System

Abstract

A system is provided.SOLUTION: A system comprising: means for registering different generative artificial intelligences; cooperation means for the generative artificial intelligences to cooperate with each other to perform a task; competition means for the generative artificial intelligences to compete with each other to perform a task; means for integrating results of the task; and means for evaluating a result of the competition.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 an instruction sentence related to a description of 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] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] Conventional generative AI has the problem of only being able to produce monotonous output. Furthermore, there is insufficient interaction between generative AIs learning the same data, and a certain amount of data is required. This limits the diversity and creativity of generative AI, making it difficult to generate more advanced expressions and ideas. Furthermore, the lack of efficient learning may slow the evolution of generative AI. [Means for solving the problem]

[0005] The present invention provides a system including a means for registering different generative artificial intelligences, a cooperation means for generative artificial intelligences to cooperate with each other to perform tasks, and a competition means for generative artificial intelligences to compete with each other to perform tasks. This system promotes interactions between generative artificial intelligences, leading to the creation of new ideas and methods of expression. Furthermore, a means for integrating task results improves the results of cooperation, and a means for evaluating the results of competition allows the selection of the optimal generative artificial intelligence. This promotes diversity and creativity in generative artificial intelligences and realizes efficient learning.

[0006] "Generative AI" refers to AI that has the ability to generate new content and information based on data.

[0007] A "task" is a specific instruction or task performed by generative artificial intelligence.

[0008] A "means" is a method or tool used to achieve a particular goal.

[0009] A "system" is a structure or network that includes multiple elements or means that are interrelated in order to achieve a specific purpose.

[0010] A "collaboration means" is a method or tool that allows multiple generative artificial intelligences to work together to perform a task.

[0011] A "competition method" is a method or tool for multiple generative artificial intelligences to compete with each other to perform tasks and evaluate the results.

[0012] A "means for integrating results" refers to a method or tool for combining the results of tasks generated by multiple generative artificial intelligences.

[0013] A "means for evaluating results" refers to a method or tool for comparing and evaluating the results of tasks generated by a generative artificial intelligence.

[0014] "Results" refer to the information or content obtained after generative artificial intelligence performs a task. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 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. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 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. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 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. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0016] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0017] First, the terms used in the following description will be explained.

[0018] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single 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), and an APU (Accelerated Processing Unit).

[0019] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0020] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, 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.

[0021] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0022] 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. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0023] [First embodiment]

[0024] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0025] 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.

[0026] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. 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).

[0027] 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.

[0028] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. 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 acquires the data indicating the user input.

[0029] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The 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 CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0030] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0031] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. 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 process 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.

[0033] 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.

[0034] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0035] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0036] The present invention provides a system for enabling different generative artificial intelligences to perform tasks through cooperation or competition, and for integrating or evaluating the generated results. This system promotes diversity and creativity in generative artificial intelligences and enables efficient learning. Specific embodiments of the system are described below.

[0037] First, a user creates a generative AI and registers it on the server. Each generative AI has a different name and characteristics and is designed to perform various tasks. Based on instructions from the user, the server initiates a procedure to have the generative AI cooperate or compete.

[0038] In the case of collaboration, the server selects two GAIs to perform a specific task. Each GAI generates its own unique result for the task. The server receives these results and uses a means to integrate them to produce a single integrated result. This result is a rich work that leverages the knowledge and creativity of the different GAIs.

[0039] In the case of a competition, the server similarly selects two generative AIs to perform a specific task. Each generative AI generates a result for the competition. The server receives these results and uses a means to evaluate them to determine the winner. This evaluation process encourages competition between generative AIs, resulting in more creative and efficient results.

[0040] As a concrete example, consider the case where a user requests a generative artificial intelligence to create a poem. The user registers two generative artificial intelligences on the server and requests the task of "writing a poem" in cooperative mode. The server instructs generative artificial intelligence 1 and generative artificial intelligence 2 to perform the task of "writing a poem." If generative artificial intelligence 1 generates a poem "Autumn evening, trees shining golden," and generative artificial intelligence 2 generates a poem "A calm lake surface, the wind rippling the water," the server integrates the results of both and generates the integrated result "Autumn evening, trees shining golden, a calm lake surface, the wind rippling the water." This integrated result is provided to the user.

[0041] Consider also the case where a user requests the task of generating a slogan from a generative AI in a competitive mode. The user assigns the task of "creating a slogan" to two generative AIs. If generative AI 1 generates the "power to create the future" and generative AI 2 generates the "key to open a new era," the server evaluates these results and determines the winner based on, for example, the length or creativity of the result. In this case, the "key to open a new era" is declared the winner based on an evaluation criterion that, for example, the longer the result, the better.

[0042] In this way, the system of the present invention provides a platform for generating new ideas and expressions through cooperation and competition between generative AIs, promoting interaction among generative AIs and the generation of new ideas, and achieving more advanced output.

[0043] The processing flow will be explained below.

[0044] Step 1:

[0045] The user creates a generative AI. The user creates an instance of the generative AI and sets its name and characteristics. For example, the user creates instances ai_agent1 and ai_agent2 using the AIAgent class.

[0046] Step 2:

[0047] The user registers the generative AI on the server. The user calls the register_ai_agent method on the server to add the generative AI to the server's database. This allows the server to manage and track the generative AI.

[0048] Step 3:

[0049] The user submits a task to the server. The user selects either collaboration or competition mode and specifies the task content. For example, to submit the task "write a poem," the user calls the create_collaboration or create_competition method.

