System and method for providing comprehensive consultation service in various fields on basis of virtual environment by using generative artificial intelligence

Abstract

A system for providing a comprehensive consultation service in various fields on basis of virtual reality is disclosed. A consultation service providing system according to an embodiment includes: a user terminal worn by a user; and a server for receiving a query of the user from the user terminal and providing virtual content to the user terminal in response to receiving the query. The server includes: a query analysis module for receiving user information and a user query of the user from the user terminal and determining at least one target consultation field corresponding to the received user query; and a virtual content providing module for generating virtual content corresponding to the target consultation field and providing the virtual content to the user terminal in response to receiving the target consultation field from the query analysis module. The virtual content includes a user query and an answer to the target consultation field generated from user information by using a generative artificial intelligence model corresponding to the target consultation field among a plurality of generative artificial intelligence models trained on the basis of training data of different consultation fields, and includes a virtual environment configured using a virtual space corresponding to the target consultation field.

Description

A system and method for providing a multi-disciplinary comprehensive consulting service based on a virtual environment using generative artificial intelligence.

[0001] The examples below relate to a technology for providing a multi-disciplinary comprehensive consulting service based on a virtual environment using generative artificial intelligence.

[0002] In the past, clients had to visit a specialist in their field to receive counseling, but with the recent development of virtual reality and metaverse technologies, technologies that provide counseling spaces through these technologies are emerging.

[0003] As an example of related prior art, Korean Patent No. 10-2406285 provides a system that enables real-time consultations between clients and counselors in a virtual environment via the metaverse. However, counselors are still required to conduct consultations.

[0004] Meanwhile, with the recent development of generative artificial intelligence technology, AI-based chatbots are being introduced.

[0005] The examples are intended to provide a multi-disciplinary, comprehensive consulting service based on a virtual environment using generative artificial intelligence. However, the technical challenges are not limited to the aforementioned technical challenges, and other technical challenges may also exist.

[0006] A system for providing a comprehensive counseling service in various fields based on virtual reality according to one embodiment includes: a user terminal worn by a user; and a server for receiving a user's query from the user terminal and providing virtual content to the user terminal in response to receiving the query; the server includes: a query analysis module for receiving user information and a user query from the user terminal and determining at least one target counseling field corresponding to the received user query; and a virtual content provision module for generating virtual content corresponding to the target counseling field and providing the virtual content to the user terminal in response to receiving the target counseling field from the query analysis module, wherein the virtual content includes an answer to the target counseling field generated from the user query and the user information using a generative artificial intelligence model corresponding to the target counseling field among a plurality of generative artificial intelligence models learned based on learning data of different counseling fields, and a virtual environment configured using a virtual space corresponding to the target counseling field.

[0007] The above virtual content provision module may be characterized in that it determines whether there are multiple target consultation fields determined by the query analysis module, and, based on the result of the determination of whether there are multiple fields, selects a generative artificial intelligence model used to generate the answer from among the plurality of generative artificial intelligence models, and configures the virtual environment in a different manner.

[0008] The virtual content providing module, in response to determining that the plurality of target counseling fields have been determined, generates an answer for each target counseling field from the user query and the user information using different generative artificial intelligence models corresponding to each of the plurality of target counseling fields among the plurality of generative artificial intelligence models, configures a first virtual environment by connecting different virtual spaces corresponding to each of the target counseling fields, and generates the virtual content including the answers for each of the generated target counseling fields and the first virtual environment, and in response to determining that one target counseling field has been determined, generates a first answer for the target counseling field from the user query and the user information using a generative artificial intelligence model corresponding to the target counseling field among the plurality of generative artificial intelligence models, and generates a query for the user in the other field using a generative artificial intelligence model corresponding to another field preset to be highly related to the target counseling field among the plurality of generative artificial intelligence models, and configures a second virtual environment using a virtual space corresponding to the target counseling field, and, the virtual environment including the first answer, the query for the user, and the second virtual environment. It can be characterized by creating content.

[0009] The generating of an answer for each target counseling field from the user query and the user information may include inputting the user query and the user information into a first generative artificial intelligence model corresponding to a first counseling field having the highest priority among the plurality of target counseling fields to generate a second answer, and inputting the user query, the user information, and the second answer into a second generative artificial intelligence model corresponding to a second counseling field having a lower priority than the first counseling field among the plurality of target counseling fields to generate a third answer based on the premise that the second answer to the user query has been provided to the user, and the generating of the query for the user may include inputting the user query, the user information, and the first answer into a third generative artificial intelligence model corresponding to another field preset as having a high relevance to the one target counseling field to generate a query for the user including a suggestion for a query in the other field.

[0010] The above query analysis module may be characterized in that it determines a score for each of a plurality of counseling fields based on the user query, and determines counseling fields having a score greater than a threshold value based on the determined score as at least one target counseling field.

