Computing device and method for providing conversation with a target character generated using an artificial neural network.

Abstract

To provide a conversation with a target character generated by using a neural network in order to achieve dialogue with a user who cannot converse.SOLUTION: A computing device comprises a communication part and a character generation part. The communication part receives, from a user terminal, a first request message for scanning one of a plurality of information codes to generate an identifiable target character, and transmits a second request message related to second generation information for generating a target character by finely tuning a basic character according to a user related to the user terminal. The character generation part generates a basic character related to decedent by using a basic command in response to the first request message, and when receiving the second generation information as response to the second request message, generates a target character customized according to the user related to the user terminal by using generative neural network.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present disclosure relates to a computing device and method for providing a conversation with a target character, and more particularly to a computing device and method for providing a conversation with a target character generated using an artificial neural network.

Background Art

[0002] A virtual assistant service such as a chatbot may be a service that infers and understands the intention of a human question and provides a response phrase thereto. Further, the chatbot can infer and understand the intention of a human request and perform an operation based thereon.

[0003] For example, when a chatbot receives a request message such as an order for food from a user, reservation / change / cancellation of a medical appointment, etc., it can infer the user's intention from the input message and perform an operation accordingly. Such a chatbot service has the effect of reducing the cost for a company to arrange consultation staff.

[0004] Recently, not only chatbots for consultation as in the above example, but also chatbot services that provide a feeling of directly talking to a person have emerged. However, since conversation can be directly carried out between humans, this type of service has not gained much popularity.

Summary of the Invention

Problems to be Solved by the Invention

[0005] The present disclosure has been devised in response to the above-described background art, and aims to provide a computing device and method for providing a conversation with a target character generated using an artificial neural network for the purpose of interacting with a user who is unable to converse.

[0006] The technical challenges described herein are not limited to those mentioned above, and other technical challenges not mentioned will be clearly understood by those skilled in the art from the following description. [Means for solving the problem]

[0007] This disclosure relates to a computing device that provides interaction with a target character generated using an artificial neural network, and is devised to solve the aforementioned problems. The computing device includes a storage unit that stores basic commands for generating a basic character related to a deceased person based on first generated information of the deceased person acquired in advance, and maps and stores the basic commands to a pre-specified information code. It also includes a communication unit that receives a first request message for generating a target character identified by scanning one of several information codes from a user terminal, and sends a second request message for second generated information for generating the target character by fine-tuning the basic character to correspond to the user associated with the user terminal. Furthermore, the character generation unit generates a basic character related to the deceased person using the basic command in response to the first request message, and when second generated information is received in response to the second request message, it can generate a target character customized to the user associated with the user terminal using a generative artificial neural network.

[0008] Furthermore, the second generated information may be characterized by including at least one of the following: a first command that provides first background knowledge regarding the basic background of the target character; a second command that determines the role of the target character; a third command that determines the values ​​of the target character; a fourth command that indicates the person with whom the target character interacts; and a fifth command that provides memories between the target character and that person.

[0009] Furthermore, the communication unit receives a history of conversations in which the deceased and the user participated within at least one chat application used by the user from the user terminal, and the character generation unit generates at least one of the following using the conversation history: a sixth command to determine the way the target character speaks, a seventh command to provide second background knowledge related to the relationship between the target character and the user, an eighth command to provide values ​​to the target character, and a ninth command to provide memories shared between the target character and the user, and generates a target character using the generated at least one command, the basic character, and a generative artificial neural network.

[0010] Furthermore, the character generation unit may be characterized by deleting non-standard language words included in the dialogue history, determining the average dialogue speed of the deceased person through the dialogue history, determining multiple dialogues generated by the deceased person that are below the average dialogue speed, generating multiple corpora (language resources) using these multiple dialogues, and generating at least one command using the generated multiple corpora.

