System

Abstract

To provide a system according to an embodiment that generates a natural reply on the basis of past conversation data.SOLUTION: A system according to an embodiment comprises: a providing unit; an analysis unit; a learning unit; and a reply unit. The providing unit is configured to provide conversation data. The analysis unit is configured to analyze the conversation data provided by the providing unit. The learning unit is configured to learn a conversation style on the basis of data analyzed by the analysis unit. The reply unit is configured to generate a reply on the basis of conversation style learned by the learning unit.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including a description and related instruction sentence regarding the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2022-180282 A Summary of the Invention [Problem to be solved by the invention]

[0004] Conventional techniques do not adequately generate natural responses based on past conversation data, and there is room for improvement.

[0005] The system according to the embodiment aims to generate natural responses based on past conversation data. [Means for solving the problem]

[0006] A system according to an embodiment includes a providing unit, an analyzing unit, a learning unit, and a responding unit. The providing unit provides conversation data. The analyzing unit analyzes the conversation data provided by the providing unit. The learning unit learns a conversation style based on the data analyzed by the analyzing unit. The responding unit generates a response based on the conversation style learned by the learning unit. Effect of the Invention

[0007] The system according to the embodiment can generate natural responses based on past conversation data. [Brief description of the drawings]

[0008] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Diagram 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. FIG. [Diagram 3] FIG. 11 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Diagram 5] FIG. 13 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 13 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 13 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0010] First, the terms used in the following description will be explained.

[0011] In the following embodiments, a signed processor (hereinafter simply referred to as a "processor") may be one arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be one type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), or a TPU (Tensor Processing Unit).

[0012] In the following embodiments, a signed RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by the processor.

[0013] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0014] In the following embodiments, a communication I / F (Interface) with a code is an interface including a communication processor and an antenna. The communication I / F controls communication between multiple computers. An example of a communication standard applied to the communication I / F is a wireless communication standard including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0015] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. In addition, in this specification, the same idea as "A and / or B" is also applied when three or more things are expressed by connecting them with "and / or."

[0016] [First embodiment] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0017] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0018] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0019] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0020] The reception device 38 includes a touch panel 38A and a microphone 38B, and receives user input. The touch panel 38A detects contact with a pointer (e.g., a pen or a finger) to receive user input by the touch of the pointer. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 (see FIG. 2) acquires the data indicating the user input.

[0021] Output device 40 includes a display 40A and a speaker 40B, and presents data to a user by outputting the data in a form of expression that the user can perceive (e.g., voice and / or text). Display 40A displays visible information such as text and images according to instructions from processor 46. Speaker 40B outputs voice according to instructions from processor 46. Camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a Complementary Metal-Oxide-Semiconductor (CMOS) image sensor or a Charge Coupled Device (CCD) image sensor.

[0022] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for transmitting and receiving various types of information between the processor 46 and the processor 28 via the network 54.

[0023] FIG. 2 shows an example of main functions of the data processing device 12 and the smart device 14.

[0024] As shown in Fig. 2, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32. The specific process program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific process program 56 from the storage 32, and executes the read specific process program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific process program 56 executed on the RAM 30.

[0025] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the emotion of the user using the emotion identification model 59 and perform identification processing using the user's emotion. In the emotion estimation function (emotion identification function) using the emotion identification model 59, various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, are performed, but are not limited to such examples. In addition, the estimation and prediction of the emotion also include, for example, analysis of the emotion.

[0026] In the smart device 14, the specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. The specific processing program 60 is used together with the specific processing program 56 by the data processing system 10. The processor 46 reads the specific processing program 60 from the storage 50 and executes the specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific processing program 60 executed on the RAM 48. Note that the smart device 14 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the specific processing unit 290 using these models.

[0027] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device (e.g., a generation server) may have the data generation model 58. In this case, the data processing device 12 obtains a processing result (prediction result, etc.) using the data generation model 58 by communicating with the server device having the data generation model 58. Also, the data processing device 12 may be a server device, or may be a terminal device (e.g., a mobile phone, a robot, a home appliance, etc.) owned by a user. Next, an example of processing by the data processing system 10 according to the first embodiment will be described.

[0028] (Example 1) A conversation simulation system according to an embodiment of the present invention is a system that allows an AI to learn past conversations on a messenger app, thereby enabling the AI ​​to converse as if it were a specific person. In this system, a user provides the AI ​​with past conversation data on a messenger app, and the AI ​​learns the conversation data and imitates the conversation style and language of a specific person, thereby providing an experience as if the user were talking to the AI ​​when conversing with the AI. For example, if an AI learns a conversation on a messenger app with a deceased grandfather, an AI grandfather is generated, and when the AI ​​speaks to the AI, a conversation as if the AI ​​was talking to the grandfather is possible. Also, if the AI ​​learns interactions with a boss that the user dislikes, the AI ​​provides feedback that is likely to be given in real life. First, a user provides the AI ​​with past conversation data on a messenger app. At this time, the conversation data is provided in text format and input to the AI. Next, the AI ​​analyzes the provided conversation data and learns the conversation style and language of a specific person. For example, if the AI ​​learns the conversation data of a grandfather, it imitates the grandfather's unique phraseology and way of speaking. After the learning is completed, the user starts a conversation with the AI. When a user talks to the AI, the AI ​​responds based on the conversation style it has learned. For example, if the user says, "Grandpa, how was your day?", the AI ​​Grandpa will respond with, "I was looking at flowers in the garden today." Also, when learning interactions with a boss that the AI ​​dislikes, the AI ​​will mimic the boss's feedback style. For example, if the user asks the AI, "What do you think of this presentation?", the AI ​​boss will provide feedback such as, "You should add a little more specific data." This system allows users to recreate conversations with specific people and feel an emotional connection. Also, by simulating interactions with the boss, the AI ​​can prepare for real situations. This allows the conversation simulation system to allow users to recreate conversations with specific people and feel an emotional connection. Also, by simulating interactions with the boss, the AI ​​can prepare for real situations.

[0029] The conversation simulation system according to the embodiment includes a providing unit, an analyzing unit, a learning unit, and a responding unit. The providing unit provides past conversation data of a user in a messenger app. The conversation data includes, for example, text data, voice data, video data, and the like, but is not limited to such examples. The providing unit can provide, for example, past conversation data of a user in a messenger app in text format. The providing unit can also provide voice data and video data. The analyzing unit analyzes the conversation data provided by the providing unit. The analysis is performed, for example, by a method such as natural language processing, emotion analysis, topic modeling, and the like, but is not limited to such examples. For example, the analyzing unit analyzes the conversation data using a natural language processing technique to extract a conversation style and a wording of a specific person. The analyzing unit can also analyze the emotional tone of the conversation data using emotion analysis technology. Furthermore, the analyzing unit can analyze the topic of the conversation data using topic modeling technology. The learning unit learns the conversation style based on the data analyzed by the analyzing unit. The learning is performed, for example, by a method such as supervised learning, unsupervised learning, reinforcement learning, and the like, but is not limited to such examples. For example, the learning unit learns the conversation style of a specific person using supervised learning. The learning unit can also learn patterns of conversation data using unsupervised learning. Furthermore, the learning unit can optimize the conversation style using reinforcement learning. The response unit generates a response based on the conversation style learned by the learning unit. The response is generated by a method such as, for example, template-based response generation, response generation using a generative model, etc., but is not limited to such examples. For example, the response unit generates a response based on the conversation style of a specific person using template-based response generation. Furthermore, the response unit can also generate a response based on the conversation style of a specific person using a generative model. This allows the conversation simulation system according to the embodiment to allow the user to reproduce a conversation with a specific person and feel an emotional connection. Furthermore, by simulating an exchange with a boss, the user can prepare for an actual situation.Some or all of the above-mentioned processes in the providing unit, the analyzing unit, the learning unit, and the responding unit may be performed using, for example, AI, or may be performed without using AI. For example, the providing unit inputs conversation data provided by a user to the generating AI, the analyzing unit analyzes the conversation data using the generating AI, the learning unit learns the conversation style using the generating AI, and the responding unit generates a response using the generating AI.

[0030] The providing unit provides past conversation data of a user in a messenger app. The conversation data includes, for example, text data, voice data, video data, etc., but is not limited to such examples. For example, the providing unit can provide the past conversation data of a user in a messenger app in text format. The providing unit can also provide voice data and video data. Specifically, the user can upload text data exported from the messenger app to the providing unit. In the case of voice data, the user can upload a recorded voice file to the providing unit, and in the case of video data, the user can upload a recorded video file to the providing unit. The providing unit receives these data, converts them into an appropriate format, and passes them to the analyzing unit. For example, the voice data is converted into text data using voice recognition technology, and the video data is converted into text data by extracting the voice portion using video analysis technology. In this way, the providing unit can centrally manage conversation data in various formats, and the analyzing unit can efficiently analyze the data. In addition, the providing unit can anonymize and encrypt the data in order to protect the privacy of the data provided by the user. This allows the user to provide data with peace of mind. Furthermore, the provision unit has a function for checking the quality of data provided by the user, detecting missing or inappropriate data, and notifying the user. In this way, the provision unit supports the analysis unit in using high-quality data.

[0031] The analysis unit analyzes the conversation data provided by the provision unit. The analysis is performed by, for example, natural language processing, emotion analysis, topic modeling, or other methods, but is not limited to these examples. For example, the analysis unit analyzes the conversation data using natural language processing technology to extract the conversation style and language of a specific person. The analysis unit can also analyze the emotional tone of the conversation data using emotion analysis technology. Furthermore, the analysis unit can analyze the topic of the conversation data using topic modeling technology. Specifically, natural language processing technology is used to extract keywords and phrases from the conversation data to grasp the characteristics of the language and speaking style of a specific person. Emotion analysis technology is used to analyze the emotional tone of each statement in the conversation data and classify them into emotion categories such as positive, negative, and neutral. Topic modeling technology is used to identify topics that are frequently discussed in the conversation data and grasp the overall theme and flow of the conversation. In this way, the analysis unit can analyze the provided conversation data from multiple angles and grasp the conversation style, emotional tendencies, and topic tendencies of a specific person in detail. Furthermore, the analysis unit stores the analysis results in a database so that the learning unit and the response unit can access them. The analysis unit has the ability to analyze data in real time and can perform analysis quickly every time a user provides new conversation data. This enables the analysis unit to always provide analysis results based on the latest data, improving the accuracy and reliability of the entire system.

[0032] The learning unit learns the conversation style based on the data analyzed by the analysis unit. The learning is performed by, for example, supervised learning, unsupervised learning, reinforcement learning, and the like, but is not limited to such examples. For example, the learning unit learns the conversation style of a specific person using supervised learning. The learning unit can also learn the patterns of the conversation data using unsupervised learning. Furthermore, the learning unit can optimize the conversation style using reinforcement learning. Specifically, the learning unit uses supervised learning to learn the conversation style and language of a specific person extracted by the analysis unit, and reflects the characteristics of the person in the model. The learning unit uses unsupervised learning to automatically extract common patterns and features from the conversation data and grasp the general tendency of the conversation. The learning unit uses reinforcement learning to learn a strategy for generating optimal responses through a conversation in a simulation environment. This allows the learning unit to reproduce the conversation style of a specific person with high accuracy and generate more natural and realistic responses when a user simulates a conversation with that person. Furthermore, the learning unit periodically updates the learning results and improves the model based on new data and feedback. This enables the learning unit to constantly provide highly accurate conversational styles based on the latest information, improving the performance of the entire system.

[0033] The response unit generates a response based on the conversation style learned by the learning unit. The response is generated by, for example, a method such as template-based response generation or response generation using a generative model, but is not limited to such examples. For example, the response unit generates a response based on the conversation style of a specific person using template-based response generation. The response unit can also generate a response based on the conversation style of a specific person using a generative model. Specifically, the template-based response generation is used to generate a response that matches the language and speaking style of a specific person based on a response template prepared in advance. The generative model is used to generate a natural response that reflects the conversation style learned by the learning unit. The generative model is constructed using, for example, a transformer model or a recurrent neural network (RNN), and generates an appropriate response based on input conversation data. This allows the response unit to provide a realistic and natural response when the user simulates a conversation with a specific person. Furthermore, the response unit also has a function of evaluating the quality of the generated response and making corrections or improvements as necessary. For example, if the generated response is inappropriate or does not meet the user's expectations, the response unit generates other response candidates and selects the optimal response. In addition, the response unit can collect feedback from users and use it to improve the response generation model, thereby enabling the response unit to consistently provide high-quality responses and improve user satisfaction.

[0034] The conversation simulation system includes a pre-processing unit that performs pre-processing of the conversation data. The pre-processing unit performs pre-processing of the conversation data. The pre-processing includes, but is not limited to, noise removal, tokenization, normalization, and the like. For example, the pre-processing unit removes noise from the conversation data. The pre-processing unit can also tokenize the conversation data. Furthermore, the pre-processing unit can also normalize the conversation data. As a result, the accuracy of the analysis is improved by pre-processing the conversation data. A part or all of the above-mentioned processing in the pre-processing unit may be performed using, for example, an AI, or may be performed without using an AI. For example, the pre-processing unit can input the conversation data to a generation AI and perform noise removal, tokenization, and normalization using the generation AI.

[0035] The conversation simulation system includes an imitation unit that imitates the conversation style of a specific person. The imitation unit imitates the conversation style of a specific person. Imitation includes, for example, feature extraction of a specific phrase or way of speaking, an imitation algorithm, and the like, but is not limited to such examples. For example, the imitation unit extracts frequent phrases of a specific person. The imitation unit can also analyze the grammar pattern of the specific person. Furthermore, the imitation unit can also apply an algorithm that imitates the conversation style of a specific person. This makes it possible to have a more realistic conversation by imitating the conversation style of a specific person. A part or all of the above-mentioned processing in the imitation unit may be performed, for example, using AI, or may be performed without using AI. For example, the imitation unit can input conversation data of a specific person to a generation AI and imitate the conversation style using the generation AI.

[0036] The conversation simulation system includes a feedback unit that provides feedback. The feedback unit provides the feedback. The feedback includes, for example, feedback based on the user's reaction, presentation of improvements to the system, and the like, but is not limited to such examples. For example, the feedback unit analyzes the user's reaction and provides appropriate feedback. The feedback unit can also present improvements to the system. As a result, the feedback is provided to improve the user's conversation experience. A part or all of the above-mentioned processing in the feedback unit may be performed, for example, using AI, or may be performed without using AI. For example, the feedback unit can input the user's reaction data to a generation AI and provide feedback using the generation AI.

[0037] The preprocessing unit can perform preprocessing of the text data (e.g., noise removal and tokenization). The preprocessing unit, for example, performs noise removal of the text data. Noise removal includes specific pattern removal and filtering techniques. For example, the preprocessing unit removes unnecessary characters and symbols from the conversation data. The preprocessing unit also tokenizes the text data. Tokenization includes tokenization by word and by sentence. For example, the preprocessing unit divides the conversation data into words. The preprocessing unit can also divide the conversation data into sentences. This improves the quality of the data by cleaning and tokenizing the text data. Some or all of the above-mentioned processing in the preprocessing unit may be performed using, for example, AI, or may be performed without using AI. For example, the preprocessing unit can input the conversation data to a generation AI and perform noise removal and tokenization using the generation AI.