[0011] The above virtual content provision module may be characterized by determining a regional dialect corresponding to an answer in the target counseling field based on the gender, age, region, language, and voice data of the user query included in the user information, and generating voice data corresponding to the answer based on the determined regional dialect.

[0012] The first virtual environment may be characterized by including avatars that are each placed in a different virtual space and speak answers in a target counseling field corresponding to each virtual space, and including an answer scenario composed of a pair of identification information of an avatar to speak and a speaking order for each of the second and third answers.

[0013] The above query analysis module may further receive a user addition query from the user, and re-determine at least one target consultation field from the user addition query, and the virtual content provision module may determine whether the re-determined target consultation field is included in the target consultation field determined for the user query, and may change or maintain a virtual space within the virtual content according to the determination of whether it is included, thereby generating virtual content for the user addition query.

[0014] The re-determining of at least one target consultation field from the user-added query may be characterized in that it is determined by considering the similarity between the query or answer included in the virtual content and the user-added query.

[0015] A method for providing a comprehensive counseling service in various fields based on virtual reality performed by a server according to one embodiment includes the steps of: receiving user information and a user query from a user terminal; determining at least one target counseling field corresponding to the received user query; and generating virtual content corresponding to the target counseling field and providing the virtual content to the user terminal in response to receiving the target counseling field from the query analysis module, wherein the virtual content includes an answer to the target counseling field generated from the user query and the user information using a generative artificial intelligence model corresponding to the target counseling field among a plurality of generative artificial intelligence models learned based on learning data of different counseling fields, and may include a virtual environment configured using a virtual space corresponding to the target counseling field.

[0016] Embodiments can realistically provide users with answers to their questions from various fields in a virtual environment in the form of an immersive experience.

[0017] FIG. 1 is a diagram illustrating a virtual environment for providing a counseling service according to one embodiment.

[0018] Figure 2 is a diagram schematically illustrating the configuration of a system according to one embodiment.

[0019] Figure 3 is a flowchart of a method for providing a comprehensive counseling service according to one embodiment.

[0020] FIG. 4 is a drawing for explaining a method for creating a virtual space according to one embodiment.

[0021] FIG. 5 is a flowchart of a method for generating answers in multiple fields according to one embodiment.

[0022] Specific structural or functional descriptions of embodiments according to the concept of the present invention disclosed in this specification are merely illustrative for the purpose of explaining embodiments according to the concept of the present invention, and embodiments according to the concept of the present invention may be implemented in various forms and are not limited to the embodiments described in this specification.

[0023] Embodiments according to the concept of the present invention may have various modifications and take various forms, and thus, embodiments are illustrated in the drawings and described in detail in this specification. However, this is not intended to limit embodiments according to the concept of the present invention to specific disclosed forms, but rather includes modifications, equivalents, or alternatives that fall within the spirit and technical scope of the present invention.

[0024] While terms such as "first" or "second" may be used to describe various components, these components should not be limited by these terms. These terms are intended solely to distinguish one component from another. For example, a first component may be referred to as a "second component," and similarly, a second component may also be referred to as a "first component," without departing from the scope of the invention.

[0025] When a component is referred to as being "connected" or "connected" to another component, it should be understood that it may be directly connected or connected to that other component, but that there may be other components in between. Conversely, when a component is referred to as being "directly connected" or "directly connected" to another component, it should be understood that there are no other components in between. Expressions that describe relationships between components, such as "between," "immediately between," or "directly adjacent to," should be interpreted similarly.

[0026] The terminology used herein is for the purpose of describing specific embodiments only and is not intended to limit the present invention. The singular expressions include plural expressions unless the context clearly indicates otherwise. In this specification, it should be understood that the terms "comprises" or "has" are intended to specify the presence of a described feature, number, step, operation, component, part, or combination thereof, but do not preclude the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

[0027] Unless otherwise defined, 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 this invention pertains. Terms defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and will not be interpreted in an idealized or overly formal sense unless explicitly defined herein.

[0028] A module in this specification may mean hardware capable of performing functions and operations according to each name described in this specification, may mean computer program code capable of performing specific functions and operations, or may mean an electronic recording medium, such as a processor or microprocessor, on which computer program code capable of performing specific functions and operations is loaded.

[0029] In other words, a module may mean a functional and / or structural combination of hardware for carrying out the technical idea of ​​the present invention and / or software for driving the hardware.

[0030] Hereinafter, embodiments will be described in detail with reference to the attached drawings. However, the scope of the patent application is not limited or restricted by these embodiments. The same reference numerals in each drawing represent the same components.

[0031] FIG. 1 is a diagram illustrating a virtual environment for providing a counseling service according to one embodiment.