[0011] Furthermore, the character generation unit may, when multiple corpora are generated, determine a first value indicating the number of words contained in each of the multiple corpora and a second value indicating the number of deleted non-standard words, sort the multiple corpora by rank based on the first and second values, and generate at least one command using the corpora of a predetermined rank among the sorted multiple corpora.

[0012] The technical solutions in this disclosure are not limited to those described above, and other solutions not described will be clearly understood by those skilled in the art. [Effects of the Invention]

[0013] According to some embodiments of this disclosure, computing devices and methods are provided that enable conversation with a target character generated using an artificial neural network, making it feel as if one is interacting with a user who is unable to converse, such as a deceased person. The effects obtained by this disclosure are not limited to those described above, and other effects not described will be clearly understood by those skilled in the art to which this disclosure pertains. [Brief explanation of the drawing]

[0014] [Figure 1] This is a block diagram illustrating an example of a computing device according to some embodiments of the present disclosure. [Figure 2] This is a flowchart illustrating an example of how a computing device according to some embodiments of this disclosure generates a target character. [Figure 3] This flowchart illustrates an example of how a computing device according to some embodiments of the present disclosure generates at least one command based on a conversation history. [Figure 4] This figure illustrates an example of a chatbot service according to some embodiments of the present disclosure. [Figure 5] This figure illustrates another example of a chatbot service according to some embodiments of the present disclosure. [Figure 6] This figure illustrates yet another example of a chatbot service according to some embodiments of the present disclosure. [Modes for carrying out the invention]

[0015] Since the present invention is subject to various modifications and can have multiple embodiments, specific embodiments are illustrated in the drawings and described in detail in the detailed description.

[0016] However, this is not intended to limit the present invention to specific embodiments, and it should be understood that it includes all modifications, equivalents, and alternatives within the spirit and technical scope of the present invention.

[0017] In the description of each drawing, similar reference numerals are used for similar components.

[0018] Terms such as first, second, A, B, etc. are used to describe various components, but these components should not be limited by these terms.

[0019] These terms are used only for the purpose of distinguishing one component from another.

[0020] For example, without departing from the scope of the present invention, the first component can be named the second component, and similarly, the second component can also be named the first component.

[0021] The term "and / or" includes combinations of multiple related described items or any of the multiple related described items.

[0022] When a component is referred to as being "connected" or "connecting" to another component, it should be understood that it may be directly connected to the other component or there may be other components between them.

[0023] On the other hand, when a component is referred to as being "directly connected" or "directly connecting" to another component, it should be understood that there are no other components between them.

[0024] The terms used in this application are only used to describe specific embodiments and are not intended to limit the present invention.

[0025] A singular expression can include multiple expressions unless the context clearly indicates otherwise.

[0026] In this application, terms such as “includes” or “having” are intended to specify the presence of features, numbers, stages, operations, components, parts, or combinations thereof as described in the specification, and should be understood not to preemptively exclude the possibility of the presence or addition of one or more other features, numbers, stages, operations, components, parts, or combinations thereof.

[0027] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as those generally understood by a person of ordinary skill in the art to which this invention pertains.

[0028] Terms should be interpreted in a way that is consistent with their meaning in the context of the relevant technology, as they are defined in commonly used dictionaries, and should not be interpreted in an ideal or overly formal sense unless explicitly defined in this application.

[0029] In this disclosure, the computing device can provide a chatbot service for communicating with entities that are unable to communicate verbally, such as deceased persons, patients in a vegetative state, or animals.

[0030] For example, a computing device can obtain generated information related to a deceased person from the user of the user terminal.

[0031] The computing device can use a generative artificial neural network to generate characters related to the deceased from the acquired generated information.

[0032] On the other hand, if you create characters related to the deceased in advance, you may incur significant costs for maintaining the computing equipment.

[0033] For example, there may be costs associated with expanding the storage capacity of the computing device.

[0034] Therefore, the computing device can store commands for generating characters related to the deceased based on generated information obtained from the user terminal.