[0038] The imitation unit can imitate a specific person's unique phrase or speaking style. The imitation unit, for example, extracts frequent phrases of a specific person. Extracting frequent phrases includes a technique for analyzing the characteristics of a specific phrase or speaking style. For example, the imitation unit extracts frequent phrases from the conversation data of a specific person. The imitation unit can also analyze the grammar pattern of a specific person. Analyzing the grammar pattern includes a technique for analyzing a specific grammatical structure or a pattern of phrases. For example, the imitation unit analyzes the grammar pattern from the conversation data of a specific person. This enables a more realistic conversation by imitating a specific phrase or speaking style. A part or all of the above-mentioned processing in the imitation unit may be performed, for example, using AI, or may be performed without using AI. For example, the imitation unit can input the conversation data of a specific person to a generation AI and imitate the conversation style using the generation AI.

[0039] The providing unit can analyze the user's past conversation data providing history and select a providing method. The providing unit, for example, analyzes the user's past conversation data providing history. The analysis includes an analysis of past usage patterns and an analysis of feedback. For example, the providing unit preferentially provides conversation data that the user frequently accessed in the past. The providing unit can also analyze the user's past usage patterns and provide conversation data at an optimal timing. Furthermore, the providing unit can adjust the content of the conversation data to be provided based on the user's past feedback. Next, the providing unit selects an optimal providing method based on the analysis result. The providing method includes real-time provision and batch provision. For example, the providing unit provides conversation data in real time based on the user's usage patterns. Furthermore, the providing unit can perform batch provision based on the user's feedback. Thereby, the optimal providing method is selected by analyzing the past conversation data providing history. A part or all of the above-mentioned processing in the providing unit may be performed using, for example, AI, or may be performed without using AI. For example, the providing unit can input the user's past conversation data providing history to a generating AI, analyze it using the generating AI, and select an optimal providing method.

[0040] The providing unit can perform filtering based on the user's current field of interest when providing the conversation data. The providing unit, for example, identifies the user's current field of interest. Identifying the field of interest includes keyword-based filtering and topic-based filtering. For example, the providing unit identifies the field of interest based on the user's recent search history. The providing unit can also analyze the user's social media activity and identify the field of interest. Next, the providing unit filters the conversation data based on the identified field of interest. For example, conversation data related to topics in which the user is currently interested is preferentially provided. The providing unit can also filter the conversation data based on the user's field of interest. As a result, highly relevant conversation data is provided by filtering based on the user's field of interest. A part or all of the above-mentioned processing in the providing unit may be performed using, for example, AI, or may be performed without using AI. For example, the providing unit can input the user's field of interest data to a generating AI and perform filtering using the generating AI.

[0041] The providing unit can provide highly relevant data preferentially based on the geographical location information of the user when providing the conversation data. The providing unit, for example, acquires the geographical location information of the user. The acquisition of the geographical location information includes the use of GPS data and the use of a location information service. For example, the providing unit acquires GPS data from the user's smartphone. The providing unit can also acquire the geographical location information of the user by using a location information service. Next, the providing unit provides highly relevant conversation data based on the acquired geographical location information. For example, when the user is in a specific area, conversation data related to the area is provided. In addition, when the user is traveling, conversation data related to the travel destination can be provided. Furthermore, when the user is at home, conversation data related to the home can be provided. As a result, by providing highly relevant data based on the geographical location information, conversation data according to the user's situation is provided. A part or all of the above-mentioned processing in the providing unit may be performed, for example, using AI, or may be performed without using AI. For example, the providing unit can input the geographical location information data of the user to a generating AI and provide highly relevant conversation data using the generating AI.

[0042] The providing unit can analyze the user's social media activity and provide related data when providing the conversation data. The providing unit, for example, analyzes the user's social media activity. The analysis includes analysis of the posted content and analysis of followers. For example, the providing unit provides conversation data related to a topic recently shared by the user on social media. The providing unit can also provide related conversation data based on the user's friendships on social media. Furthermore, the providing unit can analyze the user's social media activity history and provide interesting conversation data. In this way, conversation data based on the user's interests is provided by analyzing social media activities. A part or all of the above-mentioned processing in the providing unit may be performed, for example, using AI, or may be performed without using AI. For example, the providing unit can input the user's social media activity data into a generating AI, analyze it using the generating AI, and provide related conversation data.

[0043] The analysis unit can adjust the level of detail of the analysis based on the importance of the conversation data during the analysis. The analysis unit, for example, evaluates the importance of the conversation data. The evaluation of the importance includes frequency analysis and analysis of the user's interest. For example, the analysis unit analyzes the frequency of the conversation data and evaluates the importance. The analysis unit can also analyze the user's interest and evaluate the importance. Next, the analysis unit adjusts the level of detail of the analysis based on the evaluated importance. For example, a detailed analysis is performed on important conversation data. Also, a simplified analysis can be performed on general conversation data. Furthermore, an analysis can be focused on conversation data in which the user is particularly interested. As a result, an efficient analysis is performed by adjusting the level of detail of the analysis based on the importance of the conversation data. A part or all of the above-mentioned processing in the analysis unit may be performed using, for example, an AI, or may be performed without using an AI. For example, the analysis unit can input the importance data of the conversation data to a generation AI and adjust the level of detail of the analysis using the generation AI.

[0044] The analysis unit can apply different analysis algorithms depending on the category of the conversation data during analysis. The analysis unit, for example, classifies the category of the conversation data. The category classification includes topic-based classification and content-based classification. For example, the analysis unit classifies the conversation data by topic. The analysis unit can also classify the conversation data by content. Next, the analysis unit applies different analysis algorithms depending on the classified category. For example, an analysis algorithm that emphasizes emotional tone can be applied to conversation data with family members. An analysis algorithm that emphasizes efficiency can also be applied to work-related conversation data. Furthermore, an analysis algorithm that emphasizes familiarity can also be applied to conversation data with friends. In this way, by applying different analysis algorithms depending on the category of the conversation data, more appropriate analysis can be performed. Part or all of the above-mentioned processing in the analysis unit may be performed using, for example, AI, or may be performed without using AI. For example, the analysis unit can input category data of the conversation data to a generation AI and apply different analysis algorithms using the generation AI.

[0045] The analysis unit can adjust the order of analysis based on the submission time of the conversation data during analysis. The analysis unit, for example, evaluates the submission time of the conversation data. The evaluation of the submission time includes recording the submission date and time and managing the submission order. For example, the analysis unit records the submission date and time of the conversation data. The analysis unit can also manage the submission order of the conversation data. Next, the analysis unit adjusts the order of analysis based on the evaluated submission time. For example, recent conversation data is preferentially analyzed. Conversation data that was concentrated in a specific period can also be preferentially analyzed. Furthermore, conversation data from a period in which the user is particularly interested can also be preferentially analyzed. As a result, efficient analysis is performed by adjusting the order of analysis based on the submission time of the conversation data. Part or all of the above-mentioned processing in the analysis unit may be performed using, for example, AI, or may be performed without using AI. For example, the analysis unit can input the submission time data of the conversation data to the generation AI and adjust the order of analysis using the generation AI.

[0046] The analysis unit can adjust the order of analysis based on the relevance of the conversation data during analysis. The analysis unit, for example, evaluates the relevance of the conversation data. The evaluation of the relevance includes co-occurrence network analysis and topic modeling. For example, the analysis unit analyzes the co-occurrence network of the conversation data and evaluates the relevance. The analysis unit can also model the topic of the conversation data and evaluate the relevance. Next, the analysis unit adjusts the order of analysis based on the evaluated relevance. For example, highly relevant conversation data is preferentially analyzed. Also, conversation data related to a topic in which the user is particularly interested can be preferentially analyzed. Furthermore, based on the user's past usage history, highly relevant conversation data can be preferentially analyzed. As a result, efficient analysis is performed by adjusting the order of analysis based on the relevance of the conversation data. Part or all of the above-mentioned processing in the analysis unit may be performed using, for example, AI, or may be performed without using AI. For example, the analysis unit can input the relevance data of the conversation data to the generation AI and adjust the order of analysis using the generation AI.

[0047] The learning unit can optimize the learning algorithm by referring to past learning data during learning. The learning unit, for example, refers to past learning data. The reference includes feedback of past learning results and extraction of effective learning patterns. For example, the learning unit selects an optimal learning algorithm based on past learning data. The learning unit can also extract effective learning patterns from past learning data and optimize the algorithm. Furthermore, the learning unit can analyze past learning data and adjust parameters of the learning algorithm. As a result, the learning algorithm is optimized by referring to the past learning data. A part or all of the above-mentioned processing in the learning unit may be performed, for example, using AI or may be performed without using AI. For example, the learning unit can input past learning data to a generation AI and optimize the learning algorithm using the generation AI.

[0048] The learning unit can apply different learning methods to each category of conversation data during learning. The learning unit, for example, classifies the categories of conversation data. Category classification includes topic-based classification and content-based classification. For example, the learning unit classifies the conversation data by topic. The learning unit can also classify the conversation data by content. Next, the learning unit applies different learning methods to each classified category. For example, a learning method that emphasizes emotional tone can be applied to conversation data with family members. A learning method that emphasizes efficiency can also be applied to work-related conversation data. Furthermore, a learning method that emphasizes familiarity can also be applied to conversation data with friends. In this way, more appropriate learning is performed by applying different learning methods to each category of conversation data. Part or all of the above-mentioned processing in the learning unit may be performed using, for example, AI, or may be performed without using AI. For example, the learning unit can input category data of the conversation data to a generation AI and apply different learning methods using the generation AI.

[0049] The learning unit can weight the learning data based on the submission time of the conversation data during learning. The learning unit, for example, evaluates the submission time of the conversation data. The evaluation of the submission time includes recording the submission date and time and managing the submission order. For example, the learning unit records the submission date and time of the conversation data. The learning unit can also manage the submission order of the conversation data. Next, the learning unit weights the learning data based on the evaluated submission time. For example, a high weight is set for recent conversation data. Also, a high weight can be set for conversation data that was concentrated in a specific period. Furthermore, a high weight can be set for conversation data from a period in which the user is particularly interested. As a result, efficient learning is performed by weighting the learning data based on the submission time of the conversation data. A part or all of the above-mentioned processing in the learning unit may be performed using, for example, an AI, or may be performed without using an AI. For example, the learning unit can input the submission time data of the conversation data to a generation AI and weight the learning data using the generation AI.

[0050] The learning unit can improve the accuracy of learning by referring to literature related to the conversation data during learning. The learning unit, for example, refers to literature related to the conversation data. The reference includes the use of a literature database and the analysis of citation relationships. For example, the learning unit optimizes a learning algorithm based on the related literature. The learning unit can also extract effective learning patterns from the related literature to improve the accuracy of learning. Furthermore, the learning unit can also select learning data by referring to the related literature. As a result, the accuracy of learning is improved by referring to the literature related to the conversation data. A part or all of the above-mentioned processing in the learning unit may be performed, for example, using AI, or may be performed without using AI. For example, the learning unit can input literature data related to the conversation data to a generation AI and improve the accuracy of learning using the generation AI.

[0051] The response unit can adjust the level of detail of the response based on the importance of the conversation when generating a response. The response unit, for example, evaluates the importance of the conversation. The evaluation of the importance includes frequency analysis and analysis of the user's interest. For example, the response unit analyzes the frequency of the conversation data and evaluates the importance. The response unit can also analyze the user's interest and evaluate the importance. Next, the response unit adjusts the level of detail of the response based on the evaluated importance. For example, a detailed response is generated for an important conversation. Also, a simple response can be generated for a general conversation. Furthermore, a detailed response can be generated with a focus on a conversation in which the user is particularly interested. As a result, an efficient response is generated by adjusting the level of detail of the response based on the importance of the conversation. A part or all of the above-mentioned processing in the response unit may be performed, for example, using AI, or may be performed without using AI. For example, the response unit can input importance data of the conversation data to the generation AI and adjust the level of detail of the response using the generation AI.

[0052] The response unit can apply different response algorithms according to the category of the conversation when generating a response. The response unit, for example, classifies the category of the conversation data. The category classification includes topic-based classification and content-based classification. For example, the response unit classifies the conversation data by topic. The response unit can also classify the conversation data by content. Next, the response unit applies different response algorithms according to the classified category. For example, a response algorithm that emphasizes emotional tone can be applied to a conversation with family members. A response algorithm that emphasizes efficiency can also be applied to a work-related conversation. Furthermore, a response algorithm that emphasizes familiarity can also be applied to a conversation with friends. In this way, a more appropriate response is generated by applying different response algorithms according to the category of the conversation. Part or all of the above-mentioned processing in the response unit may be performed, for example, using AI, or may be performed without using AI. For example, the response unit can input category data of the conversation data to a generation AI and apply different response algorithms using the generation AI.

[0053] The response unit can determine the priority of responses based on the submission time of the conversation when generating a response. The response unit, for example, evaluates the submission time of the conversation data. The evaluation of the submission time includes recording the submission date and time and managing the submission order. For example, the response unit records the submission date and time of the conversation data. The response unit can also manage the submission order of the conversation data. Next, the response unit determines the priority of responses based on the evaluated submission time. For example, responses are generated preferentially for recent conversations. Responses can also be generated preferentially for conversations that were concentrated in a specific period. Furthermore, responses can also be generated preferentially for conversations in a period in which the user is particularly interested. In this way, an efficient response is generated by determining the priority of responses based on the submission time of the conversation. A part or all of the above-mentioned processing in the response unit may be performed, for example, using AI, or may be performed without using AI. For example, the response unit can input submission time data of the conversation data to the generation AI and determine the priority of responses using the generation AI.

[0054] The response unit can adjust the order of responses based on the relevance of the conversation when generating a response. The response unit, for example, evaluates the relevance of the conversation data. The evaluation of the relevance includes co-occurrence network analysis and topic modeling. For example, the response unit analyzes the co-occurrence network of the conversation data and evaluates the relevance. The response unit can also model the topic of the conversation data and evaluate the relevance. Next, the response unit adjusts the order of responses based on the evaluated relevance. For example, a response is preferentially generated for a conversation that is highly relevant. A response can also be preferentially generated for a conversation related to a topic in which the user is particularly interested. Furthermore, a response can also be preferentially generated for a conversation that is highly relevant based on the user's past usage history. As a result, an efficient response is generated by adjusting the order of responses based on the relevance of the conversation. A part or all of the above-mentioned processing in the response unit may be performed using, for example, AI, or may be performed without using AI. For example, the response unit can input the relevance data of the conversation data to the generation AI and adjust the order of responses using the generation AI.

