Server for communicating with electronic device and ai server

Abstract

A server is provided for communicating with electronic devices and artificial intelligence (AI) servers. More specifically, an interactive system is provided that assists a user in resolving his / her curiosity related to the details of the media, understanding the media, or successfully performing the activity when the user reads the media or performs a defined activity through the media.

Description

Interactive systems that assist users in using media This disclosure relates to an interactive system that assists users in using media. More specifically, this disclosure relates to a method and a system for implementing the method, in which the method assists users in resolving their curiosity about the details of the media, understanding the media, or successfully performing the activities when a user reads the media or performs a predefined activity through the media. Recently, sound pens have been widely used for young children (e.g., those aged 7 or under) to read storybooks or picture books independently. When a child uses a sound pen to touch a book, the sound pen converts the text in the touch area into speech and outputs the speech. The advantage of this type of voice pen is that it allows children to read books to each other without a guardian, but it has the following problems: children can only listen to the speech output by the voice pen and cannot engage in dialogue about the details of the book as a guardian can when reading to a child. The applicant developed the invention disclosed herein in consideration of a method that enables children to satisfy their curiosity about the details of a book while reading it, or enables children to understand the details of a book more clearly and become more interested in reading it. [The Problem to be Solved] This disclosure is about providing answers to what children say when they are reading books alone. This disclosure is about guiding children to a deeper understanding of the details of books when they are reading them alone. This disclosure relates to conveying information to children and guiding their learning through cards. The technical problems to be solved are not limited to those described above, and those skilled in this art will clearly understand other technical problems not mentioned based on this specification and the accompanying drawings. [Solution to the Problem] According to one embodiment, a server for communicating with an electronic device and an artificial intelligence (AI) server is provided, the server comprising: a memory; a communication unit; and a control unit configured to control the memory and the communication unit; wherein the control unit is configured to: [1] obtain reference information encoded in code data obtained by the electronic device, wherein the reference information is received from the electronic device via the communication unit, or the reference information is obtained by the controller after receiving the code data from the electronic device via the communication unit and decoding the code data, wherein the reference information is related to a medium with which a user interacts using the electronic device, and wherein the reference information includes media identifier (ID) information of the medium, page information indicated by the electronic device, and position information within the page indicated by the electronic device; [2] based on the reference information The system obtains content of interest from the memory and generates media context information, the content of interest being related to the printed content in the media and / or the portion of the printed content in the media that the user is interested in; [3] obtains speech text corresponding to the user's speech obtained by the electronic device, wherein the speech text is obtained by converting the speech using speech-to-text (STT) conversion; [4] generates a prompt based on the obtained media context information and the speech text, wherein the prompt includes the media context information and the speech text; [5] transmits the prompt to the AI ​​server via the communication unit; [6] receives the AI ​​server's response to the transmitted prompt via the communication unit; [7] obtains a response voice corresponding to the received response; [8] transmits the response voice to the electronic device via the communication unit. However, the solutions to the problem are not limited to those described above, and those skilled in the art will clearly understand other solutions not mentioned based on this specification and the accompanying drawings. [Effects of the Invention] According to an example, children can use an AI model (e.g., chat GPT) to converse about details of a book while reading it alone. According to an example, the response of the AI ​​model provided to children can be related to the details printed in the book. According to an embodiment, the response of the AI ​​model provided to the child can be related to the parts of the details printed in the book that the child is interested in. According to the embodiments, children can develop a greater understanding or interest in the content of books than they could gain simply by reading them. According to the example, children can experience various activities using cards. The effects disclosed herein are not limited to those described above, and those skilled in the art will clearly understand other effects not mentioned based on this specification and the accompanying drawings. According to one embodiment, a server for communicating with an electronic device and an artificial intelligence (AI) server is provided, the server including: a memory; a communication unit; and a control unit configured to control the memory and the communication unit; wherein the control unit is configured to: [1] obtain reference information encoded in code data obtained by the electronic device, wherein the reference information is received from the electronic device via the communication unit, or the reference information is obtained by the controller after receiving the code data from the electronic device via the communication unit and decoding the code data, wherein the reference information is related to a medium with which the user interacts using the electronic device, and wherein the reference information includes media ID information of the medium, page information indicated by the electronic device, and position information within the page indicated by the electronic device; [2] based on the reference information The system obtains content of interest from memory and generates media context information, the content of interest being related to the printed content in the media and / or the user's interest in the printed content in the media; [3] obtains speech text corresponding to the user's speech obtained by the electronic device, wherein the speech text is obtained by converting the speech using speech-to-text (STT) conversion; [4] generates a prompt based on the obtained media context information and speech text, wherein the prompt includes the media context information and speech text; [5] transmits the prompt to the AI ​​server via the communication unit; [6] receives the AI ​​server's response to the transmitted prompt via the communication unit; [7] obtains the response speech corresponding to the received response; [8] transmits the response speech to the electronic device via the communication unit. The memory stores a plurality of media texts, each of which corresponds to at least a portion of a specific medium, and wherein the control unit loads from the memory the media text corresponding to the media ID information of the reference information and obtains the media text as media context information. The memory stores multiple page-specific texts, each of which corresponds to at least a portion of the text printed on a specific page of a particular medium. The control unit loads the page-specific text from the memory that corresponds to the page information of the medium ID information and the page information of the reference information, and obtains the page-specific text as medium context information. The memory stores multiple location-specific texts, each of which corresponds to at least a portion of the text at a specific location on a specific page of a specific medium. The control unit loads the location-specific text from the memory that corresponds to the location information of the medium ID information, page information, and reference information, and obtains the location-specific text as medium context information. The prompts are generated by linking the audio text sequence to media context information. The prompts also include explanatory text containing media context information. The prompt also includes pre-stored user information, which includes at least one of the user's age and gender. The prompts also include pre-stored guide text, which includes considerations for the AI ​​server when generating a response. The control unit obtains the prompt template and uses media context information and voice text to modify the prompt template to generate a prompt. The prompt includes at least a portion of previous prompts that were transmitted to the AI ​​server before the electronic device obtained the coded data. The control unit transmits previous prompts to the AI ​​server via the communication unit and receives previous responses to the previous prompts from the AI ​​server, wherein the prompts include at least a portion of the previous responses. The reference information further includes area identification information related to questions or instructions printed on media that the user is interacting with using an electronic device, and the area identification information identifies a preset area in the media. The control unit obtains human context information corresponding to questions or instructions printed on the preset area from memory based on the reference information, and the prompt further includes human context information. The control unit obtains the subsequent voice text corresponding to the user's subsequent voice obtained by the electronic device, wherein the subsequent voice is obtained by converting the subsequent voice using STT conversion; generates a subsequent prompt based on the subsequent voice text; transmits the subsequent prompt to the AI ​​server via the communication unit; receives the subsequent response to the subsequent prompt transmitted by the AI ​​server via the communication unit; obtains the subsequent response voice corresponding to the subsequent response; and transmits the subsequent response voice to the electronic device via the communication unit. Subsequent prompts also include media context information. AI servers include large language models (LLMs). The electronic device includes an image sensor and is configured to capture areas of a medium by means of user manipulation and obtain coded data from the captured image. The electronic device has a pen shape with one end and another end, and in order to allow a user to indicate a medium area using the one end of the electronic device, the electronic device is configured to capture the medium using an image sensor arranged at the one end. According to another embodiment, a server for communicating with an electronic device and an artificial intelligence (AI) server is provided, the server comprising: a memory; a communication unit; and a control unit configured to control the memory and the communication unit; wherein the control unit is configured to: [1] obtain reference information encoded in code data obtained by the electronic device, wherein the reference information is received from the electronic device via the communication unit, or the reference information is obtained by the controller after receiving the code data from the electronic device via the communication unit and decoding the code data, wherein the reference information is related to a medium with which the user interacts using the electronic device, and wherein the reference information includes media ID information of the medium, and instructions from the electronic device. [1] Page information and position information within the page indicated by the electronic device; [2] Obtain content of interest from memory based on reference information and generate media context information, the content of interest being related to the printed content in the media and / or the user's interest in the printed content in the media; [3] Obtain voice data corresponding to the user's voice obtained by the electronic device; [4] Generate prompts based on the obtained media context information and voice data; [5] Transmit the prompts to the AI ​​server via the communication unit; [6] Receive the AI ​​server's response to the transmitted prompts via the communication unit; [7] Transmit the received response or the corresponding voice response to the response to the electronic device via the communication unit. The data type of media context information is any of text, sound, image, or video. When the data type of the media context information is text and the data type supported by the AI ​​server is sound, the control unit converts the media context information to sound before generating a prompt. According to another embodiment, a server for communicating with an electronic device and an artificial intelligence (AI) server is provided, the server including: a memory; a communication unit; and a control unit configured to control the memory and the communication unit; wherein the control unit is configured to: [1] obtain reference information encoded in code data obtained by the electronic device, wherein the reference information is received from the electronic device via the communication unit, or the reference information is obtained by the controller after receiving the code data from the electronic device via the communication unit and decoding the code data, and wherein the reference information includes area identification information for identifying a predetermined area in a medium in which a user interacts with the electronic device, and symbols, questions and pointers are printed on the predetermined area. [1] At least one of the following: [2] Obtaining pre-stored artificial context information corresponding to a predetermined area from memory based on reference information; [3] Obtaining speech text corresponding to the user's speech obtained by the electronic device, wherein the speech text is obtained by converting the speech using speech-to-text (STT) conversion; [4] Generating a prompt based on the obtained artificial context information and speech text, wherein the prompt includes artificial context information and speech text; [5] Transmitting the prompt to the AI ​​server via a communication unit; [6] Receiving a response from the AI ​​server to the transmitted prompt from the server; [7] Obtaining a response voice corresponding to the received response; [8] Transmitting the response voice to the electronic device via a communication unit. The memory stores multiple questions, each of which corresponds to a question printed on a specific medium, and the artificial context information is the question among the multiple questions that corresponds to the identification information of the reference information. The memory stores multiple instructions, each of which corresponds to an instruction printed on a specific medium, and the artificial context information is the instruction among the multiple instructions that corresponds to the identification information of the reference information. The reference information also includes the media ID, and the artificial context information is the information stored in memory that corresponds to the media ID and the region identification information. Reference information also includes page information, and the artificial context information is information stored in memory that corresponds to the media ID, page information, and region identification information. The prompts are generated by linking the spoken text sequence to human contextual information. The prompts also include explanatory text containing human-generated contextual information. The prompt also includes pre-stored user information, which includes at least one of the user's age and gender. The prompts also include pre-stored guide text, which includes considerations for the AI ​​server when generating a response. The control unit obtains the prompt template and uses media context information and voice text to modify the prompt template to generate a prompt. The prompt includes at least a portion of previous prompts that were transmitted to the AI ​​server before the electronic device obtained the coded data. The control unit transmits previous prompts to the AI ​​server via the communication unit and receives previous responses to the previous prompts from the AI ​​server, wherein the prompts include at least a portion of the previous responses. The reference information includes the media ID information of the media, page information indicated by the electronic device, and position information within the page indicated by the electronic device. The control unit obtains content of interest from memory based on the reference information and generates media context information. The content of interest is related to the printed content in the media and / or the part of the printed content in the media that the user is interested in. The prompt also includes media context information. The control unit: obtains the subsequent speech text corresponding to the user's subsequent speech obtained by the electronic device, wherein the subsequent speech is obtained by converting the subsequent speech using STT conversion; generates a subsequent prompt based on the subsequent speech text and human context information; transmits the subsequent prompt to the AI ​​server via the communication unit; receives the subsequent response to the subsequent prompt transmitted by the AI ​​server via the communication unit; obtains the subsequent response speech corresponding to the subsequent response; and transmits the subsequent response speech to the electronic device via the communication unit. Subsequent prompts also include media context information. AI servers include large language models (LLMs). The electronic device includes an image sensor and is configured to capture areas of a medium by means of user manipulation and obtain coded data from the captured image. The electronic device has a pen shape with one end and another end, and in order to allow a user to indicate a medium area using the one end of the electronic device, the electronic device is configured to capture the medium using an image sensor arranged at the one end. According to another embodiment, a server for communicating with an electronic device and an artificial intelligence (AI) server is provided, the server comprising: a memory; a communication unit; and a control unit configured to control the memory and the communication unit; wherein the control unit is configured to: [1] obtain reference information encoded in code data obtained by the electronic device, wherein the reference information is received from the electronic device via the communication unit, or the reference information is obtained by the controller decoding the code data after receiving the code data from the electronic device via the communication unit, and wherein the reference information is printed on the electronic device and... The interactive medium relates to a question or instruction, wherein the reference information includes area identification information for identifying a predetermined area in the medium; [2] obtains pre-stored artificial context information corresponding to the predetermined area from memory based on the reference information; [3] generates a prompt based on the obtained artificial context information, wherein the prompt includes artificial context information; [4] transmits the prompt to the AI ​​server via a communication unit; [5] receives a response from the AI ​​server to the transmitted prompt from the server; [6] obtains a response voice corresponding to the received response; [7] transmits the response voice to an electronic device via a communication unit. The control unit: obtains voice text corresponding to the user's voice obtained by the electronic device, wherein the user's voice is obtained after the response voice is transmitted to the electronic device and output, and the voice text is obtained by converting the subsequent voice using speech-to-text (STT) conversion; generates a subsequent prompt based on the voice text and transmits the subsequent prompt to the AI ​​server via the communication unit; obtains the subsequent response from the AI ​​server in response to the transmitted subsequent prompt via the communication unit; obtains the subsequent response voice corresponding to the received subsequent response; and transmits the subsequent response voice to the electronic device via the communication unit. According to another embodiment, a server for communicating with an electronic device and an artificial intelligence (AI) server is provided, the server comprising: a memory; a communication unit; and a control unit configured to control the memory and the communication unit; wherein the control unit is configured to: [1] obtain reference information encoded in code data obtained by the electronic device, wherein the reference information is received from the electronic device via the communication unit, or the reference information is obtained by the controller after receiving the code data from the electronic device via the communication unit and decoding the code data, and wherein the reference information includes media for identifying a user interacting with the electronic device. [1] The predefined area identification information in the interface, and at least one of the symbols, questions and instructions is printed on the predefined area; [2] Based on the reference information, the pre-stored artificial context information corresponding to the predefined area is obtained from memory; [3] The voice data corresponding to the user's voice obtained by the electronic device is obtained; [4] The prompt is generated based on the obtained artificial context information and voice data; [5] The prompt is transmitted to the AI ​​server via the communication unit; [6] The AI ​​server receives the response of the transmitted prompt from the server; [7] The received response or the response voice corresponding to the received response is transmitted to the electronic device via the communication unit. The data type of human contextual information can be any of the following: text, audio, video, or audio. When the data type of the human context information is text and the data type supported by the AI ​​server is audio, the control unit converts the human context information to audio before generating a prompt. The above-described objects, features, and advantages of this disclosure will become more apparent from the following detailed description with reference to the accompanying drawings. However, various modifications can be made to this disclosure, and various embodiments of this disclosure can be practiced. Therefore, specific embodiments are shown in the accompanying drawings and will be described in detail below. In the drawings, the thickness of layers and regions is exaggerated for clarity. Furthermore, when an element or layer is referred to as being located on another element or layer, the element or layer may be directly located on the other element or layer, or there may be intermediate layers or elements. Throughout the specification, the same reference numerals generally denote the same elements. Additionally, the same reference numerals are used to describe elements within the same category that have the same function shown in the drawings of the embodiments, and will not be described again. The numbers used in this specification (e.g., first, second, etc.) are merely identifiers used to distinguish the individual components. Additionally, the terms "module" and "component" used to name components in the following description are used only for ease of writing this specification. These terms are not intended to have different specific meanings or functions, and therefore may be used individually or interchangeably. In the following embodiments, unless the meaning is clearly different in the context, the singular expression encompasses the plural expression. In the following embodiments, it should be understood that terms such as "comprising" or "having" are intended to indicate the presence of features or elements disclosed in the specification, rather than to exclude the possibility of adding one or more other features or elements. For ease of illustration, the size of the elements in the drawings may be enlarged or reduced. For example, any size and thickness of each element shown in the drawings may be shown for ease of illustration, and this disclosure is not limited thereto. In cases where a particular embodiment is achieved by other means, the specific processes may not be performed in the order described. For example, two processes described consecutively may be performed substantially simultaneously, or may be performed in the reverse order described. In the following embodiments, when a membrane, region, or element is referred to as being connected to each other, the membrane, region, or element is directly connected to each other, or the membrane, region, or element is indirectly connected to each other via an intermediate intervening membrane, region, or element. For example, in this specification, when a membrane, region, or element is referred to as being electrically connected to each other, the membrane, region, or element is directly electrically connected to each other, or the membrane, region, or element is indirectly electrically connected to each other through an intermediate intervening membrane, region, or element. Unless otherwise specifically stated or obvious from the context, the term “about” in relation to a numerical value may be understood to mean the numerical value mentioned and up to + / - 10% of the numerical value, and the term “about” in relation to a numerical range may be understood to mean a range from 10% below the lower limit of the numerical range to 10% above the upper limit of the numerical range. 