[0050] Step 4:

[0051] The server assigns a task to the specified generative AI. The server passes the task to the specified generative AI, and each executes the task. The generative AI generates output for the task using the generate method.

[0052] Step 5:

[0053] The server receives the output of the generative AI. Each generative AI sends the results of its task to the server. The server collects and stores these results.

[0054] Step 6:

[0055] In the collaborative case, the server combines the outputs of the generative AI. The server uses the combine_results method to combine multiple results into a single combined result, for example by adding "AND" to the beginning of the sentence.

[0056] Step 7:

[0057] In the case of a competition, the server evaluates the outputs of the generative AI. The server uses the evaluate_results method to compare the results of the competition between the generative AIs and determine the winner. Evaluation criteria can include the length of the output, creativity, etc.

[0058] Step 8:

[0059] The server provides the synthesized or evaluated results to the user. The server returns the final result to the user and displays it to the user, which may be a synthesized poem or a slogan of the competition winner.

[0060] Through the above processing steps, users can take advantage of cooperation and competition between generative AIs to obtain a variety of ideas and methods of expression.

[0061] Example 1

[0062] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0063] In conventional generative AI systems, different generative AIs perform tasks through cooperation or competition, but there is a lack of means to efficiently integrate or evaluate the results. It is also difficult to maximize the diversity and creativity of generative AI. As a result, task results tend to fall into a uniform pattern, and more advanced and effective output is required.

[0064] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0065] In this invention, the server includes means for registering generative AIs, cooperation means for generative AIs to cooperate with each other to perform tasks, competition means for generative AIs to compete with each other to perform tasks, means for issuing instructions to the generative AIs based on a specified task, means for receiving the results of the tasks generated by the generative AIs, means for integrating the results of the tasks, means for evaluating the results of the competition, and means for providing the generated results to a user. This makes it possible for different generative AIs to perform tasks through cooperation or competition, and for the results to be efficiently integrated or evaluated, thereby maximizing the diversity and creativity of the generative AIs.

[0066] "Generative AI" is an AI model that has the ability to generate and answer questions for given tasks.

[0067] A "server" is a computer system that registers, manages, and instructs generative artificial intelligences, integrates or evaluates results, and provides results to users.

[0068] "Registration means" refers to a method or process for registering a generative artificial intelligence on a server, and includes the function of storing necessary data.

[0069] A "collaboration means" is a method or process by which multiple generative artificial intelligences cooperate to perform a task and integrate the results.

[0070] A "competition method" is a method or process by which multiple generative artificial intelligences compete to perform tasks and evaluate the results.

[0071] "Instruction means" refers to a method or process by which the server instructs the generative artificial intelligence on tasks and prompt sentences.

[0072] "Result receiving means" refers to a method or process for returning the results generated by the generative artificial intelligence to the server.

[0073] An "integration means" is a method or process for combining the results obtained from multiple generative artificial intelligences.

[0074] An "evaluation means" is a method or process for comparing and evaluating multiple results generated by a generative artificial intelligence and determining a winner.

[0075] A "results delivery means" is a method or process for delivering the aggregated results or selected results of an evaluation to a user.

[0076] This invention provides a system for different generative artificial intelligences to perform tasks through cooperation or competition, and then integrate or evaluate the generated results. This system promotes diversity and creativity in generative artificial intelligences and enables efficient learning.

[0077] First, the user creates a generative AI and registers it on the server. The generative AI is built using a generative AI model such as GPT-3 or BERT. The user uses a device to input the name and characteristics of the generative AI, as well as information about the generative AI model to be used, and sends this information to the server. The server then stores this information in a database.

[0078] The user then submits a specific task to the server, such as writing a poem or generating a slogan. Using the terminal, the user enters the details of the task, selects cooperative or competitive mode, and specifies a prompt to be used for execution. Examples of prompts include:

[0079] Co-op prompt:

[0080] "Write a poem on the theme of autumn evenings."

[0081] Competitive mode prompt:

[0082] "Create a slogan for a new product."

[0083] The server selects two of the registered generative AIs based on the mode specified by the user and instructs each to perform a task. At this time, the server provides the specified prompt sentence to each generative AI.

[0084] The generative AIs execute tasks based on prompts provided by the server and generate results. For example, for the task of "writing a poem," generative AI 1 generates a poem about an autumn evening, with trees shining golden, while generative AI 2 generates a poem about a calm lake surface, with the wind rippling the water. These results are then sent back to the server.

[0085] In cooperative mode, the server runs an algorithm to combine the generated results and generate a single combined result, such as "Autumn evening, the trees shine golden, the surface of the lake is calm, the wind ripples the surface of the water."

[0086] In competitive mode, the server evaluates each generated result and determines the winner. Evaluation criteria include the creativity and length of the result. For example, the server evaluates "Power to Create the Future" and "Key to Open a New Era" and determines the winner as "Key to Open a New Era" based on the length and creativity of the result.

[0087] Finally, the server provides the integrated results or the results selected by the evaluation to the user, who can receive the results through the terminal and decide on the next action.

[0088] The following specific hardware and software is used to implement this system:

[0089] Hardware used:

[0090] Server: AWS EC2 instance

[0091] Software used:

[0092] Generative AI: Generative AI models like GPT-3 and BERT

[0093] Integration and evaluation algorithms: Unique integration algorithms and evaluation logic implemented in Python

[0094] In this way, the present invention promotes new ideas and expressions through cooperation and competition between different generative artificial intelligences, making it possible to provide more advanced output to users.

[0095] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0096] Step 1:

[0097] The user registers the generative artificial intelligence on the server.