[0055] The preprocessing unit can adjust the level of detail of the preprocessing based on the importance of the conversation data during preprocessing. The preprocessing unit, for example, evaluates the importance of the conversation data. The evaluation of the importance includes frequency analysis and analysis of the user's interest level. For example, the preprocessing unit analyzes the frequency of the conversation data and evaluates the importance. The preprocessing unit can also analyze the user's interest level and evaluate the importance. Next, the preprocessing unit adjusts the level of detail of the preprocessing based on the evaluated importance. For example, detailed preprocessing is performed for important conversation data. Also, simple preprocessing can be performed for general conversation data. Furthermore, preprocessing can be focused on conversation data in which the user is particularly interested. As a result, efficient preprocessing is performed by adjusting the level of detail of the preprocessing based on the importance of the conversation data. Part or all of the above-mentioned processing in the preprocessing unit may be performed using, for example, AI, or may be performed without using AI. For example, the preprocessing unit can input importance data of the conversation data to a generation AI and adjust the level of detail of the preprocessing using the generation AI.

[0056] The preprocessing unit can adjust the order of preprocessing based on the submission time of the conversation data during preprocessing. The preprocessing unit, for example, evaluates the submission time of the conversation data. The evaluation of the submission time includes recording the submission date and time and managing the submission order. For example, the preprocessing unit records the submission date and time of the conversation data. The preprocessing unit can also manage the submission order of the conversation data. Next, the preprocessing unit adjusts the order of preprocessing based on the evaluated submission time. For example, recent conversation data is preprocessed preferentially. Conversation data that was concentrated in a specific period can also be preprocessed preferentially. Furthermore, conversation data from a period in which the user is particularly interested can also be preprocessed preferentially. As a result, efficient preprocessing is performed by adjusting the order of preprocessing based on the submission time of the conversation data. Part or all of the above-mentioned processing in the preprocessing unit may be performed using, for example, AI, or may be performed without using AI. For example, the preprocessing unit can input submission time data of the conversation data to a generation AI and adjust the order of preprocessing using the generation AI.

[0057] The imitation unit can adjust the level of detail of the imitation based on the importance of the conversation data during imitation. The imitation unit, for example, evaluates the importance of the conversation data. The evaluation of the importance includes frequency analysis and analysis of the user's interest. For example, the imitation unit analyzes the frequency of the conversation data and evaluates the importance. The imitation unit can also analyze the user's interest and evaluate the importance. Next, the imitation unit adjusts the level of detail of the imitation based on the evaluated importance. For example, detailed imitation is performed for important conversation data. Also, simple imitation can be performed for general conversation data. Furthermore, imitation can be focused on conversation data in which the user is particularly interested. As a result, efficient imitation is performed by adjusting the level of detail of the imitation based on the importance of the conversation data. Part or all of the above-mentioned processing in the imitation unit may be performed using, for example, AI, or may be performed without using AI. For example, the imitation unit can input the importance data of the conversation data to the generation AI and adjust the level of detail of the imitation using the generation AI.

[0058] The imitation unit can apply different imitation algorithms depending on the category of the conversation data during imitation. The imitation unit, for example, classifies the category of the conversation data. The category classification includes topic-based classification and content-based classification. For example, the imitation unit classifies the conversation data by topic. The imitation unit can also classify the conversation data by content. Next, the imitation unit applies different imitation algorithms depending on the classified category. For example, an imitation algorithm that emphasizes emotional tone can be applied to conversation data with family members. An imitation algorithm that emphasizes efficiency can also be applied to work-related conversation data. Furthermore, an imitation algorithm that emphasizes familiarity can also be applied to conversation data with friends. In this way, by applying different imitation algorithms depending on the category of the conversation data, more appropriate imitation is performed. Part or all of the above-mentioned processing in the imitation unit may be performed using, for example, AI, or may be performed without using AI. For example, the imitation unit can input category data of the conversation data to a generation AI and apply different imitation algorithms using the generation AI.

[0059] The imitation unit can adjust the order of imitation based on the submission time of the conversation data during imitation. The imitation unit, for example, evaluates the submission time of the conversation data. The evaluation of the submission time includes recording the submission date and time and managing the submission order. For example, the imitation unit records the submission date and time of the conversation data. The imitation unit can also manage the submission order of the conversation data. Next, the imitation unit adjusts the order of imitation based on the evaluated submission time. For example, recent conversation data is preferentially imitated. Also, conversation data that was concentrated in a specific period can be preferentially imitated. Furthermore, conversation data from a period in which the user is particularly interested can be preferentially imitated. As a result, efficient imitation is performed by adjusting the order of imitation based on the submission time of the conversation data. Part or all of the above-mentioned processing in the imitation unit may be performed using, for example, AI, or may be performed without using AI. For example, the imitation unit can input the submission time data of the conversation data to the generation AI and adjust the order of imitation using the generation AI.

[0060] The imitation unit can adjust the order of imitation based on the relevance of the conversation data during imitation. The imitation unit, for example, evaluates the relevance of the conversation data. The evaluation of the relevance includes co-occurrence network analysis and topic modeling. For example, the imitation unit analyzes the co-occurrence network of the conversation data and evaluates the relevance. The imitation unit can also model the topic of the conversation data and evaluate the relevance. Next, the imitation unit adjusts the order of imitation based on the evaluated relevance. For example, highly relevant conversation data is preferentially imitated. Also, conversation data related to a topic in which the user is particularly interested can be preferentially imitated. Furthermore, based on the user's past usage history, highly relevant conversation data can be preferentially imitated. As a result, efficient imitation is performed by adjusting the order of imitation based on the relevance of the conversation data. A part or all of the above-mentioned processing in the imitation unit may be performed using, for example, AI, or may be performed without using AI. For example, the imitation unit can input the relevance data of the conversation data to the generation AI and adjust the order of imitation using the generation AI.

[0061] The feedback unit can adjust the level of detail of the feedback based on the importance of the conversation data at the time of feedback. The feedback unit, for example, evaluates the importance of the conversation data. The evaluation of the importance includes frequency analysis and analysis of the user's interest. For example, the feedback unit analyzes the frequency of the conversation data and evaluates the importance. The feedback unit can also analyze the user's interest and evaluate the importance. Next, the feedback unit adjusts the level of detail of the feedback based on the evaluated importance. For example, detailed feedback is provided for important conversation data. Also, brief feedback can be provided for general conversation data. Furthermore, detailed feedback can be provided with a focus on conversation data in which the user is particularly interested. As a result, efficient feedback is provided by adjusting the level of detail of the feedback based on the importance of the conversation data. A part or all of the above-mentioned processing in the feedback unit may be performed, for example, using AI, or may be performed without using AI. For example, the feedback unit can input importance data of the conversation data to a generation AI and adjust the level of detail of the feedback using the generation AI.

[0062] The feedback unit can adjust the order of feedback based on the submission time of the conversation data at the time of feedback. The feedback unit, for example, evaluates the submission time of the conversation data. The evaluation of the submission time includes recording the submission date and time and managing the submission order. For example, the feedback unit records the submission date and time of the conversation data. The feedback unit can also manage the submission order of the conversation data. Next, the feedback unit adjusts the order of feedback based on the evaluated submission time. For example, recent conversation data is preferentially fed back. Also, conversation data that was concentrated in a specific period can be preferentially fed back. Furthermore, conversation data from a period in which the user is particularly interested can be preferentially fed back. As a result, efficient feedback is provided by adjusting the order of feedback based on the submission time of the conversation data. A part or all of the above-mentioned processing in the feedback unit may be performed, for example, using AI, or may be performed without using AI. For example, the feedback unit can input the submission time data of the conversation data to a generation AI and adjust the order of feedback using the generation AI.

[0063] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0064] The conversation simulation system can estimate the user's preferences and interests based on the user's past conversation data, and customize the conversation content based on the estimated preferences and interests. For example, it can identify hobbies and interests that the user has frequently talked about in the past, and develop the conversation based on that information. In addition, if the user shows a strong interest in a specific topic, it can provide information related to that topic with priority. Furthermore, it can analyze the user's tendency to prefer specific phrases and expressions from the user's past conversation data, and adopt a conversation style that matches that tendency. This provides the user with a more familiar and interesting conversation experience.

[0065] The conversation simulation system can monitor the user's current activity status in real time and provide conversation content according to the status. For example, when the user is exercising, advice and words of encouragement related to the exercise can be provided. When the user is working, work-related information and light topics for refreshing can be provided. Furthermore, when the user is relaxing, relaxing topics and information about hobbies can be provided. This provides appropriate conversation content according to the user's activity status, realizing a more personalized experience.

[0066] The conversation simulation system can estimate a user's learning style based on the user's past conversation data, and provide conversation content based on the estimated learning style. For example, if the user has a visual learning style, explanations using diagrams and graphs can be provided. Also, if the user has an auditory learning style, emphasis can be placed on audio explanations. Furthermore, if the user has an experiential learning style, topics that encourage learning through actual experience can be provided. This provides appropriate conversation content according to the user's learning style, improving learning effectiveness.

[0067] The conversation simulation system can estimate the user's health condition based on the user's past conversation data, and provide conversation content based on the estimated health condition. For example, if the user has a health problem, advice and support information for that problem can be provided. Also, if the user is living a healthy lifestyle, information and motivation for maintaining that lifestyle can be provided. Furthermore, the tone and tempo of the conversation can be adjusted according to the user's health condition. In this way, appropriate conversation content according to the user's health condition can be provided, supporting the user's health management.

[0068] The conversation simulation system can estimate the user's goals and dreams based on the user's past conversation data, and provide conversation content based on the estimated goals and dreams. For example, if the user has a specific career goal, advice and information toward that goal can be provided. Also, if the user is pursuing a specific hobby or interest, information related to the hobby or interest can be provided. Furthermore, the tone and tempo of the conversation can be adjusted according to the user's goals and dreams. This provides appropriate conversation content according to the user's goals and dreams, improving the user's motivation.

[0069] The conversation simulation system can estimate the sociability of a user based on the user's past conversation data, and provide conversation content based on the estimated sociability. For example, if the user has a sociable personality, positive topics and topics that encourage meeting new people can be provided. Also, if the user has an introverted personality, relaxing topics and information about hobbies that can be enjoyed alone can be provided. Furthermore, the tone and tempo of the conversation can be adjusted according to the user's sociability. This provides appropriate conversation content according to the user's sociability, improving the user's communication experience.

[0070] The conversation simulation system can estimate the cultural background of a user based on the user's past conversation data, and provide conversation content based on the estimated cultural background. For example, if the user has a specific culture or tradition, topics related to that culture or tradition can be provided. Also, if the user has a multicultural background, information and topics related to different cultures can be provided. Furthermore, the tone and tempo of the conversation can be adjusted according to the user's cultural background. This provides appropriate conversation content according to the user's cultural background, strengthening the user's cultural connection.

[0071] The process flow of the first embodiment will be briefly described below.

[0072] Step 1: The providing unit provides past conversation data from a user on a messenger app. The conversation data includes text data, audio data, video data, and the like. For example, the user can provide past conversation data from a messenger app in text format. Audio data and video data can also be provided. Step 2: The analysis unit analyzes the conversation data provided by the provision unit. The analysis is performed using methods such as natural language processing, sentiment analysis, and topic modeling. For example, the conversation data is analyzed using natural language processing technology to extract the conversation style and language of a particular person. In addition, the emotional tone of the conversation data can be analyzed using sentiment analysis technology. Furthermore, the topic of the conversation data can be analyzed using topic modeling technology. Step 3: The learning unit learns the conversation style based on the data analyzed by the analysis unit. The learning is performed by a method such as supervised learning, unsupervised learning, or reinforcement learning. For example, supervised learning is used to learn the conversation style of a specific person. Also, unsupervised learning can be used to learn patterns in conversation data. Furthermore, reinforcement learning can be used to optimize the conversation style. Step 4: The response unit generates a response based on the conversation style learned by the learning unit. The response is generated by a method such as template-based response generation or response generation using a generative model. For example, a response based on a conversation style of a specific person can be generated using template-based response generation. A response based on a conversation style of a specific person can also be generated using a generative model.

[0073] (Example 2) A conversation simulation system according to an embodiment of the present invention is a system that allows an AI to learn past conversations on a messenger app, thereby enabling the AI ​​to converse as if it were a specific person. In this system, a user provides the AI ​​with past conversation data on a messenger app, and the AI ​​learns the conversation data and imitates the conversation style and language of a specific person, thereby providing an experience as if the user were talking to the AI ​​when conversing with the AI. For example, if an AI learns a conversation on a messenger app with a deceased grandfather, an AI grandfather is generated, and when the AI ​​speaks to the AI, it becomes possible to have a conversation as if the AI ​​was talking to the grandfather. Also, if the AI ​​learns interactions with a boss that the AI ​​dislikes, the AI ​​provides feedback that is likely to be given in real life. First, a user provides the AI ​​with past conversation data on a messenger app. At this time, the conversation data is provided in text format and input to the AI. Next, the AI ​​analyzes the provided conversation data and learns the conversation style and language of a specific person. For example, if the AI ​​learns the conversation data of a grandfather, it imitates the grandfather's unique phraseology and way of speaking. After the learning is completed, the user starts a conversation with the AI. When a user talks to the AI, the AI ​​responds based on the conversation style it has learned. For example, if the user says, "Grandpa, how was your day?", the AI ​​Grandpa will respond with, "I was looking at flowers in the garden today." Also, when learning interactions with a boss that the AI ​​dislikes, the AI ​​will mimic the boss's feedback style. For example, if the user asks the AI, "What do you think of this presentation?", the AI ​​boss will provide feedback such as, "You should add a little more specific data." This system allows users to recreate conversations with specific people and feel an emotional connection. Also, by simulating interactions with the boss, the AI ​​can prepare for real situations. This allows the conversation simulation system to allow users to recreate conversations with specific people and feel an emotional connection. Also, by simulating interactions with the boss, the AI ​​can prepare for real situations.

[0074] The conversation simulation system according to the embodiment includes a providing unit, an analyzing unit, a learning unit, and a responding unit. The providing unit provides past conversation data of a user in a messenger app. The conversation data includes, for example, text data, voice data, video data, and the like, but is not limited to such examples. The providing unit can provide, for example, past conversation data of a user in a messenger app in text format. The providing unit can also provide voice data and video data. The analyzing unit analyzes the conversation data provided by the providing unit. The analysis is performed, for example, by a method such as natural language processing, emotion analysis, topic modeling, and the like, but is not limited to such examples. For example, the analyzing unit analyzes the conversation data using a natural language processing technique to extract a conversation style and a wording of a specific person. The analyzing unit can also analyze the emotional tone of the conversation data using emotion analysis technology. Furthermore, the analyzing unit can analyze the topic of the conversation data using topic modeling technology. The learning unit learns the conversation style based on the data analyzed by the analyzing unit. The learning is performed, for example, by a method such as supervised learning, unsupervised learning, reinforcement learning, and the like, but is not limited to such examples. For example, the learning unit learns the conversation style of a specific person using supervised learning. The learning unit can also learn patterns of conversation data using unsupervised learning. Furthermore, the learning unit can optimize the conversation style using reinforcement learning. The response unit generates a response based on the conversation style learned by the learning unit. The response is generated by a method such as, for example, template-based response generation, response generation using a generative model, etc., but is not limited to such examples. For example, the response unit generates a response based on the conversation style of a specific person using template-based response generation. Furthermore, the response unit can also generate a response based on the conversation style of a specific person using a generative model. This allows the conversation simulation system according to the embodiment to allow the user to reproduce a conversation with a specific person and feel an emotional connection. Furthermore, by simulating an exchange with a boss, the user can prepare for an actual situation.Some or all of the above-mentioned processes in the providing unit, the analyzing unit, the learning unit, and the responding unit may be performed using, for example, AI, or may be performed without using AI. For example, the providing unit inputs conversation data provided by a user to the generating AI, the analyzing unit analyzes the conversation data using the generating AI, the learning unit learns the conversation style using the generating AI, and the responding unit generates a response using the generating AI.

