itself.

Here, the data for each user is the user's individual data temporarily obtained in a step prior to generation of the secondary personality data obtained by analyzing the avatar's actions or words. Here, the data for each user may be updated on a time axis. For example, in the simple selection step for the user's personality, data analyzed in terms of physiognomy, such as the user's voice or face, can be updated while continuously updating the personality data on a time axis.

In the personality analysis system, in step S170, a user's avatar is created using data for each user. The appearance or attire of the avatar can be determined according to the data for each user, and the user can live in the metaverse through the created avatar.

In the personality analysis system, in step S190, the personality of the user is analyzed in real time using the avatar, and artificial intelligence is trained. That is, the personality of the user analyzed by artificial intelligence in step S130 is not fixed. Initial data is generated by simply screening the user's personality with minimal voice information and/or visual information, and then the user's personality data is updated in real time or periodically by reflecting the user's avatar's actions or words within the metaverse. It can be.

As an example of reflecting the avatar's behavior, the user's personality data may be updated according to whether the avatar runs or walks to a specific place. Alternatively, the user's personality data may be updated according to whether the avatar moves to a specific place through a straight path or detour through a detour route to the specific place.

As an example of reflecting the avatar's speech, a user who uses the avatar after extracting words (subject, object, adverb, particle, verb) used repeatedly or repeatedly from the words, conversations, or texts used by the avatar through redundancy check of interests and tendencies can be analyzed.

From the multimodal behavioral pattern perspective, it is possible to analyze the interests and tendencies of users who use avatars by considering the words, conversations, or texts used by the avatars and the behaviors of the avatars (movements of hands, head, gaze). For example, if the avatar's hand or head is shaken or moved more actively at a specific word used by the avatar, whether the avatar's gaze moves to a specific location or a specific target in the virtual space, whether the gaze stops, or continues a specific action It is possible to analyze the field of interest and tendency of the user who uses the avatar in consideration of the time and the like.

On the other hand, since the personality data is updated using the actions or words of the avatar, not the actions or words of the user, if the user wants to know his or her actual personality, he or she can act or speak the same as he or she actually acts or speaks within the metaverse. there is. On the other hand, in order to create an avatar with the personality desired by the user, the user may act or speak within the metaverse differently from what he or she normally does or say, and the personality may be updated to reflect this.

Communication services can be provided through methods such as matching with other users with similar personalities to users within the metaverse or matching with other users with opposite personalities to users within the metaverse through analyzed user-specific data or user avatars. there is.

In addition, a personality test may be conducted as a part of a company's recruitment process through the analyzed user-specific data or the user's avatar. For example, a company can set up a work environment, configure a metaverse according to the settings, and then check whether the user's avatar can perform work while adapting well to the work environment set by the company. For example, if you change the workspace setting of the company where the user's avatar works from a wide space to a narrow space (e.g., a crowded space or a space without light), the user's work progress You can check if it falls. Alternatively, when the company's affiliated department is changed to a sales department rather than a development department for the avatar of a user having a developer job, the degree of adaptation of the user to work may be examined.

In addition, for example, for a prospective new hire, the effect of hiring may be verified by simulating how an artificially intelligent avatar reflecting the personality of the prospective new hire operates in a virtual work environment.

In addition, for example, for new hires, the effect of hiring can be verified by evaluating the work performance of the new hires in the metaverse for each department in the company in terms of personnel evaluation when the avatar of the new hires performs work.

In addition, for example, by putting the avatars of already hired personnel or new recruits in the company into a virtual team to carry out a new project in the company, and having them carry out the new project in the metaverse space, what kind of synergy effect the project has can be verified.

Compared to the case of taking a personality test or hiring process in reality, the user does not have to go to the test in person, so an efficient personality test or recruitment process can be performed, and therefore, from the standpoint of a company, ESG (Environmental, Social and Governance) management can contribute

When an avatar is created, the analyzed user-specific data is reflected, but whether to disclose the user-specific data itself to others may be determined by the user's choice.

The personality analysis method according to an embodiment of the present invention may be used not only to analyze the personality of a user (ie, a person) but also to analyze the personality of an animal or object.

The aforementioned steps S130, S150, S170 and S190 may be performed by the artificial intelligence engine of the personality analysis system, and detailed operations will be described later with reference to FIGS. 2 to 6.

2 schematically illustrates an overall process for generating an avatar according to an embodiment of the present invention.

The personality analysis method according to an embodiment of the present invention can be serviced to users as a metaverse personality solution. The metaverse personality solution can provide a personality analysis service if a user accesses it in real time on an online basis, or if online access is not possible, it can provide a personality analysis service offline based on previously stored data. Later, the offline data may be synchronized with the online data to reflect the results analyzed offline.