[0098] Specific operation: The user accesses the server from their device and enters information such as the name and characteristics of the generative AI, the generative AI model to be used (e.g., GPT-3), etc. The information entered by the user is sent to the server, which then stores it in a database.

[0099] Input: Name of generative artificial intelligence, characteristics, information of the generative AI model to be used

[0100] Output: Generative artificial intelligence information stored in a database

[0101] Step 2:

[0102] A user requests a specific task from the server.

[0103] Specific operation: The user uses the terminal to enter the details of the task, select cooperative or competitive mode, and send a prompt to the server to be used for execution.

[0104] Input: Task details, cooperative or competitive mode selection, prompt

[0105] Output: Task request information saved on the server

[0106] Step 3:

[0107] The server selects a generative AI and gives it instructions for the task.

[0108] Specific operation: The server selects two generative AIs from the database and provides each with a designated prompt sentence, which the server then sends to the generative AIs via an API call.

[0109] Input: Task request information, prompt text

[0110] Output: Task instructions for generative AI

[0111] Step 4:

[0112] Generative artificial intelligence performs tasks and generates results.

[0113] Specific operation: The generative artificial intelligence performs tasks based on prompts provided by the server and generates results, which are then sent back to the server.

[0114] Input: prompt statement

[0115] Output: The results generated by generative artificial intelligence

[0116] Step 5:

[0117] The server aggregates or evaluates the results.

[0118] Specific operation: In cooperative mode, the server runs an algorithm that combines multiple generated results to produce a single combined result. In competitive mode, the server evaluates the generated results and determines a winner. Evaluation criteria include the creativity and length of the results.

[0119] Input: Results received from the generative artificial intelligence

[0120] Output: Consolidated results or winners by rating

[0121] Step 6:

[0122] The results are presented to the user.

[0123] Specific operation: The server sends the integrated results or the results selected by the evaluation to the user and displays them on the user's device, so that the user can decide on the next action based on them.

[0124] Input: Winners by combined results or ratings

[0125] Output: The final result provided to the user

[0126] (Application example 1)

[0127] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0128] The modern advertising industry requires the rapid generation of more effective and creative slogans and advertising materials. However, conventional methods require a large amount of human resources and time, making efficient generation difficult. Furthermore, limited means for generating new ideas and expressions often result in a lack of quality and variety in advertisements. Therefore, there is a need for a system that utilizes generative artificial intelligence to generate efficient and creative advertising slogans.

[0129] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0130] In this invention, the server includes means for registering different generative AIs, cooperation means for the generative AIs to cooperate with each other to perform tasks, competition means for the generative AIs to compete with each other to perform tasks, means for integrating the results of the tasks, means for evaluating the results of the competition, means for inputting a prompt sentence from the user, and means for generating catchy slogans generated by the generative AIs for advertising media. This enables the rapid generation of high-quality and diverse advertising copy by having different generative AIs generate advertising catchy slogans through cooperation or competition and integrating or evaluating the results.

[0131] "Generative artificial intelligence" is an artificial intelligence system that generates information based on user prompts.

[0132] The "registration means" is a means having a function for registering different generative artificial intelligences in the server.

[0133] A "cooperative means" is a means having the function of allowing generative artificial intelligences to cooperate with each other to carry out tasks.

[0134] A "competition means" is a means that has the function of allowing generative artificial intelligences to compete with each other to perform tasks.

[0135] An "integration means" is a means having the function of integrating multiple results generated by a generative artificial intelligence into one.

[0136] The "evaluation means" is a means having a function for evaluating the results generated by the competition and selecting the superior results.

[0137] The "input means" is a means having a function for a user to input a prompt sentence to the server.

[0138] The "generation means" is a means having the function of generating a catchy slogan generated by the generative artificial intelligence for an advertising medium.

[0139] This invention relates to a system for generating effective advertising slogans through cooperation and competition between different generative artificial intelligences. This system can improve the quality and efficiency of advertising campaigns by having generative artificial intelligences cooperate to generate advertising materials or compete to select the best results.

[0140] System Program

[0141] The server contains the following main facilities:

[0142] 1. A means of registering different generative AIs

[0143] 2. A means of cooperation for generative AIs to work together to accomplish tasks

[0144] 3. A means of competition for generative AIs to perform tasks in competition with each other

[0145] 4. Means of integrating task results

[0146] 5. Means of assessing the results of the competition

[0147] 6. A means of inputting a prompt from the user

[0148] 7. A method for generating catchphrases generated by generative AI for advertising media

[0149] Program processing

[0150] The user inputs a prompt sentence, for example, a prompt sentence that generates an "advertising catchphrase for a new smartphone." This prompt sentence is sent to the server via the user terminal.

[0151] The server uses different generative AI models (such as GPT-2) to generate catchphrases based on the input prompt. Each generative AI model generates a unique catchphrase, resulting in a wide variety of ideas.

[0152] In the collaborative mode, a means for integrating these catchphrases is activated, combining or arranging multiple catchphrases to generate a single integrated result. For example, the integrated result "Experience the next generation of smart life. The technology of the future is here" may be generated as an "advertising catchphrase for a new smartphone."

[0153] In the competitive mode, the server evaluates multiple generated slogans and selects the most effective one. For example, between the slogans "Experience the next generation of smart life" and "The technology of the future is here," the server selects the best one based on the evaluation criteria.

[0154] This allows users to obtain high-quality advertising copy generated through the cooperation and competition of different generative AIs. The main software used is a generative AI model (e.g., GPT-2), TensorFlow, and the Hugging Face Transformers library. For hardware, a server equipped with a high-performance CPU or GPU is used.