[0032] Referring to Figure 1, an avatar (105) for counseling and a user avatar (110) located in a virtual space (115) within a virtual environment (100) are illustrated. Currently, users seeking counseling must visit a counselor in real life to receive counseling, which hinders accessibility to counseling services. In particular, many people, such as those living in rural or island areas or those with disabilities, find it difficult to utilize counseling services in real life.

[0033] According to a system for providing a multi-disciplinary comprehensive counseling service based on a virtual environment (100) using a generative artificial intelligence of one embodiment, a user can receive an immersive and natural counseling service online within a virtual environment (100) as shown in FIG. 1. The user can access the virtual environment (100) as shown in FIG. 1 using a virtual reality device and receive the counseling service.

[0034] Simply using text for consultations makes it difficult to provide in-depth consultations because information beyond the text cannot be obtained. According to a comprehensive consultation service system according to one embodiment, various information, such as the user's behavior and voice, can be obtained during the process of providing consultation services through a virtual environment (100) as shown in FIG. 1, and this information can be used to provide more realistic feedback to the user.

[0035] When counseling is conducted by real-life counselors, time and personnel are limited. As the number of people seeking counseling increases, the quality of the consultation deteriorates, and the time required to receive counseling increases, leading to inefficiencies. According to one embodiment, a comprehensive counseling service system utilizes generative artificial intelligence to conduct counseling, thereby increasing accessibility and efficiency.

[0036] Furthermore, in reality, conducting comprehensive counseling with counselors from various fields is often more challenging. According to one embodiment, a comprehensive counseling service system can provide comprehensive answers to a user's inquiry across various fields based on multiple generative AI models trained on learning data from different fields.

[0037] Since the counseling service is provided through a virtual environment (100), the service must operate naturally, as if conversing with a real person, so that users can have confidence in the counseling results. According to an exemplary embodiment, a comprehensive counseling service provision system can generate natural answers by strengthening the connectivity between multiple generative artificial intelligence models, provide a virtual environment (100) including a virtual space (115) suitable for the user's inquiry, and provide a more realistic experience by utilizing user information.

[0038] Figure 2 is a diagram schematically illustrating the configuration of a system according to one embodiment.

[0039] Referring to FIG. 2, a user terminal (205) worn by a user to receive counseling services and a server (200) providing counseling services are illustrated. The server (200) receives a user's query from the user terminal (205) and, in response to the query, provides virtual content to the user terminal (205), thereby providing the counseling service. In one embodiment, the server (200) may include a query analysis module (210) and a virtual content provision module (215). The server (200) may further include a database (220).

[0040] In one embodiment, the server (200) may be a computer device including a processor (not shown) and a memory (not shown) that stores instructions to be executed by the processor (not shown). The processor may be a hardware-implemented data processing device having a circuit having a physical structure for executing desired operations. For example, the desired operations may include code or instructions included in a program. For example, the hardware-implemented data processing device may include a microprocessor, a central processing unit, a processor core, a multi-core processor, a multiprocessor, an application-specific integrated circuit (ASIC), or a field programmable gate array (FPGA).

[0041] The memory may be implemented as a volatile memory device or a nonvolatile memory device. The volatile memory device may be implemented as a dynamic random access memory (DRAM), a static random access memory (SRAM), a thyristor RAM (T-RAM), a zero capacitor RAM (Z-RAM), or a twin transistor RAM (TTRAM). The nonvolatile memory device can be implemented as an Electrically Erasable Programmable Read-Only Memory (EEPROM), flash memory, Magnetic RAM (MRAM), Spin-Transfer Torque (STT)-MRAM, Conductive Bridging RAM (CBRAM), Ferroelectric RAM (FeRAM), Phase change RAM (PRAM), Resistive RAM (RRAM), Nanotube RRAM, Polymer RAM (PoRAM), Nano Floating Gate Memory (NFGM), holographic memory, Molecular Electronic Memory Device, or Insulator Resistance Change Memory.

[0042] The query analysis module (210) and the virtual content provision module (215) may refer to functional units performed by a processor. That is, the query analysis module (210) and the virtual content provision module (215) may refer to computer program codes, and the functions performed by the query analysis module (210) and the virtual content provision module (215) may all be performed by the processor. Alternatively, the server (200) may include hardware for each of the query analysis module (210) and the virtual content provision module (215). In this case, the query analysis module (210) and the virtual content provision module (215) may each include a processor and a memory. Hereinafter, a method for providing a comprehensive counseling service according to one embodiment will be described with further reference to FIG. 3.

[0043] Figure 3 is a flowchart of a method for providing a comprehensive counseling service according to one embodiment.

[0044] In step (305), the query analysis module (210) may receive a user query and user information from the user. The user information may include the user's identification information, gender, age, region of residence, and language used by the user. The user information may be stored in a database (220) and transmitted to the virtual content provision module (215) to be used for virtual content generation.