The metaverse can be a virtual world copied from the real world, or it can be a digital twin that operates the same as the real world. It may also be a world created for a special purpose or imagination rather than a reflection of reality.

Users can access the metaverse space in a variety of ways, such as accessing the metaverse space through VR equipment such as HMD (Head Mounted Display), or accessing the metaverse space through applications or browsers such as smartphones and PCs.

As shown in FIG. 2, when a user accesses the metaverse space and logs in, a registration process 203 for creating an avatar is performed. The registration process 203 may consist of a basic mode 210 and an expert mode 291, and the basic mode 210 includes voice information 205 and/or face information 207 Avatar is created by simple personality screening using, and the expert mode (291) is a detailed test, such as strong test, MLST learning strategy test, MBTI, enneagram personality type test, bio information analysis (brain wave, DNA), etc. It is to analyze the personality more precisely by. The expert mode is optional, and whether to perform it can be determined according to the user's choice.

Personality analysis may be performed by the artificial intelligence engine 200 .

In the basic mode 210, voice information 205 and/or face information 207 are analyzed (211, 213), personality is analyzed by an artificial intelligence process (231, 233, 235), and as a result, the user's Personality is recommended (250) and an avatar is recommended (270). A detailed description will be given later.

In the expert mode (291), personality is analyzed by an expert personality process (293) or personality is analyzed through an actual personality test (295), and personality is recommended according to the result (297) and an avatar is recommended (270). ). A detailed description will be given later.

That is, the artificial intelligence engine 200 analyzes the user's personality using voice information 205 and/or face information 207 and then recommends an avatar. The user may use the avatar 270 recommended by the artificial intelligence engine 200 as his or her own avatar, or may customize it.

When the avatar is created, the user enters the metaverse world and lives in the metaverse with his/her avatar.

3 illustrates a process of analyzing voice information according to an embodiment of the present invention.

Referring to FIG. 3, first, noise is canceled and normalized from voice information 205 (211-1). That is, other sounds other than the user's voice are removed from the acquired voice information 205 and the voice section is corrected to the maximum level.

Next, based on the voice information 205, voice data 211-3 that can be marked on a timeline is generated. Voice data is input data for personality prediction, and may provide data that can be marked on the time axis based on the length of the entire voice, and may be a part of a specific region of the voice information 205 or the entire region of the voice information 205. there is. The voice data 211-3 that can be marked on the timeline is, for example, whether the user speaks fast, slows down, or emphasizes speech. It can be used to distinguish through voice pattern recognition. The voice data 211-3 that can be marked on the timeline can be automatically analyzed and marked to determine which voice pattern corresponds to which point in time, rather than manually. The case where the speaker speaks quickly and then speaks slowly can be judged as a section emphasizing speech. Alternatively, even if the user speaks quickly and then speaks slowly, it may not be the section where the user emphasizes the speech, so whether the user's intonation or tone rises or the user's gaze is also considered to determine whether the user emphasizes the speech.

From a multimodal point of view, in a specific section on the timeline, not only the user's words but also behavioral data such as whether or not the user's hand has been operated and the position of the gaze can be marked along the timeline.

Next, based on the voice information 205, a voice style including at least one of a voice tempo, a voice idle time in the middle of speech, a tone, and a volume Generates (211-5). The voice style may be input data for predicting personality or input data for predicting personality in a form marked on the time axis.

4 illustrates a process of analyzing personality using voice information according to an embodiment of the present invention.

Referring to FIG. 4 , personality prediction 231-3 may be performed based on at least one of voice data 211-3 and voice style 211-5, and a machine learning (or deep learning) engine ( 231-5) predicts personality (231-3) and learns in real time. The voice personality dataset 231-7 includes at least one of data whose personality is classified according to the voice data 211-3 and data whose personality is classified according to a voice style.

When the user's personality is predicted (231-3) based on the voice data (211-3), the voice data (211-3) is used as an input and the first engine learned by learning voice data classified by personality type It is possible to predict the user's personality by (231-3). For example, a similarity may be measured for each personality type, and a personality having a high similarity may be predicted as a personality result to be recommended to a user. Based on the voice data 211-3, the predicted personality of the user is recommended to the user (250), and an avatar corresponding to the predicted personality of the user is recommended (270).

When the user's personality is predicted (231-3) based on the voice style (211-5), the average value of the voice style (211-5) for a specific part that is easy to read is input, and classified by personality type The personality of the user may be predicted by the second engine learned by the learning voice style 211-5 (231-3). For example, a similarity may be measured for each personality type, and a personality having a high similarity may be predicted as a personality result to be recommended to a user. Based on the voice style 211-5, the predicted personality of the user is recommended to the user (250), and an avatar corresponding to the predicted personality of the user is recommended (270).