[0155] As an example, suppose the user enters the following prompt:

[0156] Example prompt: "What beautiful hair can you achieve with your new hair care products?"

[0157] In this way, generative AI models can be used cooperatively and competitively to generate sophisticated advertising slogans.

[0158] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0159] Step 1:

[0160] The user inputs a prompt sentence. For example, a prompt sentence that generates an "advertising catchphrase for a new smartphone" is input from the user terminal and sent to the server. In this step, the input data is the prompt sentence, and the output is the prompt sentence sent to the server.

[0161] Step 2:

[0162] The server analyzes the received prompt sentence and prepares it to be passed to the generative AI model. Specifically, it tokenizes the prompt sentence and converts it into a format that the generative AI model can understand. In this step, the input data is the prompt sentence sent by the user, and the output is the tokenized data.

[0163] Step 3:

[0164] The server inputs the tokenized data into different generative artificial intelligence models and generates catchphrases based on each model. For example, Model1 and Model2 are used, each of which generates its own catchphrase. In this step, the input data is the tokenized prompt sentence, and the output is the generated multiple catchphrases.

[0165] Step 4:

[0166] In the collaborative mode, the server integrates the generated catchphrases. Specifically, it combines or organizes multiple catchphrases to generate a single integrated result. For example, the integrated result is "Experience the next generation of smart life. The technology of the future is here." In this step, the input data are multiple catchphrases, and the output is the integrated catchphrase.

[0167] Step 5:

[0168] In the competitive mode, the server evaluates the generated copy and selects the most effective one. It compares each copy based on evaluation criteria, such as copy length or creativity, and selects the best copy. In this step, the input data are multiple copy titles, and the output is the copy that is evaluated as the best.

[0169] Step 6:

[0170] The server sends the final tagline to the user terminal, so that the user can receive the generated advertising tagline. In this step, the input data is the synthesized tagline or the evaluated tagline, and the output is the tagline sent to the user terminal.

[0171] As a specific example of operation, when the prompt sentence "What kind of beautiful hair can you get by using new hair care products?" is input and processing is performed in collaborative mode, the server receives the output from both generative artificial intelligence models, combines them, and generates an integrated result "Use new hair care products to get vibrant, shiny, beautiful hair," which is provided to the user.

[0172] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0173] The present invention provides a more advanced experience by combining a system in which generative artificial intelligences execute tasks through cooperation and competition with each other and integrate or evaluate the generated results with an emotion engine that recognizes the user's emotions. This system promotes diversity and creativity in generative artificial intelligences and is capable of generating output in response to the user's emotions. Specific embodiments of the system are described below.

[0174] First, the user creates a generative AI and registers it on the server. Each generative AI has a different name and characteristics and is designed to perform various tasks. Based on instructions from the user, the server initiates procedures to make the generative AI cooperate or compete. Furthermore, the emotion engine recognizes the user's emotions and reflects this information in the generative AI's task execution.

[0175] In the case of collaboration, the server selects two GAIs to perform a specific task. Each GAI generates its own unique result for the task. The server receives these results and uses a means to integrate them to produce a single integrated result. This result is a rich work that leverages the knowledge and creativity of the different GAIs.

[0176] In the case of a competition, the server similarly selects two generative AIs to perform a specific task. Each generative AI generates a result for the competition. The server receives these results and uses a means to evaluate them to determine the winner. This evaluation process encourages competition between generative AIs, resulting in more creative and efficient results.

[0177] The emotion engine recognizes the user's emotions and adjusts the generative AI's output accordingly. This process generates content appropriate to the user's current emotional state. For example, if the user is recognized as sad, the emotion engine can instruct the generative AI to generate a comforting poem or message.

[0178] As a concrete example, consider the case where a user requests a generative AI to write a poem, while at the same time the emotion engine recognizes the emotion as "sad." The user registers two generative AIs on a server and requests the task of "writing a poem" in cooperative mode. The emotion engine sends the emotion "sad" to the server, and the server takes this into consideration and has the generative AIs execute the task. If generative AI 1 generates "a poem like tears falling on a quiet lakeside" and generative AI 2 generates "a poem that conveys the melancholy of the stars disappearing in the night sky," the server will integrate the results of both and generate a combined result: "a poem like tears falling on a quiet lakeside and a poem that conveys the melancholy of the stars disappearing in the night sky." This integrated result takes into account the user's emotional state.

[0179] Let's also consider the case where a user requests the task of generating a slogan from a generative AI in competitive mode, and at the same time, the emotion engine recognizes "joy." In this case, the user assigns the task of "creating a slogan" to two generative AIs, and the emotion engines send the emotion "joy" to the server. If generative AI 1 generates the slogan "The power to spread happiness" and generative AI 2 generates the slogan "The key to spreading smiles," the server evaluates these results and declares "The key to spreading smiles" as the winner. In this way, the emotion engines provide the optimal result according to the user's emotions.

[0180] By combining generative artificial intelligence with an emotion engine, this system can provide a more personalized experience for users and further enhance the diversity and creativity of generative artificial intelligence.

[0181] The processing flow will be explained below.

[0182] Step 1:

[0183] Users create generative AI. Users create instances of generative AI and set their names and characteristics, which allows the AI ​​to perform specific tasks.

[0184] Step 2:

[0185] The user registers the generative AI on the server. The user calls the register_ai_agent method on the server to add the generative AI to the server's database. This allows the server to manage and track the generative AI.