[0035] Furthermore, the computing device can generate a character related to the deceased if a user interaction occurs that would generate a character related to the deceased.

[0036] This can reduce the costs associated with maintaining computing equipment.

[0037] The following illustrates an example of how a computing device pursuant to this disclosure generates a character, as shown in Figures 1 through 6.

[0038] Figure 1 is a block diagram illustrating an example of a computing device according to some embodiments of the present disclosure.

[0039] Referring to Figure 1, the computing device (100) may include a character generation unit (110), a storage unit (120), and a communication unit (130).

[0040] However, the above-mentioned components are not essential for implementing the computing device (100), and the computing device (100) may have more or fewer components than those listed above.

[0041] The computing device (100) may include any type of computer system or computer device, such as a microprocessor, a mainframe computer, a digital processor, a portable device, or a device controller.

[0042] A computing device (100) can achieve desired system performance by utilizing a combination of typical computer hardware (e.g., devices that may include computer processors, memory, storage, input and output devices, and other components of existing computing devices, electronic communication devices such as routers and switches, network-attached storage (NAS), and electronic information storage systems such as storage area networks (SANs)) and computer software (i.e., commands to make the computing device function in a particular way).

[0043] The character generation unit (110) can typically handle the overall operation of the computing device (100).

[0044] The character generation unit (110) can provide or process appropriate information or functions to the user by processing signals, data, information, etc. that are input or output through the components of the computing device (100), or by executing applications stored in the storage unit (120).

[0045] The character generation unit (110) can consist of one or more cores and may include data analysis processors such as a central processing unit (CPU), a general-purpose graphics processing unit (GPGPU), and a tensor processing unit (TPU).

[0046] In this disclosure, the character generation unit (110) can generate a basic command to generate a basic character related to a deceased person based on first generation information of the deceased person obtained in advance.

[0047] The first generated information may be information obtained from the user's terminal.

[0048] The first generated information may include basic personal details of the deceased.

[0049] For example, primary generated information may include the deceased's photograph, video data, audio data, name, gender, date of birth, age at death, MBTI, blood type, place of birth, or religion.

[0050] The base character may be a character generated using only basic information related to the deceased, in order to conserve resources on the computing device (100).

[0051] The character generation unit (110) can map basic commands to pre-specified information codes and save them in the storage unit (120).

[0052] The information code may be a QR code (Quick Response) (registered trademark) or a barcode, among other things.

[0053] The character generation unit (110) can generate a basic character based on interaction from a user terminal that scans an information code.

[0054] The character generation unit (110) can then generate a target character by customizing the basic character based on the second generation information received from the user terminal.

[0055] An example of how the character generation unit (110) generates a target character will be explained below through Figure 2.

[0056] The storage unit (120) may include memory and / or permanent storage media.

[0057] Memory may include at least one type of storage medium from among flash memory type, hard disk type, multimedia card micro type, card type memory (e.g., SD or XD memory), RAM (Random Access Memory), SRAM (Static Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), PROM (Programmable Read-Only Memory), magnetic memory, magnetic disk, and optical disk.

[0058] In this disclosure, the storage unit (120) can store basic commands for generating a basic character related to a deceased person based on first generated information of the deceased person obtained in advance.

[0059] The storage unit (120) can store basic commands mapped to pre-specified information codes.

[0060] The communication unit (130) may include one or more modules that enable communication between the computing device (100) and the communication system, between the computing device (100) and the user terminal (200), or between the computing device (100) and the network.

[0061] In this disclosure, the communications unit (130) can receive a first request message from the user terminal.

[0062] The first request message may be a message intended to generate a target character that is identified through scanning one of several information codes.

[0063] The communications unit (130) can send a second request message to the user terminal.

[0064] The second request message may be a message requesting second generation information to generate the target character by fine-tuning the base character to correspond to the user associated with the user terminal.

[0065] The following describes an example of how a computing device (100) generates a target character.

[0066] Figure 2 is a flowchart illustrating an example of how a computing device according to some embodiments of this disclosure generates a target character.