[0075] The providing unit provides past conversation data of a user in a messenger app. The conversation data includes, for example, text data, voice data, video data, etc., but is not limited to such examples. For example, the providing unit can provide the past conversation data of a user in a messenger app in text format. The providing unit can also provide voice data and video data. Specifically, the user can upload text data exported from the messenger app to the providing unit. In the case of voice data, the user can upload a recorded voice file to the providing unit, and in the case of video data, the user can upload a recorded video file to the providing unit. The providing unit receives these data, converts them into an appropriate format, and passes them to the analyzing unit. For example, the voice data is converted into text data using voice recognition technology, and the video data is converted into text data by extracting the voice portion using video analysis technology. In this way, the providing unit can centrally manage conversation data in various formats, and the analyzing unit can efficiently analyze the data. In addition, the providing unit can anonymize and encrypt the data in order to protect the privacy of the data provided by the user. This allows the user to provide data with peace of mind. Furthermore, the provision unit has a function for checking the quality of data provided by the user, detecting missing or inappropriate data, and notifying the user. In this way, the provision unit supports the analysis unit in using high-quality data.

[0076] The analysis unit analyzes the conversation data provided by the provision unit. The analysis is performed by, for example, natural language processing, emotion analysis, topic modeling, or other methods, but is not limited to these examples. For example, the analysis unit analyzes the conversation data using natural language processing technology to extract the conversation style and language of a specific person. The analysis unit can also analyze the emotional tone of the conversation data using emotion analysis technology. Furthermore, the analysis unit can analyze the topic of the conversation data using topic modeling technology. Specifically, natural language processing technology is used to extract keywords and phrases from the conversation data to grasp the characteristics of the language and speaking style of a specific person. Emotion analysis technology is used to analyze the emotional tone of each statement in the conversation data and classify them into emotion categories such as positive, negative, and neutral. Topic modeling technology is used to identify topics that are frequently discussed in the conversation data and grasp the overall theme and flow of the conversation. In this way, the analysis unit can analyze the provided conversation data from multiple angles and grasp the conversation style, emotional tendencies, and topic tendencies of a specific person in detail. Furthermore, the analysis unit stores the analysis results in a database so that the learning unit and the response unit can access them. The analysis unit has the ability to analyze data in real time and can perform analysis quickly every time a user provides new conversation data. This enables the analysis unit to always provide analysis results based on the latest data, improving the accuracy and reliability of the entire system.

[0077] The learning unit learns the conversation style based on the data analyzed by the analysis unit. The learning is performed by, for example, supervised learning, unsupervised learning, reinforcement learning, and the like, but is not limited to such examples. For example, the learning unit learns the conversation style of a specific person using supervised learning. The learning unit can also learn the patterns of the conversation data using unsupervised learning. Furthermore, the learning unit can optimize the conversation style using reinforcement learning. Specifically, the learning unit uses supervised learning to learn the conversation style and language of a specific person extracted by the analysis unit, and reflects the characteristics of the person in the model. The learning unit uses unsupervised learning to automatically extract common patterns and features from the conversation data and grasp the general tendency of the conversation. The learning unit uses reinforcement learning to learn a strategy for generating optimal responses through a conversation in a simulation environment. This allows the learning unit to reproduce the conversation style of a specific person with high accuracy and generate more natural and realistic responses when a user simulates a conversation with that person. Furthermore, the learning unit periodically updates the learning results and improves the model based on new data and feedback. This enables the learning unit to constantly provide highly accurate conversational styles based on the latest information, improving the performance of the entire system.

[0078] The response unit generates a response based on the conversation style learned by the learning unit. The response is generated by, for example, a method such as template-based response generation or response generation using a generative model, but is not limited to such examples. For example, the response unit generates a response based on the conversation style of a specific person using template-based response generation. The response unit can also generate a response based on the conversation style of a specific person using a generative model. Specifically, the template-based response generation is used to generate a response that matches the language and speaking style of a specific person based on a response template prepared in advance. The generative model is used to generate a natural response that reflects the conversation style learned by the learning unit. The generative model is constructed using, for example, a transformer model or a recurrent neural network (RNN), and generates an appropriate response based on input conversation data. This allows the response unit to provide a realistic and natural response when the user simulates a conversation with a specific person. Furthermore, the response unit also has a function of evaluating the quality of the generated response and making corrections or improvements as necessary. For example, if the generated response is inappropriate or does not meet the user's expectations, the response unit generates other response candidates and selects the optimal response. In addition, the response unit can collect feedback from users and use it to improve the response generation model, thereby enabling the response unit to consistently provide high-quality responses and improve user satisfaction.

[0079] The conversation simulation system includes a pre-processing unit that performs pre-processing of the conversation data. The pre-processing unit performs pre-processing of the conversation data. The pre-processing includes, but is not limited to, noise removal, tokenization, normalization, and the like. For example, the pre-processing unit removes noise from the conversation data. The pre-processing unit can also tokenize the conversation data. Furthermore, the pre-processing unit can also normalize the conversation data. As a result, the accuracy of the analysis is improved by pre-processing the conversation data. A part or all of the above-mentioned processing in the pre-processing unit may be performed using, for example, an AI, or may be performed without using an AI. For example, the pre-processing unit can input the conversation data to a generation AI and perform noise removal, tokenization, and normalization using the generation AI.

[0080] The conversation simulation system includes an imitation unit that imitates the conversation style of a specific person. The imitation unit imitates the conversation style of a specific person. Imitation includes, for example, feature extraction of a specific phrase or way of speaking, an imitation algorithm, and the like, but is not limited to such examples. For example, the imitation unit extracts frequent phrases of a specific person. The imitation unit can also analyze the grammar pattern of the specific person. Furthermore, the imitation unit can also apply an algorithm that imitates the conversation style of a specific person. This makes it possible to have a more realistic conversation by imitating the conversation style of a specific person. A part or all of the above-mentioned processing in the imitation unit may be performed, for example, using AI, or may be performed without using AI. For example, the imitation unit can input conversation data of a specific person to a generation AI and imitate the conversation style using the generation AI.

[0081] The conversation simulation system includes a feedback unit that provides feedback. The feedback unit provides the feedback. The feedback includes, for example, feedback based on the user's reaction, presentation of improvements to the system, and the like, but is not limited to such examples. For example, the feedback unit analyzes the user's reaction and provides appropriate feedback. The feedback unit can also present improvements to the system. As a result, the feedback is provided to improve the user's conversation experience. A part or all of the above-mentioned processing in the feedback unit may be performed, for example, using AI, or may be performed without using AI. For example, the feedback unit can input the user's reaction data to a generation AI and provide feedback using the generation AI.

[0082] The preprocessing unit can perform preprocessing of the text data (e.g., noise removal and tokenization). The preprocessing unit, for example, performs noise removal of the text data. Noise removal includes specific pattern removal and filtering techniques. For example, the preprocessing unit removes unnecessary characters and symbols from the conversation data. The preprocessing unit also tokenizes the text data. Tokenization includes tokenization by word and by sentence. For example, the preprocessing unit divides the conversation data into words. The preprocessing unit can also divide the conversation data into sentences. This improves the quality of the data by cleaning and tokenizing the text data. Some or all of the above-mentioned processing in the preprocessing unit may be performed using, for example, AI, or may be performed without using AI. For example, the preprocessing unit can input the conversation data to a generation AI and perform noise removal and tokenization using the generation AI.

[0083] The imitation unit can imitate a specific person's unique phrase or speaking style. The imitation unit, for example, extracts frequent phrases of a specific person. Extracting frequent phrases includes a technique for analyzing the characteristics of a specific phrase or speaking style. For example, the imitation unit extracts frequent phrases from the conversation data of a specific person. The imitation unit can also analyze the grammar pattern of a specific person. Analyzing the grammar pattern includes a technique for analyzing a specific grammatical structure or a pattern of phrases. For example, the imitation unit analyzes the grammar pattern from the conversation data of a specific person. This enables a more realistic conversation by imitating a specific phrase or speaking style. A part or all of the above-mentioned processing in the imitation unit may be performed, for example, using AI, or may be performed without using AI. For example, the imitation unit can input the conversation data of a specific person to a generation AI and imitate the conversation style using the generation AI.

[0084] The providing unit can estimate the user's emotion and adjust the timing of providing the conversation data based on the estimated user's emotion. The providing unit, for example, estimates the user's emotion. The estimation of emotion includes an emotion analysis algorithm and a facial expression recognition technology. For example, the providing unit analyzes the user's facial expression data and estimates the emotion. The providing unit can also analyze the user's voice data and estimate the emotion. Furthermore, the providing unit can analyze the user's biometric data and estimate the emotion. Next, the providing unit adjusts the timing of providing the conversation data based on the estimated user's emotion. For example, if the user is sad, past fun conversation data is preferentially provided. Furthermore, if the user is stressed, conversation data that can relax the user can be provided. Furthermore, if the user is excited, conversation data that attracts the user's interest can be provided. As a result, a more appropriate conversation experience is provided by adjusting the timing of providing the conversation data according to the user's emotion. A part or all of the above-mentioned processing in the providing unit may be performed using, for example, AI, or may be performed without using AI. For example, the providing unit can input the user's emotion data to the generating AI, estimate the emotion using the generating AI, and adjust the timing of providing the conversation data.

[0085] The providing unit can analyze the user's past conversation data providing history and select a providing method. The providing unit, for example, analyzes the user's past conversation data providing history. The analysis includes an analysis of past usage patterns and an analysis of feedback. For example, the providing unit preferentially provides conversation data that the user frequently accessed in the past. The providing unit can also analyze the user's past usage patterns and provide conversation data at an optimal timing. Furthermore, the providing unit can adjust the content of the conversation data to be provided based on the user's past feedback. Next, the providing unit selects an optimal providing method based on the analysis result. The providing method includes real-time provision and batch provision. For example, the providing unit provides conversation data in real time based on the user's usage patterns. Furthermore, the providing unit can perform batch provision based on the user's feedback. Thereby, the optimal providing method is selected by analyzing the past conversation data providing history. A part or all of the above-mentioned processing in the providing unit may be performed using, for example, AI, or may be performed without using AI. For example, the providing unit can input the user's past conversation data providing history to a generating AI, analyze it using the generating AI, and select an optimal providing method.

[0086] The providing unit can perform filtering based on the user's current field of interest when providing the conversation data. The providing unit, for example, identifies the user's current field of interest. Identifying the field of interest includes keyword-based filtering and topic-based filtering. For example, the providing unit identifies the field of interest based on the user's recent search history. The providing unit can also analyze the user's social media activity and identify the field of interest. Next, the providing unit filters the conversation data based on the identified field of interest. For example, conversation data related to topics in which the user is currently interested is preferentially provided. The providing unit can also filter the conversation data based on the user's field of interest. As a result, highly relevant conversation data is provided by filtering based on the user's field of interest. A part or all of the above-mentioned processing in the providing unit may be performed using, for example, AI, or may be performed without using AI. For example, the providing unit can input the user's field of interest data to a generating AI and perform filtering using the generating AI.

[0087] The providing unit can estimate the user's emotion and determine the priority of the conversation data to be provided based on the estimated user's emotion. The providing unit, for example, estimates the user's emotion. The estimation of emotion includes an emotion analysis algorithm and a facial expression recognition technology. For example, the providing unit analyzes the user's facial expression data and estimates the emotion. The providing unit can also analyze the user's voice data and estimate the emotion. Furthermore, the providing unit can analyze the user's biometric data and estimate the emotion. Next, the providing unit determines the priority of the conversation data to be provided based on the estimated user's emotion. For example, if the user is sad, past fun conversation data is preferentially provided. Furthermore, if the user is stressed, conversation data that can relax the user can be provided. Furthermore, if the user is excited, conversation data that attracts the user's interest can be provided. In this way, by determining the priority of the conversation data according to the user's emotion, more appropriate conversation data is provided. The estimation of emotion is realized using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is, for example, a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. A part or all of the above-mentioned processing in the providing unit may be performed using, for example, AI, or may be performed without using AI. For example, the providing unit can input the user's emotion data to the generating AI, estimate the emotion using the generating AI, and determine the priority order of the conversation data.

[0088] The providing unit can provide highly relevant data preferentially based on the geographical location information of the user when providing the conversation data. The providing unit, for example, acquires the geographical location information of the user. The acquisition of the geographical location information includes the use of GPS data and the use of a location information service. For example, the providing unit acquires GPS data from the user's smartphone. The providing unit can also acquire the geographical location information of the user by using a location information service. Next, the providing unit provides highly relevant conversation data based on the acquired geographical location information. For example, when the user is in a specific area, conversation data related to the area is provided. In addition, when the user is traveling, conversation data related to the travel destination can be provided. Furthermore, when the user is at home, conversation data related to the home can be provided. As a result, by providing highly relevant data based on the geographical location information, conversation data according to the user's situation is provided. A part or all of the above-mentioned processing in the providing unit may be performed, for example, using AI, or may be performed without using AI. For example, the providing unit can input the geographical location information data of the user to a generating AI and provide highly relevant conversation data using the generating AI.

[0089] The providing unit can analyze the user's social media activity and provide related data when providing the conversation data. The providing unit, for example, analyzes the user's social media activity. The analysis includes analysis of the posted content and analysis of followers. For example, the providing unit provides conversation data related to a topic recently shared by the user on social media. The providing unit can also provide related conversation data based on the user's friendships on social media. Furthermore, the providing unit can analyze the user's social media activity history and provide interesting conversation data. In this way, conversation data based on the user's interests is provided by analyzing social media activities. A part or all of the above-mentioned processing in the providing unit may be performed, for example, using AI, or may be performed without using AI. For example, the providing unit can input the user's social media activity data into a generating AI, analyze it using the generating AI, and provide related conversation data.