[0186] Step 3:

[0187] The user activates the emotion engine and recognizes emotions. The emotion engine connected to the device analyzes the user's facial expressions and voice to determine their current emotional state. This information is then sent to the server.

[0188] Step 4:

[0189] The user submits a task to the server. The user selects either collaboration or competition mode and specifies the task content, along with the emotional information recognized by the emotion engine. For example, to submit the task of "writing a poem," the user calls the create_collaboration or create_competition method.

[0190] Step 5:

[0191] The server receives the emotion information and assigns a task to the generative AI. The server takes into account the emotion information received from the emotion engine and instructs the generative AI to perform the task. The generative AI uses the generate method to generate an output that matches the emotion.

[0192] Step 6:

[0193] The server receives the output of the generative AI. Each generative AI sends the results of its task to the server. The server collects and stores these results.

[0194] Step 7:

[0195] In the collaborative case, the server combines the outputs of the generative AI. The server uses the combine_results method to combine multiple results into a single combined result, for example by adding "AND" to the beginning of the sentence.

[0196] Step 8:

[0197] In the case of a competition, the server evaluates the outputs of the generative AI. The server uses the evaluate_results method to compare the results of the competition between the generative AIs and determine the winner. Evaluation criteria can include the length of the output, creativity, etc.

[0198] Step 9:

[0199] The server provides the integrated or evaluated results to the user, and the server returns the final results to the user and displays them to the user, allowing the user to enjoy personalized content generated through the cooperation of generative artificial intelligence and the emotion engine.

[0200] For example, if a user feels sad, the emotion engine sends that information to the server. When the user requests the task of "writing a poem" in collaborative mode, the server uses this emotion information to have the generative AI execute the task. The generative AI outputs results such as "a poem like tears falling on a quiet lakeside" or "a poem that conveys the melancholy of stars disappearing in the night sky," and the server integrates these and provides them to the user. In this way, appropriate content is generated according to the user's emotions.

[0201] Example 2

[0202] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0203] Conventional generative AI systems provide fixed outputs without considering the user's emotional state, making it difficult to obtain results that are fully satisfying to the user. In addition, there is a lack of a mechanism for enabling generative AIs to cooperate or compete to complete tasks, which makes it difficult to maximize diversity and creativity.

[0204] The identification processing by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a means for registering different generative AIs, a cooperation means for the generative AIs to cooperate with each other to perform tasks, a competition means for the generative AIs to compete with each other to perform tasks, a means for integrating the results of the tasks, a means for evaluating the results of the competition, an emotion recognition means for recognizing the emotional state of the user, and a means for adjusting the output of the generative AI based on the emotional state. This allows the results of task execution to be output in a form appropriate to the user's emotional state, making it possible to provide an advanced experience that makes use of diversity and creativity.

[0205] "Generative AI" refers to an AI system that generates creative output based on user input and instructions.

[0206] "Cooperative means" refers to a means by which two or more generative artificial intelligences work together to accomplish a task.

[0207] "Competitive means" refers to a means by which two or more generative artificial intelligences independently perform the same task and compete with each other over the results they generate.

[0208] "Means for integrating results" refers to a means for combining outputs generated by multiple generative artificial intelligences into a single integrated output.

[0209] "Means for evaluating results" refers to a means for comparing outputs generated by multiple generative AIs and determining their relative merits.

[0210] "Emotion recognition means" refers to a means for recognizing the user's emotional state in real time and transmitting that information to a server.

[0211] "Means for adjusting output" refers to means for appropriately adjusting the output of the generative artificial intelligence based on the recognized emotional state of the user.

[0212] A "task" refers to a specific task or instruction that a generative artificial intelligence performs.

[0213] A "prompt sentence" refers to text input that allows a user to specify the task content and conditions in detail to a generative artificial intelligence.

[0214] MODE FOR CARRYING OUT THE INVENTION

[0215] The present invention combines an emotion recognition means for recognizing the emotions of a user with a system in which different generative artificial intelligence (hereinafter referred to as generative AI models) perform tasks through cooperation or competition and integrate or evaluate the results. Specific embodiments are described below.

[0216] 1. Creating and registering a generative AI model

[0217] Users create generative AI models to perform specific tasks and register them on a server. Generative AI models are developed using dedicated software (e.g., GPT-4 API or similar AI model development tools). The created generative AI models are assigned metadata that describes their name, characteristics, and task suitability.

[0218] 2. Recognizing user emotions using emotion recognition means

[0219] The device used by the user has a built-in camera and microphone, and emotion recognition means (e.g., Emotion API) analyzes the user's facial expressions and tone of voice through these devices. This analysis recognizes the user's emotional state in real time and transmits it to the server.

[0220] 3. Assigning tasks and specifying prompts

[0221] The user requests a task from the generative AI model through the server's web interface. The details of the task and the desired conditions are specified in a prompt. For example, the prompt could be set in the form, "When the user is feeling sad, please generate a poem to comfort them."

[0222] 4. Server selects generative AI model and assigns tasks

[0223] The server selects the most suitable model to execute the specified task from the registered generative AI models. In cooperative mode, the server assigns the same task to two generative AI models, and in competitive mode, the server assigns the task to two generative AI models in the same way.

[0224] 5. Generative AI model executes tasks and generates results

[0225] The selected generative AI model performs the task and sends its results to the server, which adjusts the output based on the user's emotional state. For example, if the user is recognized as sad, the generative AI model can be instructed to generate a comforting poem or message.

[0226] 6. Synthesis or evaluation of results

[0227] In collaborative mode, the server integrates the results received from the generative AI models, for example by generating different poems and combining them into a single integrated poem using NLP (Natural Language Processing) tools. In competitive mode, the server compares the generated results using an evaluation algorithm (e.g., a ranking model) and selects the best result.