[0045] In step (310), the query analysis module (210) may determine at least one target counseling field corresponding to the received user query to provide comprehensive counseling across various fields. In one embodiment, the fields of counseling services that the server (200) may provide may be pre-determined. For example, the server (200) may provide counseling services in the fields of psychology, education, law, and health, but is not limited thereto.

[0046] In one embodiment, the query analysis module (210) may receive a user query and determine a score for the input user query in relation to a plurality of preset consultation fields. The query analysis module (210) may determine the score using, for example, a query analysis artificial intelligence model trained to output scores for the input user query in relation to a plurality of preset consultation fields, but is not limited thereto.

[0047] Throughout this disclosure, an artificial intelligence model may refer to a deep neural network model. A deep neural network may refer to a neural network that includes multiple hidden layers in addition to input and output layers. Using a deep neural network, latent structures in data can be identified. For example, the latent structures of images, text, video, speech, protein sequence structures, gene sequence structures, peptide sequence structures, and / or music can be identified through a deep neural network.

[0048] In the present disclosure, the AI ​​model may be trained using at least one of supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning. Training the AI ​​model may be a process of applying knowledge to the AI ​​model to perform a specific action.

[0049] The query analysis module (210) can determine at least one target counseling area from the determined score. For example, the query analysis module (210) can determine all counseling areas with a score above a threshold value as target counseling areas. Accordingly, one or more target counseling areas can be determined.

[0050] In step (315), the virtual content provision module (215) can generate virtual content corresponding to the target counseling field in response to receiving the target counseling field from the query analysis module (210) and provide the virtual content to the user terminal (205).

[0051] The virtual content provision module (215) can generate an answer for the target counseling field from a user query and user information by using a generative artificial intelligence model corresponding to the target counseling field among a plurality of generative artificial intelligence models learned based on learning data of different counseling fields.

[0052] In one embodiment, the virtual content provision module (215) can check whether the user has previously received a consultation. The virtual content provided by the virtual content provision module (215) can be stored in a database (220) in conjunction with user information, and the virtual content provision module (215) can compare the database (220) with the user information to check whether the user has previously received a consultation. If the user has a consultation history, the virtual content provision module (215) can input the user query included in the virtual content provided during the previous consultation process and the answer generated by the virtual content provision module (215) into the generative artificial intelligence model, along with the user query and user information, only once when generating an answer to the current user query. As a result, the generative artificial intelligence model can generate an answer to the current user query by considering the previous consultation history.

[0053] If the virtual content provision module (215) provides answers for all consultation fields using a single generative artificial intelligence model, the accuracy of the answers may be reduced. In one embodiment, the server (200) may store generative artificial intelligence models trained differently based on learning data for each field in a database (220), and may use the generative artificial intelligence model corresponding to a user query when generating virtual content. Here, the generative artificial intelligence model may be a generative language model such as a Generative Pre-trained Transformer (GPT), but is not limited thereto. For example, the database (220) may store generative artificial intelligence models trained based on learning data for each field, or may store generative artificial intelligence models that are fine-tuned based on learning data for each field, and generative artificial intelligence models that are pre-trained based on other data.

[0054] The data input to generative artificial intelligence models may include not only user queries but also user information. The training data used for training or fine-tuning the generative artificial intelligence models may include information about the user who performed the question-and-answer session (e.g., information about the user's gender, age, region of residence, and language used by the user) for each consultation field's question-and-answer session. The generative artificial intelligence models stored in the database (220) are trained or fine-tuned using training data containing user information, thereby learning the inherent correlation between user information and user queries, and can infer answers to user queries based on the learned information.

[0055] The virtual content provision module (215) can configure a virtual environment using virtual spaces corresponding to target counseling fields. For example, base virtual spaces for each of multiple counseling fields may be stored in the database (220). The virtual content provision module (215) can use the stored virtual spaces to create a virtual space appropriate for a user's query.

[0056] Referring to FIG. 4, base virtual spaces (405, 410, 415, 420) for each of a plurality of counseling fields are illustrated. For example, the counseling fields that the server (200) can provide may be psychology, education, law, and health, and each base virtual space (405, 410, 415, 420) may be a three-dimensionally modeled virtual space with the concept of a psychological counseling room, an educational counseling room, a legal counseling room, and a health counseling room. The virtual content providing module (215) may create a virtual space (425) to be included in the virtual content by combining the base virtual spaces (405, 410, 415, 420). For example, the virtual content providing module (215) may create a virtual space (425) by combining two or more virtual spaces corresponding to a target counseling field among the first virtual space, the second virtual space, the third virtual space, and the fourth virtual space. For example, the virtual content provision module (215) can create a virtual space (425) by connecting two or more virtual spaces corresponding to a target counseling field among the first virtual space, the second virtual space, the third virtual space, and the fourth virtual space.