1. Overview As described in the background section of this invention, using an AI server can be considered a solution to the problem of voice pens only reading books but not engaging in conversation with children or answering their questions. That is, a system could be considered that records a child's speech and inputs the recorded speech (or text obtained through speech-to-text conversion) into a large language model (LLM) (e.g., ChatGPT (Generative Pre-trained Transformer)) to generate a response and output the response to the child. In this text, the child's young age and relatively low language skills make it difficult for him / her to form definite sentences. To compensate for this lack of language skills, the child frequently uses his / her finger and demonstrative pronouns (such as "this" or "that") while reading. This characteristic should be taken into consideration. In other words, large language models can only receive children’s speech (or text obtained through speech conversion) as input, but cannot receive information such as what the child is pointing to with his / her finger as input. Therefore, it is difficult to know exactly what the child is trying to say or what the child wants to know. For example, when a child is reading the Snow White storybook and asks questions (such as "Does the princess live here?" or "Why is a certain character bullying the princess?"), and the questions are passed to a large language model that has no context (or no context about the Snow White story), the large language model may have difficulty understanding the child's intention and therefore output a response that is not very relevant to the Snow White storybook. To address these issues, this disclosure proposes a method in which, when a user's voice (e.g., a user question) is input, details of content related to the user's voice are extracted and used. In this case, to specify content related to the user's voice, a camera-readable code is printed on the medium along with the content, and the code is identified by the user's electronic device. Therefore, according to the method provided in this disclosure, when a user simply converses with a generative and conversational AI model such as GPT, information about the user's voice and content of interest printed on the media becomes more readily available. Using content of interest printed on the media when conversing with the AI ​​model helps to form the basic "context" for the dialogue between the AI ​​model and the user. That is, further generating information about the "context" of the dialogue between the user and the AI ​​model, and using this contextual information in the dialogue, is a key concept of the method provided in this disclosure, in order to address the problems in the aforementioned technology. The following sections will describe a specific system for achieving the method provided in this disclosure. Furthermore, various embodiments of the method provided in this disclosure will be described. Specifically, the specific methods for generating contextual information and for passing this contextual information to the AI ​​model will be clearly understood from the following description. 2. Interactive System The configuration and operation of the interactive system will be described below with reference to Figure 1. Figure 1 is a diagram illustrating an interactive system 10 according to an embodiment. Referring to Figure 1, the interactive system 100 may include an electronic device 1000, a main server 2000, and an AI server 3000. When using the media, users can interact with the media using the interactive system 100. In this text, "medium" refers to a physical medium on which information or content is recorded. For example, information or content recorded on a medium can be a story, explanation, news, question, event, or instruction. Furthermore, a physical medium can be a book, paper, or film, as well as a portion of the surface of an object, a portion of the surface of a sculpture, a portion of the surface of a piece of furniture, a portion of the surface of an electronic product, or a portion of the surface of a building. In this text, the manner in which information is recorded on a medium does not refer to an "electronic" method, but rather to the user's visual perception of the information recorded on the medium. For example, information or content such as images or text can be printed on a medium. Simultaneously, the medium can be a panel such as a liquid crystal display (LCD), an organic light-emitting diode (OLED), or an LED, or a display that includes a panel, and information or content such as images or text can be output electronically on the medium. The use of media by users can be, for example, by users reading stories or information recorded on the media, or by users performing activities or instructions recorded on the media. Users can use the interactive system 100 to interact with the media. As a representative example, interaction refers to engaging in dialogue with the interactive system 100 regarding the medium. Specifically, while using the medium, the user can ask questions and receive answers via the electronic device 1000. For example, interaction can be implemented by having the user ask questions or express opinions via voice while reading a book, and the electronic device 1000 outputs the answer to the question or the response to the opinion via voice. As another example, interaction can be implemented by having the electronic device 1000 output activities or instructions printed on cards via voice, and the user performing the activity or responding to the instruction. The process by which a user interacts with the media using an electronic device 1000 can be carried out as follows. First, the electronic device 1000 can acquire image data of the medium. Image data refers to an image obtained by the electronic device 1000 capturing a portion of the medium through user manipulation. For example, while the user positions the electronic device 1000 to touch or approach said portion of the medium, the electronic device 1000 can capture said portion of the medium to obtain image data. Information about the medium can be encoded in the image data. The electronic device 1000 can transmit the image data to the main server 2000. Electronic device 1000 can obtain user voice data. Electronic device 1000 can obtain user voice data by recording user voice. Electronic device 1000 can transmit user voice data to main server 2000. Electronic device 1000 can transmit its electronic device ID to the main server 2000. The electronic device ID is information used to identify electronic device 1000 and can be used to generate session information, which will be explained later. The main server 2000 can use video and audio data to generate prompts. Specifically, the main server 2000 can analyze video data to obtain contextual information, and can use the contextual information and audio data to generate prompts. The main server 2000 can transmit prompts to the AI ​​server 3000. The AI ​​server 3000 may include a large language model (LLM) and receive the prompts to output a response. The AI ​​server 3000 can transmit the response to the main server 2000, and the main server can generate response voice data for the response and transmit the response voice data to the electronic device 1000. The electronic device 1000 can output the received response voice data. When a user interacts with the medium using the interactive system 100, the aforementioned contextual information can be used to specify the context of the interaction. For example, when a user expresses a question or opinion while reading (or pointing to) a portion of a book, the details of the book or the details of that portion of the book can serve as the context of the interaction. In this document, information about the portion the user is reading (or pointing to) while expressing a question or opinion can be obtained as contextual information, and the obtained contextual information can be used to generate prompts. In this scenario, the response generated by the AI ​​server 3000 can also take contextual information into account, and the response voice data output by the electronic device 1000 includes details corresponding to the user's intent regarding the question or opinion expressed. Contextual information may include media contextual information, human contextual information, or both media contextual information and human contextual information. Media context information refers to the information among multiple pieces of information about a medium that forms the context contained within the medium. For example, when the medium is a book with a specific story printed on it, the media context may refer to the specific story, and the media context information is information about a part of the specific story and may refer to words, sentences, paragraphs, or images that constitute the specific story. Media context information may be information about the part of the multiple pieces of information about the medium that the user interacts with. For example, media context information may be information about words, sentences, paragraphs, or portions of images that the user points to or touches using electronic device 1000 within the media context. Methods for obtaining media context information will be described later. Artificial contextual information refers to information among multiple pieces of information about a medium that does not form the context contained within the medium. For example, artificial contextual information may include information about instructions, questions, or activities among multiple pieces of information about a medium. More specifically, when the medium is a book with a specific story printed on it, the context of the medium may refer to the specific story, and the media contextual information may refer to questions, instructions, or statements that elicit specific actions related to the specific story. Each element of the interactive system 100 will be described in detail below with reference to Figures 2 through 5. Figure 2 is a diagram showing the configuration of an electronic device 1000 according to an embodiment. Figure 4 is a diagram illustrating the configuration of the main server 2000 according to an embodiment. Figure 5 is a diagram illustrating the configuration of the AI ​​server 3000 according to an embodiment. Referring to FIG2, the electronic device 1000 may include a sensing unit 1100, an electronic device memory 1200, an electronic device input unit 1300, an electronic device output unit 1400, an electronic device communication unit 1500, and an electronic device control unit 1600. The electronic device 1000 can capture details printed on a medium using the sensing unit 1100. For example, the sensing unit 1100 may include an image sensor (such as a camera), and the electronic device 1000 can capture at least a portion of the medium to obtain an image based on the user's operation. The image acquired by the sensing unit 1100 may include a code. Specifically, the code may be printed on the medium according to a preset rule, and the image acquired by capturing the medium may include the code. In this document, the ink used to print the code on the medium may be different from the ink used to print content on the medium. For example, an ink that absorbs infrared light may be used to print the code on the medium, and the sensing unit 1100 of the electronic device 1000 may include an infrared image sensor, and the image captured by the sensing unit 1100 may include the code printed on the medium rather than the content printed on the medium (or may include both content and code). As will be explained later, the electronic device 1000 can capture the code image on the medium printed thereon, and the electronic device 1000 or the main server 2000 can analyze the acquired code image to obtain reference information indicating information about the medium or to obtain information about the medium. The code printed on the medium will be described in detail below with reference to Figures 3A and 3B. Figures 3A and 3B are diagrams illustrating codes printed on a medium according to embodiments. Figure 3(a) is a diagram illustrating the unit cells constituting the code. Figure 3(b) is a diagram illustrating a method of encoding information in information segments. Referring to Figure 3, coding can be achieved by repeatedly printing a unit cell in a two-dimensional manner according to the size of the medium. Each unit cell includes multiple line segments (or points) placed according to a specific rule, which can be determined based on information encoded in the unit cell. Referring to Figure 3(a), the line segments included in the unit cell are divided into reference line segments and information line segments, and the reference line segments include horizontal reference line segments, vertical reference line segments and intersecting reference line segments. When analyzing the coded image in the electronic device 1000 or the main server 2000, the area of ​​the unit cell is divided by reference line segments, and the information encoded in the unit cell can be obtained by information line segments. Simultaneously, within a unit cell, virtual lines can be defined, including multiple horizontal virtual lines extending in the horizontal direction and multiple vertical virtual lines extending in the vertical direction. Virtual lines often (but not necessarily) refer to lines not printed on the media. Virtual lines can be printed on the media. Multiple horizontal and vertical virtual lines intersect to form multiple intersection points, which are referred to below as virtual reference points. Virtual reference points are often (but not always) printed on media. Referring to Figures 3A and 3B, a unit cell is shown to include a total of seven horizontal virtual lines and a total of seven vertical virtual lines. A horizontal reference line segment is placed on the uppermost horizontal virtual line, and a vertical reference line segment is placed on the leftmost vertical virtual line. Depending on the method of placing the information segments, a specific number of binary numbers can be represented. In the following text, for ease of explanation, the case where the information segments can represent four binary numbers (i.e., 2 bits) will be described. For clarity, we assume that the virtual reference point described above is used as the origin and that a rectangular coordinate system is defined in the directions of the horizontal and vertical reference line segments. When the information to be encoded by the information segment is "00", the information segment can be placed in the following manner (see the leftmost part of Figure 3(b)): one end of the information segment (the end positioned closer to the virtual reference point) is positioned in the third quadrant of the Cartesian coordinate system, and the other end (the end positioned further away from the virtual reference point) is positioned in the first quadrant of the Cartesian coordinate system. When the information to be encoded by the information segment is "01", the information segment can be placed in the following manner (see the second part from the left in Figure 3(b)): one end of the information segment (the end positioned closer to the virtual reference point) is positioned in the fourth quadrant of the Cartesian coordinate system, and the other end (the end positioned further away from the virtual reference point) is positioned in the second quadrant of the Cartesian coordinate system. Additionally, when the information to be encoded by the information segment is "10", the information segment can be placed in the following manner (see the third part from the left in Figure 3(b)): one end of the information segment (the end positioned closer to the virtual reference point) is positioned in the first quadrant of the Cartesian coordinate system, and the other end (the end positioned further away from the virtual reference point) is positioned in the third quadrant of the Cartesian coordinate system. Additionally, when the information to be encoded by the information segment is "11", the information segment can be placed in the following manner (see the fourth part from the left in Figure 3(b)): one end of the information segment (the end positioned closer to the virtual reference point) is positioned in the second quadrant of the Cartesian coordinate system, and the other end (the end positioned further away from the virtual reference point) is positioned in the fourth quadrant of the Cartesian coordinate system. When 2 bits of information are encoded in the information segments shown in Figure 3(b) and a unit cell includes 36 information segments as shown in Figure 3(a), the information that can be encoded in a unit cell is 2 bits * 36 = 72 bits. However, the way information segments are placed involves positioning one end and the other end of the information segment. Therefore, this method can be designed by adjusting the angle of the information segment or by spacing the information segment from a reference point. In this case, the information that can be encoded in the information segment is not limited to 2 bits, but can be 3 bits or more. At the same time, codes can be achieved through various methods other than those described above. For example, a code can take the form of an image in which specific information (such as a QR code) is encoded. The sensing unit 1100 acquires an image (hereinafter referred to as a code image) of a portion of a plurality of codes pre-printed on a medium. The electronic device 1000 analyzes the acquired code image to obtain information encoded in the code image. According to some embodiments, multiple pieces of information (such as the title of the medium, the page of the medium, and the X and Y coordinates within a page of the medium) can be encoded in a code pre-printed on the medium. For example, (book title, page, X and Y coordinates) can be encoded in a single code unit. In this case, when a user points to a specific location on the medium via electronic device 1000, electronic device 1000 knows the specific location on the medium pointed to by the user. According to some other embodiments, predetermined area identification information can be encoded in a code pre-printed on the medium. In this case, the user can know the type of content (e.g., text, illustration, or icon) printed in a specific area of ​​the medium via the electronic device 1000. According to some embodiments, electronic device 1000 may transmit multiple pieces of information (hereinafter referred to as code data) obtained by analyzing code images to main server 2000, and main server 2000 may use the received code data to specify the text or image (i.e., a portion of the printed content) to be printed at a specific location on a medium indicated by the user. According to some other embodiments, the electronic device 1000 can transmit the obtained code image to the main server 2000. In this case, the main server 2000 can analyze the code image to obtain the aforementioned code data. As described above, text or images (i.e., a portion of the printed content) can be specified to be printed at a specific location on a medium indicated by the user. Referring back to FIG2, the electronic device memory 1200 may store information processed by the electronic device 1000 and programs executed by the electronic device 1000. For example, the electronic device memory 1200 may store image data acquired by the sensing unit 1100. As another example, the electronic device memory 1200 may store code data obtained by analyzing the image data. As yet another example, the electronic device memory 1200 may store data about the medium obtained using the code data. In terms of hardware, the electronic device memory 1200 can be implemented in the form of various storage devices, such as read-only memory (ROM), random access memory (RAM), erasable and programmable read-only memory (EPROM), flash memory, or hard drive. The electronic device input unit 1300 may include a microphone. The electronic device input unit 1300 may record the user's voice to generate a voice signal. The voice signal may be stored as voice data in the electronic device memory 1200. The electronic device input unit 1300 may include a trigger button. When a user manipulates (e.g., presses or touches) the trigger button, the microphone included in the electronic device input unit 1300 is activated to begin voice recording. The user can press the trigger button and ask a question or speak what he / she wishes to say, and the electronic device input unit 1300 can record the user's voice to generate a voice signal. The electronic device input unit 1300 may be a keyboard, button, mouse, microphone, camera, sensor, touch screen or any combination thereof, but is not limited thereto. The electronic device output unit 1400 may include a speaker. The electronic device output unit 1400 may output response voice data obtained by the autonomous server 2000. The electronic device output unit 1400 may be a display, a speaker, an indicator, or any combination thereof, but is not limited thereto. The electronic device communication unit 1500 enables data communication between the electronic device 1000 and external devices. For example, the electronic device communication unit 1500 can transmit code images (or code data or context information) to the main server 2000. As another example, the electronic device communication unit 1500 can independently receive response voice data from the server 2000. The electronic device communication unit 1500 may be, for example, a wired / wireless local area network (LAN) module, a wide area network (WAN) module, an Ethernet module, a Bluetooth module, a VPN module, a Universal Serial Bus (USB) module, an IEEE 1394 module, a Wi-Fi module, a mobile communication module, a satellite communication module, or any combination thereof, but is not limited thereto. The electronic device control unit 1600 can control the components of the electronic device 1000 or execute programs stored in the electronic device memory 1200. The electronic device control unit 1600 may be, for example, a central processing unit (CPU), a graphics processing unit (GPU), a neural processing unit (NPU), a digital signal processor (DSP), a state machine, an application-specific integrated circuit (ASIC), a radio-frequency integrated circuit (RFIC), or any combination thereof, but is not limited thereto. For ease of explanation, the electronic device 1000 will be described as a pen-shaped device. When a user positions the end of the electronic device 1000 to contact or approach a medium on which a code is printed, the sensing unit 1100 captures the medium to obtain a code image, the electronic device 1000 analyzes the code image to obtain code data, and the electronic device 1000 transmits the code data to the main server 2000. However, the technical concept disclosed herein is not limited to this. For example, the electronic device 1000 may be a smartphone or a personal computer (PC). In this document, applications for performing the functions of the electronic device 1000 described above (e.g., capturing code images, obtaining code data, performing analysis, and transmitting) may be stored in a smartphone or PC. The main server 2000 may include a main server memory 2100, a main server communication unit 2200, and a main server control unit 2300. The main server memory 2100 can store information processed by the main server 2000 and programs executed by the main server 2000. Referring to FIG4, the main server memory 2100 can store at least a speech-to-text (STT) model 2110, a text-to-speech (TTS) model 2130 and a database 2150. STT model 2110 refers to a program that converts voice data (or voice signals) into text data (or voice text). The main server 2000 can use the STT model to convert the user's voice data received from the electronic device 1000 into text data. The main server 2000 can use the converted text data to generate prompts, which will be explained later. Simultaneously, the main server 2000 can transmit voice data to an external server, and the external server can convert the voice data into text data and provide the text data to the main server 2000. Alternatively, the main server 2000 can use voice data to generate prompts without converting the voice data into text data. In this case, the main server 2000 may not include the STT model 2110. TTS model 