[0090] The analysis unit can estimate the user's emotion and adjust the analysis method based on the estimated user's emotion. The analysis unit, for example, estimates the user's emotion. The estimation of emotion includes emotion analysis algorithms and facial expression recognition technology. For example, the analysis unit analyzes the user's facial expression data and estimates the emotion. The analysis unit can also analyze the user's voice data and estimate the emotion. Furthermore, the analysis unit can analyze the user's biometric data and estimate the emotion. Next, the analysis unit adjusts the analysis method based on the estimated user's emotion. For example, if the user is sad, the analysis is performed with emphasis on emotional tone. Furthermore, if the user is stressed, the analysis can be performed with emphasis on relaxing content. Furthermore, if the user is excited, the analysis can be performed with emphasis on interesting content. In this way, a more appropriate analysis is performed by adjusting the analysis method according to the user's emotion. The estimation of emotion is realized using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is, for example, a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. A part or all of the above-mentioned processing in the analysis unit may be performed using, for example, AI, or may be performed without using AI. For example, the analysis unit can input the user's emotion data to the generation AI, estimate the emotion using the generation AI, and adjust the analysis method.

[0091] The analysis unit can adjust the level of detail of the analysis based on the importance of the conversation data during the analysis. The analysis unit, for example, evaluates the importance of the conversation data. The evaluation of the importance includes frequency analysis and analysis of the user's interest. For example, the analysis unit analyzes the frequency of the conversation data and evaluates the importance. The analysis unit can also analyze the user's interest and evaluate the importance. Next, the analysis unit adjusts the level of detail of the analysis based on the evaluated importance. For example, a detailed analysis is performed on important conversation data. Also, a simplified analysis can be performed on general conversation data. Furthermore, an analysis can be focused on conversation data in which the user is particularly interested. As a result, an efficient analysis is performed by adjusting the level of detail of the analysis based on the importance of the conversation data. A part or all of the above-mentioned processing in the analysis unit may be performed using, for example, an AI, or may be performed without using an AI. For example, the analysis unit can input the importance data of the conversation data to a generation AI and adjust the level of detail of the analysis using the generation AI.

[0092] The analysis unit can apply different analysis algorithms depending on the category of the conversation data during analysis. The analysis unit, for example, classifies the category of the conversation data. The category classification includes topic-based classification and content-based classification. For example, the analysis unit classifies the conversation data by topic. The analysis unit can also classify the conversation data by content. Next, the analysis unit applies different analysis algorithms depending on the classified category. For example, an analysis algorithm that emphasizes emotional tone can be applied to conversation data with family members. An analysis algorithm that emphasizes efficiency can also be applied to work-related conversation data. Furthermore, an analysis algorithm that emphasizes familiarity can also be applied to conversation data with friends. In this way, by applying different analysis algorithms depending on the category of the conversation data, more appropriate analysis can be performed. Part or all of the above-mentioned processing in the analysis unit may be performed using, for example, AI, or may be performed without using AI. For example, the analysis unit can input category data of the conversation data to a generation AI and apply different analysis algorithms using the generation AI.

[0093] The analysis unit can estimate the user's emotion and determine the priority of analysis based on the estimated user's emotion. The analysis unit, for example, estimates the user's emotion. The estimation of emotion includes an emotion analysis algorithm and a facial expression recognition technology. For example, the analysis unit analyzes the user's facial expression data and estimates the emotion. The analysis unit can also analyze the user's voice data and estimate the emotion. Furthermore, the analysis unit can analyze the user's biometric data and estimate the emotion. Next, the analysis unit determines the priority of analysis based on the estimated user's emotion. For example, if the user is sad, conversation data that emphasizes emotional tones can be analyzed preferentially. Furthermore, if the user is stressed, conversation data that emphasizes relaxing content can be analyzed preferentially. Furthermore, if the user is excited, conversation data that emphasizes interesting content can be analyzed preferentially. As a result, by determining the priority of analysis according to the user's emotion, more appropriate analysis is performed. The estimation of emotion is realized using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI may be, for example, a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples. Some or all of the above-mentioned processing in the analysis unit may be performed, for example, using AI, or may be performed without using AI. For example, the analysis unit may input the user's emotion data to the generation AI, estimate the emotion using the generation AI, and determine the priority of the analysis.

[0094] The analysis unit can adjust the order of analysis based on the submission time of the conversation data during analysis. The analysis unit, for example, evaluates the submission time of the conversation data. The evaluation of the submission time includes recording the submission date and time and managing the submission order. For example, the analysis unit records the submission date and time of the conversation data. The analysis unit can also manage the submission order of the conversation data. Next, the analysis unit adjusts the order of analysis based on the evaluated submission time. For example, recent conversation data is preferentially analyzed. Conversation data that was concentrated in a specific period can also be preferentially analyzed. Furthermore, conversation data from a period in which the user is particularly interested can also be preferentially analyzed. As a result, efficient analysis is performed by adjusting the order of analysis based on the submission time of the conversation data. Part or all of the above-mentioned processing in the analysis unit may be performed using, for example, AI, or may be performed without using AI. For example, the analysis unit can input the submission time data of the conversation data to the generation AI and adjust the order of analysis using the generation AI.

[0095] The analysis unit can adjust the order of analysis based on the relevance of the conversation data during analysis. The analysis unit, for example, evaluates the relevance of the conversation data. The evaluation of the relevance includes co-occurrence network analysis and topic modeling. For example, the analysis unit analyzes the co-occurrence network of the conversation data and evaluates the relevance. The analysis unit can also model the topic of the conversation data and evaluate the relevance. Next, the analysis unit adjusts the order of analysis based on the evaluated relevance. For example, highly relevant conversation data is preferentially analyzed. Also, conversation data related to a topic in which the user is particularly interested can be preferentially analyzed. Furthermore, based on the user's past usage history, highly relevant conversation data can be preferentially analyzed. As a result, efficient analysis is performed by adjusting the order of analysis based on the relevance of the conversation data. Part or all of the above-mentioned processing in the analysis unit may be performed using, for example, AI, or may be performed without using AI. For example, the analysis unit can input the relevance data of the conversation data to the generation AI and adjust the order of analysis using the generation AI.

[0096] The learning unit can estimate the user's emotion and select learning data based on the estimated user's emotion. The learning unit, for example, estimates the user's emotion. The estimation of emotion includes an emotion analysis algorithm and a facial expression recognition technology. For example, the learning unit analyzes the user's facial expression data and estimates the emotion. The learning unit can also analyze the user's voice data and estimate the emotion. Furthermore, the learning unit can analyze the user's biometric data and estimate the emotion. Next, the learning unit selects learning data based on the estimated user's emotion. For example, if the user is sad, learning data that emphasizes emotional tones is selected. Furthermore, if the user is stressed, learning data that emphasizes relaxing content can be selected. Furthermore, if the user is excited, learning data that emphasizes interesting content can be selected. In this way, more appropriate learning is performed by selecting learning data according to the user's emotion. The estimation of emotion is realized using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is, for example, a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. A part or all of the above-mentioned processing in the learning unit may be performed using, for example, an AI, or may be performed without using an AI. For example, the learning unit may input the user's emotion data to a generation AI, estimate the emotion using the generation AI, and select the learning data.

[0097] The learning unit can optimize the learning algorithm by referring to past learning data during learning. The learning unit, for example, refers to past learning data. The reference includes feedback of past learning results and extraction of effective learning patterns. For example, the learning unit selects an optimal learning algorithm based on past learning data. The learning unit can also extract effective learning patterns from past learning data and optimize the algorithm. Furthermore, the learning unit can analyze past learning data and adjust parameters of the learning algorithm. As a result, the learning algorithm is optimized by referring to the past learning data. A part or all of the above-mentioned processing in the learning unit may be performed, for example, using AI or may be performed without using AI. For example, the learning unit can input past learning data to a generation AI and optimize the learning algorithm using the generation AI.

[0098] The learning unit can apply different learning methods to each category of conversation data during learning. The learning unit, for example, classifies the categories of conversation data. Category classification includes topic-based classification and content-based classification. For example, the learning unit classifies the conversation data by topic. The learning unit can also classify the conversation data by content. Next, the learning unit applies different learning methods to each classified category. For example, a learning method that emphasizes emotional tone can be applied to conversation data with family members. A learning method that emphasizes efficiency can also be applied to work-related conversation data. Furthermore, a learning method that emphasizes familiarity can also be applied to conversation data with friends. In this way, more appropriate learning is performed by applying different learning methods to each category of conversation data. Part or all of the above-mentioned processing in the learning unit may be performed using, for example, AI, or may be performed without using AI. For example, the learning unit can input category data of the conversation data to a generation AI and apply different learning methods using the generation AI.

[0099] The learning unit can estimate the user's emotion and adjust the frequency of learning based on the estimated user's emotion. The learning unit, for example, estimates the user's emotion. The estimation of emotion includes an emotion analysis algorithm and a facial expression recognition technology. For example, the learning unit analyzes the user's facial expression data and estimates the emotion. The learning unit can also analyze the user's voice data and estimate the emotion. Furthermore, the learning unit can analyze the user's biometric data and estimate the emotion. Next, the learning unit adjusts the frequency of learning based on the estimated user's emotion. For example, when the user is sad, learning that emphasizes emotional tones can be performed frequently. Furthermore, when the user is stressed, learning that emphasizes relaxing content can be performed frequently. Furthermore, when the user is excited, learning that emphasizes interesting content can be performed frequently. In this way, more appropriate learning can be performed by adjusting the frequency of learning according to the user's emotion. The estimation of emotion is realized using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is, for example, a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. A part or all of the above-mentioned processing in the learning unit may be performed using, for example, AI, or may be performed without using AI. For example, the learning unit can input the user's emotion data to the generation AI, estimate the emotion using the generation AI, and adjust the frequency of learning.

[0100] The learning unit can weight the learning data based on the submission time of the conversation data during learning. The learning unit, for example, evaluates the submission time of the conversation data. The evaluation of the submission time includes recording the submission date and time and managing the submission order. For example, the learning unit records the submission date and time of the conversation data. The learning unit can also manage the submission order of the conversation data. Next, the learning unit weights the learning data based on the evaluated submission time. For example, a high weight is set for recent conversation data. Also, a high weight can be set for conversation data that was concentrated in a specific period. Furthermore, a high weight can be set for conversation data from a period in which the user is particularly interested. As a result, efficient learning is performed by weighting the learning data based on the submission time of the conversation data. A part or all of the above-mentioned processing in the learning unit may be performed using, for example, an AI, or may be performed without using an AI. For example, the learning unit can input the submission time data of the conversation data to a generation AI and weight the learning data using the generation AI.

[0101] The learning unit can improve the accuracy of learning by referring to literature related to the conversation data during learning. The learning unit, for example, refers to literature related to the conversation data. The reference includes the use of a literature database and the analysis of citation relationships. For example, the learning unit optimizes a learning algorithm based on the related literature. The learning unit can also extract effective learning patterns from the related literature to improve the accuracy of learning. Furthermore, the learning unit can also select learning data by referring to the related literature. As a result, the accuracy of learning is improved by referring to the literature related to the conversation data. A part or all of the above-mentioned processing in the learning unit may be performed, for example, using AI, or may be performed without using AI. For example, the learning unit can input literature data related to the conversation data to a generation AI and improve the accuracy of learning using the generation AI.

[0102] The response unit can estimate the user's emotion and adjust the way of expressing the response based on the estimated user's emotion. The response unit, for example, estimates the user's emotion. The estimation of the emotion includes an emotion analysis algorithm and a facial expression recognition technology. For example, the response unit analyzes the user's facial expression data and estimates the emotion. The response unit can also analyze the user's voice data and estimate the emotion. Furthermore, the response unit can analyze the user's biometric data and estimate the emotion. Next, the response unit adjusts the way of expressing the response based on the estimated user's emotion. For example, if the user is sad, the response unit responds in a gentle tone. Furthermore, if the user is stressed, the response unit can respond in a relaxing tone. Furthermore, if the user is excited, the response unit can respond in an interesting tone. In this way, a more appropriate response is generated by adjusting the way of expressing the response according to the user's emotion. The estimation of the emotion is realized using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is, for example, a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. A part or all of the above-mentioned processing in the response unit may be performed using, for example, AI, or may be performed without using AI. For example, the response unit can input the user's emotion data to a generation AI, estimate the emotion using the generation AI, and adjust the way the response is expressed.

[0103] The response unit can adjust the level of detail of the response based on the importance of the conversation when generating a response. The response unit, for example, evaluates the importance of the conversation. The evaluation of the importance includes frequency analysis and analysis of the user's interest. For example, the response unit analyzes the frequency of the conversation data and evaluates the importance. The response unit can also analyze the user's interest and evaluate the importance. Next, the response unit adjusts the level of detail of the response based on the evaluated importance. For example, a detailed response is generated for an important conversation. Also, a simple response can be generated for a general conversation. Furthermore, a detailed response can be generated with a focus on a conversation in which the user is particularly interested. As a result, an efficient response is generated by adjusting the level of detail of the response based on the importance of the conversation. A part or all of the above-mentioned processing in the response unit may be performed, for example, using AI, or may be performed without using AI. For example, the response unit can input importance data of the conversation data to the generation AI and adjust the level of detail of the response using the generation AI.

[0104] The response unit can apply different response algorithms according to the category of the conversation when generating a response. The response unit, for example, classifies the category of the conversation data. The category classification includes topic-based classification and content-based classification. For example, the response unit classifies the conversation data by topic. The response unit can also classify the conversation data by content. Next, the response unit applies different response algorithms according to the classified category. For example, a response algorithm that emphasizes emotional tone can be applied to a conversation with family members. A response algorithm that emphasizes efficiency can also be applied to a work-related conversation. Furthermore, a response algorithm that emphasizes familiarity can also be applied to a conversation with friends. In this way, a more appropriate response is generated by applying different response algorithms according to the category of the conversation. Part or all of the above-mentioned processing in the response unit may be performed, for example, using AI, or may be performed without using AI. For example, the response unit can input category data of the conversation data to a generation AI and apply different response algorithms using the generation AI.

[0105] The response unit can estimate the user's emotion and adjust the length of the response based on the estimated user's emotion. The response unit, for example, estimates the user's emotion. The estimation of emotion includes an emotion analysis algorithm and a facial expression recognition technology. For example, the response unit analyzes the user's facial expression data and estimates the emotion. The response unit can also analyze the user's voice data and estimate the emotion. Furthermore, the response unit can analyze the user's biometric data and estimate the emotion. Next, the response unit adjusts the length of the response based on the estimated user's emotion. For example, if the user is sad, a short and gentle response is generated. Also, if the user is stressed, a longer response that relaxes the user can be generated. Furthermore, if the user is excited, a longer response that attracts interest can be generated. In this way, a more appropriate response is generated by adjusting the length of the response according to the user's emotion. The estimation of emotion is realized using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is, for example, a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. A part or all of the above-mentioned processing in the response unit may be performed using, for example, AI, or may be performed without using AI. For example, the response unit can input the user's emotion data to a generation AI, estimate the emotion using the generation AI, and adjust the length of the response.