As another method, when the user's personality 231-3 is predicted based on the voice style 211-5, the voice style 211-5 marked at regular time intervals is used as an input, and data for each marking or marking The personality of the user may be predicted by a third engine learned from the continuous pattern data of the interval (231-3). For example, the degree of similarity may be measured for each personality type, the measurement values of all markings may be statisticized, or continuous pattern classification may be predicted as a result of a personality to be recommended to a user.

When the user's personality is predicted (231-3) based on both the voice data (211-3) and the voice style (211-5), the predicted result by the first engine and the predicted by the second engine The result may be statistically processed to finally predict the personality to be recommended to the user, or the predicted result by the first engine and the predicted result by the third engine may be statistically processed to predict the final personality to be recommended to the user.

5 illustrates a process of analyzing visual information according to an embodiment of the present invention.

Referring to FIG. 5 , first, a user's face 270 is recognized from visual information (Face recognition) and the user's face 270 is extracted (Crop). If there are two or more people in the visual information, other people except the subject can be deleted, and parts other than the face can also be deleted.

Next, based on the time information, face data 213-3 that can be marked on a timeline as one scene (Still) of a specific image or the entire specific video (Video) is generated.

Next, based on the visual information, the shape and size of the face (Face Shape / Size), the shape and position of the eyes (Eyes Shape / pts), the shape and position of the nose (Nose Shape / pts), the shape and position of the mouth ( Mouth Shape/pts), ear shape and position (Ears Shape/pts), eyebrow shape and position (Eyebrows Shape/pts), facial expression (Facial Expression), and a facial style including at least one of a hairstyle (Hair Shape). (Face Style) (213-5) is created. In addition, the face style may further include separately recognized information such as glasses, piercings, and beards. The face style may be input data for predicting personality or input data for predicting personality in a form marked on the time axis.

6 illustrates a process of analyzing personality using visual information according to an embodiment of the present invention.

Referring to FIG. 6 , personality prediction may be performed based on at least one of face data 213-3 and face style 213-5, and a machine learning (or deep learning) engine 233-5 Through this, it is possible to predict personality and learn in real time (233-3). The face personality dataset 233-7 includes at least one of data whose personality is classified according to the face data 213-3 and data whose personality is classified according to the voice style 211-5.

In the case of predicting the user's personality based on the face data 213-3 (233-3), the fourth engine learned by the face data for learning classified by personality type using the face data 213-3 as an input It is possible to predict the personality of the user, for example, measure the similarity for each personality type, and predict the personality with a high similarity as the personality result to be recommended to the user. Based on the face data 213-3, the predicted personality of the user is recommended to the user (250), and an avatar corresponding to the predicted personality of the user is recommended (270).

When the user's personality is predicted (233-3) based on the face style 213-5, a 2D or 3D face pattern generated based on the face style of a scene of a specific image that is easy to read is input, The user's personality can be predicted by the fifth engine learned from the learning face style classified by personality type (233-3). For example, a similarity may be measured for each personality type, and a personality having a high similarity may be predicted as a personality result to be recommended to a user. Based on the face style 213-5, the predicted personality of the user is recommended (250) to the user, and an avatar corresponding to the predicted personality of the user is recommended (270).

As another method, when the user's personality is predicted (233-3) based on the face style (213-5), a 2D or 3D face pattern generated based on the face style marked by time for the entirety of a specific image is As an input, the user's personality can be predicted by the sixth engine learned from the face style for learning classified by personality type (233-3). For example, it is possible to predict a personality result to be recommended to a user by measuring similarity for each marking and making final statistics.

When the user's personality is predicted (233-3) based on both the face data (213-3) and the face style (213-5), the result predicted by the fourth engine and the predicted result by the fifth engine The result may be statistically processed to finally predict the personality to be recommended to the user, or the result predicted by the fourth engine and the predicted result by the sixth engine may be statistically processed to predict the final personality to be recommended to the user.

Although it has been described with reference to the drawings and examples above, it does not mean that the scope of protection of the present invention is limited by the drawings or examples, and those skilled in the art can understand the spirit and spirit of the present invention described in the claims below. It will be understood that the present invention can be variously modified and changed without departing from the scope.