2130 refers to a program that converts text data into speech data. The main server 2000 can convert the response text data obtained from the AI ​​server 3000 into response speech data. The main server 2000 can transmit the response speech data to the electronic device 1000. Simultaneously, the main server 2000 can transmit the response text data to an external server, and the external server can convert the response text data into response voice data and provide the response voice data to the main server 2000. Alternatively, the main server 2000 can receive the response voice data from the AI ​​server 3000. In this case, the main server 2000 may not include the TTS model 2130. Database 2150 stores information about various content printed on multiple media. For example, database 2150 can store such content by mapping all text printed on the multiple media and / or images printed on the multiple media to the corresponding media. For each medium, the database 2150 can further store information about the pages containing the text and images. For example, the database 2150 can store multiple pieces of necessary information so that, within all the content on a particular medium, the details of the content printed on the first page are distinguished from the details of the content printed on the second page. For each page, the database 2150 can further store information about the position of the text and images printed on the page. For example, the database 2150 can store multiple pieces of necessary information so that the details of the content printed at the first position (first coordinate) and the details of the content printed at the second position (second coordinate) are distinguished from each other in the content printed on a specific page. Simultaneously, database 2150 can store user information (e.g., a user's unique identifier, name, age, gender, personality, or family relationship). User information can be recorded or stored by an administrator or guardian. User information can be used to generate prompts, as will be explained later. When user information is used to generate prompts, customized responses can be generated for those prompts. Additionally, user information can be used to generate session information, which will be explained later. In terms of hardware, the main server memory 2100 can be implemented in the form of various storage devices, such as ROM, RAM, EPROM, flash memory, or hard drive. The main server communication unit 2200 enables data communication between the main server 2000 and external devices. For example, the main server communication unit 2200 can receive code images (or code data or context information) from the electronic device 1000. As another example, the main server communication unit 2200 can receive response data (response text data or response voice data) from the AI ​​server 3000. As yet another example, the main server communication unit 2200 can transmit response voice data to the electronic device 1000. The main server communication unit 2200 may be, for example, a wired / wireless local area network (LAN) module, a WAN module, an Ethernet module, a Bluetooth module, a VPN module, a Universal Serial Bus (USB) module, an IEEE 1394 module, a Wi-Fi module, a mobile communication module, a satellite communication module, or any combination thereof, but is not limited thereto. The main server control unit 2300 can control the components of the main server 2000 or execute programs stored in the main server memory 2100. The main server control unit 2300 may be, for example, a central processing unit (CPU), a graphics processing unit (GPU), a neural processing unit (NPU), a digital signal processor (DSP), a state machine, an application-specific integrated circuit (ASIC), a radio frequency integrated circuit (RFIC), or any combination thereof, but is not limited thereto. Although not shown in Figure 4, the main server 2000 may include a main server input unit and a main server output unit. The main server input unit may receive input from a manager that manages the main server 2000, and may be a keyboard, buttons, a mouse, a microphone, a camera, a sensor, a touch screen, or any combination thereof, but is not limited thereto. The main server output unit may output information processed by the main server 2000 to the manager, and may be a display, a speaker, an indicator, or any combination thereof, but is not limited thereto. The AI ​​server 3000 may include an AI server memory 3100, an AI server communication unit 3200, and an AI server control unit 3300. AI server memory 3100 can store information processed by AI server 3000 and programs executed by AI server 3000. AI server 3000 can generate response data in response to user questions or voice. For this purpose, generative models can be stored in AI server memory 3100. In this paper, generative models can be natural language processing models (NLP) or large language models (LLMs) trained with a large amount of data. Generative models can be generated by training a generative pre-trained transformer (GPT) model using a transformer model structure. Examples of generative models include ChatGPT, Google's Bard or Gemini, and Microsoft's Copilot, and these models can be stored in AI server memory 3100. Furthermore, the generative model can vary depending on the type of data that the AI ​​server 3000 wishes to generate. For example, when the AI ​​server 3000 wishes the autonomous server 2000 to receive voice prompts to generate voice response data, the generative model can be a model for generating voice, such as WaveNet. As another example, when the master server 2000 provides image data to the AI ​​server 3000 and the AI ​​server 3000 wishes to generate image response data, the generative model can be a model for generating images, such as a generative adversarial network (GAN) or a diffusion model. That is, the generative model can generate response data of the same type as the received data (e.g., text, sound, image, or video), such as receiving text data and generating text response data, or receiving voice data and generating voice response data. As an alternative, generative models can generate response data of a different type than the data received, such as receiving text data and generating speech response data, or receiving speech data and generating text response data. As an alternative, generative models can simultaneously receive multiple data types and produce response data of any data type. For example, a generative model can receive data including text and audio and output text response data or audio response data. In terms of hardware, the AI ​​server memory 3100 can be implemented in the form of various storage devices, such as ROM, RAM, EPROM, flash memory, or hard drive. The AI ​​server communication unit 3200 enables data communication between the AI ​​server 3000 and external devices. For example, the AI ​​server communication unit 3200 can receive prompts from the host server 2000. As another example, the AI ​​server communication unit 3200 can transmit response data (response text data or response voice data) to the host server 2000. The AI ​​server communication unit 3200 may be, for example, a wired / wireless local area network (LAN) module, a WAN module, an Ethernet module, a Bluetooth module, a Bee module, a universal serial bus (USB) module, an IEEE 1394 module, a Wi-Fi module, a mobile communication module, a satellite communication module, or any combination thereof, but is not limited thereto. The AI ​​server control unit 3300 can control the components of the AI ​​server 3000 or execute programs stored in the AI ​​server memory 3100. The AI ​​server control unit 3300 may be, for example, a central processing unit (CPU), a graphics processing unit (GPU), a neural processing unit (NPU), a digital signal processor (DSP), a state machine, an application-specific integrated circuit (ASIC), a radio frequency integrated circuit (RFIC), or any combination thereof, but is not limited thereto. Although not shown in Figure 5, the AI ​​server 3000 may include an AI server input unit and an AI server output unit. The AI ​​server input unit may receive input from a manager that manages the AI ​​server 3000, and may be a keyboard, button, mouse, microphone, camera, sensor, touchscreen, or any combination thereof, but is not limited thereto. The AI ​​server output unit may output information processed by the AI ​​server 3000 to the manager, and may be a display, speaker, indicator, or any combination thereof, but is not limited thereto. Another embodiment of the interactive system 100 described above will be described below with reference to Figures 6 and 7. Figure 6 is a block diagram illustrating an electronic pen service system using conversational AI services according to an embodiment. Referring to Figure 6, the electronic pen service system 101 using conversational AI services includes an electronic pen 110, a first server 120, and a conversational AI server 130. In this document, the electronic pen service system 101 corresponds to the interactive system 100 illustrated with reference to Figures 1 to 5, the electronic pen 110 corresponds to the electronic device 1000, the first server 120 corresponds to the main server 2000, and the conversational AI server 130 corresponds to the AI ​​server 3000. The electronic pen 110 includes a first communication unit 111, a code recognition unit 112, a first control unit 113, a TTS unit 114, and a speaker unit 115. The first communication unit 111 can communicate with the first server 120 using a wired or wireless interface. The first communication unit 111 is substantially the same as the electronic device communication unit 1500 illustrated with reference to FIG2. The code recognition unit 112 recognizes a predetermined code from a printed item on which a predetermined code is printed. The predetermined code includes position information, i.e., coordinate information, on the printed item. The code recognition unit 112 includes a camera. When the camera approaches within a predetermined distance of the printed item on which the predetermined code is printed, the camera captures the predetermined code within its recognizable area. The code recognition unit 112 then transmits the image captured by the camera to the first control unit 113. The code recognition unit 112 is substantially the same as the sensing unit 1100 illustrated with reference to FIG2. The first control unit 113 reads coordinate information from a predetermined code in the image received by the code identification unit 112 (obtaining coordinate information corresponding to the predetermined code in the image from a pre-stored database). The first control unit 113 is substantially the same as the electronic device control unit 1600 illustrated with reference to FIG2. Subsequently, the first control unit 113 transmits the read coordinate information to the first server 120 via the first communication unit 111. As will be explained later, the first communication unit 111 receives the result value of the coordinate information transmitted from the first server 120. Meanwhile, the first control unit 1130 can identify information from the predetermined code reading area in the image received by the code identification unit 112. Zone identification information is information used to identify zones within a medium. For example, the medium may include at least one page, and each page may contain printed content. In this document, zone identification information may be set for specific zones within a page. In this case, the zone identification information may be encoded in a coded image obtained by capturing the specific zone. In this document, a specific zone may be a zone with a specific graphic printed on it, a zone with text printed on it, or a zone with an illustration printed on it. Region identification information can be used instead of coordinate information. For example, the first control unit 113 can transmit the read region identification information to the first server 120 via the first communication unit 111. In this case, the first server 120 can transmit a prompt corresponding to the region identification information to the conversational AI server 130, and the conversational AI server 130 can generate a result value for the received prompt and transmit the result value to the first server 120. The TTS unit 114 is an element equipped with text-to-speech (TTS) functionality and performs the function of converting text into speech. The TTS unit 114 converts the result value received by the first communication unit 111 into speech and outputs the converted speech to the user of the electronic pen 110 via the speaker unit 115. The TTS unit 114 is an element corresponding to the TTS model 2130 illustrated with reference to FIG4. The TTS unit 114 may be included in the first server 120 rather than in the electronic pen 110. In this case, the electronic pen 110 may receive the speech obtained by converting the result value from the first server 120 and may output the converted speech to the speaker unit 115. In an additional embodiment, the electronic pen 110 may further include a microphone 116 and an STT unit 117. The electronic pen 110 receives voice input from the user via the microphone 116. STT unit 117 is a component equipped with a speech-to-text (STT) function, which performs the function of converting speech into text. STT unit 117 converts speech information received by microphone 116 into character information and transmits the converted character information to first server 120. Then, first communication unit 111 receives the result value of the transmitted character as a character from first server 120, and can convert the character received by first communication unit 111 into speech and output the speech to the user of electronic pen 110 via speaker unit 115. STT unit 117 is a component corresponding to the STT model 2110 illustrated with reference to FIG4. STT unit 117 may be included in first server 120, but not in electronic pen 110. In this case, electronic pen 110 may transmit speech signals to first server 120, and first server 120 may use STT unit 117 to convert speech signals into character information. In another embodiment, the electronic pen 110 may further include a storage unit 118. Storage unit 118 includes the content specified in the coordinate information. Storage unit 118 may be included in the first server 120 rather than in the electronic pen 110. The first control unit 113 can read coordinate information from the predetermined code captured by the image received from the code identification unit 112, and can extract the content stored in the read coordinate information, and can output the extracted content to the user through the speaker unit 115. In the embodiments disclosed herein, the coordinate information and character information are transmitted to the first server 120, including the user's ID information and the ID information of the electronic pen 110. The first server 120 includes a second communication unit 121, a first database 122, and a second control unit 123. When the second communication unit 121 receives coordinate information from the electronic pen 110, the second control unit 123 stores the coordinate information in the first database 122. The second communication unit 121 is essentially the same as the main server communication unit 2200 described with reference to FIG4. The first database 122 can pre-store prompts corresponding to coordinate information. A prompt refers to input data used to give commands or instructions to the conversational AI server 130. The first database 122 is an element corresponding to the database 2150 illustrated with reference to FIG4. The first database 122 can store information about the media. For example, the first database 122 may include at least one of the following groups of media IDs, media titles, all text printed on the media, text printed on a specific page of the media, and text printed at a specific coordinate on the media: for each of the plurality of media. Additionally, the first database 122 can store predetermined prompts and / or prompt templates. The second control unit 123 extracts the prompt corresponding to the received coordinate information and transmits the extracted prompt to the conversational AI server 130. The second control unit 123 is substantially the same as the main server control unit 2300 described with reference to FIG4. The second control unit 133 may load prompts from the first database 122. Alternatively, the second control unit 123 may load prompt templates and information about the medium from the first database 122 to generate prompts. When the second communication unit 121 receives character information from the electronic pen 110, the second control unit 123 stores the character information in the first database 122 and transmits the received character information to the conversational AI server 130. Conversational AI Server 130 is a server in which a language processing model for using AI is embedded, and it means a server for providing conversational AI services. Examples of conversational AI services include OpenAI's ChatGPT, Google's Bard or Gemini, and Microsoft's Copilot. The conversational AI server 130 uses a language processing model with prompts or character information from the first server 120 as input to extract a result value, and transmits the extracted result value to the first server 120. The result value consists of string data. The second communication unit 121 receives the result value from the conversational AI server 130 and transmits the result value to the electronic pen 110. A session refers to a series of processes in which character information or prompts are transmitted to the conversational AI server 130 and the result values ​​are received. When a session terminates, the second control unit 123 associates the received result value with the session ID, the character information or prompts transmitted to the conversational AI server 130, and the IDs of the user and the electronic pen 110 on which the character information or prompts are based to generate session information, and stores the session information in the first database 122. The second control unit 123 stores this information in the form of a JavaScript Object Notation (JSON) file, and may also store the time information of the session ID. The conversation information may include the main text obtained by converting the user's voice signal, prompts generated by the first server 120, and result values ​​generated by the conversational AI server 130. The second control unit 123 retrieves at least one piece of conversation information stored in the first database 122 for a predetermined period based on the user ID or electronic pen ID. The second control unit 123 transmits predetermined conversation prompt information and the retrieved at least one piece of conversation information to the conversational AI server 130 via the second communication unit 121. The predetermined conversation prompt may include details related to a summary of the at least one piece of conversation information. Additionally, the predetermined conversation prompt may include details related to the analysis of the at least one piece of conversation information. The conversational AI server 130 uses a language processing model with conversational prompts received from the first server 120 as input to extract a result value and transmits the extracted result value to the first server 120. More specifically, the first server 120 uses session information to generate session prompts and transmits the generated session prompts to the conversational AI server 130. The conversational AI server 130 can receive session prompts and output at least one of the following: conversation log, conversation summary, pronunciation evaluation, frequently used keywords by the user, user's media usage time, user's media type-specific usage time, and user interest groups. The conversational AI server 130 can provide output analysis information to the first server 120. The first server 120 stores the received session prompt result value (or analysis information) in the first database 122. The first server 120 provides the user with at least one session information and session prompt result value according to the user's request. Specifically, the first server 120 can transmit analytics information to the terminal (e.g., laptop, desktop computer, or smartphone) of a guardian (e.g., the child's parent if the user is a child). The guardian can then view the analytics information on his / her terminal. Simultaneously, the first server 120 can use session information instead of the conversational AI server 130 to generate analysis information. The analysis information may include at least one of the following: a conversation log indicating the conversation between the user and the electronic pen 110; a conversation summary outlining the conversation between the user and the electronic pen 110; pronunciation evaluation information indicating evaluations of the user's pronunciation; frequently used keywords by the user; the user's media usage time; the user's media type-specific usage time; and the user's areas of interest. Figure 7 is a block diagram illustrating an electronic pen service system using a conversational AI service according to another embodiment. Referring to Figure 7, the electronic pen service system 101 using conversational AI services includes an electronic pen 110, a first server 120, a conversational AI server 130, and a second server 140. In this document, the electronic pen service system 101 corresponds to the interactive system 100, the electronic pen 110 corresponds to the electronic device 1000, the first server 120 corresponds to the main server 2000, and the conversational AI server 130 corresponds to the AI ​​server 3000. The difference between Figure 6 and Figure 7 is that Figure 7 further includes the second server 140. The second server 140 can perform the functions described above: obtaining conversation information, generating conversation prompts, and obtaining analysis information. The electronic pen 110 includes a first communication unit 111, a code recognition unit 112, a first control unit 113, a TTS unit 114, and a speaker unit 115. The first communication unit 111 can communicate with the first server 120 using a wired or wireless interface. The first communication unit 111 is substantially the same as the electronic device communication unit 1500 illustrated with reference to FIG2. The code recognition unit 112 recognizes a predetermined code from a printed item on which a predetermined code is printed. The predetermined code includes position information, i.e., coordinate information, on the printed item. The code recognition unit 112 includes a camera. When the camera approaches within a predetermined distance of the printed item on which the predetermined code is printed, the camera captures the predetermined code within its recognizable area. The code recognition unit 112 then transmits the image captured by the camera to the first control unit 113. The code recognition unit 112 is substantially the same as the sensing unit 1100 illustrated with reference to FIG2. The first control unit 113 reads coordinate information from a predetermined code in the image received by the code identification unit 112. The first control unit 113 is substantially the same as the electronic device control unit 1600 illustrated with reference to FIG2. Subsequently, the first control unit 113 transmits the read coordinate information to the first server 120 via the first communication unit 111. As will be explained later, the first communication unit 111 receives the result value of the coordinate information transmitted from the first server 120. Alternatively, region identification information can be used instead of coordinate information. For example, the first control unit 113 can transmit the read region identification information to the first server 120 via the first communication unit 111. In this case, the first server 120 can transmit a prompt corresponding to the region identification information to the conversational AI server 130, and the conversational AI server 130 can generate a result value for the received prompt and transmit the result value to the first server 120. The TTS unit 114 is an element equipped with text-to-speech (TTS) functionality and performs the function of converting text into speech. The TTS unit 114 converts the result value received by the first communication unit 111 into speech and outputs the converted speech to the user of the electronic pen 110 via the speaker unit 115. The TTS unit 114 is an element corresponding to the TTS model 2130 illustrated with reference to FIG4. The TTS unit 114 may be included in the first server 120 rather than in the electronic pen 110. In this case, the electronic pen 110 may receive the speech obtained by converting the result value from the first server 120 and may output the converted speech to the speaker unit 115. In an additional embodiment, the electronic pen 110 may further include a microphone 116 and an STT unit 117. The electronic pen 110 receives voice input from the user via the microphone 116. STT unit 117 is an element equipped with speech-to-text (STT) functionality and performs the function of converting speech into text. STT unit 117 converts speech information received by microphone 116 into character information and transmits