[0057] As another example, the base virtual spaces (405, 410, 415, 420) may include combination rules that allow them to be combined with each other. The virtual content provision module (215) may combine two or more virtual spaces based on the combination rules included in the base virtual spaces (405, 410, 415, 420) and the target consultation field determined by the query analysis module (210) to configure a virtual space (425) to be included in the virtual content.

[0058] The virtual content provision module (215) may consider keywords of a user query when combining base virtual spaces (405, 410, 415, 420). Keywords of a user query may be determined by a query analysis module (210). For example, the virtual content provision module (215) may modify at least a portion of the base virtual spaces (405, 410, 415, 420) based on keywords of a user query and then combine two or more virtual spaces corresponding to a target counseling field. For example, the virtual content provision module (215) may change at least one virtual object of the base virtual spaces (405, 410, 415, 420) into a virtual object corresponding to keywords of a user query and then combine two or more virtual spaces corresponding to a target counseling field. By reflecting keywords of a user query in the virtual space (425), the reliability of a user experiencing the virtual space may be increased.

[0059] When a virtual space (425) is created in response to a user query, the virtual content provision module (215) can create a virtual environment using the virtual space (425). The virtual environment can include the virtual space (425) created in response to the user query and at least one avatar.

[0060] Returning to Figure 3, the virtual content provision module (215) can generate virtual content including the target counseling field's answer and virtual environment from the user query and the user information at step (315). The generated virtual content can be stored in the database (220) in conjunction with the user information.

[0061] In step (320), the virtual content provision module (215) can transmit the generated virtual content to the user terminal (205). When the virtual content is transmitted to the user terminal (205) and output, the user can hear an answer to the user's query through an avatar speaking within the virtual environment.

[0062] The virtual content provision module (215) can generate voice data for a user query and an answer in a target counseling field generated from the user information, and include the generated voice data in the virtual content.

[0063] The virtual content provision module (215) determines a regional dialect corresponding to a response in the target consultation field based on the user information, including gender, age, region, language, and voice data of the user's query, and generates voice data corresponding to the response based on the determined regional dialect. This allows voice data included in the virtual content to be delivered to the user in a more familiar form, thereby enhancing the user experience.

[0064] For example, the virtual content provision module (215) can determine whether the user uses a regional dialect from the user's voice data. If the user is determined to use a regional dialect, the gender and age of the user information satisfy preset conditions, and the region of the user information is the same as the region corresponding to the regional dialect of the user's query, the virtual content provision module (215) can generate voice data of the regional dialect from the response in the target consultation field. If the user is determined not to use a regional dialect, or the gender and age of the user information do not satisfy preset conditions, or the region of the user information is not the same as the region corresponding to the regional dialect of the user's query, the virtual content provision module (215) can generate voice data of the standard language of the language from the response in the target consultation field.

[0065] FIG. 5 is a flowchart of a method for generating answers in multiple fields according to one embodiment.

[0066] Steps (505, 510, 515, 520, 525, 530, 535) of FIG. 5 may be included in step (315) of FIG. 3.

[0067] In step (505), the virtual content provision module (215) can determine whether there are multiple target counseling fields determined by the query analysis module (210). Based on the determination result of whether there are multiple fields, the virtual content provision module (215) can select a generative artificial intelligence model among the multiple generative artificial intelligence models to be used for generating answers to user queries and can configure a virtual environment in different ways. For example, if there are multiple target counseling fields, a generative artificial intelligence model corresponding to each target counseling field may be selected, but if there is only one target counseling field, in addition to the generative artificial intelligence model corresponding to the target counseling field, a generative artificial intelligence model corresponding to another field that is preset to be highly related to the target counseling field may be further selected.

[0068] In response to determining that multiple target counseling fields have been determined in step (505), the virtual content provision module (215) can, in step (510), generate answers for each target counseling field from the user query and user information using different generative artificial intelligence models corresponding to each target counseling field among the multiple generative artificial intelligence models.

[0069] In one embodiment, generating an answer for each target counseling field from a user query and user information may include inputting the user query and user information into a first generative artificial intelligence model corresponding to a first counseling field having the highest priority among a plurality of target counseling fields to generate a second answer, and inputting the user query, user information, and second answer into a second generative artificial intelligence model corresponding to a second counseling field having a lower priority than the first counseling field among the plurality of target counseling fields to generate a third answer on the premise that the second answer to the user query has been provided to the user. By inputting the output data of the first generative artificial intelligence model into the second generative artificial intelligence model to generate an answer, the connectivity of each generative artificial intelligence model is strengthened, so that more natural answers can be generated.

[0070] In step (515), the virtual content provision module (215) may configure a first virtual environment using different virtual spaces corresponding to each target counseling field. For example, the virtual content provision module (215) may configure the first virtual environment by connecting different virtual spaces corresponding to each target counseling field. As another example, as described above, the virtual content provision module (215) may configure the first virtual environment by combining two or more virtual spaces based on combination rules included in base virtual spaces (e.g., 405, 410, 415, 420 of FIG. 4) and the target counseling field.