[0067] Referring to Figure 2, the character generation unit (110) of the computing device (100) can store in the storage unit (120) basic commands for generating a basic character related to the deceased based on the first generation information of the deceased acquired in advance (S110).

[0068] Specifically, the character generation unit (110) can obtain the first generation information of the deceased person through the communication unit (130).

[0069] Alternatively, the first generated information may be obtained based on user input through a service platform provided by a computing device (100).

[0070] The first generated information may include basic personal details of the deceased.

[0071] For example, primary generated information may include the deceased's photograph, video data, audio data, name, gender, date of birth, age at death, MBTI, blood type, place of birth, language or religion used by the deceased, etc.

[0072] The character generation unit (110) can use the first generation information to generate basic commands for generating a basic character.

[0073] The character generation unit (110) can save the generated basic commands to the storage unit (120).

[0074] The command can be understood as a system prompt to generate a base character or to train a base character.

[0075] The character generation unit (110) can map basic commands to pre-specified information codes and save them in the storage unit (120) (S120).

[0076] The information code could be a QR code (Quick Response) or a barcode, among other things.

[0077] If the character generation unit (110) generates a basic character every time it acquires the first generation information, there is a possibility of wasting resources used to utilize the character generation unit (110).

[0078] Furthermore, there is a possibility of wasted resources in the storage unit (120) used to save the generated base character.

[0079] To prevent this, when the first generation information is obtained, the character generation unit (110) can map the basic command for generating the basic character to a pre-specified information code and save it in the storage unit (120).

[0080] The character generation unit (110) can generate a basic character when it receives an interaction from the user terminal to generate a target character.

[0081] The character generation unit (110) can receive a first request message via the communication unit (130) for generating a target character identified from a user terminal through scanning one of several information codes (S130).

[0082] The character generation unit (110) can generate a basic character for the deceased using basic commands in response to the first request message (S140). Specifically, the user terminal can scan for information codes related to the deceased person that the user attempts to generate, based on user interaction.

[0083] The user terminal can send a first request message to the computing device (100) to generate a target character based on the scanning of the information code.

[0084] The character generation unit (110) can generate a basic character when it receives a first request message through the communication unit (130).

[0085] The character generation unit (110) can send a second request message regarding second generation information for generating a target character to the user terminal via the communication unit (130) by fine-tuning the basic character to correspond to the user associated with the user terminal (S150).

[0086] Fine-tuning can be understood as the operation of updating the parameters of a base character using additional training data.

[0087] In other words, fine-tuning in this disclosure can be understood as an operation to generate a target character by updating the parameters of the base character.

[0088] The second generation information can contain at least one command.

[0089] The second generated information may include a first command that provides first background knowledge related to the target character's basic background, a second command that determines the target character's role, a third command that determines the target character's values, a fourth command that indicates who the target character interacts with, and a fifth command that provides memories of the target character and that person.

[0090] While not strictly limited by examples, the first command could be something like "Character: Lee Jung-jae (My missed dad), Age: 41".

[0091] The age may be the age at the time of death, or it may be an age arbitrarily selected by the user of the user's device.

[0092] The second command could be something like "Occupation: My eternal dad and my passed away parent."

[0093] The third command is ``Personality Traits: Lee Jung-jae (Dad) consistently offers warm consolation as a parent to the user, always listening to the user's concerns and supporting them with a warm heart. He knows he has passed away and is not beside the user but always comforts and supports the user from heaven. a dad, he is always supporting and being a strength to his child (user).''

[0094] The fourth command could be something like "Character: Lee Jung-jae (My missed dad)".

[0095] A fifth command might be, "Relationship with User: As the user's dad, he always supports and prays for the user from heaven. He watches over the user's dreams and cheers for their fulfillment. When the user seems to be facing tough times, he listens to the user's stories, always caring for the user's heart. He dearly misses the user, watching over him from heaven, curious about the user's daily life and always proud of the user as his dad."