the converted character information to first server 120. Then, first communication unit 111 receives the result value of the transmitted character as a character from first server 120, and can convert the character received by first communication unit 111 into speech and output the speech to the user of electronic pen 110 via speaker unit 115. STT unit 117 is an element corresponding to the STT model 2110 illustrated with reference to FIG4. STT unit 117 may be included in first server 120 but not in electronic pen 110. In this case, electronic pen 110 may transmit speech signals to first server 120, and first server 120 may use STT unit 117 to convert speech signals into character information. In another embodiment, the electronic pen 110 may further include a storage unit 118. Storage unit 118 includes the content specified in the coordinate information. Storage unit 118 may be included in the first server 120 rather than in the electronic pen 110. The first control unit 113 can read coordinate information from the predetermined code captured by the image received from the code identification unit 112, and can extract the content stored in the read coordinate information, and can output the extracted content to the user through the speaker unit 115. In the embodiments disclosed herein, the coordinate information and character information are transmitted to the first server 120, including the user's ID information and the ID information of the electronic pen 110. The first server 120 includes a second communication unit 121, a first database 122, and a second control unit 123. When the second communication unit 121 receives coordinate information from the electronic pen 110, the second control unit 123 stores the coordinate information in the first database 122. The second communication unit 121 is essentially the same as the main server communication unit 2200 described with reference to FIG4. The first database 122 can pre-store prompts corresponding to coordinate information. A prompt refers to input data used to give commands or instructions to the conversational AI server 130. The first database 122 is an element corresponding to the database 2150 illustrated with reference to FIG4. The first database 122 can store information about the media. For example, the first database 122 may include at least one of the following groups of media IDs, media titles, all text printed on the media, text printed on a specific page of the media, and text printed at a specific coordinate on the media: for each of the plurality of media. Additionally, the database 2150 can store predetermined prompts and / or prompt templates. The second control unit 123 extracts the prompt corresponding to the received coordinate information and transmits the extracted prompt to the conversational AI server 130. The second control unit 123 is substantially the same as the main server control unit 2300 illustrated with reference to FIG4. The second control unit 133 may load prompts from the first database 122. Alternatively, the second control unit 123 may load prompt templates and information about the medium from the first database 122 to generate prompts. When the second communication unit 121 receives character information from the electronic pen 110, the second control unit 123 stores the character information in the first database 122 and transmits the received character information to the conversational AI server 130. Conversational AI Server 130 is a server in which a language processing model for using AI is embedded, and it means a server for providing conversational AI services. Examples of conversational AI services include OpenAI’s ChatGPT, Google’s Bard or Gemini, and Microsoft’s Copilot. The conversational AI server 130 uses a language processing model with prompts or character information from the first server 120 as input to extract a result value and transmits the extracted result value to the first server 120. The result value consists of string data. The second communication unit 121 receives the result value from the conversational AI server 130 and transmits the result value to the electronic pen 110. In addition, the second communication unit 121 transmits the result value to the second server 140. The second server 140 includes a third communication unit 141, a second database 142, and a third control unit 143. The third communication unit 141 receives the result value from the first server 120 and stores the result value in the second database 142. A session refers to a series of processes in which the conversational AI server 130 transmits character information or prompts and receives the resulting values. When a session terminates, the third control unit 143 associates the received result value with the session ID, the character information or prompts transmitted to the conversational AI server 130, and the IDs of the user and the electronic pen 110 on which the character information or prompts were based to generate session information, and stores the session information in the second database 142. The third control unit 143 stores this information in the form of a JSON file, and may also store the time information of the session ID. The third control unit 143 extracts at least one session information that has been stored in the second database 142 for a predetermined period based on the user ID or electronic pen ID. The third control unit 143 transmits predetermined session prompt information and the extracted at least one session message to the conversational AI server 130 via the third communication unit 141. The predetermined session prompt may include details related to a summary of the at least one session message. In addition, the predetermined session prompt may include details related to the analysis of the at least one session message. The conversational AI server 130 uses a language processing model with conversational prompts received from the second server 140 as input to extract the result value and transmits the extracted result value to the second server 140. The second server 140 stores the received session prompt result value in the second database 142. The second server 140 provides the user with at least one session information and session prompt result value according to the user's request. 3. Information used in the interactive system The information stored in the database 2150 of the interactive system 100 will be described below with reference to Figure 8. Figure 8 is a diagram illustrating multiple pieces of information stored in database 2150 according to an embodiment. The multiple pieces of information stored in database 2150 can be used by the main server 2000, and specifically, can be used by the main server 2000 to generate prompts. Referring to Figure 8, the database 2150 can store information about the medium (information about the first medium to information about the nth medium), prompts (first prompt to nth prompt) and prompt templates (first prompt template to nth prompt template). As will be explained later, information about the medium can be used to generate prompts. Information about the medium can be understood as information used to identify what media a user is using or what parts of the user are interested in. Information about the first medium refers to information about the first medium among the plurality of media. Information about the first medium may include the first medium ID, the first medium type, the first medium name, the first medium summary, the first medium body text, the first medium page-specific body text, the first medium area-specific body text, the first medium page-specific image, the first medium area-specific image, first medium issues, and first medium instructions. The first media ID is information used to identify a first media among the plurality of media. The first media may be assigned a first unique identifier, and the first unique identifier may be the first media ID (i.e., the first media ID). The first type of medium refers to the form of the medium and can be books, cards, or maps. The primary medium name is information that names the primary medium. When the primary medium is a book, the primary medium name is the title of the book. When the primary medium is a card, the primary medium name refers to the name of the card. A primary medium summary is information that provides an overview of the content of the primary medium. When the primary medium is a book, the primary medium summary is a summary of the story printed on the book. When the primary medium is a card, the primary medium summary is omitted or is a summary of the content printed on the card. The primary medium text is information about the content of the primary medium. When the primary medium is a book, the primary medium text means the entire text (or story or document) or part of the text of the book printed on the primary medium. When the primary medium is a book, the primary medium page-specific text is the text printed on every page of the book. When the primary medium is a card, the primary medium page-specific text is the image printed on the front or back of the card. The primary medium-specific text is the text printed in a specific area on a specific page (or side) of the primary medium. When the primary medium is a book, the primary medium-specific text is the text (paragraph or sentence) printed in a specific area on a specific page of the primary medium. When the primary medium is a card, the primary medium-specific text is the text printed in a specific area on the front or back of the card. In this document, when information about each medium is stored, the specific area can be preset, and the specific area can have various shapes, such as polygons, circles, ovals, or any graphic. When the primary medium is a book, the primary medium page-specific image is an illustration printed on each page of the book. When the primary medium is a card, the primary medium page-specific image is an illustration printed on the front or back of the card. A first-medium area-specific image is an illustration printed in a specific area on a specific page (or specific side) of the first medium. When the first medium is a book, the first-medium area-specific image is an illustration printed in a specific area on a specific page of the first medium. When the first medium is a card, the first-medium area-specific image is an illustration printed in a specific area on the front or back of the card. In this document, when information about each medium is stored, the specific area can be preset, and the specific area can have various shapes, such as polygons, circles, ovals, or any graphic. The text specific to the first media page, the text specific to the first media area, the image specific to the first media page, and the image specific to the first media area can be understood as media context information. For example, as will be explained later, when a user speaks or asks a question while touching or pointing to a specific page of the first media or the text or image on that specific page, the main server 2000 can obtain at least one of the following groups of text specific to the first media page, text specific to the first media area, image specific to the first media page, and image specific to the first media area as media context information, and can use the media context information to generate prompts. The problem of the primary medium is the problem of what is printed on the primary medium. When the primary medium is a book, the questions for the primary medium may include questions about the story of the primary medium. For example, the questions for the primary medium may include quizzes about the story of the primary medium or questions asking about the feelings of the main characters. In the primary medium, different questions may be printed on each page. Specifically, on each page of the primary medium, questions about the text printed on that page may be printed. When the primary medium is a card, the primary medium questions may include questions related to the content of the primary medium. For example, when the content of the primary medium is a "self-introduction," the primary medium questions may include "What is your name?" and / or "What are your hobbies?" First-medium instructions are instructions printed on a first medium. When the primary medium is a book, the instructions for the primary medium may include instructions about the story of the primary medium. For example, the instructions for the primary medium may include "expressing the user's thoughts" and "discussing the user's experience" regarding the story of the primary medium. In the primary medium, different instructions may be printed on each page. Specifically, on each page of the primary medium, instructions regarding the text printed on said page may be printed. When the primary medium is a card, the instructions for the primary medium may include instructions related to the content of the primary medium. For example, when the content of the primary medium is "future wishes," the instructions for the primary medium may include "talk about who you want to be" and "talk about what you are good at." The first media question or first media instruction can be understood as human context information. For example, as will be explained later, when a user touches or points to the area in the first media where the question is printed, the main server 2000 can obtain the first media question or first media instruction, or both, as human context information and use the human context information to generate a prompt. In addition to the text, information about the media may also include sound, image, or video. For example, database 2150 may include audio about a summary of the media, audio about the text printed on the media (text printed on each page or each area), and audio about questions or instructions printed on the media. Furthermore, database 2150 may include video related to the media, video related to each page of the media, and video related to areas or printed images of the media. Therefore, the contextual information described below may include at least one selected from the group consisting of text, sound, image, and video. A prompt refers to information input to the AI ​​server 3000. The AI ​​server 3000 can receive prompts and generate a response to them. Prompts may include contextual information. Prompts stored in the database 2150 may include contextual information related to a specific medium. The following section will provide specific examples of the prompts. The prompt template may include details used to generate the prompt. Specifically, the prompt template may include statements or paragraphs that enable the AI ​​server 3000 to understand the context. The prompt template may be modified using additional information such as context information (e.g., adding context information) to become a prompt to be input to the AI ​​server 3000. The main server 2000 can generate prompts using prompt templates and text data obtained from converting user speech data. Alternatively, the main server 2000 can generate prompts using prompt templates and media context information. Alternatively, the main server 2000 can generate prompts using prompt templates and human context information. Alternatively, the main server 2000 can generate prompts using prompt templates, human context information, and media context information. The following section will provide specific examples of the prompt template. In addition to the aforementioned multiple pieces of information, database 2150 may include other multiple pieces of information. Furthermore, the form in which multiple pieces of information are stored in database 2150 is not limited to the forms described above, and various methods of storing multiple pieces of information may exist (e.g., categories and relationships between multiple pieces of information). The process of obtaining contextual information will be described below. As mentioned above, contextual information refers to information related to the area where the user interacts with the medium using the electronic device 1000. Contextual information can be used to generate prompts. When a user interacts with the medium using the electronic device 1000, the user can use the electronic device 1000 to touch or point at a specific area within the medium. For example, when the user positions the electronic device 1000 to touch or approach a specific area within the medium, the sensing unit 1100 of the electronic device 1000 can capture the specific area to obtain a coded image. The code image can be analyzed by electronic device 1000 or main server 2000, and the code data can be obtained accordingly. The main server 2000 can obtain reference information from the database. In this document, reference information means information used to load required information from the database 2150. For example, reference information may include at least one of the following groups: media ID, page information, and location information (coordinate information and / or zone identification information). The main server 2000 can use reference information to search for required information in the database 2150 and obtain the required information as context information. Context information may include information corresponding to at least a portion of the text or illustrations printed on the medium. The main server 2000 may use at least the context information to generate prompts. The information included in the context information can be determined based on the area of ​​the medium in which the electronic device 1000 is touched or positioned. In other words, when the user positions the electronic device 1000 in different areas within the medium, the multiple pieces of context information obtained by the main server 2000 need to be different from each other. This means that different codes are printed in the corresponding areas within the medium, and the code images obtained by the electronic device 1000 are also different from each other. The following text will illustrate the pattern of separating the various areas within the medium with reference to Figures 9 and 10. Figure 9 is a diagram showing the pattern of separating the various regions in the medium according to an embodiment. Referring to Figure 9, the medium may include an illustration area, a text area, a question area, page numbers, and symbols. The codes printed on the corresponding areas may have different patterns. For example, when decoding code data obtained by capturing codes printed in the corresponding areas, multiple different area identification information can be obtained. The illustration area is a region in which a specific illustration is printed. When decoding the code data obtained by touching the illustration area via the electronic device 1000, at least the media ID, page information, and region identification information indicating the illustration area can be obtained. The main server 2000 can use the obtained media ID, page information, and region identification information to obtain the image corresponding to the illustration printed in the area from the database 2150. The text area is the area where the text (or document, written content, story) is printed. When decoding the code data obtained by touching the text area via the electronic device 1000, at least the media ID, page information, and area identification information indicating the text area can be obtained. The main server 2000 can use the obtained media ID, page information, and area identification information to obtain the text corresponding to the text printed in the area from the database 2150. At the same time, a text area can be set for each of the sentences that make up the main text. In this case, multiple text areas can be set within a single page. The problem area is the area where problems are printed. When decoding the code data obtained by touching the problem area via the electronic device 1000, at least the media ID, page information, and area identification information indicating the problem area can be obtained. The main server 2000 can use the obtained media ID, page information, and area identification information to obtain the text corresponding to the problem printed in the area from the database 2150. Additionally, at least one of the groups selected from symbols, questions, and instructions may be printed in the question area, which may be named differently depending on the printing details. For example, when instructions are printed, the question area may be named the instruction area. Alternatively, when only symbols are printed, the question area may be named the symbol area. The areas that distinguish each other in the media are not limited to the areas mentioned above, and the units and rules used to distinguish each area can be determined in various ways. Figure 10 is a diagram illustrating the pattern of separating different areas in a medium according to another embodiment. Referring to Figure 10, in the medium, the question area and the interpretation area can be distinguished from each other. An image can be printed in the question area, and text can be printed in the interpretation area. The codes printed on the corresponding areas can have different patterns. For example, when decoding code data obtained by capturing codes printed in the corresponding areas, different multiple area identification information can be obtained. When decoding the code data obtained by touching the problem area via the electronic device 1000, at least the media ID and area identification information indicating the problem area can be obtained. The main server 2000 can use the obtained media ID and area identification information to obtain the text corresponding to the problem related to the media from the database 2150. When decoding the code data obtained by touching the interpreter area of ​​the electronic device 1000, at least the media ID and area identification information indicating the interpreter area can be obtained. The main server 2000 can use the obtained media ID and area identification information to obtain the text corresponding to the interpreter related to the media from the database 2150. The zones that can be distinguished from each other in a medium are not limited to those mentioned above, and the type of zone, the shape of the graphic of the zone, and the pattern of the code printed in the zone can be determined in various ways. 4. Interaction using human contextual information The following will describe, with reference to Figures 11 to 22, how the interactive system 100 uses artificial contextual information to assist the user in using the medium. The primary purpose of media is to inform users of specific stories or information, such as fairy tales, nature, animals, or science. In addition to simply conveying specific stories or information, media can also be generated to include details (such as questions, instructions, exercises, or exploration activities), allowing users to gain a deeper understanding of the media's nuances. In this scenario, with the assistance of a guardian, the user can only answer questions or follow instructions, and cannot engage in further learning or exploration. Therefore, assisting the user in using the media on behalf of the guardian can deepen the user's understanding of the story or content printed on the media and increase the user's interest. The following section will describe the interactive system 100 used to assist users in using media on behalf of their guardians, with particular emphasis on how the interactive system 100 generates interactive prompts when stories or information, along with additional questions or instructions, are printed on the media. Using human contextual information to assist users in using media is considered in scenarios such as the following: A child is reading a storybook without a guardian, the storybook contains the story and quizzes or additional questions about the story, and the child wants to use the interactive system 100 to have a conversation about the quiz or additional questions. Figure 11 is a flowchart illustrating a method for assisting a user in using a medium using artificial contextual information according to a first embodiment. Hereinafter, as described above with reference to FIG6, the electronic device 1000 is named electronic pen 110, the main server 2000 is named first server 120, and the AI ​​server 3000 is named conversational AI server 130. The details illustrated with reference to FIG6 can be applied to each element in the same manner. Referring to Figure 11, in step 301, the electronic pen identifies the predetermined code from the printed item on which the predetermined code is printed and extracts coordinate information. The predetermined code includes position information (i.e., coordinate information) on the printed item. The electronic pen includes a camera. When the camera approaches within a predetermined distance from the printed item on which the predetermined code is printed, the camera captures the predetermined code within its recognizable area. Then, the electronic pen reads the coordinate information from the predetermined code in the captured image. In step 302, the electronic pen transmits the read coordinate information to the first server. Alternatively, the electronic pen transmits the captured image to the first server, and the first server analyzes the captured image to obtain code data and from the obtained code data, at least