[0071] In one embodiment, the first virtual environment may include avatars, each positioned in a different virtual space connected within the first virtual environment, that utter answers corresponding to the target counseling field in each virtual space. Avatars corresponding to each target counseling field are positioned within a single virtual environment, and answers for the corresponding target counseling field are delivered to the user through each avatar. This allows the user to clearly recognize which field the answer they are currently hearing is from during a comprehensive counseling process across multiple fields, and provides an experience similar to receiving counseling from an actual expert.

[0072] The first virtual environment may include a response scenario, consisting of pairs of avatar identification information for each response in each target counseling area and the order in which responses are to be delivered in each target counseling area. By including the response scenarios in the virtual content provision module (215), comprehensive counseling across various fields can be conducted naturally, as if face-to-face with an actual expert.

[0073] In step (520), the virtual content provision module (215) can generate virtual content including answers for each generated target counseling field and the first virtual environment.

[0074] In response to determining that a target counseling field has been determined in step (505), the virtual content provision module (215) can, in step (525), generate a first answer in the target counseling field from the user query and user information using a generative artificial intelligence model corresponding to the target counseling field among a plurality of generative artificial intelligence models.

[0075] In step (530), the virtual content provision module (215) can generate a query for a user in another field by using a generative artificial intelligence model corresponding to another field that is preset to have a high relevance to the target counseling field among a plurality of generative artificial intelligence models.

[0076] For example, generating a query for a user may include inputting the user query, user information, and a first answer into a third generative artificial intelligence model corresponding to another field that is pre-determined to be highly relevant to one target counseling field, thereby generating a query for the user that includes suggestions for the query in the other field.

[0077] The virtual content provision module (215) can recommend additional inquiries about other consultation areas that the user was not aware of, and the user can make additional inquiries based on these inquiries and receive more comprehensive consultation services.

[0078] The virtual content provision module (215) can configure a second virtual environment using a virtual space corresponding to a target counseling field. Unlike the first virtual environment, the second virtual environment can be configured with only a virtual space corresponding to a single target counseling field.

[0079] The virtual content provision module (215) can generate virtual content including the first answer generated in step (525), the query for the user generated in step (530), and the second virtual environment generated in step (535), in step (535).

[0080] The second virtual environment may include an avatar that utters answers in the target counseling field, and may include an answer scenario consisting of a first answer generated in step (525) and an order of utterance of a question to the user generated in step (530).

[0081] In one embodiment, the query analysis module (210) may receive an additional user query from the user terminal (205) after the virtual content is transmitted to the user terminal (205) (e.g., step (320) of FIG. 3).

[0082] The query analysis module (210) may re-determine at least one target counseling field from the user's additional query. When re-determining at least one target counseling field, the query analysis module (210) may determine the target counseling field by further considering the similarity between the user's additional query and at least one of the answers (e.g., the first answer, the second answer, and the third answer) or the query (the user's query of step (305) of FIG. 3, the query to the user of step (530) of FIG. 5) included in the virtual content provided immediately before, in addition to the scores for the preset multiple counseling fields.

[0083] The virtual content provision module (215) can generate an answer to a user additional query through steps (505, 510 or 505, 525) of FIG. 5 in providing virtual content for a user additional query.

[0084] The virtual content provision module (215) may, in providing virtual content for additional user queries, determine whether the target counseling field re-determined by the query analysis module (210) for the additional user query is included in the target counseling field determined for the previous user query, and may generate virtual content for the additional user query by changing or maintaining the virtual space within the previous virtual content based on the determination of whether or not it is included. For example, if the target counseling field re-determined by the query analysis module (210) is included in the target counseling field determined for the previous user query, the virtual content provision module (215) may generate virtual content for the additional user query by maintaining the virtual space within the previous virtual content, and if the target counseling field re-determined by the query analysis module (210) is not included in the target counseling field determined for the previous user query, the virtual content provision module (215) may generate virtual content for the additional user query by changing the virtual space within the previous virtual content.

[0085] Through the above embodiments, the virtual content provision module (215) not only generates a second answer through the first generative artificial intelligence model of the first counseling field with the highest priority as described above, and inputs this into the second generative artificial intelligence model to generate a third answer, but also inputs the third answer again into the first generative artificial intelligence model to generate a fourth answer, and inputs the fourth answer again into the second generative artificial intelligence model to obtain and provide a fifth answer, thereby performing a recursive operation to provide the user with more in-depth answers in each field, and the above recursive operation can be repeatedly performed until the user requests to stop.