[0096] In addition to the first through fifth commands, the user can provide various other information related to the deceased to the computing device (100) as second-generation information.

[0097] According to one embodiment, the second generated information may be text.

[0098] The character generation unit (110) can generate at least one command based on the second generation information received as text.

[0099] The second request message may include messages such as, "What was your relationship with the deceased? What did the deceased call you? Please provide any records of conversations with the deceased that may be helpful."

[0100] The user can generate second generated information corresponding to the second request message through the user terminal and provide it to the computing device (100).

[0101] The character generation unit (110) can receive second generation information as a response to the second request message via the communication unit (130).

[0102] When the character generation unit (110) receives the second generation information, it can generate a target character customized to the user associated with the user terminal using a generative artificial neural network (S160).

[0103] Generative artificial neural networks can include neural network models that utilize convolutional neural networks (CNNs), recurrent neural networks (RNNs), autoencoders, generative adversarial networks (GANs), restricted Boltzmann machines (RBMs), deep belief networks (DBNs), Q networks, U networks, Siam networks, generative adversarial networks (GANs), large language models (LMMs), stable diffusion, Dreambooth, Neural Radiance Fields (NeRFs), Nerfies, or DragGAN.

[0104] Generative artificial neural networks are models that take a basic character and second-generation information as input values ​​and generate a target character as the output value.

[0105] Depending on the implementation, a generative artificial neural network can also generate basic characters as output values, using basic commands as input values.

[0106] The character generation unit (110) can provide a chatbot service using the generated target character.

[0107] A chatbot service may be a service that provides features such as chat, voice interaction, and video interaction.

[0108] The target character generated through the above configuration can be created to have specific characteristics depending on the deceased's appearance, personality, or relationship with the person they were interacting with.

[0109] This allows users to interact with the target character through chat, voice, or video interaction functions, enabling them to use the service in an immersive way while recalling the deceased's appearance during their lifetime.

[0110] On the other hand, according to some embodiments of this disclosure, a computing device (100) can receive a record of conversations in which the deceased and the user participated, within at least one chat application used by the user from the user terminal.

[0111] In this case, the computing device (100) can generate commands to generate a target character based on the dialogue record.

[0112] An example of how a computing device conforming to this disclosure generates at least one command based on a conversation record is illustrated below through Figures 3 and 4.

[0113] Figure 3 is a flowchart illustrating an example of how a computing device according to some embodiments of the present disclosure generates at least one command based on an interaction record.

[0114] Referring to Figure 3, the communication unit (130) of the computing device (100) can receive a record of the conversation in which the deceased and the user participated, from the user terminal within at least one chat application used by the user (S210).

[0115] For example, the communications unit (130) can receive a dialogue record as a response to the second request message.

[0116] The character generation unit (110) can delete non-standard language words included in the received dialogue record (S220).

[0117] Non-standard language words may be words that are not registered in standard Japanese dictionaries provided on servers such as those of the National Institute of Korean Language.

[0118] The character generation unit (110) can determine the average conversation speed of the deceased through the conversation record (S230).

[0119] For example, the character generation unit (110) can determine the average dialogue speed of the deceased as a response to the other party's dialogue, or the average dialogue speed of the deceased when speaking alone, regardless of the other party's dialogue.

[0120] The character generation unit (110) can determine multiple dialogues generated by the deceased at a speed less than the average dialogue speed (S240).

[0121] The character generation unit (110) can generate multiple corpora using multiple dialogues (S250).

[0122] A corpus can be a collection of words, such as verbs, nouns, adjectives, or adverbs.

[0123] The character generation unit (110) can generate at least one command using the generated corpora (S260).

[0124] For example, the character generation unit (110) can generate a sixth command that determines the target character's way of speaking, a seventh command that provides second background knowledge related to the relationship between the target character and the user, an eighth command that provides the target character with values, and a ninth command that provides memories shared between the target character and the user.