the media ID and coordinate information. In this document, page information may also be obtained from the code data. The electronic pen can further transmit the user's ID information and the electronic pen's ID information to the first server. The first server associates and stores the received coordinate information, the electronic pen user's ID information, and the electronic pen's ID information. In step 303, the first server extracts the prompts corresponding to the coordinate information. Prompts corresponding to coordinate information can be pre-stored in the first server. A prompt refers to input data used to give commands or instructions to the conversational AI server. The first server can obtain prompts corresponding to the media ID, prompts corresponding to coordinate information, or prompts corresponding to both the media ID and coordinate information from the first database. Alternatively, the first server can search the first database for multiple pieces of information about the media and coordinate information corresponding to the media ID, and obtain the found information as context information. In this paper, the first server may give more consideration to page information when obtaining context information. The first server can use contextual information to generate a prompt. For example, the first server can add details to the contextual information to generate a prompt. As another example, the first server can obtain a prompt template corresponding to at least one of a group selected from media ID, page information, and coordinate information, and can use the contextual information and the prompt template to generate a prompt. In step 304, the first server transmits the extracted prompts to the conversational AI server. Conversational AI servers are servers that embed language processing models for using AI, and are intended to provide conversational AI services. Examples of conversational AI services include OpenAI's ChatGPT, Google's Bard or Gemini, and Microsoft's Copilot. In step 305, the conversational AI server uses a language processing model to take the prompts received from the first server as input values ​​and generate and extract result values. The result value consists of string data. In step 306, the conversational AI server transmits the extracted result values ​​to the first server. In step 307, the first server transmits the received result value to the electronic pen. In step 308, the electronic pen uses its text-to-speech function to convert the received result value into speech. In step 309, the electronic pen outputs the converted voice to the user of the electronic pen. In step 310, the first server generates session information. A session refers to a series of processes in which a conversational AI server sends prompts and receives the resulting values. When a session terminates, the first server associates the received result value with the session ID, the prompt sent to the conversational AI server, and the IDs of the user and the electronic pen on which the prompt was based, to generate session information. The first server also stores this session information. This information is stored as a JSON file, and may also store the time information of the session ID. In step 311, the first server extracts at least one session information that has been stored for a predetermined period based on the user ID or electronic pen ID, and transmits the predetermined session prompt information and the extracted at least one session information to the conversational AI server. The scheduled session notification may include details related to a summary of at least one session message. Additionally, the scheduled session notification may include details related to the analysis of at least one session message. In step 312, the conversational AI server uses a language processing model to take the conversational prompts received from the first server as input values ​​and generate result values. In step 313, the conversational AI server transmits the resulting value to the first server. The first server stores the received session prompt results. Based on the user's request, the first server provides the user with at least one session message and the session prompt results. Figure 12 is a flowchart illustrating a method for assisting a user in using a medium using artificial context information according to a second embodiment of this disclosure. Hereinafter, as described above with reference to FIG6, the electronic device 1000 is named electronic pen 110, the main server 2000 is named first server 120, and the AI ​​server 3000 is named conversational AI server 130. The details illustrated with reference to FIG6 can be applied to each element in the same manner. Referring to Figure 12, in step 501, the electronic pen receives the user's voice input via a microphone and converts the voice information into character information. In step 502, the electronic pen transmits the character information obtained through conversion to the first server. The first server associates and stores the received character information, the electronic pen user's ID information, and the electronic pen's ID information. The first server can use character information to generate a prompt. For example, the first server can obtain a prompt template corresponding to a media ID from a first database, and can use the prompt template and character information to generate a prompt. As another example, the first server can add details to the character information or modify the character information to generate a prompt. As yet another example, the first server can use a prompt generated before obtaining the character information, along with the character information, to generate a new prompt. In step 503, the first server transmits character information to the conversational AI server. The first server can then transmit prompts to the AI ​​server. Conversational AI servers are servers that embed language processing models for using AI, and are intended to provide conversational AI services. Examples of conversational AI services include OpenAI's ChatGPT, Google's Bard or Gemini, and Microsoft's Copilot. In step 504, the conversational AI server uses a language processing model to take the character information received from the first server as input and generates and extracts the result value. Alternatively, the conversational AI server can use prompts received from the first server as input and generate the result value. The result value consists of string data. In step 505, the conversational AI server transmits the extracted result value to the first server. In step 506, the first server transmits the received result value to the electronic pen. In step 507, the electronic pen uses its text-to-speech function to convert the received result value into speech. In step 508, the electronic pen outputs the converted voice to the user of the electronic pen. In step 509, the first server generates session information. A session refers to a series of processes in which a conversational AI server transmits character information and receives the resulting values. When a session terminates, the first server associates the received result value with the session ID, the character information transmitted to the conversational AI server, and the IDs of the user and the electronic pen on which the input character information was based, to generate session information. Furthermore, the first server stores this session information. The first server stores this information in the form of a JSON file, and may also store the time information of the session ID. In step 510, the first server extracts at least one session information that has been stored for a predetermined period based on the user ID or electronic pen ID, and transmits the predetermined session prompt information and the extracted at least one session information to the conversational AI server. The scheduled session notification may include details related to a summary of at least one session message. Additionally, the scheduled session notification may include details related to the analysis of at least one session message. In step 511, the conversational AI server uses a language processing model to take the conversational prompts received from the first server as input values ​​and generate result values. In step 512, the conversational AI server transmits the resulting value to the first server. The first server stores the received session prompt results. Based on the user's request, the first server provides the user with at least one session message and the session prompt results. Figure 13 is a flowchart illustrating a method for assisting a user in using a medium using artificial context information according to a third embodiment of this disclosure. Hereinafter, as described above with reference to FIG7, the electronic device 1000 is named electronic pen 110, the main server 2000 is named first server 120, and the AI ​​server 3000 is named conversational AI server 130. The details illustrated with reference to FIG7 can be applied to each element in the same manner. Referring to Figure 13, in step 401, the electronic pen identifies the predetermined code from the printed item on which the predetermined code is printed and extracts coordinate information. The predetermined code includes position information (i.e., coordinate information) on the printed item. The electronic pen includes a camera. When the camera approaches within a predetermined distance from the printed item on which the predetermined code is printed, the camera captures the predetermined code within its recognizable area. Then, the electronic pen reads the coordinate information from the predetermined code in the captured image. In step 402, the electronic pen transmits the read coordinate information to the first server. The electronic pen can further transmit the user's ID information and the electronic pen's ID information to the first server. The first server associates and stores the received coordinate information, the electronic pen user's ID information, and the electronic pen's ID information. In step 403, the first server extracts the prompt corresponding to the coordinate information. The prompts corresponding to the coordinate information are pre-stored in the first server. The prompts refer to the input data used to give commands or instructions to the conversational AI server. In step 404, the first server transmits the extracted prompts to the conversational AI server. Conversational AI servers are servers that embed language processing models for leveraging AI, and also refer to servers that provide conversational AI services. Examples of conversational AI services include OpenAI's ChatGPT, Google's Bard or Gemini, and Microsoft's Copilot. In step 405, the conversational AI server uses a language processing model to take the prompts received from the first server as input values ​​and generates and extracts the result values. The result value consists of string data. In step 406, the conversational AI server transmits the extracted result value to the first server. In step 407, the first server transmits the received result value to the electronic pen. In step 408, the electronic pen uses its text-to-speech function to convert the received result value into speech. In step 409, the electronic pen outputs the converted voice to the user of the electronic pen. In step 410, the first server transmits the received result value to the second server. In step 411, the second server generates session information. A session refers to a series of processes in which a conversational AI server sends prompts and receives result values ​​in response. When a session terminates, the first server associates the received result values ​​with the session ID, the prompt sent to the conversational AI server, and the IDs of the user and the electronic pen on which the prompt was based, to generate session information. The first server also stores this session information. The first server stores this information in the form of a JSON file, and may also store the time information of the session ID. In step 412, the second server extracts at least one session information that has been stored for a predetermined period based on the user ID or electronic pen ID, and transmits the predetermined session prompt information and the extracted at least one session information to the conversational AI server. The scheduled session notification may include details related to a summary of at least one session message. Additionally, the scheduled session notification may include details related to the analysis of at least one session message. In step 413, the conversational AI server uses a language processing model to take the conversational prompts received from the second server as input values ​​and generate result values. In step 414, the conversational AI server transmits the resulting value to the second server. The second server stores the received session prompt results. Based on the user's request, the second server provides the user with at least one session message and the session prompt results. Figure 14 is a flowchart illustrating a method for assisting a user in using a medium using artificial contextual information according to a fourth embodiment. In the following text, as described above with reference to FIG7, the electronic device 1000 is named electronic pen 110, the main server 2000 is named first server 120, and the AI ​​server 3000 is named conversational AI server 130. The details illustrated with reference to FIG7 can be applied to each element in the same manner. In step 601, the electronic pen receives the user's voice input via a microphone and converts the voice information into character information. In step 602, the electronic pen transmits the character information obtained through conversion to the first server. The first server associates and stores the received character information, the electronic pen user's ID information, and the electronic pen's ID information. In step 603, the first server transmits character information to the conversational AI server. Conversational AI servers are servers that embed language processing models for using AI, and are intended to provide conversational AI services. Examples of conversational AI services include OpenAI's ChatGPT, Google's Bard or Gemini, and Microsoft's Copilot. In step 604, the conversational AI server uses a language processing model to take the character information received from the first server as input value, and generates and extracts the result value. The result value consists of string data. In step 605, the conversational AI server transmits the extracted result value to the first server. In step 606, the first server transmits the received result value to the electronic pen. In step 607, the electronic pen uses its text-to-speech function to convert the received result value into speech. In step 608, the electronic pen outputs the converted voice to the user of the electronic pen. In step 609, the first server transmits the received result value to the second server. In step 610, the second server generates session information. A session refers to a series of processes involving the transmission of character information or prompts to a conversational AI server and the receipt of the resulting values. When a session terminates, the first server associates the received result values ​​with the session ID, the character information transmitted to the conversational AI server, and the IDs of the user and the electronic pen upon which the transmitted character information was based, to generate session information. Additionally, a second server stores this session information. The second server stores this information in the form of a JSON file, and may also store the time information of the session ID. In step 611, the second server extracts at least one session information that has been stored for a predetermined period based on the user ID or electronic pen ID, and transmits the predetermined session prompt information and the extracted at least one session information to the conversational AI server. The scheduled session notification may include details related to a summary of at least one session message. Additionally, the scheduled session notification may include details related to the analysis of at least one session message. In step 612, the conversational AI server uses a language processing model to take the conversational prompts received from the second server as input values ​​and generate result values. In step 613, the conversational AI server transmits the resulting value to the second server. The second server stores the received session prompt results. Based on the user's request, the second server provides the user with at least one session message and the session prompt results. Figure 15 is a flowchart illustrating a method for assisting a user in using media using artificial contextual information according to a fifth embodiment. Figure 16 is a diagram illustrating the process of obtaining artificial contextual information according to the fifth embodiment. Figure 17 is a diagram illustrating a prompt generated using human contextual information according to the fifth embodiment. Referring to FIG15, the method for assisting in the use of a medium may include: obtaining a code image in step S1100, analyzing the code image in step S1200 to obtain code data, obtaining reference information from the code data in step S1300, obtaining artificial context information based on the reference information in step S1400, using the artificial context information to generate a prompt in step S1500, using the prompt to generate response data in step S1600, obtaining response voice data corresponding to the response data in step S1700, and outputting the response voice data in step S1800. Each step will be explained in detail below. In step S1100, a code image can be obtained. Specifically, the code is printed on the medium along with the content (text or illustration), and when the user positions the electronic device 1000 to touch or approach the area of ​​interest on the medium, the electronic device 1000 can use the sensing unit 1100 to capture at least a portion of the area of ​​interest and obtain a code image. Here, the sensing unit 1100 can be activated periodically to obtain an image. Alternatively, the sensing unit 1100 can be activated and an image obtained only when the user manipulates a button on the electronic device 1000. Referring to FIG16, for example, a user can use electronic device 1000 to touch a portion of the question area (or the graphic portion of the question area) on page 14 of a book, and electronic device 1000 can use sensing unit 1100 to obtain a code image of the code printed in the touched area. In step S1200, the code image can be analyzed to obtain code data. Specifically, the electronic device 1000 or the main server 2000 can analyze the code image to obtain code data. For example, the electronic device 1000 can transmit the code image to the main server 2000, and the main server 2000 can obtain code data based on the arrangement of multiple fine points included in the code image. As another example, the electronic device 1000 can obtain code data from the code image and transmit the obtained code data to the main server 2000. In this document, code data can refer to data that encodes specific information among multiple pieces of information on a medium. In step S1300, reference information can be obtained from the coded data. Specifically, referring to FIG16, the main server 2000 can decode the coded data to obtain at least one selected from the group of media type, media ID, page information, and location information (coordinate information and / or zone identification information). For example, the reference information obtained in FIG16 may include {book (media type), BOOK000 (media ID), 14 (page information), A1 (zone identification information)}. In step S1400, artificial contextual information can be obtained based on reference information. Specifically, referring to Figure 16, the main server 2000 can search the database 2150 for information corresponding to the reference information among multiple pieces of information on the media to obtain the media title and question. As mentioned above, the question is information that is not very relevant to the flow of the story printed on the media, and since the information is added by the content creator to improve the user's understanding or interest in the story, it can be understood as artificial contextual information. For example, in Figure 16, the obtained media title is "Snow White and the Seven Dwarfs," and the question is "Why is the Queen jealous of the Princess?" In step S1500, human context information can be used to generate prompts. Specifically, the main server 2000 can use at least human context information to generate prompts. For example, the main server 2000 can load a prompt template from the database 2150 and can use human context information to modify the prompt template to generate a prompt. The main server 2000 can load a prompt template corresponding to reference information or context information from the database 2150. Furthermore, when modifying the prompt template, the main server 2000 can use multiple pieces of information other than human context information, such as context information or user information. For example, referring to Figure 17, the main server 2000 can load a first prompt template from the database 2150. The first prompt template is the body text that includes the media title, user context information, and additional information. The main server 2000 can modify the first prompt template using a pre-obtained media title "Snow White and the Seven Dwarfs" and the question "Why is the Queen jealous of the Princess?", as well as pre-stored user information "Child is 5 years old" (other user information, such as the user's gender, personality, or family relationship, can also be used). The main server 2000 can modify the first prompt template to generate the first prompt. As another example, the master server 2000 can modify the human context information or add details to generate a prompt. As yet another example, the master server 2000 can generate a prompt that only includes the human context information. The prompt may include explanatory text that clarifies the artificial context information. Additionally, the prompt may further include guidance text outlining the factors the AI ​​server 3000 should consider when generating a response. In step S1600, the prompt can be used to generate response data. Specifically, the main server 2000 can transmit the prompt generated in step S1500 to the AI ​​server 3000, and the AI ​​server 3000 can input the received prompt into a large language model to generate response data. For example, the AI ​​server 3000 can receive the first prompt shown in Figure 17 and generate response data, such as "Why is the Queen jealous of the Princess?" or "I think the Queen is jealous of Snow White. Why is the Queen jealous of the Princess?" The response data can be understood as data generated by the AI ​​server 2000 taking into account human context information. That is, when a user uses the medium, the interactive system 100 can guide the user to engage in dialogue, and the topic of the dialogue can be related to human context information. The response data can be text or audio. Specifically, the large language model included in the AI ​​server 3000 can be based on either text or audio. When the input prompt is in text form, the large language model can output the response data in text form. When the input prompt is in audio form, the large language model can output the audio response data. In step S1700, response voice data corresponding to the response data can be obtained. Specifically, the main server 2000 can receive response data from the AI ​​server 3000 and can use the TTS model 2130 to convert the response data into response voice data. Alternatively, the main server 2000 may not include the TTS model 2130. In this case, the main server 2000 can use an external text-to-speech conversion service to obtain the response voice data corresponding to the response data. Furthermore, when the main server 2000 receives voice response data from the AI ​​server 3000, the main server 2000 can use the data as is without conversion. In step S1800, response voice data can be output. Specifically, the main server 2000 can transmit the voice response data to the electronic device 1000, and the electronic device 1000 can output the voice response data via the electronic device output unit 1400. The method for generating follow-up prompts will be described below with reference to Figures 18 and 19. Follow-up prompts can be generated when the response data generated considering human context information is transmitted to the user and the user responds. The main server 2000 generates the follow-up prompt and transmits it to the AI ​​server 3000. The AI ​​server 3000 outputs the follow-up response data of the follow-up prompt, and the follow-up response data is transmitted to the user, thereby enabling further dialogue. Figure 18 is a flowchart illustrating a method for generating subsequent prompts according to a fifth embodiment. Figure 19 is a diagram illustrating a subsequent prompt generated using human contextual information according to the fifth embodiment. Referring to