[0086] For example, if a user inputs a query regarding whether the use of a specific technology constitutes a patent infringement issue, the virtual content provision module (215) can input the query into a first generative artificial intelligence model pre-trained in the patent field and provide a response regarding the judgment result regarding infringement. Furthermore, the virtual content provision module (215) can input the existing query and the response of the first generative artificial intelligence model into a second generative artificial intelligence model pre-trained in a legal field other than patents, thereby generating an additional second response and providing it to the user. Through additional review by the second generative artificial intelligence model, a comprehensive judgment can be provided regarding not only patent infringement issues but also violations of the Unfair Competition Prevention Act and copyright infringement.

[0087] According to another embodiment, when a question about points to note in knee surgery is input, the virtual content provision module (215) may input a first generative artificial intelligence model pre-trained in the field of orthopedics to generate a first answer about points to note in orthopedics, input the first answer into a second generative artificial intelligence model pre-trained in the field of internal medicine to generate a second answer about points to note in internal medicine, input the generated first and second answers into a third generative artificial intelligence model pre-trained in the field of law to generate a third answer about points to note in medical accidents during surgery and provide the third answer to the user, and input the third answer again into the first generative artificial intelligence model and the second generative artificial intelligence model to additionally provide the user with a fourth answer about orthopedic and internal medicine checkpoints related to medical accidents.

[0088] The devices described above may be implemented as hardware components, software components, and / or a combination of hardware components and software components. For example, the devices and components described in the embodiments may be implemented using one or more general-purpose computers or special-purpose computers, such as, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA), a programmable logic unit (PLU), a microprocessor, or any other device capable of executing instructions and responding to them. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. The processing device may also access, store, manipulate, process, and generate data in response to the execution of the software. For ease of understanding, the processing device is sometimes described as being used alone; however, one of ordinary skill in the art will recognize that the processing device may include multiple processing elements and / or multiple types of processing elements. For example, a processing unit may include multiple processors, or a processor and a controller. Other processing configurations, such as parallel processors, are also possible.

[0089] Software may include a computer program, code, instructions, or a combination of one or more of these, and may configure a processing device to perform a desired operation or, independently or collectively, command the processing device. The software and / or data may be permanently or temporarily embodied in any type of machine, component, physical device, virtual equipment, computer storage medium or device, or transmitted signal wave, for interpretation by the processing device or for providing instructions or data to the processing device. The software may also be distributed over networked computer systems and stored or executed in a distributed manner. The software and data may be stored on one or more computer-readable recording media.

[0090] The method according to the embodiment may be implemented in the form of program commands that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program commands, data files, data structures, etc., alone or in combination. The program commands recorded on the medium may be those specially designed and configured for the embodiment or may be those known and available to those skilled in the art of computer software. Examples of the computer-readable recording medium include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, and hardware devices specially configured to store and execute program commands, such as ROMs, RAMs, and flash memories. Examples of the program commands include not only machine language codes generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter, etc. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiment, and vice versa.

[0091] Although the embodiments described above have been described by way of limited examples and drawings, those skilled in the art will appreciate that various modifications and variations can be made based on the above teachings. For example, appropriate results can still be achieved even if the described techniques are performed in a different order than described, and / or components of the described systems, structures, devices, circuits, etc. are combined or combined in a different manner than described, or are replaced or substituted with other components or equivalents.

[0092] Therefore, other implementations, other embodiments, and equivalents to the claims also fall within the scope of the claims described below.

Claims

1. In a system that provides comprehensive consulting services in various fields based on virtual reality, a user terminal worn by the user; and A server that receives a user's query from the user terminal and provides virtual content to the user terminal in response to receiving the query; Including, The above server, A query analysis module that receives user information and user queries of the user from the user terminal and determines at least one target consultation field corresponding to the received user query; and In response to receiving the target consultation field from the query analysis module, a virtual content provision module is included that generates the virtual content corresponding to the target consultation field and provides it to the user terminal. The above virtual content is, A method of generating a response to a target counseling field generated from a user query and user information using a generative artificial intelligence model corresponding to the target counseling field among multiple generative artificial intelligence models learned based on learning data of different counseling fields, and including a virtual environment configured using a virtual space corresponding to the target counseling field. The above virtual content provision module, The method of determining whether there are multiple target consultation fields determined by the above query analysis module and selecting a generative artificial intelligence model used to generate the answer among the multiple generative artificial intelligence models based on the result of determining whether there are multiple target consultation fields and the method of configuring the virtual environment are different. The above virtual content provision module, In response to determining that the plurality of target counseling fields are determined, different generative artificial intelligence models corresponding to each of the plurality of target counseling fields are used among the plurality of generative artificial intelligence models to generate answers for each target counseling field from the user query and the user information, different virtual spaces corresponding to each of the plurality of target counseling fields are connected to form a first virtual environment, and the virtual content including the answers for each of the generated target counseling fields and the first virtual environment is generated, and in response to determining that one target counseling field is determined, a generative artificial intelligence model corresponding to the target counseling field among the plurality of generative artificial intelligence models is used to generate a first answer for the target counseling field from the user query and the user information, and a generative artificial intelligence model corresponding to another field preset as having a high relevance to the target counseling field among the plurality of generative artificial intelligence models is used to generate a query for the user in the other field, and a second virtual environment is used to form the virtual space corresponding to the target counseling field, and the virtual content including the first answer, the query for the user, and the second virtual environment is generated, Generating answers for each target consultation field from the above user queries and user information is as follows: A method of generating a second answer by inputting the user query and the user information into a first generative artificial intelligence model corresponding to a first counseling field having the highest priority among the plurality of target counseling fields, and then generating a third answer on the premise that the second answer to the user query has been provided to the user by inputting the user query, the user information, and the second answer into a second generative artificial intelligence model corresponding to a second counseling field having a lower priority than the first counseling field among the plurality of target counseling fields. Generating a query for the above user, A system characterized in that the system generates a query for the user including a suggestion for the query in the other field by inputting the user query, the user information, and the first answer into a third generative artificial intelligence model corresponding to another field that is preset as having a high relevance to the one target counseling field.