[0106] The response unit can determine the priority of responses based on the submission time of the conversation when generating a response. The response unit, for example, evaluates the submission time of the conversation data. The evaluation of the submission time includes recording the submission date and time and managing the submission order. For example, the response unit records the submission date and time of the conversation data. The response unit can also manage the submission order of the conversation data. Next, the response unit determines the priority of responses based on the evaluated submission time. For example, responses are generated preferentially for recent conversations. Responses can also be generated preferentially for conversations that were concentrated in a specific period. Furthermore, responses can also be generated preferentially for conversations in a period in which the user is particularly interested. In this way, an efficient response is generated by determining the priority of responses based on the submission time of the conversation. A part or all of the above-mentioned processing in the response unit may be performed, for example, using AI, or may be performed without using AI. For example, the response unit can input submission time data of the conversation data to the generation AI and determine the priority of responses using the generation AI.

[0107] The response unit can adjust the order of responses based on the relevance of the conversation when generating a response. The response unit, for example, evaluates the relevance of the conversation data. The evaluation of the relevance includes co-occurrence network analysis and topic modeling. For example, the response unit analyzes the co-occurrence network of the conversation data and evaluates the relevance. The response unit can also model the topic of the conversation data and evaluate the relevance. Next, the response unit adjusts the order of responses based on the evaluated relevance. For example, a response is preferentially generated for a conversation that is highly relevant. A response can also be preferentially generated for a conversation related to a topic in which the user is particularly interested. Furthermore, a response can also be preferentially generated for a conversation that is highly relevant based on the user's past usage history. As a result, an efficient response is generated by adjusting the order of responses based on the relevance of the conversation. A part or all of the above-mentioned processing in the response unit may be performed using, for example, AI, or may be performed without using AI. For example, the response unit can input the relevance data of the conversation data to the generation AI and adjust the order of responses using the generation AI.

[0108] The preprocessing unit can estimate the user's emotion and adjust the preprocessing method based on the estimated user's emotion. The preprocessing unit, for example, estimates the user's emotion. The emotion estimation includes an emotion analysis algorithm and a facial expression recognition technology. For example, the preprocessing unit analyzes the user's facial expression data and estimates the emotion. The preprocessing unit can also analyze the user's voice data and estimate the emotion. Furthermore, the preprocessing unit can analyze the user's biometric data and estimate the emotion. Next, the preprocessing unit adjusts the preprocessing method based on the estimated user's emotion. For example, if the user is sad, the preprocessing unit can prioritize emotional tones and perform preprocessing. If the user is stressed, the preprocessing unit can prioritize relaxing content and perform preprocessing. Furthermore, if the user is excited, the preprocessing unit can prioritize interesting content and perform preprocessing. In this way, more appropriate preprocessing is performed by adjusting the preprocessing method according to the user's emotion. The emotion estimation is realized using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is, for example, a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. Some or all of the above-mentioned processing in the preprocessing unit may be performed using AI or without AI. For example, the preprocessing unit may input the user's emotion data to the generation AI, estimate the emotion using the generation AI, and adjust the preprocessing method.

[0109] The preprocessing unit can adjust the level of detail of the preprocessing based on the importance of the conversation data during preprocessing. The preprocessing unit, for example, evaluates the importance of the conversation data. The evaluation of the importance includes frequency analysis and analysis of the user's interest level. For example, the preprocessing unit analyzes the frequency of the conversation data and evaluates the importance. The preprocessing unit can also analyze the user's interest level and evaluate the importance. Next, the preprocessing unit adjusts the level of detail of the preprocessing based on the evaluated importance. For example, detailed preprocessing is performed for important conversation data. Also, simple preprocessing can be performed for general conversation data. Furthermore, preprocessing can be focused on conversation data in which the user is particularly interested. As a result, efficient preprocessing is performed by adjusting the level of detail of the preprocessing based on the importance of the conversation data. Part or all of the above-mentioned processing in the preprocessing unit may be performed using, for example, AI, or may be performed without using AI. For example, the preprocessing unit can input importance data of the conversation data to a generation AI and adjust the level of detail of the preprocessing using the generation AI.

[0110] The preprocessing unit can estimate the user's emotion and determine the priority of preprocessing based on the estimated user's emotion. The preprocessing unit, for example, estimates the user's emotion. The estimation of emotion includes an emotion analysis algorithm and a facial expression recognition technology. For example, the preprocessing unit analyzes the user's facial expression data and estimates the emotion. The preprocessing unit can also analyze the user's voice data and estimate the emotion. Furthermore, the preprocessing unit can analyze the user's biometric data and estimate the emotion. Next, the preprocessing unit determines the priority of preprocessing based on the estimated user's emotion. For example, if the user is sad, conversation data that emphasizes emotional tones can be preprocessed preferentially. Furthermore, if the user is stressed, conversation data that emphasizes relaxing content can be preprocessed preferentially. Furthermore, if the user is excited, conversation data that emphasizes interesting content can be preprocessed preferentially. As a result, more appropriate preprocessing is performed by determining the priority of preprocessing according to the user's emotion. The emotion estimation is realized using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI may be, for example, a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples. Some or all of the above-mentioned processing in the preprocessing unit may be performed, for example, using AI, or may be performed without using AI. For example, the preprocessing unit may input user emotion data to the generation AI, estimate the emotion using the generation AI, and determine the priority of preprocessing.

[0111] The preprocessing unit can adjust the order of preprocessing based on the submission time of the conversation data during preprocessing. The preprocessing unit, for example, evaluates the submission time of the conversation data. The evaluation of the submission time includes recording the submission date and time and managing the submission order. For example, the preprocessing unit records the submission date and time of the conversation data. The preprocessing unit can also manage the submission order of the conversation data. Next, the preprocessing unit adjusts the order of preprocessing based on the evaluated submission time. For example, recent conversation data is preprocessed preferentially. Conversation data that was concentrated in a specific period can also be preprocessed preferentially. Furthermore, conversation data from a period in which the user is particularly interested can also be preprocessed preferentially. As a result, efficient preprocessing is performed by adjusting the order of preprocessing based on the submission time of the conversation data. Part or all of the above-mentioned processing in the preprocessing unit may be performed using, for example, AI, or may be performed without using AI. For example, the preprocessing unit can input submission time data of the conversation data to a generation AI and adjust the order of preprocessing using the generation AI.

[0112] The mimicry unit can estimate the user's emotion and adjust the mimicry method based on the estimated user's emotion. The mimicry unit, for example, estimates the user's emotion. The estimation of emotion includes emotion analysis algorithms and facial expression recognition technology. For example, the mimicry unit analyzes the user's facial expression data and estimates the emotion. The mimicry unit can also analyze the user's voice data and estimate the emotion. Furthermore, the mimicry unit can analyze the user's biometric data and estimate the emotion. Next, the mimicry unit adjusts the mimicry method based on the estimated user's emotion. For example, when the user is sad, the mimicry is performed with emphasis on emotional tones. When the user is stressed, the mimicry can be performed with emphasis on relaxing content. Furthermore, when the user is excited, the mimicry can be performed with emphasis on interesting content. In this way, by adjusting the mimicry method according to the user's emotion, more appropriate mimicry is performed. The estimation of emotion is realized using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is, for example, a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. A part or all of the above-mentioned processing in the mimicry unit may be performed using, for example, AI, or may be performed without using AI. For example, the mimicry unit can input the user's emotion data to the generation AI, estimate the emotion using the generation AI, and adjust the mimicry method.

[0113] The imitation unit can adjust the level of detail of imitation based on the importance of the conversation data during imitation. The imitation unit, for example, evaluates the importance of the conversation data. The evaluation of the importance includes frequency analysis and analysis of the user's interest. For example, the imitation unit analyzes the frequency of the conversation data and evaluates the importance. The imitation unit can also analyze the user's interest and evaluate the importance. Next, the imitation unit adjusts the level of detail of imitation based on the evaluated importance. For example, detailed imitation is performed for important conversation data. Also, simple imitation can be performed for general conversation data. Furthermore, imitation can be focused on conversation data in which the user is particularly interested. As a result, efficient imitation is performed by adjusting the level of detail of imitation based on the importance of the conversation data. Part or all of the above-mentioned processing in the imitation unit may be performed using, for example, AI, or may be performed without using AI. For example, the imitation unit can input the importance data of the conversation data to the generation AI and adjust the level of detail of imitation using the generation AI.

[0114] The imitation unit can apply different imitation algorithms depending on the category of the conversation data during imitation. The imitation unit, for example, classifies the category of the conversation data. The category classification includes topic-based classification and content-based classification. For example, the imitation unit classifies the conversation data by topic. The imitation unit can also classify the conversation data by content. Next, the imitation unit applies different imitation algorithms depending on the classified category. For example, an imitation algorithm that emphasizes emotional tone can be applied to conversation data with family members. An imitation algorithm that emphasizes efficiency can also be applied to work-related conversation data. Furthermore, an imitation algorithm that emphasizes familiarity can also be applied to conversation data with friends. In this way, by applying different imitation algorithms depending on the category of the conversation data, more appropriate imitation is performed. Part or all of the above-mentioned processing in the imitation unit may be performed using, for example, AI, or may be performed without using AI. For example, the imitation unit can input category data of the conversation data to a generation AI and apply different imitation algorithms using the generation AI.

[0115] The mimicry unit can estimate the user's emotion and determine the priority of mimicry based on the estimated user's emotion. The mimicry unit, for example, estimates the user's emotion. The estimation of emotion includes an emotion analysis algorithm and a facial expression recognition technology. For example, the mimicry unit analyzes the user's facial expression data and estimates the emotion. The mimicry unit can also analyze the user's voice data and estimate the emotion. Furthermore, the mimicry unit can analyze the user's biometric data and estimate the emotion. Next, the mimicry unit determines the priority of mimicry based on the estimated user's emotion. For example, if the user is sad, conversation data that emphasizes emotional tones can be preferentially imitated. Furthermore, if the user is stressed, conversation data that emphasizes relaxing content can be preferentially imitated. Furthermore, if the user is excited, conversation data that emphasizes interesting content can be preferentially imitated. As a result, by determining the priority of mimicry according to the user's emotion, more appropriate mimicry can be performed. The estimation of emotion is realized using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI may be, for example, a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples. Some or all of the above-mentioned processing in the mimicry unit may be performed, for example, using AI, or may be performed without using AI. For example, the mimicry unit may input the user's emotion data to the generation AI, estimate the emotion using the generation AI, and determine the priority of the mimicry.

[0116] The imitation unit can adjust the order of imitation based on the submission time of the conversation data during imitation. The imitation unit, for example, evaluates the submission time of the conversation data. The evaluation of the submission time includes recording the submission date and time and managing the submission order. For example, the imitation unit records the submission date and time of the conversation data. The imitation unit can also manage the submission order of the conversation data. Next, the imitation unit adjusts the order of imitation based on the evaluated submission time. For example, recent conversation data is preferentially imitated. Also, conversation data that was concentrated in a specific period can be preferentially imitated. Furthermore, conversation data from a period in which the user is particularly interested can be preferentially imitated. As a result, efficient imitation is performed by adjusting the order of imitation based on the submission time of the conversation data. Part or all of the above-mentioned processing in the imitation unit may be performed using, for example, AI, or may be performed without using AI. For example, the imitation unit can input the submission time data of the conversation data to the generation AI and adjust the order of imitation using the generation AI.

[0117] The imitation unit can adjust the order of imitation based on the relevance of the conversation data during imitation. The imitation unit, for example, evaluates the relevance of the conversation data. The evaluation of the relevance includes co-occurrence network analysis and topic modeling. For example, the imitation unit analyzes the co-occurrence network of the conversation data and evaluates the relevance. The imitation unit can also model the topic of the conversation data and evaluate the relevance. Next, the imitation unit adjusts the order of imitation based on the evaluated relevance. For example, highly relevant conversation data is preferentially imitated. Also, conversation data related to a topic in which the user is particularly interested can be preferentially imitated. Furthermore, based on the user's past usage history, highly relevant conversation data can be preferentially imitated. As a result, efficient imitation is performed by adjusting the order of imitation based on the relevance of the conversation data. A part or all of the above-mentioned processing in the imitation unit may be performed using, for example, AI, or may be performed without using AI. For example, the imitation unit can input the relevance data of the conversation data to the generation AI and adjust the order of imitation using the generation AI.

[0118] The feedback unit can estimate the user's emotion and adjust the feedback method based on the estimated user's emotion. The feedback unit, for example, estimates the user's emotion. The emotion estimation includes an emotion analysis algorithm and a facial expression recognition technology. For example, the feedback unit analyzes the user's facial expression data and estimates the emotion. The feedback unit can also analyze the user's voice data and estimate the emotion. Furthermore, the feedback unit can analyze the user's biometric data and estimate the emotion. Next, the feedback unit adjusts the feedback method based on the estimated user's emotion. For example, if the user is sad, the feedback is provided in a gentle tone. If the user is stressed, the feedback can be provided in a relaxing tone. Furthermore, if the user is excited, the feedback can be provided in an interesting tone. In this way, by adjusting the feedback method according to the user's emotion, more appropriate feedback is provided. The emotion estimation is realized using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is, for example, a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. A part or all of the above-mentioned processing in the feedback unit may be performed using, for example, AI, or may be performed without using AI. For example, the feedback unit can input the user's emotion data to the generation AI, estimate the emotion using the generation AI, and adjust the feedback method.

[0119] The feedback unit can adjust the level of detail of the feedback based on the importance of the conversation data at the time of feedback. The feedback unit, for example, evaluates the importance of the conversation data. The evaluation of the importance includes frequency analysis and analysis of the user's interest. For example, the feedback unit analyzes the frequency of the conversation data and evaluates the importance. The feedback unit can also analyze the user's interest and evaluate the importance. Next, the feedback unit adjusts the level of detail of the feedback based on the evaluated importance. For example, detailed feedback is provided for important conversation data. Also, brief feedback can be provided for general conversation data. Furthermore, detailed feedback can be provided with a focus on conversation data in which the user is particularly interested. As a result, efficient feedback is provided by adjusting the level of detail of the feedback based on the importance of the conversation data. A part or all of the above-mentioned processing in the feedback unit may be performed, for example, using AI, or may be performed without using AI. For example, the feedback unit can input importance data of the conversation data to a generation AI and adjust the level of detail of the feedback using the generation AI.

[0120] The feedback unit can estimate the user's emotion and determine the priority of feedback based on the estimated user's emotion. The feedback unit, for example, estimates the user's emotion. The estimation of emotion includes an emotion analysis algorithm and a facial expression recognition technology. For example, the feedback unit analyzes the user's facial expression data and estimates the emotion. The feedback unit can also analyze the user's voice data and estimate the emotion. Furthermore, the feedback unit can analyze the user's biometric data and estimate the emotion. Next, the feedback unit determines the priority of feedback based on the estimated user's emotion. For example, when the user is sad, conversation data that emphasizes emotional tones can be preferentially fed back. Furthermore, when the user is stressed, conversation data that emphasizes relaxing content can be preferentially fed back. Furthermore, when the user is excited, conversation data that emphasizes interesting content can be preferentially fed back. In this way, by determining the priority of feedback according to the user's emotion, more appropriate feedback can be provided. The estimation of emotion is realized using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI may be, for example, a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples. Some or all of the above-mentioned processing in the feedback unit may be performed, for example, using AI, or may be performed without using AI. For example, the feedback unit may input the user's emotion data to the generation AI, estimate the emotion using the generation AI, and determine the priority of the feedback.