FIG18, the method for generating subsequent prompts may include obtaining the speech text corresponding to the user's speech in step S2100, using human context information and speech text to generate subsequent prompts in step S2200, using subsequent prompts to generate subsequent response data in step S2300, obtaining subsequent response speech data corresponding to the subsequent response data in step S2400, and outputting the subsequent response speech data in step S2500. Each step will be explained in detail below. In step S2100, the speech text corresponding to the user's voice can be obtained. Specifically, the electronic device 1000 can activate the microphone of the electronic device input unit 1300 by the user operating a button to record the user's voice. The electronic device 1000 can transmit the recorded user voice data to the main server 2000, and the main server 2000 can use the STT model 2110 to convert the user voice data into speech text. Alternatively, the main server 2000 may not include the STT model 2110. In this case, the main server 2000 can use an external speech-to-text conversion service to obtain the speech text corresponding to the user's voice data. In step S2200, human contextual information and spoken text can be used to generate subsequent prompts. Specifically, the main server 2000 can use at least human contextual information and spoken text to generate subsequent prompts. For example, the main server 2000 can load follow-up prompt templates from the database 2150 and can modify the follow-up prompt templates to generate prompts using human context information and voice text. In this document, the main server 2000 can load follow-up prompt templates corresponding to reference information or context information from the database 2150. Furthermore, when modifying the follow-up prompt template, the main server 2000 can use multiple pieces of information other than human context information, such as context information or user information. For example, referring to Figure 19, the main server 2000 can load a first follow-up prompt template from the database 2150. The first follow-up prompt template can be a body text that includes portions of human context information and user voice information. The main server 2000 can modify the first follow-up prompt template using the question "Why is the queen jealous of the princess?" obtained in step S1400 above and the voice text "Because the mirror says the princess is prettier" obtained in step S2100 above. The main server 2000 can modify the first follow-up prompt template to generate the first follow-up prompt. As another example, the main server 2000 can modify the speech text or add details to generate follow-up prompts. As yet another example, the main server 2000 can generate follow-up prompts that only include the speech text. Follow-up prompts may include explanatory text that elaborates on the speech content. Additionally, follow-up prompts may include guidance text outlining the factors that the AI ​​Server 3000 should consider when generating a response. In step S2300, a follow-up prompt can be used to generate subsequent response data. Specifically, the main server 2000 can transmit the follow-up prompt generated in step S2200 to the AI ​​server 3000, and the AI ​​server 3000 can input the received follow-up prompt into a large language model to generate subsequent response data. For example, the AI ​​server 3000 can receive the first follow-up prompt shown in FIG19 and generate subsequent response data, such as "Yes, the queen is jealous of the princess because of the mirror" or "Yes, the queen is jealous of the princess because of the mirror. Have you ever been jealous of someone?". Subsequent response data can be understood as data generated by the AI ​​server 2000 taking into account human context information. That is, after the interactive system 100 guides the user into a dialogue, the topic of the dialogue can continue to be related to human context information. Subsequent response data can be text or audio. Specifically, the large language model included in the AI ​​server 3000 can be based on either text or audio. When the input prompt is in text form, the large language model can output the subsequent response data in text form. When the input prompt is in audio form, the large language model can output the subsequent audio response data. In step S2400, subsequent response voice data corresponding to the subsequent response data can be obtained. Specifically, the main server 2000 can receive subsequent response data from the AI ​​server 3000 and can use the TTS model 2130 to convert the subsequent response data into subsequent response voice data. Alternatively, the main server 2000 may not include the TTS model 2130. In this case, the main server 2000 can use an external text-to-speech conversion service to obtain the subsequent response voice data corresponding to the subsequent response data. Furthermore, when the main server 2000 receives subsequent voice response data from the AI ​​server 3000, the main server 2000 can use the data as is without conversion. In step S2500, subsequent voice response data can be output. Specifically, the main server 2000 can transmit the subsequent voice response data to the electronic device 1000, and the electronic device 1000 can output the subsequent voice response data via the electronic device output unit 1400. After outputting the subsequent response voice data, the method for generating subsequent prompts can be implemented again. Specifically, when the user uses the electronic device 1000 to record his / her voice, the obtained voice text and human context information can be used to generate subsequent prompts, and subsequent voice response data for the subsequent prompts can be generated and output to the user (in this document, the subsequent prompt template may vary). This means that the dialogue between the interactive system 100 and the user continues. Simultaneously, when the human context information in use changes (e.g., when the user touches or points to another problem area using the electronic device 1000 and thus obtains another human context information), the session can be terminated and a new session can begin. In this document, a session refers to a dialogue about a piece of human context information. Starting a new session means re-implementing the method shown in Figure 15 of using human context information to assist the user in using the medium. The main server 2000 can store prompts and responses generated during the session as session information. As described above, session information can be used to generate analytical information. For example, the main server 2000 or the AI ​​server 3000 can use session information to generate at least one of the following groups: a dialogue log indicating the conversation between the user and the electronic device 1000, a dialogue summary summarizing the conversation between the user and the electronic device 1000, a pronunciation evaluation indicating the evaluation of the user's pronunciation, frequently used keywords of the user, the user's media usage time, the user's media type-specific usage time, and the user's areas of interest. In the following text, the method of using artificial contextual information to assist the user in using the medium in the interactive system 100 will be described with reference to Figures 20 to 22. Figure 20 is a diagram illustrating the process of obtaining a pre-stored prompt according to the sixth embodiment. Figure 21 is a diagram showing a pre-stored prompt according to the sixth embodiment. Figure 22 is a diagram illustrating a subsequent prompt generated using human contextual information according to the sixth embodiment. The method of using pre-stored prompts to assist users in using the medium may include the steps illustrated in Figures 15 and 18. Therefore, parts that are repeated with the above details will be omitted, and the details illustrated with reference to Figures 15 and 18 can be applied in the same manner. Referring to Figure 20, the medium is a card with specific information printed on it, and the user can position the electronic device 1000 to contact or be positioned on an area of ​​the card. In the medium, the question area and the explanation area can be distinguished from each other. When the electronic device 1000 contacts or is positioned on a region of the card, reference information can be obtained as described in steps S1100 to S1300. In this document, the reference information may include media type, media ID, and location information (coordinate information and / or region identification information). For example, referring to FIG20, the reference information may be {card (media type), CARD000 (media ID), (1308.59, 393.417) (coordinate information)}. The main server 2000 can load hints corresponding to reference information from the database 2150. The hints can be pre-stored in the database 2150 and can be input into the large language model of the AI ​​server 3000 without additional modification or processing. Meanwhile, the location information of the reference information may change depending on the area touched by the electronic device 1000, and the loaded prompt may change depending on whether the location information is included in the question area or the explanation area (when the location information is coordinate information) or whether the location information indicates the question area or the explanation area (when the location information is area identification information). For example, when location information is included in or indicates a question area, the main server 2000 can load the second prompt shown in Figure 21. The second prompt may include information about the content printed on the card and guidelines for generating a question to begin a conversation. The main server 2000 can transmit the second prompt to the AI ​​server 3000, and the AI ​​server 3000 can use the second prompt to generate response data. For example, the response data could be "Have you heard of flying squirrels?". Then, according to steps S1700 and S1800, the voice corresponding to the response data can be output to the user. After the voice corresponding to the response data is output to the user, the user can activate the recording function of the electronic device 1000 to answer the output voice, thereby obtaining the user's voice data. Then, step S2100 can be performed to obtain the voice text. The main server 2000 can use response text and voice text to generate subsequent prompts. In this document, response text refers to the response data generated using the aforementioned pre-stored prompts. For example, referring to Figure 22, the main server 2000 can load a second follow-up prompt template from the database 2150. The second follow-up prompt template can be a body text that includes a portion containing response information and user voice information. The main server 2000 can modify the second follow-up prompt template using the response text "Have you heard of flying squirrels?" and the previously obtained voice text "I know about squirrels". The main server 2000 can modify the second follow-up prompt template to generate a second follow-up prompt. As another example, the main server 2000 can modify the speech text or add details to generate follow-up prompts. As yet another example, the main server 2000 can generate follow-up prompts that only include the speech text. Subsequent prompts may include explanatory text that clarifies the speech content. Additionally, the prompts may further include guidance text outlining the factors the AI ​​server 3000 should consider when generating a response. Then, steps S2300 to S2500 can be performed, and the subsequent response voice data corresponding to the subsequent prompts can be output to the user. After outputting the subsequent response voice data, the method for generating subsequent prompts can be repeated. Specifically, when the user further records his / her voice using the electronic device 1000, the obtained voice text can be used to generate subsequent prompts, and subsequent voice response data for the subsequent prompts can be generated and output to the user. This means that the dialogue between the interactive system 100 and the user continues. In this document, the subsequent prompt template may be changed. Simultaneously, when the media ID changes (e.g., when the user touches another card using electronic device 1000), the session can be terminated and a new session can begin. In this document, a session refers to a dialogue conducted with a medium. Starting a new session means resuming the method of using the aforementioned pre-stored prompts to assist the user in using the medium. The main server 2000 can store prompts and responses generated during the session as session information. As described above, session information can be used to generate analytical information. For example, the main server 2000 or the AI ​​server 3000 can use session information to generate at least one of the following groups: a dialogue log indicating the conversation between the user and the electronic device 1000, a dialogue summary summarizing the conversation between the user and the electronic device 1000, a pronunciation evaluation indicating the evaluation of the user's pronunciation, frequently used keywords of the user, the user's media usage time, the user's media type-specific usage time, and the user's areas of interest. 5. Interactions using media context information The following will describe, with reference to Figures 23 to 27, how the interactive system 100 uses media context information to assist the user in using the media. Users (especially children) may talk about foods they are curious about or ideas that come to mind while reading a book. In such cases, the user's voice can be input into a large language model, and a response can be provided. However, in this case, the user's questions or ideas arise while reading a book. To generate more appropriate responses, the large language model needs to understand the context of the user's speech, such as the user's intention or purpose. As a method for enabling large language models to understand user context, content information printed on media can be provided to the large language model. More specifically, by inputting the user's voice along with information about the title or details of the book the user is reading, or details that the user is interested in, into the large language model, it is expected that the desired response data will be generated in the user's context. Figure 23 is a flowchart illustrating a method for assisting a user in using media by using media context information according to the seventh embodiment. Figure 24 is a diagram illustrating the process of obtaining media context information according to the seventh embodiment. Figure 25 is a diagram illustrating a prompt generated by using media context information according to the seventh embodiment. Referring to FIG23, the method for assisting in the use of a medium may include: obtaining a code image in step S3100, analyzing the code image in step S3200 to obtain code data, obtaining reference information from the code data in step S3300, obtaining media context information based on the reference information in step S3400, obtaining speech text corresponding to the user's speech in step S3500, generating a prompt using the speech text and media context information in step S3600, generating response data using the prompt in step S3700, obtaining response speech data corresponding to the response data in step S3800, and outputting the response speech data in step S3900. Each step will be described in detail below. However, step S3100 is the same as step S1100, step S3200 is the same as step S1200, step S3800 is the same as step S1700, and step S3900 is the same as step S1800. Therefore, the details described with reference to FIG15 can be applied in the same manner. After performing steps S3100 and S3200, reference information can be obtained from the code data in step S3300. In this document, the code image obtained in step S3100 can be obtained by a user using an electronic device 1000 to touch the text area on a specific page of the medium in which the text is printed. Referring to Figure 24, the main server 2000 can decode the code data to obtain at least one selected from the group of media type, media ID, page information, and location information (coordinate information and / or zone identification information). For example, the reference information obtained by the main server 2000 in Figure 24 may include {book (media type), BOOK000 (media ID), 14 (page information), A2 (zone identification information)}. In step S3400, media context information can be obtained based on reference information. Specifically, referring to Figure 24, the main server 2000 can search the database 2150 for information corresponding to the reference information among multiple pieces of information about the media to obtain the media title and page-specific text. As mentioned above, page-specific text is information about the flow of the story printed on the media and can be understood as media context information. For example, the media title obtained in Figure 24 is "Snow White and the Seven Dwarfs" and the page-specific text is "One day the magic mirror's answer began to differ. 'Snow White is the fairest of them all.' 'What? She is fairer than me?' the Queen asked again and again, but the magic mirror kept repeating that Snow White was the fairest of them all. Jealousy gleamed in the Queen's eyes." (Hereinafter referred to as "One day...jealousy gleamed in the Queen's eyes"). While media context information has been described as page-specific text, the technical concepts disclosed herein are not limited to this. Media context information may be the entire text or illustration printed on the media, or a portion of the text or illustration that is of interest to the user. That is, media context information may include text or a summary or variation thereof related to the portion of the media that the user touches using the electronic device 1000. Media context information may be phrases, sentences, or paragraphs, and may include illustrations. In step S3500, the speech text corresponding to the user's voice can be obtained. Specifically, the electronic device 1000 can activate the microphone of the electronic device input unit 1300 to record the user's voice by having the user operate a button. The electronic device 1000 can transmit the recorded user voice data to the main server 2000, and the main server 2000 can use the STT model 2110 to convert the user voice data into speech text. Alternatively, the main server 2000 may not include the STT model 2110. In this case, the main server 2000 can use an external speech-to-text conversion service to obtain the speech text corresponding to the user's voice data. In step S3600, voice text and media context information can be used to generate prompts. Specifically, the main server 2000 can use at least voice text and media context information to generate prompts. For example, the main server 2000 can load a prompt template from the database 2150 and can use voice text and media context information to modify the prompt template to generate prompts. The main server 2000 can load prompt templates from the database 2150 that correspond to reference information (e.g., media type, media ID, page information, and area identification information) or contextual information (e.g., media title). Furthermore, when modifying a prompt template, the main server 2000 can use multiple pieces of information other than media contextual information, such as contextual information or user information. For example, referring to Figure 25, the main server 2000 can load a third prompt template from the database 2150. The third prompt template is a body text that includes a media title, media context information, and user voice information. The main server 2000 can modify the third prompt template using a pre-obtained media title "Snow White and the Seven Dwarfs," a page-specific text "One day…shining with jealousy," and a voice text "Where does the princess live?" The main server 2000 can modify the third prompt template to generate a third prompt. As another example, the main server 2000 can partially modify the media context information and voice text or add details to generate prompts. As yet another example, the main server 2000 can generate prompts that only include media context information and prompts that only include voice text separately, and transmit these prompts sequentially to the AI ​​server 3000. The prompt may include explanatory text that clarifies the contextual information. Additionally, the prompt may further include guidance text outlining the factors the AI ​​server 3000 should consider when generating a response. In step S3700, the prompt can be used to generate response data. Specifically, the main server 2000 can transmit the prompt generated in step S3600 to the AI ​​server 3000, and the AI ​​server 3000 can input the received prompt into a large language model to generate response data. For example, the AI ​​server 3000 can receive the third prompt shown in FIG25 and generate response data, such as "The princess and queen live in a castle" or "The princess and queen live in a castle, what else would you like to know?". The response data can be understood as data generated by the AI ​​server 2000 considering media context information. That is, when a user uses the media, the interactive system 100 can guide the user to engage in dialogue, and the topic of the dialogue can be related to the part of the story printed on the media that the user is reading (or is interested in). The response data can be text or audio. Specifically, the large language model included in the AI ​​server 3000 can be based on either text or audio. When the input prompt is in text form, the large language model can output the response data in text form. When the input prompt is in audio form, the large language model can output the audio response data. Then, steps S3800 and S3900 are performed, and the corresponding response voice data can be output to the user. After the voice corresponding to the response data is output to the user, the user can activate the recording function of the electronic device 1000 to answer the output voice, so that the user's voice data can be obtained. In this scenario, the main server 2000 generates a follow-up prompt and transmits it to the AI ​​server 3000. The AI ​​server 3000 then outputs the follow-up response data for the prompt, which is then transmitted to the user, allowing the conversation to continue. The method for generating subsequent prompts will be described below with reference to Figure 26. Figure 26 is a flowchart illustrating a method for generating subsequent prompts according to the seventh embodiment. Referring to FIG26, the method for generating subsequent prompts may include obtaining speech text corresponding to the user's speech in step S4100, using media context information and speech text to generate subsequent prompts in step S4200, using subsequent prompts to generate subsequent response data in step S4300, obtaining subsequent response speech data corresponding to the subsequent response data in step S4400, and outputting subsequent response speech data in step S4500. In this document, step S4100 is the same as step S2100, step S4300 is the same as step S2300, step S4400 is the same as step S2400, and step S4500 is the same as step S2500. Therefore, the details illustrated with reference to FIG18 can be applied in the same manner. Therefore, only step S4200 will be described in detail. After obtaining the voice text in step S4100, subsequent prompts can be generated in step S4200 using the response data and the voice text. Specifically, the main server 2000 can generate subsequent prompts using at least the response data and the voice text. In this document, the response data refers to the data generated in step S3700. For example, the main server 2000 can load follow-up prompt templates from the database 2150 and can modify the follow-up prompt templates using response data and voice text to generate prompts. In this document, the main server 2000 can load follow-up prompt templates corresponding to reference information or context information from the database 2150. Furthermore, when modifying the follow-up prompt template, the main server 2000 can use multiple pieces of information other than response data and voice text, such as media context information or user information. For another example, the main server 2000 can generate follow-up prompts that include the previously obtained response data and voice text. In this document, the response data and voice data can be partially modified or details can be added. For example, the main server 2000 can modify the voice text or add details to generate follow-up prompts. As yet another example, the main server 2000 can generate follow-up prompts that only include the voice text. Then, steps S4300 to S4500 are performed, and subsequent response voice data can be output to the user via electronic device 1000. 