[0125] In one embodiment, the character generation unit (110) can also use multiple corpora to update at least one command included in the second generation information.

[0126] The second generation information may include commands generated by the user on the user terminal.

[0127] However, users may not have an objective memory of what the deceased person was like when they were alive.

[0128] Therefore, the character generation unit (110) can generate commands such as the 1-1 command, which is an updated version of the 1st command, the 2-1 command, which is an updated version of the 2nd command, the 3-1 command, which is an updated version of the 3rd command, the 4-1 command, which is an updated version of the 4th command, and the 5-1 command, which is an updated version of the 5th command, using multiple corpora.

[0129] The character generation unit (110) can generate a target character using at least one command, a base character, and a generative artificial neural network if at least one command is generated.

[0130] On the other hand, according to some embodiments of this disclosure, when multiple corpora are generated, the character generation unit (110) can determine a first value indicating the number of words contained in each of the multiple corpora, and a second value indicating the number of non-standard words that have been deleted.

[0131] The character generation unit (110) can rank multiple corpora based on the first and second values.

[0132] For example, the character generation unit (110) can rank multiple corpora in a system where a larger primary value results in a higher rank.

[0133] The character generation unit (110) can rank multiple corpora using a system where a smaller secondary value results in a higher rank.

[0134] The character generation unit (110) can rank multiple corpora using a system where a higher primary value and a lower secondary value result in a higher rank.

[0135] In other words, the character generation unit (110) can arrange multiple corpora in order of the number of words in each corpus, followed by the number of words that have been deleted.

[0136] The character generation unit (110) can generate at least one command using a corpus of a predetermined rank from among multiple corpora arranged in order of rank.

[0137] While not limiting in this way, the character generation unit (110) can generate commands 5 through 9 using the top 10 corpora from among several corpora arranged in order of rank.

[0138] According to the above configuration, the computing device (100) can generate at least one command to customize the basic character based on the dialogue records in which the deceased and the user participated.

[0139] This allows target characters generated using at least one command to interact with users in a more realistic way.

[0140] Figure 4 is a diagram illustrating an example of a chatbot service that follows some embodiments of the present disclosure.

[0141] The diagram shown in Figure 4 may represent the first interface (210) of a chatbot service provided by the computing device (100).

[0142] The first interface (210) can be displayed on the display unit of the user terminal.

[0143] Referring to the first interface (210), the computing device (100) can provide a chat service between the user and the target character.

[0144] The computing device (100) may have generated the target character (211) based on its interaction with the user terminal.

[0145] The user of the user terminal can provide the dialogue (212) they wanted to have with the target character (211).

[0146] The computing device (100) can use the target character (211) to provide a response (213) to a dialogue (212).

[0147] The target character (211) is generated using the deceased person's personality, temperament, speaking style, values, religious views, and background knowledge, allowing it to interact with the user as if the deceased person were speaking in response to their dialogue.

[0148] Figure 5 illustrates another example of a chatbot service that follows some embodiments of the present disclosure.

[0149] The diagram shown in Figure 5 may represent a second interface (220) of a chatbot service provided by a computing device (100).

[0150] The second interface (220) can be displayed on the user terminal's display.

[0151] Referring to the second interface (220), the computing device (100) can provide voice call services between the user and the target character.

[0152] The target character can be generated to speak in the deceased person's voice using video data or audio data included in the first generation information.

[0153] Users can have voice calls with the deceased through a chatbot service provided by the computing device (100).

[0154] In one embodiment, a user's voice may be converted to text through Speech-to-Text (STT) technology.

[0155] The computing device (100) can generate response messages to text.

[0156] The computing device (100) can convert text into speech through TTS (Text To Speech) technology.

[0157] The computing device (100) can enable voice messages to be played on the user terminal via a target character.

[0158] Figure 6 illustrates another example of a chatbot service that follows some embodiments of the present disclosure.