2. In paragraph 1, The above query analysis module, Based on the above user query, a score is determined for each of a plurality of counseling fields, and counseling fields having a score higher than a threshold value based on the determined score are determined as at least one target counseling field. The above virtual content provision module, Based on the gender, age, region, language, and voice data of the user's query included in the user information, a regional dialect corresponding to the answer in the target consultation field is determined, and voice data corresponding to the answer is generated based on the determined regional dialect. The above first virtual environment is, A system characterized in that it includes avatars that are respectively placed in the different virtual spaces and speak answers in target counseling fields corresponding to each virtual space, and includes an answer scenario composed of a pair of identification information of the avatar to speak and the speaking order for each of the second and third answers.

3. In paragraph 1, The above query analysis module, Receive additional user queries from the user, and re-determine at least one target consultation field from the user query, The above virtual content provision module, Determine whether the above re-determined target consultation field is included in the target consultation field determined for the above user query, It is characterized in that the virtual space within the virtual content is changed or maintained according to the determination of whether the above is included or not to generate the virtual content for the additional user query, and the at least one target consultation field is determined again from the additional user query. A system characterized in that the determination is made by considering the similarity between the query or answer included in the virtual content and the user's additional query.

4. A method for providing a comprehensive consulting service in various fields based on virtual reality performed by a server, A step of receiving user information and user query from a user terminal; A step of determining at least one target consultation field corresponding to the received user query; In response to receiving the target consultation field from the query analysis module, a step of generating virtual content corresponding to the target consultation field and providing it to the user terminal is included. The above virtual content is, A method of generating a response to a target counseling field generated from a user query and user information using a generative artificial intelligence model corresponding to the target counseling field among multiple generative artificial intelligence models learned based on learning data of different counseling fields, and including a virtual environment configured using a virtual space corresponding to the target counseling field. The steps provided above are: The method of determining whether there are multiple target consultation fields determined by the above query analysis module and selecting a generative artificial intelligence model used to generate the answer among the multiple generative artificial intelligence models based on the result of determining whether there are multiple target consultation fields and the method of configuring the virtual environment are different. The steps provided above are: In response to determining that the plurality of target counseling fields are determined, different generative artificial intelligence models corresponding to each of the plurality of target counseling fields are used among the plurality of generative artificial intelligence models to generate answers for each target counseling field from the user query and the user information, different virtual spaces corresponding to each of the plurality of target counseling fields are connected to form a first virtual environment, and the virtual content including the answers for each of the generated target counseling fields and the first virtual environment is generated, and in response to determining that one target counseling field is determined, a generative artificial intelligence model corresponding to the target counseling field among the plurality of generative artificial intelligence models is used to generate a first answer for the target counseling field from the user query and the user information, and a generative artificial intelligence model corresponding to another field preset as having a high relevance to the target counseling field among the plurality of generative artificial intelligence models is used to generate a query for the user in the other field, and a second virtual environment is used to form the virtual space corresponding to the target counseling field, and the virtual content including the first answer, the query for the user, and the second virtual environment is generated, Generating answers for each target consultation field from the above user queries and user information is as follows: A method of generating a second answer by inputting the user query and the user information into a first generative artificial intelligence model corresponding to a first counseling field having the highest priority among the plurality of target counseling fields, and then generating a third answer on the premise that the second answer to the user query has been provided to the user by inputting the user query, the user information, and the second answer into a second generative artificial intelligence model corresponding to a second counseling field having a lower priority than the first counseling field among the plurality of target counseling fields. Generating a query for the above user, A method characterized in that the user query, the user information, and the first answer are input into a third generative artificial intelligence model corresponding to another field that is preset as having a high relevance to the one target counseling field, thereby generating a query for the user that includes a suggestion for the query in the other field.

Images