6. Interactions using human contextual information and media contextual information The following will illustrate, with reference to Figures 27 to 29, how the interactive system 100 uses human context information and media context information to assist users in using media. Figure 27 is a flowchart illustrating a method for assisting a user in using media by using human context information and media context information according to the eighth embodiment. Figure 28 is a diagram illustrating the process of obtaining artificial context information and media context information according to the eighth embodiment. Figure 29 is a diagram illustrating a prompt generated using human context information and media context information according to the eighth embodiment. Referring to FIG27, the method for assisting in the use of media may include: obtaining a code image in step S5100, analyzing the code image in step S5200 to obtain code data, obtaining reference information from the code data in step S5300, obtaining human context information and media context information based on the reference information in step S5400, generating a prompt using the human context information and media context information in step S5500, generating response data using the prompt in step S5600, obtaining response voice data corresponding to the response data in step S5700, and outputting the response voice data in step S5800. Each step will be described in detail below. However, step S5100 is the same as step S1100, step S5200 is the same as step S1200, step S5700 is the same as step S1700, and step S5800 is the same as step S1800. Therefore, the details described with reference to FIG15 can be applied in the same manner. After performing steps S5100 and S5200, reference information can be obtained from the code data in step S5300. In this document, the code image obtained in step S5100 can be obtained by the user using a question area on a specific page of the touch medium of the electronic device 1000. Referring to Figure 28, the main server 2000 can decode the code data to obtain at least one selected from the group of media type, media ID, page information, and location information (coordinate information and / or zone identification information). For example, the reference information obtained by the main server 2000 in Figure 28 may include {book (media type), BOOK000 (media ID), 14 (page information), A1 (zone identification information)}. In step S5400, artificial context information and media context information can be obtained based on reference information. Specifically, referring to Figure 28, the main server 2000 can search the database 2150 for information corresponding to the reference information among multiple pieces of information about the media to obtain the media title, page-specific text, and questions. As mentioned above, the page-specific text is information about the flow of the story printed on the media and can be understood as media context information. As mentioned above, the questions are not very relevant to the flow of the story printed on the media, and since the questions are created by the content creator to enhance user interest and understanding, they can be understood as artificial context information. For example, in Figure 28, the obtained media title is "Snow White and the Seven Dwarfs," the page-specific text is "One day...shining with the light of jealousy," and the question is "Why was the Queen jealous of the Princess?" While the media context information has been described as page-specific text, the technical ideas disclosed herein are not limited to this. Media context information can be page-specific text as well as text or illustration information printed throughout the media. Alternatively, media context information can be a specific statement within a specific page. Furthermore, although artificial context information has been described as a question, the technical ideas disclosed herein are not limited to this. Details that help understand the media without affecting the story flow (e.g., instructions for specific learning) can be obtained as artificial context information. In step S5500, human context information and media context information can be used to generate prompts. Specifically, the main server 2000 can use at least human context information and media context information to generate prompts. For example, the main server 2000 can load a prompt template from the database 2150 and can use human context information and media context information to modify the prompt template to generate prompts. The main server 2000 can load prompt templates from the database 2150 that correspond to reference information (e.g., media type, media ID, page information, and area identification information) or contextual information (e.g., media title). Furthermore, when modifying a prompt template, the main server 2000 can use additional information besides human contextual information and media contextual information, such as contextual information or user information. For example, referring to Figure 29, the main server 2000 can load a fourth prompt template from the database 2150. The fourth prompt template is a body text that includes a media title, media context information, and human context information. The main server 2000 can modify the fourth prompt template using a pre-obtained media title "Snow White and the Seven Dwarfs," a page-specific body text "One day…shining with the light of jealousy," and the question information "Why is the Queen jealous of the Princess?" The main server 2000 can modify the fourth prompt template to generate a fourth prompt. As another example, the main server 2000 may partially modify the human context information and the media context information or add details to generate prompts. As another example, the main server 2000 may generate prompts that include only human context information or only media context information separately, and transmit these prompts sequentially to the AI ​​server 3000. The prompt may include explanatory text that clarifies the artificial context information. More specifically, the prompt may include guidance text that explains the factors the AI ​​server 3000 should consider when generating a response. In step S5600, the prompt can be used to generate response data. Specifically, the main server 2000 can transmit the prompt generated in step S5500 to the AI ​​server 3000, and the AI ​​server 3000 can input the received prompt into a large language model to generate response data. For example, the AI ​​server 3000 can receive the fourth prompt shown in FIG29 and generate response data, such as "Why is the queen jealous of the princess?" or "Why is the queen jealous of the princess? What did the mirror say?". The response data can be understood as data generated by the AI ​​server 2000 considering both human contextual information and media contextual information. That is, when a user uses the media, the interactive system 100 can guide the user to engage in dialogue, and the topic of the dialogue can be related to a part of the story that the user is reading (or is interested in) printed on the media. The dialogue can be understood as taking into account deep learning guided by the content creator (to improve the understanding of the media). The response data can be text or audio. Specifically, the large language model included in the AI ​​server 3000 can be based on either text or audio. When the input prompt is in text form, the large language model can output the response data in text form. When the input prompt is in audio form, the large language model can output the audio response data. Then, steps S5700 and S5800 are performed, and the corresponding response voice data can be output to the user. After the voice corresponding to the response data is output to the user, the user can activate the recording function of the electronic device 1000 to answer the output voice, so that the user's voice data can be obtained. In this scenario, the method for generating follow-up prompts illustrated with reference to FIG18 can be implemented, and follow-up response data corresponding to the follow-up prompts can be generated. This follow-up response data is then transmitted to the user, allowing the continuation of the subsequent dialogue. It should be understood herein that human contextual information and media contextual information can be used when generating follow-up prompts. 7. Generative models process various types of data. The following text will illustrate, with reference to Figure 30, how to use contextual information to assist users in using the medium when the generative model stored in the AI ​​server 3000 processes data in forms other than text data. Figure 30 is a flowchart illustrating a method for using context information to assist a user in using a medium according to the ninth embodiment. Referring to FIG30, the method for assisting in the use of a medium may include: obtaining a code image in step S6100, analyzing the code image in step S6200 to obtain code data, obtaining reference information from the code data in step S6300, obtaining context information based on the reference information in step S6400, converting the context information according to a specific data type in step S6500, obtaining voice data by recording the user's voice in step S6600, converting the voice data according to a specific data type in step S6700, generating a prompt using the converted context information and the converted voice data in step S6800, generating response data using the prompt in step S6900, and outputting the response data in step S7000. Each step will be described in detail below. However, step S6100 is the same as step S1100, and step S6200 is the same as step S1200, so the details described with reference to FIG15 can be applied in the same manner. After performing steps S6100 and S6200, reference information can be obtained from the code data in step S6300. In this document, the code image obtained in step S6100 can be obtained by the user using a specific area on a page of the touch medium of the electronic device 1000 (e.g., the text area, the illustration area, or the question area). Reference information may include at least one of the following groups: media type, media ID, page information, and location information (coordinate information and / or zone identification information). In step S6400, context information can be obtained based on reference information. In this document, context information may include human context information and / or media context information. As an example of a method for obtaining human context information, the method described in step S1400 can be applied in the same manner. As an example of a method for obtaining media context information, the method described in step S3400 can be applied in the same manner. As a method for obtaining both human and media context information, the method described in step S5500 can be applied in the same manner. Additionally, contextual information may include at least one of the groups selected from text, sound, image, or video. For example, contextual information may include text printed on a specific page of the media, audio reading of printed text, sound associated with printed text, image of printed text, or video associated with printed text. As another example, contextual information may include an image printed on a specific page of the media, text describing an image, audio describing an image, sound associated with an image, or video associated with an image. In step S6500, context information can be transformed based on a specific data type. For example, the main server 2000 can transform context information based on the data types that the generative model stored in the AI ​​server 3000 can process. In this document, the data types that the generative model can process mean the data types of input data supported by the generative model (or the data types input to output an appropriate response), and may include at least one selected from the group of text, sound, image, or video. For example, when the context information includes text and the data type that the generative model can receive as input is sound, the main server 2000 can convert the text of the context information into sound (e.g., speech). In this paper, the TTS model 2130 described above can be used, or an external server capable of performing the function of converting text into speech can be used. As another example, when the context information includes images or videos and the data type that the generative model can receive as input is text, the main server 2000 can convert the images or videos in the context information into text. In this document, text refers to an interpretation of an image or a interpretation of a video. Meanwhile, the generative model can process multiple data types, and the data types of the context information are included among the data types that the generative model can process. Therefore, the main server 2000 does not need to convert the context information. For example, when the context information includes audio and the data types that the generative model can receive as input are text and audio, the main server 2000 does not need to convert the context information. That is, step S6500 can be omitted. In step S6600, the main server 2000 can obtain voice data by recording the user's voice. Specifically, the electronic device 1000 can activate the microphone of the electronic device input unit 1300 by the user operating a button to record the user's voice. The electronic device 1000 can then transmit the recorded user voice data to the main server 2000. In this document, voice data can be understood as data containing voice information or voice signals. In step S6700, speech data can be converted based on a specific data type. For example, the main server 2000 can convert speech data based on the data types that the generative model stored in the AI ​​server 3000 can process. Here, the data types that the generative model can process refer to the data types of input data supported by the generative model (or the data types input to produce an appropriate response), and may include at least one selected from the group consisting of text, audio, video, or audio. For example, when the data type that the generative model can receive as input is text, the main server 2000 can convert speech data into text. In this paper, the STT model 2110 described above can be used, or an external server capable of converting speech into text can be used. Meanwhile, the data types that the generative model can receive as input may include audio. In this paper, the main server 2000 may not convert the audio data. That is, step S6700 can be omitted. In step S6800, the context information obtained through the conversion and the voice data obtained through the conversion can be used to generate prompts. Specifically, the main server 2000 can use the context information obtained through the conversion in step S6500 and the voice data obtained through the conversion in step S6700 to generate prompts. For example, the main server 2000 can generate prompts by sequentially linking the context information obtained through conversion with the voice data obtained through conversion. As another example, the main server 2000 can use the above prompt template, but can use the context information obtained by conversion and the voice data obtained by conversion to process the prompt template and generate prompts. Furthermore, the data type of the context information obtained through conversion and the data type of the voice data obtained through conversion can be the same or different. The main server 2000 can generate a prompt after unifying the data type of the context information obtained through conversion and the data type of the voice data obtained through conversion into any one type. In step S6900, the prompt can be used to generate response data. For example, the main server 2000 can transmit the generated prompt to the AI ​​server 3000, and the AI ​​server 300 can use a generative model to generate response data based on the prompt. The generated response data may include at least one of the groups selected from text, audio, video, and audio. The data type of the response data may vary depending on the training process of the generative model. The generative model may be trained to receive data including at least one data type selected from the groups of text, audio, video, and audio, and output data including at least one data type selected from the groups of text, audio, video, and audio. In step S7000, response data can be output to the user. Specifically, the main server 2000 can transmit the response data to the electronic device 1000 with or without conversion. In this document, the format in which the electronic device 1000 outputs the response data may vary depending on the data type of the response data. For example, when the response data is text, the main server 2000 can convert the response data into response audio data and provide the response audio data to the electronic device 1000, and the electronic device 1000 can output the response audio data through the speaker of the electronic device output unit 1400. In this document, the TTS model 2130 of the main server 2000 can be used. As another example, when the response data is voice (or sound), the main server 2000 can provide the response data to the electronic device 1000 without conversion, and the electronic device 1000 can output the response data via the speaker of the electronic device output unit 1400. As another example, when the response data is an image or video, the main server 2000 can convert the response data into response audio data and provide the response audio data to the electronic device 1000, and the electronic device 1000 can output the response data through the speaker of the electronic device output unit 1400. As another example, when the response data is an image or video, the main server 2000 can provide the response data to the electronic device 1000 as is without conversion, and the electronic device 1000 can output the response data via the display of the electronic device output unit 1400. Because the generative model stored in the AI ​​server 3000 can process various types of data, the response data output to the user can be understood as being generated with greater consideration for the user's context. For example, when a user asks a question while pointing to an image on the medium using the electronic device 1000, if the generative model can only process the text, it needs to follow the process of converting the image into text. The text obtained through conversion is related to the image, but is not the image itself. Therefore, the response data generated by the generative model cannot be considered as generated with a clear understanding of the user's interests or question intent. However, when the generative model can process both text and image, it can receive the original image and generate response data with a clear understanding of the user's interests or question intent. Therefore, a more user-friendly response can be provided, that is, a response that is more suitable for the user's problem objectives. The features, structures, and effects described in the embodiments are included in at least one embodiment of this specification, and are not necessarily limited to one embodiment. Furthermore, the features, structures, and effects provided in each embodiment can be combined or modified by those skilled in the art in other embodiments. Therefore, content related to combinations and modifications should be interpreted as including within the scope of this specification. Furthermore, the above description is primarily based on embodiments, which are merely illustrative and not intended to limit the technical concept of this specification. Those skilled in the art can make various modifications and alterations without departing from the fundamental characteristics of this specification. That is, each component specifically shown in the embodiments can be modified. Moreover, differences related to such modifications and applications should be interpreted as including within the scope of this specification as defined in the appended claims. 100: Interactive System; 101: Electronic Pen Service System; 110: Electronic Pen; 111: First Communication Unit; 112: Code Recognition Unit; 113: First Control Unit; 114: TTS Unit; 115: Speaker Unit; 116: Microphone; 117: STT Unit; 118: Storage Unit; 120: First Server; 121: Second Communication Unit; 122: First Database; 123: Second Control Unit; 130: Conversational AI Server; 140: Second Server; 141: Third Communication Unit; 142: Second Database; 143: Third Control Unit; 301, 302, 303, 304, 305, 306, 307, 308, 309, 3 10, 311, 312, 313, 401, 402, 403, 404, 405, 406, 407, 408, 408, 409, 410, 411, 412, 413, 414, 501, 502, 503, 504, 505, 506, 507, 508, 509, 510, 511, 512, 602, 03, 04, 605, 606, 607, 608, 609, 610, 611, 612, 613, S1100, S1200, S1300, S1400, S1500, S1600, S1700, S1800, S2100, S2200, S2300, S2400, S2500, S3100, S3200, S3300, S3400, S3500, S3600, S3700, S3800, S3900, S4100, S4200, S4300, S4400, S4500, S5100, S5200, S5300, S5400, S5500, S5600, S5700, S5800, S6100, S6200, S6300, S6400, S6500, S6600, S6700, S6800, S6900, S7000: Step 1000: Electronic Device 1100: Sensing Unit; 1200: Electronic Device Memory; 1300: Electronic Device Input Unit; 1400: Electronic Device Output Unit; 1500: Electronic Device Communication Unit; 1600: Electronic Device Control Unit; 2000: Main Server / AI Server; 2100: Main Server Memory; 2110: Speech-to-Text (STT) Model; 2130: Text-to-Speech (TTS) Model; 2150: Database; 2200: Main Server Communication Unit; 2300: Main Server Control Unit; 3000: AI Server; 3100: AI Server Memory; 3200: AI Server Communication Unit; 3300: AI Server Control Unit. Figure 1 is a diagram illustrating an interactive system according to an embodiment. Figure 2 is a diagram illustrating the configuration of an electronic device according to an embodiment. Figure 3 is a diagram illustrating a code printed on a medium according to an embodiment. Figure 4 is a diagram illustrating the configuration of a main server according to an embodiment. Figure 5 is a diagram illustrating the configuration of an AI server according to an embodiment. Figure 6 is a block diagram illustrating an electronic pen service system using conversational AI services according to an embodiment. Figure 7 is a block diagram illustrating an electronic pen service system using conversational AI services according to another embodiment. Figure 8 is a diagram illustrating multiple pieces of information stored in a database according to an embodiment. Figure 9 is a diagram illustrating the pattern of separating different areas in a medium according to an embodiment. Figure 10 is a diagram illustrating the pattern of separating different areas in a medium according to another embodiment. Figure 11 is a flowchart illustrating a method for assisting a user in using a medium using human context information according to a first embodiment. Figure 12 is a flowchart illustrating a method for assisting a user in using a medium using human context information according to a second embodiment. Figure 13 is a flowchart illustrating a method for assisting a user in using a medium using human context information according to a third embodiment. Figure 14 is a flowchart illustrating a method for assisting a user in using a medium using human context information according to a fourth embodiment. Figure 15 is a flowchart illustrating a method for assisting a user in using media using human context information according to a fifth embodiment. Figure 16 is a diagram illustrating a process for obtaining human context information according to a fifth embodiment. Figure 17 is a diagram illustrating a prompt generated using human context information according to a fifth embodiment. Figure 18 is a flowchart illustrating a method for generating subsequent prompts according to a fifth embodiment. Figure 19 is a diagram illustrating subsequent prompts generated using human context information according to a fifth embodiment. Figure 20 is a diagram illustrating a process for obtaining pre-stored prompts according to a sixth embodiment. Figure 21 is a diagram illustrating pre-stored prompts according to a sixth embodiment. Figure 22 is a diagram illustrating subsequent prompts generated using human context information according to a sixth embodiment. Figure 23 is a flowchart illustrating a method for assisting a user in using media using media context information according to a seventh embodiment. Figure 24 is a diagram illustrating a process for obtaining media context information according to a seventh embodiment. Figure 25 is a diagram illustrating a prompt generated using media context information according to a seventh embodiment. Figure 26 is a flowchart illustrating a method for generating subsequent prompts according to a seventh embodiment. Figure 27 is a flowchart illustrating a method for assisting a user in using media using both human context information and media context information according to an eighth embodiment. Figure 28 is a diagram illustrating a process for obtaining both human context information and media context information according to an eighth embodiment. Figure 29 is a diagram illustrating a prompt generated using human context information and media context information according to the eighth embodiment. Figure 30 is a flowchart illustrating a method for assisting a user in using media using context information according to the ninth embodiment. 1000: Electronic devices 2000: Main Server / AI Server 3000: AI Server