200: artificial intelligence engine

Claims (15)

The personality analysis method based on the metaverse and artificial intelligence performed in the personality analysis system,
acquiring at least one of voice information, visual information, and bio information of a user;
analyzing the user's personality using at least one of the voice information, visual information, and bio information;
generating data for each user including a result of personality analysis of the user; and
Generating an avatar of the user using the data for each user
A personality analysis method based on metaverse and artificial intelligence, including a. According to claim 1,
Analyzing the user's personality,
analyzing the voice information;
predicting the personality of the user based on an analysis result of the voice information; and
recommending the personality of the user based on the prediction;
A personality analysis method based on metaverse and artificial intelligence, including a. According to claim 2,
Analyzing the voice information,
removing noise from the voice information and normalizing the voice information;
generating voice data that can be marked on a timeline based on the voice information; and
Generating a voice style including at least one of a tempo, an idle time, a tone, and a volume based on the voice information
A personality analysis method based on metaverse and artificial intelligence, including a. According to claim 3,
Predicting the personality of the user based on the analysis result of the voice information,
Predicting the personality of the user based on the voice data,
A personality analysis method based on metaverse and artificial intelligence, which takes the voice data as an input and predicts the personality of the user by a first engine learned by learning voice data classified by personality type. According to claim 3,
Predicting the personality of the user based on the analysis result of the voice information,
Predicting the personality of the user based on the voice style,
Based on metaverse and artificial intelligence, predicting the user's personality by the second engine learned by the learning voice style classified by personality type by taking the average value of the voice style for a specific part that is easy to read as input. Personality analysis method. According to claim 3,
Predicting the personality of the user based on the analysis result of the voice information,
Predicting the personality of the user based on the voice style,
Personality analysis based on metaverse and artificial intelligence, which predicts the user's personality by a third engine learned from the marking-specific data or continuous pattern data of the marking section by taking the voice style marked at regular time intervals as an input method. According to claim 3,
Predicting the personality of the user based on the analysis result of the voice information,
Predicting the personality of the user based on the voice data and the voice style;
A personality analysis method based on metaverse and artificial intelligence, which predicts the personality of the user by statistically processing the predicted result based on the voice data and the predicted result based on the voice style. According to claim 1,
Analyzing the user's personality,
analyzing the visual information;
predicting the personality of the user based on an analysis result of the visual information; and
recommending the personality of the user based on the prediction;
A personality analysis method based on metaverse and artificial intelligence, including a. According to claim 8,
Analyzing the visual information,
Recognizing the user's face from the visual information and extracting the user's face;
generating face data that can be marked on a timeline as one scene of a specific image or the entire specific image based on the visual information; and
Based on the visual information, at least one of the shape and size of the face, the shape and position of the eyes, the shape and position of the nose, the shape and position of the mouth, the shape and position of the ears, the shape and position of the eyebrows, the expression, and the hairstyle Steps to create a face style that includes
A personality analysis method based on metaverse and artificial intelligence, including a. According to claim 9,
Predicting the personality of the user based on the analysis result of the visual information,
Predicting the personality of the user based on the face data,
A personality analysis method based on metaverse and artificial intelligence, which takes the face data as an input and predicts the personality of the user by a fourth engine learned by face data for learning classified by personality type. According to claim 9,
Predicting the personality of the user based on the analysis result of the visual information,
Predicting the personality of the user based on the face style,
Personality of the user by the fifth engine learned by the face style for learning classified by personality type by taking as an input a 2D or 3D face pattern generated based on the face style of a scene of the specific image that is easy to read A personality analysis method based on metaverse and artificial intelligence that predicts According to claim 9,
Predicting the personality of the user based on the analysis result of the visual information,
Predicting the personality of the user based on the face style,
Personality of the user by the sixth engine learned by face styles for learning classified by personality types by taking as input a 2D or 3D face pattern generated based on the face style marked by time for the entirety of the specific image A personality analysis method based on metaverse and artificial intelligence that predicts According to claim 9,
Predicting the personality of the user based on the analysis result of the visual information,
Predicting the personality of the user based on the face data and the face style,
A personality analysis method based on metaverse and artificial intelligence, which predicts the personality of the user by statistically processing the predicted result based on the face data and the predicted result based on the face style. According to claim 1,
Further comprising the step of updating the user-specific data including the user's personality analysis result in real time or periodically by reflecting the generated avatar's actions, conversations, words or writings in the metaverse, metaverse and artificial intelligence Personality analysis method based on. According to claim 1,
In terms of multimodal behavioral patterns, words, conversations, or writings used by the created avatar in the metaverse and the behavior of the avatar—the behavior includes the movements of the avatar’s hands, head, and gaze—are considered Personality analysis method based on metaverse and artificial intelligence, further comprising the step of analyzing at least one of the interests and tendencies of the user using the avatar.

Images