[0228] 7. Providing Results

[0229] The final result is sent from the server to the user's device and displayed to the user. For example, the integrated result may be "A poem like tears falling on a quiet lakeside and a poem that evokes the melancholy of stars disappearing in the night sky."

[0230] Specific examples

[0231] When creating a poem (cooperative mode)

[0232] A specific example will be given in which a user requests the task of "writing a poem" and the emotion recognition means recognizes the emotion as "sad."

[0233] Example prompt sentence:

[0234] "When the user is feeling sad, generate a comforting poem."

[0235] In this case, the user registers two generative AI models on the server and requests them to create a poem in collaborative mode. The emotion recognition means detects "sadness," and the server has the two generative AI models generate a poem. Generative AI model 1 generates a poem like "tears falling on a quiet lake," while generative AI model 2 generates a poem that conveys the melancholy of stars disappearing in the night sky. The server then combines these results and displays them on the user's device.

[0236] When generating a slogan (competitive mode)

[0237] A specific example is given in which a task of creating a slogan is requested and the emotion recognition means recognizes "joy."

[0238] Example prompt sentence:

[0239] "When the user's emotion is joy, generate a slogan that emphasizes joy."

[0240] In this case, the user requests two generative AI models to create a slogan. The emotion recognition means detects "joy," and the server assigns the task to the two generative AI models. If generative AI model 1 generates the slogan "The power to spread happiness" and generative AI model 2 generates the slogan "The key to spreading smiles," the server uses an evaluation algorithm to select the best result and displays it on the user's device.

[0241] In this way, the user can obtain personalized results according to his / her emotional state.

[0242] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0243] Step 1:

[0244] Input: A user creates a generative AI model and sends its metadata (name, features, task suitability) to the server.

[0245] How it works: Users develop generative AI models using dedicated software tools (e.g., GPT-4 API) and upload them to a server through a web interface.

[0246] Output: The generative AI model registered on the server is stored in the database and displayed in the user's generative AI model list.

[0247] Step 2:

[0248] Input: Data on the user's facial expressions and voice obtained from the camera and microphone built into the user's device.

[0249] How it works: An emotion recognizer (e.g., Emotion API) analyzes this data and recognizes the user's emotional state in real time.

[0250] Output: The recognized emotional state (e.g. sad, happy) is sent to the server.

[0251] Step 3:

[0252] Input: A user-specified prompt (e.g., "When the user feels sad, please generate a comforting poem.") and task request information.

[0253] How it works: The user enters a prompt through the server's web interface to request a specific task from the generative AI model.

[0254] Output: The prompt statement and task request information received by the server are added to the processing queue.

[0255] Step 4:

[0256] Input: A list of generative AI models registered by the user and the specified task information.

[0257] How it works: The server selects the best generative AI model to perform the task. In cooperative mode, it selects two generative AI models, and in competitive mode, it selects two generative AI models.

[0258] Output: The selected generative AI model generates allocation data for a specific task.

[0259] Step 5:

[0260] Input: Assigned task information and selected generative AI model.

[0261] How it works: The server assigns tasks to each generative AI model. For example, the server assigns the same task of "writing a poem" to two generative AI models in cooperative mode.

[0262] Output: The generative AI model receives task information to be executed.

[0263] Step 6:

[0264] Input: Task results generated by the generative AI model and the user's emotional state data.

[0265] How it works: The generative AI model performs a given task and produces a result, adjusting for the user's emotional state.

[0266] Output: The generated results are sent to the server.

[0267] Step 7:

[0268] Input: The resulting data sent from the generative AI model.

[0269] How it works: The server combines the results to produce a single combined output in collaborative mode, or selects the best results using an evaluation algorithm (e.g., a ranking model) in competitive mode.

[0270] Output: Final results data that are integrated or evaluated are produced.

[0271] Step 8:

[0272] Input: Final result data.

[0273] How it works: The server sends the final result to the user's device and displays it to the user, for example, the synthesized poem or the selected slogan.

[0274] Output: The final result is displayed on the user's terminal and the user views the result.

[0275] (Application example 2)

[0276] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0277] In advertising creation, there is a demand for providing personalized content that reflects the user's emotions. Conventional advertising generation systems have difficulty taking user emotions into account, and as a result, they often generate advertisements that fail to attract the user's interest. To solve this problem, a system is needed that can properly recognize user emotions and adjust the output of generative artificial intelligence based on those emotions.

[0278] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for registering different generative AIs, a cooperation means for the generative AIs to cooperate with each other to perform tasks, a competition means for the generative AIs to compete with each other to perform tasks, a means for integrating the results of the tasks, a means for evaluating the results of the competition, a means including an emotion engine that recognizes the user's emotions, and a means for adjusting the output of the generative AI based on the user's emotions. This makes it possible to generate personalized advertising content that is optimized for the user's emotions.

[0279] ---

[0280] "Generative AI" is an AI system that can generate useful information or results based on user input.

[0281] A "collaborative means" is a method by which multiple generative artificial intelligences work together to accomplish tasks and generate results.

[0282] A "competitive method" is a method by which multiple generative artificial intelligences each carry out a task in their own unique way and compete for the results.

[0283] "Means for integrating task results" refers to a method for combining the results output by multiple generative artificial intelligences into one to generate an integrated result.

[0284] The "means for evaluating the results of the competition" is a method for comparing the results output by multiple generative artificial intelligences and selecting the optimal result.