Claims

1. A server for communicating with electronic devices and artificial intelligence (AI) servers, the server comprising: Memory; Communication unit; And a control unit, configured to control the memory and the communication unit; The control unit is configured to: obtain reference information encoded in code data obtained by the electronic device, wherein the reference information is received from the electronic device via the communication unit, or the reference information is obtained by the controller after receiving the code data from the electronic device via the communication unit and decoding the code data, wherein the reference information is related to a medium that the user interacts with using the electronic device, and wherein the reference information includes media identifier information of the medium, page information indicated by the electronic device, and position information within the page indicated by the electronic device; obtain content of interest from the memory based on the reference information and generate media context information, wherein the content of interest is related to the printed content in the medium and / or the portion of the printed content in the medium that the user is interested in; obtain speech text corresponding to the user's speech obtained by the electronic device, wherein the speech text is obtained by converting the speech using speech-to-text (STT) conversion; and generate prompts based on the obtained media context information and the speech text, wherein the prompts include the media context information and the speech text. The system transmits the prompt to the artificial intelligence server via the communication unit; receives the response from the artificial intelligence server to the transmitted prompt via the communication unit; obtains a response voice corresponding to the received response; and transmits the response voice to the electronic device via the communication unit.

2. The server as claimed in claim 1, wherein the memory stores a plurality of media texts and each of the plurality of media texts corresponds to at least a portion of a particular medium, and wherein the control unit loads from the memory the media text corresponding to the media identifier information of the reference information and obtains the media text as the media context information.

3. The server as claimed in claim 1, wherein the memory stores a plurality of page-specific texts and each of the plurality of page-specific texts corresponds to at least a portion of a plurality of texts printed on a specific page of a specific medium, and wherein the control unit loads from the memory the page-specific texts that correspond to the media identifier information and the page information of the reference information and obtains the page-specific texts as the media context information.

4. The server as claimed in claim 1, wherein the memory stores a plurality of location-specific texts and each of the plurality of location-specific texts corresponds to at least a portion of a plurality of texts at a specific location on a specific page of a specific medium, and wherein the control unit loads from the memory the location-specific texts corresponding to the location information of the media identifier information, the page information, and the reference information and obtains the location-specific texts as the media context information.

5. The server as described in claim 1, wherein the prompt is generated by concatenating the speech text to the media context information.

6. The server as described in claim 1, wherein the prompt further includes explanatory text of the media context information.

7. The server as claimed in claim 1, wherein the prompt further includes pre-stored user information, and wherein the user information includes at least one of the user's age and gender.

8. The server as claimed in claim 1, wherein the prompt further includes pre-stored guidance text, and wherein the guidance text includes considerations of the artificial intelligence server when the response is generated.

9. The server as claimed in claim 1, wherein the control unit: obtains a prompt template, and modifies the prompt template using the media context information and the voice text to generate the prompt.

10. The server as claimed in claim 1, wherein the prompt includes at least a portion of a previous prompt that was transmitted to the AI ​​server before the code data was obtained from the electronic device.

11. The server as claimed in claim 10, wherein the control unit transmits the previous prompt to the AI ​​server via the communication unit and receives previous responses to the previous prompt from the AI ​​server, and wherein the prompt includes at least a portion of the previous response.

12. The server as claimed in claim 1, wherein the reference information further includes area identification information relating to a question or instruction printed on the medium to which the user is interacting with the electronic device, and the area identification information identifies a preset area in the medium, wherein the control unit obtains human context information corresponding to the question or instruction printed on the preset area from the memory based on the reference information, and wherein the prompt further includes the human context information.

13. The server as claimed in claim 1, wherein the control unit: obtains subsequent speech text corresponding to subsequent speech of the user obtained by the electronic device, wherein the subsequent speech is obtained by converting the subsequent speech using speech-to-text conversion; generates a subsequent prompt based on the subsequent speech text; transmits the subsequent prompt to the artificial intelligence server via the communication unit; receives a subsequent response from the artificial intelligence server to the transmitted subsequent prompt via the communication unit; obtains a subsequent response speech corresponding to the subsequent response; and transmits the subsequent response speech to the electronic device via the communication unit.

14. The server as described in request item 13, wherein the subsequent prompt further includes the media context information.

15. The server as described in claim 1, wherein the artificial intelligence server includes a large language model (LLM).

16. The server as claimed in claim 1, wherein the electronic device includes an image sensor and is configured to capture a region of the medium by means of the user's manipulation and to obtain the code data from the captured image.

17. The server as claimed in claim 1, wherein the electronic device has a pen shape having one end and another end, and wherein, in response to the user indicating an area of ​​the medium using the one end of the electronic device, the electronic device is configured to capture the medium using the image sensor disposed at the one end.

18. A server for communicating with electronic devices and artificial intelligence (AI) servers, the server comprising: Memory; Communication unit; And a control unit, configured to control the memory and the communication unit; The control unit is configured to: obtain reference information encoded in code data obtained by the electronic device, wherein the reference information is received from the electronic device via the communication unit, or the reference information is obtained by the controller after receiving the code data from the electronic device via the communication unit and decoding the code data, wherein the reference information is related to a medium that the user interacts with using the electronic device, and wherein the reference information includes media identifier information of the medium, page information indicated by the electronic device, and position information within the page indicated by the electronic device; obtain content of interest from the memory based on the reference information and generate media context information, wherein the content of interest is related to the printed content in the medium and / or the portion of the printed content in the medium that the user is interested in; obtain voice data corresponding to the user's voice obtained by the electronic device; generate a prompt based on the obtained media context information and the voice data; transmit the prompt to the artificial intelligence server via the communication unit; and receive a response from the artificial intelligence server to the transmitted prompt via the communication unit. And transmit the received response or the corresponding voice response to the electronic device via the communication unit.

19. The server as described in claim 18, wherein the data type of the media context information is any one of text, sound, image, or video.

20. The server as claimed in claim 18, wherein when the data type of the media context information is text and the data type supported by the AI ​​server is audio, the control unit converts the media context information such that the data type of the media context information becomes audio before generating the prompt.

Images