[0121] The feedback unit can adjust the order of feedback based on the submission time of the conversation data at the time of feedback. The feedback unit, for example, evaluates the submission time of the conversation data. The evaluation of the submission time includes recording the submission date and time and managing the submission order. For example, the feedback unit records the submission date and time of the conversation data. The feedback unit can also manage the submission order of the conversation data. Next, the feedback unit adjusts the order of feedback based on the evaluated submission time. For example, recent conversation data is preferentially fed back. Also, conversation data that was concentrated in a specific period can be preferentially fed back. Furthermore, conversation data from a period in which the user is particularly interested can be preferentially fed back. As a result, efficient feedback is provided by adjusting the order of feedback based on the submission time of the conversation data. A part or all of the above-mentioned processing in the feedback unit may be performed, for example, using AI, or may be performed without using AI. For example, the feedback unit can input the submission time data of the conversation data to a generation AI and adjust the order of feedback using the generation AI.

[0122] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0123] The conversation simulation system can estimate the user's preferences and interests based on the user's past conversation data, and customize the conversation content based on the estimated preferences and interests. For example, it can identify hobbies and interests that the user has frequently talked about in the past, and develop the conversation based on that information. In addition, if the user shows a strong interest in a specific topic, it can provide information related to that topic with priority. Furthermore, it can analyze the user's tendency to prefer specific phrases and expressions from the user's past conversation data, and adopt a conversation style that matches that tendency. This provides the user with a more familiar and interesting conversation experience.

[0124] The conversation simulation system can monitor the user's current activity status in real time and provide conversation content according to the status. For example, when the user is exercising, advice and words of encouragement related to the exercise can be provided. When the user is working, work-related information and light topics for refreshing can be provided. Furthermore, when the user is relaxing, relaxing topics and information about hobbies can be provided. This provides appropriate conversation content according to the user's activity status, realizing a more personalized experience.

[0125] The conversation simulation system can estimate the user's stress level based on the user's past conversation data, and adjust the conversation content based on the estimated stress level. For example, if the user shows a high stress level, it can provide relaxing topics and advice for relieving stress. Also, if the user shows a low stress level, it can provide interesting topics and new information. Furthermore, it can adjust the tone and tempo of the conversation according to the user's stress level. This provides appropriate conversation content according to the user's stress level, improving the user's comfort.

[0126] The conversation simulation system can estimate a user's learning style based on the user's past conversation data, and provide conversation content based on the estimated learning style. For example, if the user has a visual learning style, explanations using diagrams and graphs can be provided. Also, if the user has an auditory learning style, emphasis can be placed on audio explanations. Furthermore, if the user has an experiential learning style, topics that encourage learning through actual experience can be provided. This provides appropriate conversation content according to the user's learning style, improving learning effectiveness.

[0127] The conversation simulation system can track changes in a user's emotions based on the user's past conversation data and adjust the content of the conversation according to those changes. For example, if a user has recently experienced a sad event, it can provide kind words that empathize with those emotions. If the user is feeling happy, it can also provide positive topics that share that joy. Furthermore, it can adjust the tone and tempo of the conversation according to changes in the user's emotions. This provides appropriate conversation content according to the user's emotions, strengthening the user's emotional connection.

[0128] The conversation simulation system can estimate the user's health condition based on the user's past conversation data, and provide conversation content based on the estimated health condition. For example, if the user has a health problem, advice and support information for that problem can be provided. Also, if the user is living a healthy lifestyle, information and motivation for maintaining that lifestyle can be provided. Furthermore, the tone and tempo of the conversation can be adjusted according to the user's health condition. In this way, appropriate conversation content according to the user's health condition can be provided, supporting the user's health management.

[0129] The conversation simulation system can estimate the user's goals and dreams based on the user's past conversation data, and provide conversation content based on the estimated goals and dreams. For example, if the user has a specific career goal, advice and information toward that goal can be provided. Also, if the user is pursuing a specific hobby or interest, information related to the hobby or interest can be provided. Furthermore, the tone and tempo of the conversation can be adjusted according to the user's goals and dreams. This provides appropriate conversation content according to the user's goals and dreams, improving the user's motivation.

[0130] The conversation simulation system can estimate the sociability of a user based on the user's past conversation data, and provide conversation content based on the estimated sociability. For example, if the user has a sociable personality, positive topics and topics that encourage meeting new people can be provided. Also, if the user has an introverted personality, relaxing topics and information about hobbies that can be enjoyed alone can be provided. Furthermore, the tone and tempo of the conversation can be adjusted according to the user's sociability. This provides appropriate conversation content according to the user's sociability, improving the user's communication experience.

[0131] The conversation simulation system can analyze the user's emotional patterns based on the user's past conversation data, and provide conversation content based on those patterns. For example, if the user is prone to feeling stress at a certain time of day, relaxing topics can be provided at that time of day. Also, if the user is prone to feeling joy in a certain situation, topics related to that situation can be provided. Furthermore, the tone and tempo of the conversation can be adjusted according to the user's emotional patterns. This provides appropriate conversation content according to the user's emotional patterns, strengthening the user's emotional connection.

[0132] The conversation simulation system can estimate the cultural background of a user based on the user's past conversation data, and provide conversation content based on the estimated cultural background. For example, if the user has a specific culture or tradition, topics related to that culture or tradition can be provided. Also, if the user has a multicultural background, information and topics related to different cultures can be provided. Furthermore, the tone and tempo of the conversation can be adjusted according to the user's cultural background. This provides appropriate conversation content according to the user's cultural background, strengthening the user's cultural connection.

[0133] The process flow of the second embodiment will be briefly described below.

[0134] Step 1: The providing unit provides past conversation data from a user on a messenger app. The conversation data includes text data, audio data, video data, and the like. For example, the user can provide past conversation data from a messenger app in text format. Audio data and video data can also be provided. Step 2: The analysis unit analyzes the conversation data provided by the provision unit. The analysis is performed using methods such as natural language processing, sentiment analysis, and topic modeling. For example, the conversation data is analyzed using natural language processing technology to extract the conversation style and language of a particular person. In addition, the emotional tone of the conversation data can be analyzed using sentiment analysis technology. Furthermore, the topic of the conversation data can be analyzed using topic modeling technology. Step 3: The learning unit learns the conversation style based on the data analyzed by the analysis unit. The learning is performed by a method such as supervised learning, unsupervised learning, or reinforcement learning. For example, supervised learning is used to learn the conversation style of a specific person. Also, unsupervised learning can be used to learn patterns in conversation data. Furthermore, reinforcement learning can be used to optimize the conversation style. Step 4: The response unit generates a response based on the conversation style learned by the learning unit. The response is generated by a method such as template-based response generation or response generation using a generative model. For example, a response based on a conversation style of a specific person can be generated using template-based response generation. A response based on a conversation style of a specific person can also be generated using a generative model.

[0135] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires a voice indicating a user input for the result of the specific processing. The control unit 46A transmits the voice data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0136] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search<URL: https: / / openai.com / blog / chatgpt> ) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image (e.g., still image data or video data) are input. The data generation model 58 infers the input inference data according to the instruction indicated by the prompt, and outputs the inference result in one or more data formats of voice data, text data, and image data. The data generation model 58 includes, for example, a text generation AI, an image generation AI, and a multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing while using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction, and in this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. In the data processing device 12 and the like, the data generation model 58 includes a plurality of types, and the data generation model 58 includes an AI other than the generation AI. The AI ​​other than the generation AI is, for example, a linear regression, a logistic regression, a decision tree, a random forest, a support vector machine (SVM), a k-means clustering, a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network (GAN), or a naive Bayes, and can perform various processes, but is not limited to such examples. The AI ​​may also be an AI agent. In addition, when the processes of each of the above-mentioned parts are performed by an AI, the processes are performed in part or in whole by the AI, but are not limited to such examples.In addition, processing performed by AI including the generation AI may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by AI including the generation AI.

[0137] Furthermore, the above-described processing by the data processing system 10 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart device 14, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart device 14. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the smart device 14 or an external device, and the smart device 14 acquires or collects information required for processing from the data processing device 12 or an external device.

[0138] Each of the multiple elements including the above-mentioned providing unit, analyzing unit, learning unit, response unit, preprocessing unit, imitation unit, and feedback unit is realized, for example, by at least one of the smart device 14 and the data processing device 12. For example, the providing unit is realized by the control unit 46A of the smart device 14, and a user provides past conversation data in a messenger app. The analyzing unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and analyzes the provided conversation data. The learning unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and learns a conversation style based on the analyzed data. The responding unit is realized, for example, by the control unit 46A of the smart device 14, and generates a response based on the learned conversation style. The preprocessing unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and performs preprocessing of the conversation data. The imitation unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and imitates the conversation style of a specific person. The feedback unit is realized, for example, by the control unit 46A of the smart device 14, and provides feedback based on the user's reaction. The correspondence between each unit and the device or control unit is not limited to the above example, and various changes are possible.

[0139] [Second embodiment] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0140] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0141] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0142] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0143] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs the voice according to instructions from the processor 46.

[0144] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the user's surroundings (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of an average healthy person).

[0145] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is performed in a secure state.

[0146] Fig. 4 shows an example of main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0147] The processor 28 reads the specific processing program 56 from the storage 32, and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0148] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the emotion of the user using the emotion identification model 59 and perform identification processing using the user's emotion. In the emotion estimation function (emotion identification function) using the emotion identification model 59, various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, are performed, but are not limited to such examples. In addition, the estimation and prediction of the emotion also include, for example, analysis of the emotion.

[0149] In the smart glasses 214, the specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific processing program 60 executed on the RAM 48. Note that the smart glasses 214 have a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the specific processing unit 290 using these models.

[0150] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains a processing result (such as a prediction result) using the data generation model 58 by communicating with the server device having the data generation model 58. In addition, the data processing device 12 may be a server device, or may be a terminal device (such as a mobile phone, a robot, or a home appliance) owned by a user.

[0151] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires a voice indicating a user input for the result of the specific processing. The control unit 46A transmits the voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0152] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image (e.g., still image data or video data) is input. The data generation model 58 infers the input inference data according to the instruction indicated by the prompt, and outputs the inference result in one or more data formats of voice data, text data, and image data. The data generation model 58 includes, for example, a text generation AI, an image generation AI, and a multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing while using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction, and in this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. In the data processing device 12 and the like, the data generation model 58 includes a plurality of types, and the data generation model 58 includes an AI other than the generation AI. The AI ​​other than the generation AI is, for example, a linear regression, a logistic regression, a decision tree, a random forest, a support vector machine (SVM), a k-means clustering, a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network (GAN), or a naive Bayes, and can perform various processes, but is not limited to such examples. The AI ​​may be an AI agent. In addition, when the processes of each part described above are performed by an AI, the processes are performed in part or in whole by the AI, but are not limited to such examples. In addition, the processes performed by the AI ​​including the generation AI may be replaced with rule-based processes, and the rule-based processes may be replaced with processes performed by the AI ​​including the generation AI.

[0153] The data processing system 210 according to the second embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 210 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart glasses 214, but may be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart glasses 214. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the smart glasses 214 or an external device, and the smart glasses 214 acquires or collects information required for processing from the data processing device 12 or an external device.

[0154] Each of the multiple elements including the above-mentioned providing unit, analyzing unit, learning unit, responding unit, preprocessing unit, imitation unit, and feedback unit is realized, for example, by at least one of the smart glasses 214 and the data processing device 12. For example, the providing unit is realized by the control unit 46A of the smart glasses 214, and a user provides past conversation data in a messenger app. The analyzing unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and analyzes the provided conversation data. The learning unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and learns a conversation style based on the analyzed data. The responding unit is realized, for example, by the control unit 46A of the smart glasses 214, and generates a response based on the learned conversation style. The preprocessing unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and performs preprocessing of the conversation data. The imitation unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and imitates the conversation style of a specific person. The feedback unit is realized, for example, by the control unit 46A of the smart glasses 214, and provides feedback based on the user's reaction. The correspondence between each unit and the device or control unit is not limited to the above example, and various changes are possible.

[0155] [Third embodiment] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0156] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0157] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0158] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0159] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs the voice according to instructions from the processor 46.

[0160] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the user's surroundings (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of an average healthy person).

[0161] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is performed in a secure state.

[0162] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0163] The processor 28 reads the specific processing program 56 from the storage 32, and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0164] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the emotion of the user using the emotion identification model 59 and perform identification processing using the user's emotion. In the emotion estimation function (emotion identification function) using the emotion identification model 59, various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, are performed, but are not limited to such examples. In addition, the estimation and prediction of the emotion also include, for example, analysis of the emotion.

[0165] In the headset type terminal 314, the specific processing is performed by the processor 46. The storage 50 stores a specific program 60. The processor 46 reads the specific program 60 from the storage 50, and executes the read specific program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific program 60 executed on the RAM 48. Note that the headset type terminal 314 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the identification processing unit 290 using these models.

[0166] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains a processing result (such as a prediction result) using the data generation model 58 by communicating with the server device having the data generation model 58. In addition, the data processing device 12 may be a server device, or may be a terminal device (such as a mobile phone, a robot, or a home appliance) owned by a user.

[0167] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input for the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0168] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image (e.g., still image data or video data) is input. The data generation model 58 infers the input inference data according to the instruction indicated by the prompt, and outputs the inference result in one or more data formats of voice data, text data, and image data. The data generation model 58 includes, for example, a text generation AI, an image generation AI, and a multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing while using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction, and in this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. In the data processing device 12 and the like, the data generation model 58 includes a plurality of types, and the data generation model 58 includes an AI other than the generation AI. The AI ​​other than the generation AI is, for example, a linear regression, a logistic regression, a decision tree, a random forest, a support vector machine (SVM), a k-means clustering, a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network (GAN), or a naive Bayes, and can perform various processes, but is not limited to such examples. The AI ​​may be an AI agent. In addition, when the processes of each part described above are performed by an AI, the processes are performed in part or in whole by the AI, but are not limited to such examples. In addition, the processes performed by the AI ​​including the generation AI may be replaced with rule-based processes, and the rule-based processes may be replaced with processes performed by the AI ​​including the generation AI.

[0169] The data processing system 310 according to the third embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 310 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the headset type terminal 314, but may be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the headset type terminal 314. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the headset type terminal 314 or an external device, and the headset type terminal 314 acquires or collects information required for processing from the data processing device 12 or an external device.

[0170] Each of the multiple elements including the above-mentioned providing unit, analyzing unit, learning unit, responding unit, preprocessing unit, imitation unit, and feedback unit is realized, for example, by at least one of the headset type terminal 314 and the data processing device 12. For example, the providing unit is realized by the control unit 46A of the headset type terminal 314, and a user provides past conversation data in a messenger app. The analyzing unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and analyzes the provided conversation data. The learning unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and learns a conversation style based on the analyzed data. The responding unit is realized, for example, by the control unit 46A of the headset type terminal 314, and generates a response based on the learned conversation style. The preprocessing unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and performs preprocessing of the conversation data. The imitation unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and imitates the conversation style of a specific person. The feedback unit is realized, for example, by the control unit 46A of the headset type terminal 314, and provides feedback based on the user's reaction. The correspondence between each unit and the device or control unit is not limited to the above example, and various modifications are possible.