[0285] An "emotion engine" is a system that recognizes the user's emotions from their facial expressions, voice, etc.

[0286] "Means for adjusting the output of generative artificial intelligence based on user emotions" refers to a method for adjusting the results output by generative artificial intelligence based on the user's emotional information recognized by the emotion engine.

[0287] The present invention relates to a system for generating optimal advertisements for users by combining generative artificial intelligence and an emotion engine. Specific embodiments for carrying out the present invention will be described below.

[0288] The system includes a server, a user terminal (such as a smartphone), and multiple generative AI models for registering generative AI. Users can use their own terminals to operate the generative AI and perform tasks such as creating advertisements.

[0289] First, the user enters an overview of the product or service for which they want to create an ad and related keywords into the device. Next, the emotion engine recognizes the user's emotions through the camera and microphone. The emotion engine uses, for example, AWS Rekognition or Microsoft Azure Emotion API.

[0290] The server sends a prompt message to a generative AI based on the user's emotional data and the keywords entered, and begins generating ad copy. For example, a generative AI model such as OpenAI's GPT-3 is used to generate ad copy that corresponds to the user's emotions.

[0291] The generated ad copy is then evaluated and integrated by the server, and finally output in a form that best suits the user's emotional state. This process provides personalized ad content to the user.

[0292] As a concrete example, consider a user who wants to create an ad for a new camera. If the user has the emotion "happy" along with the keywords "new camera," "high quality," and "professional," the following prompt sentence will be sent to the generative AI model:

[0293] Prompt Sentence Examples

[0294] Keywords: new camera, high quality, professional

[0295] Emotion: Joy

[0296] Generate compelling ad copy.

[0297] Based on this prompt, the generative AI model generates the following ad copy:

[0298] "A high-quality professional camera! Capture special moments from a new perspective. Why not use this camera to take a photo that will double your joy?"

[0299] The advertising copy generated in this way is optimized to the user's emotions, which is expected to increase the effectiveness of the advertisement. This system allows users to create optimal advertisements very easily.

[0300] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0301] ---

[0302] Step 1:

[0303] The user inputs into the device the summary and keywords of the product or service for which they want to create an ad. The input data collects keywords such as "new camera," "high image quality," and "professional" as well as emotional information. Terminal input: product summary, keywords, emotional information. Output: collected product summary, keywords, emotional information.

[0304] Step 2:

[0305] The emotion engine recognizes the user's current emotional state using the device's camera and microphone. The emotion engine analyzes the user's facial expressions and tone of voice to extract emotion data, for example, using AWS Rekognition or Microsoft Azure Emotion API. Device input: User's facial photo or voice data. Output: Recognized emotion data (e.g., joy).

[0306] Step 3:

[0307] The device sends the operation details and emotional data to the server. Based on the received data, the server prepares information to send a prompt to the generative AI. Server input: product summary, keywords, emotional data. Output: prompt to the generative AI model.

[0308] Step 4:

[0309] Send a prompt to a generative AI model (e.g., OpenAI GPT-3) to start generating ad copy. Example prompt: "Keywords: new camera, high quality, professional.\nEmotion: joy.\nPlease generate a compelling ad copy." Server input: prompt. Output: generated ad copy.

[0310] Step 5:

[0311] The server receives the ad copy output from the generative AI model. The server evaluates and combines the results to determine the optimal ad copy. Server input: Generated ad copy (multiple variations). Output: Optimal ad copy.

[0312] Step 6:

[0313] The optimized ad copy is sent back to the device and displayed to the user. Device input: Optimized ad copy received from the server. Output: Ad copy displayed to the user.

[0314] Step 7:

[0315] The user can review the generated ad copy and modify or approve it as needed. Terminal input: Displayed ad copy. Output: User modified or approved ad copy.

[0316] ---

[0317] This allows users to easily generate and view personalized, emotion-based ads.

[0318] 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 audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio 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 audio data.

[0319] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) 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 voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0320] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0321] [Second embodiment]

[0322] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0323] 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.

[0324] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. 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).

[0325] 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.

[0326] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0327] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and 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 surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0328] 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 carried out in a secure state.

[0329] Fig. 4 shows an example of the 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.

[0330] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. 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.

[0331] 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.

[0332] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0333] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0334] The present invention provides a system for enabling different generative artificial intelligences to perform tasks through cooperation or competition, and for integrating or evaluating the generated results. This system promotes diversity and creativity in generative artificial intelligences and enables efficient learning. Specific embodiments of the system are described below.

[0335] First, a user creates a generative AI and registers it on the server. Each generative AI has a different name and characteristics and is designed to perform various tasks. Based on instructions from the user, the server initiates a procedure to have the generative AI cooperate or compete.

[0336] In the case of collaboration, the server selects two GAIs to perform a specific task. Each GAI generates its own unique result for the task. The server receives these results and uses a means to integrate them to produce a single integrated result. This result is a rich work that leverages the knowledge and creativity of the different GAIs.

[0337] In the case of a competition, the server similarly selects two generative AIs to perform a specific task. Each generative AI generates a result for the competition. The server receives these results and uses a means to evaluate them to determine the winner. This evaluation process encourages competition between generative AIs, resulting in more creative and efficient results.

[0338] As a concrete example, consider the case where a user requests a generative artificial intelligence to create a poem. The user registers two generative artificial intelligences on the server and requests the task of "writing a poem" in cooperative mode. The server instructs generative artificial intelligence 1 and generative artificial intelligence 2 to perform the task of "writing a poem." If generative artificial intelligence 1 generates a poem "Autumn evening, trees shining golden," and generative artificial intelligence 2 generates a poem "A calm lake surface, the wind rippling the water," the server integrates the results of both and generates the integrated result "Autumn evening, trees shining golden, a calm lake surface, the wind rippling the water." This integrated result is provided to the user.