[0159] The diagram shown in Figure 6 may represent a third interface (230) of a chatbot service provided by a computing device (100).

[0160] The third interface (230) can be displayed on the user terminal's display.

[0161] Referring to the third interface (230), the computing device (100) can provide a video call service between the user and the target character (231).

[0162] The target character (231) can be generated to speak in the voice of the deceased, using video data or audio data included in the first generated information.

[0163] Furthermore, the target character (231) can be generated to correspond to the deceased's appearance during their lifetime, using photographic or video data included in the first generation information.

[0164] Alternatively, if the age included in the first generated information is not the age of the deceased at the time of death, the target character (231) can be generated in a form corresponding to the age entered by the user, using the photographic or video data included in the first generated information.

[0165] Users can have video calls with the deceased through a chatbot service provided by the computing device (100).

[0166] Depending on the embodiment, the video call may include a two-way video call in which the target character (231) and the user can interact, or a one-way video call in which the target character reads a given line of dialogue.

[0167] According to one embodiment, in the case of two-way video calls, the user's voice may be converted to text through STT (Speech To Text) technology.

[0168] The computing device (100) can generate response messages to text.

[0169] The computing device (100) can convert text into speech through TTS (Text To Speech) technology.

[0170] The computing device (100) can analyze the user's emotions based on the user's voice.

[0171] The computing device (100) can determine the facial expression of the target character (231) based on the analyzed user emotions and the text converted from the user's voice.

[0172] The computing device (100) can enable voice messages to be played on the user terminal via the target character (231).

[0173] Furthermore, the computing device can change the mouth movements or facial expressions of the target character (231) in sync with the playback of the audio message. The description of the presented embodiments is provided so that a person with ordinary skill in the art can use or practice the disclosure.

[0174] Various modifications to these embodiments will be obvious to those with ordinary skill in the art, and the general principles defined herein can be applied to other embodiments without departing from the scope of this disclosure.

[0175] Therefore, this disclosure is not limited to the embodiments presented herein and should be interpreted in the broadest sense to be consistent with the principles and novel features presented herein.

Claims

1. A computer device that includes at least one processor and provides a conversation with a target character generated using an artificial neural network, A storage unit that saves basic commands for generating a basic character related to a deceased person based on pre-acquired first generation information of the deceased person, and maps and saves these basic commands to a pre-specified information code. A communication unit receives a first request message for generating a target character identified by scanning one of several information codes from a user terminal, and transmits a second request message regarding second generation information for generating the target character by fine-tuning the basic character to correspond to the user associated with the user terminal. A computer device including a character generation unit that, in response to a first request message, generates a basic character for the deceased using basic commands, and when second generation information is received as a response to a second request message, generates a target character customized to the user associated with the user terminal, based on the basic character and the second generation information, using a generative artificial neural network.

2. In claim 1, The second generation information is, A computer device characterized by including at least one of the following: a first command that provides first background knowledge related to the basic background of a target character; a second command that determines the role of the target character; a third command that determines the values ​​of the target character; a fourth command that indicates the person with whom the target character will converse; and a fifth command that provides memories of the target character with that person.

3. In claim 1, The communications department, From the user's terminal, the system receives the conversation history of the deceased and the user in which they participated within at least one chat application used by the user. The character generation unit is, Using the conversation history, at least one of the following commands is generated: a sixth command that determines the target character's way of speaking; a seventh command that provides second background knowledge related to the relationship between the target character and the user; an eighth command that assigns values ​​to the target character; and a ninth command that assigns memories shared between the target character and the user. A computer device characterized by generating a target character using at least one generated command, a base character, and a generative artificial neural network.

4. In claim 3, The character generation unit is, Remove non-standard language words from the conversation history. The average conversation speed of the deceased person is determined through the conversation history. We identify multiple conversations generated by the deceased at a speed below the average conversational speed. Multiple corpora are generated using these multiple conversations, A computer device characterized by generating at least one command using multiple generated corpora.

Images