[0171] [Fourth embodiment] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0172] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0173] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0174] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[0175] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs the voice according to instructions from the processor 46.

[0176] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a CMOS image sensor or a CCD image sensor, and captures images of the user's surroundings (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of a typical healthy person).

[0177] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is performed in a secure state.

[0178] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and legs. The posture and behavior of the robot 414 are controlled by controlling the motors of the arms, hands, and legs. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[0179] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0180] The processor 28 reads the specific processing program 56 from the storage 32, and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0181] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the emotion of the user using the emotion identification model 59 and perform identification processing using the user's emotion. In the emotion estimation function (emotion identification function) using the emotion identification model 59, various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, are performed, but are not limited to such examples. In addition, the estimation and prediction of the emotion also include, for example, analysis of the emotion.

[0182] In the robot 414, the specific processing is performed by the processor 46. The storage 50 stores a specific program 60. The processor 46 reads out the specific program 60 from the storage 50, and executes the specific program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific program 60 executed on the RAM 48. The robot 414 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the specific processing unit 290 using these models.

[0183] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains a processing result (such as a prediction result) using the data generation model 58 by communicating with the server device having the data generation model 58. In addition, the data processing device 12 may be a server device, or may be a terminal device (such as a mobile phone, a robot, or a home appliance) owned by a user.

[0184] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires a voice indicating a user input for the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0185] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image (e.g., still image data or video data) is input. The data generation model 58 infers the input inference data according to the instruction indicated by the prompt, and outputs the inference result in one or more data formats of voice data, text data, and image data. The data generation model 58 includes, for example, a text generation AI, an image generation AI, and a multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing while using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction, and in this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. In the data processing device 12 and the like, the data generation model 58 includes a plurality of types, and the data generation model 58 includes an AI other than the generation AI. The AI ​​other than the generation AI is, for example, a linear regression, a logistic regression, a decision tree, a random forest, a support vector machine (SVM), a k-means clustering, a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network (GAN), or a naive Bayes, and can perform various processes, but is not limited to such examples. The AI ​​may be an AI agent. In addition, when the processes of each part described above are performed by an AI, the processes are performed in part or in whole by the AI, but are not limited to such examples. In addition, the processes performed by the AI ​​including the generation AI may be replaced with rule-based processes, and the rule-based processes may be replaced with processes performed by the AI ​​including the generation AI.

[0186] The data processing system 410 according to the fourth embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 410 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the robot 414, but may be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the robot 414. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the robot 414 or an external device, etc., and the robot 414 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0187] Each of the multiple elements including the above-mentioned providing unit, analyzing unit, learning unit, responding unit, preprocessing unit, imitation unit, and feedback unit is realized, for example, by at least one of the robot 414 and the data processing device 12. For example, the providing unit is realized by the control unit 46A of the robot 414, and a user provides past conversation data in a messenger app. The analyzing unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and analyzes the provided conversation data. The learning unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and learns a conversation style based on the analyzed data. The responding unit is realized, for example, by the control unit 46A of the robot 414, and generates a response based on the learned conversation style. The preprocessing unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and performs preprocessing of the conversation data. The imitation unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and imitates the conversation style of a specific person. The feedback unit is realized, for example, by the control unit 46A of the robot 414, and provides feedback based on the user's reaction. The correspondence between each unit and the device or control unit is not limited to the above example, and various modifications are possible.

[0188] The emotion identification model 59 as an emotion engine may determine the emotion of the user according to a specific mapping. Specifically, the emotion identification model 59 may determine the emotion of the user according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the emotion of the robot, and the identification processing unit 290 may perform identification processing using the emotion of the robot.

[0189] FIG. 9 is a diagram showing an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive emotions are arranged. The more outside the concentric circles, the more emotions that represent states and actions that arise from a state of mind are arranged. Emotions are a concept that includes emotions and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions that occur in the brain are arranged. On the right side of the concentric circles, emotions that are generally induced by situational judgment are arranged. On the upper and lower sides of the concentric circles, emotions that are generally generated from reactions that occur in the brain and are induced by situational judgment are arranged. In addition, on the upper side of the concentric circles, emotions of "pleasure" are arranged, and on the lower side, emotions of "discomfort" are arranged. In this way, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[0190] These emotions are distributed in the 3 o'clock direction of emotion map 400 and usually fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[0191] The inside of emotion map 400 represents what is going on inside one's mind, and the outside of emotion map 400 represents behavior, so the further out you go on emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[0192] Here, human emotions are based on various balances such as posture and blood sugar level, and when these balances are far from the ideal, it indicates an unpleasant state, and when they are close to the ideal, it indicates a pleasant state. Emotions can also be created for robots, cars, motorcycles, etc., based on various balances such as posture and battery level, so that when these balances are far from the ideal, it indicates an unpleasant state, and when they are close to the ideal, it indicates a pleasant state. The emotion map may be generated, for example, based on the emotion map of Dr. Mitsuyoshi (Research on speech emotion recognition and emotion brain physiological signal analysis system, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). On the left half of the emotion map, emotions belonging to an area called "reaction" where sensation is dominant are lined up. On the right half of the emotion map, emotions belonging to an area called "situation" where situation recognition is dominant are lined up.

[0193] The emotion map defines two emotions that promote learning. The first is the negative emotion around the middle of "repentance" or "remorse" on the situation side. In other words, this is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the positive emotion around "desire" on the response side. In other words, this is when the robot has positive feelings such as "I want more" or "I want to know more."

[0194] The emotion identification model 59 inputs the user input to a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the emotion of the user. This neural network is pre-trained based on multiple learning data that are combinations of the user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in Fig. 10. Fig. 10 shows an example in which multiple emotions, "relief," "calm," and "encouraging," have similar emotion values.

[0195] In the above embodiment, an example was given in which a specific process is performed by one computer 22, but the technology disclosed herein is not limited to this, and distributed processing of the specific process may be performed by multiple computers, including computer 22.

[0196] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable computer-readable non-transitory storage medium such as a Universal Serial Bus (USB) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[0197] In addition, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 upon request from the data processing device 12.

[0198] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[0199] As the hardware resource for executing the specific process, various processors as shown below can be used. An example of the processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing the specific process by executing software, i.e., a program. Another example of the processor is a dedicated electric circuit, which is a processor having a circuit configuration designed exclusively for executing a specific process, such as a Field-Programmable Gate Array (FPGA), a Programmable Logic Device (PLD), or an Application Specific Integrated Circuit (ASIC). Each processor has a built-in or connected memory, and each processor executes the specific process by using the memory.

[0200] The hardware resource that executes the specific process may be one of these various processors, or may be a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific process may be a single processor.

[0201] As an example of a configuration using one processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a configuration using a processor that realizes the functions of the entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[0202] Furthermore, more specifically, the hardware structure of these various processors can be an electric circuit that combines circuit elements such as semiconductor elements. The specific processes described above are merely examples. It goes without saying that unnecessary steps may be deleted, new steps may be added, or the order of processes may be changed without departing from the spirit of the invention.

[0203] In the above example, the first to fourth embodiments are described separately, but some or all of these embodiments may be combined. The smart device 14, the smart glasses 214, the headset terminal 314, and the robot 414 are only examples, and they may be combined with each other, or may be other devices. In the above example, the first and second embodiments are described separately, but they may be combined.

[0204] The above description and illustrations are detailed descriptions of the parts related to the technology of the present disclosure, and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, function, action, and effect is an example of the configuration, function, action, and effect of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replaced with respect to the description and illustrations shown above, within the scope of the gist of the technology of the present disclosure. In addition, in order to avoid confusion and to facilitate understanding of the parts related to the technology of the present disclosure, the description and illustrations shown above omit explanations of technical common sense that do not require particular explanation in enabling the implementation of the technology of the present disclosure.

[0205] All publications, patent applications, and standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, and standard was specifically and individually indicated to be incorporated by reference.

[0206] (Appendix 1) a providing unit that provides conversation data; an analysis unit that analyzes the conversation data provided by the provision unit; a learning unit that learns a conversation style based on the data analyzed by the analysis unit; A response unit that generates a response based on the conversation style learned by the learning unit. A system characterized by: (Appendix 2) Equipped with a preprocessing unit that preprocesses conversation data 2. The system of claim 1, (Appendix 3) Equipped with an imitation unit that imitates the conversation style of a specific person 2. The system of claim 1, (Appendix 4) A feedback unit is provided to provide feedback. 2. The system of claim 1, (Appendix 5) The pre-treatment unit includes: Preprocessing the text data (e.g., denoising and tokenization) 3. The system of claim 2, (Appendix 6) The mimicking portion is Imitating the distinctive phrasing or speaking style of a particular person 4. The system of claim 3. (Appendix 7) The providing unit is To estimate a user's emotion and adjust the timing of providing conversation data based on the estimated user's emotion. 2. The system of claim 1, (Appendix 8) The providing unit is Analyze the user's past conversation data provision history and select the method of provision 2. The system of claim 1, (Appendix 9) The providing unit is Filter conversation data as it is provided based on the user's current interests 2. The system of claim 1, (Appendix 10) The providing unit is Estimate user emotions and determine the priority of conversation data to be provided based on the estimated user emotions. 2. The system of claim 1, (Appendix 11) The providing unit is When providing conversation data, it prioritizes providing data that is more relevant based on the user's geographic location. 2. The system of claim 1, (Appendix 12) The providing unit is When you provide conversation data, we analyze your social media activity and provide you with related data. 2. The system of claim 1, (Appendix 13) The analysis unit is Infer user emotions and adjust analysis methods based on the inferred user emotions 2. The system of claim 1, (Appendix 14) The analysis unit is During analysis, adjust the level of analysis detail based on the importance of the conversation data 2. The system of claim 1, (Appendix 15) The analysis unit is During analysis, different analysis algorithms are applied depending on the category of conversation data. 2. The system of claim 1, (Appendix 16) The analysis unit is Estimate user emotions and determine analysis priorities based on the estimated user emotions 2. The system of claim 1, (Appendix 17) The analysis unit is During analysis, the order of analysis is adjusted based on when the conversation data was submitted. 2. The system of claim 1, (Appendix 18) The analysis unit is During analysis, adjust the order of analysis based on the relevance of the conversation data 2. The system of claim 1, (Appendix 19) The learning unit is Estimate the user's emotions and select learning data based on the estimated user emotions. 2. The system of claim 1, (Appendix 20) The learning unit is During training, the learning algorithm is optimized by referencing past training data. 2. The system of claim 1, (Appendix 21) The learning unit is During training, different training methods are applied to different categories of conversation data. 2. The system of claim 1, (Appendix 22) The learning unit is Infer user emotions and adjust the frequency of learning based on the estimated user emotions. 2. The system of claim 1, (Appendix 23) The learning unit is During training, the training data is weighted based on when the conversation data was submitted. 2. The system of claim 1, (Appendix 24) The learning unit is During training, the accuracy of training is improved by referring to related literature on conversation data. 2. The system of claim 1, (Appendix 25) The response unit: Inferring a user's emotions and adjusting the way responses are expressed based on the inferred user emotions 2. The system of claim 1, (Appendix 26) The response unit: When generating responses, adjust the detail of the response based on the importance of the conversation 2. The system of claim 1, (Appendix 27) The response unit: When generating responses, apply different response algorithms depending on the conversation category 2. The system of claim 1, (Appendix 28) The response unit: Inferring a user's emotion and adjusting the length of a response based on the estimated user's emotion 2. The system of claim 1, (Appendix 29) The response unit: During response generation, prioritize responses based on when the conversation was submitted 2. The system of claim 1, (Appendix 30) The response unit: Adjust the order of responses based on conversational relevance during response generation 2. The system of claim 1, (Appendix 31) The pre-treatment unit includes: Estimate user emotions and adjust preprocessing methods based on the estimated user emotions. 3. The system of claim 2, (Appendix 32) The pre-treatment unit includes: During preprocessing, adjust the level of detail based on the importance of the conversation data 3. The system of claim 2, (Appendix 33) The pre-treatment unit includes: Estimate user emotions and determine preprocessing priorities based on the estimated user emotions. 3. The system of claim 2, (Appendix 34) The pre-treatment unit includes: During preprocessing, adjust the preprocessing order based on when the conversation data was submitted 3. The system of claim 2, (Appendix 35) The mimicking portion is Estimating a user's emotions and adjusting the mimicry method based on the estimated user's emotions 4. The system of claim 3. (Appendix 36) The mimicking portion is During imitation, adjust the level of imitation detail based on the importance of the conversation data 4. The system of claim 3. (Appendix 37) The mimicking portion is During imitation, different imitation algorithms are applied depending on the category of conversation data. 4. The system of claim 3. (Appendix 38) The mimicking portion is Estimate the user's emotions and determine the priority of imitation based on the estimated user's emotions. 4. The system of claim 3. (Appendix 39) The mimicking portion is During imitation, adjust the order of imitation based on when the conversation data was submitted 4. The system of claim 3. (Appendix 40) The mimicking portion is During imitation, adjust the order of imitation based on the relevance of the conversation data 4. The system of claim 3. (Appendix 41) The feedback unit is Estimate user's emotions and adjust the feedback method based on the estimated user's emotions 5. The system of claim 4. (Appendix 42) The feedback unit is When providing feedback, adjust the level of detail in your feedback based on the importance of the conversation data 5. The system of claim 4. (Appendix 43) The feedback unit is Estimate user emotions and prioritize feedback based on the estimated user emotions 5. The system of claim 4. (Appendix 44) The feedback unit is During feedback, adjust the order of feedback based on when conversation data was submitted 5. The system of claim 4. [Explanation of symbols]

[0207] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot

Claims

1. a providing unit that provides conversation data; an analysis unit that analyzes the conversation data provided by the provision unit; a learning unit that learns a conversation style based on the data analyzed by the analysis unit; A response unit that generates a response based on the conversation style learned by the learning unit. A system characterized by:

2. Equipped with a preprocessing unit that preprocesses conversation data 2. The system of claim 1 .

3. Equipped with an imitation unit that imitates the conversation style of a specific person 2. The system of claim 1 .

4. A feedback unit is provided to provide feedback.

2. The system of claim 1 .

5. The pre-treatment unit includes: Preprocessing text data 3. The system of claim 2.

6. The mimicking portion is Imitating the distinctive phrasing or speaking style of a particular person 4. The system of claim 3.

7. The providing unit is To estimate a user's emotion and adjust the timing of providing conversation data based on the estimated user's emotion.

2. The system of claim 1 .

8. The providing unit is Analyze the user's past conversation data provision history and select the method of provision 2. The system of claim 1 .

Images