[0339] Consider also the case where a user requests the task of generating a slogan from a generative AI in a competitive mode. The user assigns the task of "creating a slogan" to two generative AIs. If generative AI 1 generates the "power to create the future" and generative AI 2 generates the "key to open a new era," the server evaluates these results and determines the winner based on, for example, the length or creativity of the result. In this case, the "key to open a new era" is declared the winner based on an evaluation criterion that, for example, the longer the result, the better.

[0340] In this way, the system of the present invention provides a platform for generating new ideas and expressions through cooperation and competition between generative AIs, promoting interaction among generative AIs and the generation of new ideas, and achieving more advanced output.

[0341] The processing flow will be explained below.

[0342] Step 1:

[0343] The user creates a generative AI. The user creates an instance of the generative AI and sets its name and characteristics. For example, the user creates instances ai_agent1 and ai_agent2 using the AIAgent class.

[0344] Step 2:

[0345] The user registers the generative AI on the server. The user calls the register_ai_agent method on the server to add the generative AI to the server's database. This allows the server to manage and track the generative AI.

[0346] Step 3:

[0347] The user submits a task to the server. The user selects either collaboration or competition mode and specifies the task content. For example, to submit the task "write a poem," the user calls the create_collaboration or create_competition method.

[0348] Step 4:

[0349] The server assigns a task to the specified generative AI. The server passes the task to the specified generative AI, and each executes the task. The generative AI generates output for the task using the generate method.

[0350] Step 5:

[0351] The server receives the output of the generative AI. Each generative AI sends the results of its task to the server. The server collects and stores these results.

[0352] Step 6:

[0353] In the collaborative case, the server combines the outputs of the generative AI. The server uses the combine_results method to combine multiple results into a single combined result, for example by adding "AND" to the beginning of the sentence.

[0354] Step 7:

[0355] In the case of a competition, the server evaluates the outputs of the generative AI. The server uses the evaluate_results method to compare the results of the competition between the generative AIs and determine the winner. Evaluation criteria can include the length of the output, creativity, etc.

[0356] Step 8:

[0357] The server provides the synthesized or evaluated results to the user. The server returns the final result to the user and displays it to the user, which may be a synthesized poem or a slogan of the competition winner.

[0358] Through the above processing steps, users can take advantage of cooperation and competition between generative AIs to obtain a variety of ideas and methods of expression.

[0359] Example 1

[0360] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0361] In conventional generative AI systems, different generative AIs perform tasks through cooperation or competition, but there is a lack of means to efficiently integrate or evaluate the results. It is also difficult to maximize the diversity and creativity of generative AI. As a result, task results tend to fall into a uniform pattern, and more advanced and effective output is required.

[0362] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0363] In this invention, the server includes means for registering generative AIs, cooperation means for generative AIs to cooperate with each other to perform tasks, competition means for generative AIs to compete with each other to perform tasks, means for issuing instructions to the generative AIs based on a specified task, means for receiving the results of the tasks generated by the generative AIs, means for integrating the results of the tasks, means for evaluating the results of the competition, and means for providing the generated results to a user. This makes it possible for different generative AIs to perform tasks through cooperation or competition, and for the results to be efficiently integrated or evaluated, thereby maximizing the diversity and creativity of the generative AIs.

[0364] "Generative AI" is an AI model that has the ability to generate and answer questions for given tasks.

[0365] A "server" is a computer system that registers, manages, and instructs generative artificial intelligences, integrates or evaluates results, and provides results to users.

[0366] "Registration means" refers to a method or process for registering a generative artificial intelligence on a server, and includes the function of storing necessary data.

[0367] A "collaboration means" is a method or process by which multiple generative artificial intelligences cooperate to perform a task and integrate the results.

[0368] A "competition method" is a method or process by which multiple generative artificial intelligences compete to perform tasks and evaluate the results.

[0369] "Instruction means" refers to a method or process by which the server instructs the generative artificial intelligence on tasks and prompt sentences.

[0370] "Result receiving means" refers to a method or process for returning the results generated by the generative artificial intelligence to the server.

[0371] An "integration means" is a method or process for combining the results obtained from multiple generative artificial intelligences.

[0372] An "evaluation means" is a method or process for comparing and evaluating multiple results generated by a generative artificial intelligence and determining a winner.

[0373] A "results delivery means" is a method or process for delivering the aggregated results or selected results of an evaluation to a user.

[0374] This invention provides a system for different generative artificial intelligences to perform tasks through cooperation or competition, and then integrate or evaluate the generated results. This system promotes diversity and creativity in generative artificial intelligences and enables efficient learning.

[0375] First, the user creates a generative AI and registers it on the server. The generative AI is built using a generative AI model such as GPT-3 or BERT. The user uses a device to input the name and characteristics of the generative AI, as well as information about the generative AI model to be used, and sends this information to the server. The server then stores this information in a database.

[0376] The user then submits a specific task to the server, such as writing a poem or generating a slogan. Using the terminal, the user enters the details of the task, selects cooperative or competitive mode, and specifies a prompt to be used for execution. Examples of prompts include:

[0377] Co-op prompt:

[0378] "Write a poem on the theme of autumn evenings."

[0379] Competitive mode prompt:

[0380] "Create a slogan for a new product."

[0381] The server selects two of the registered generative AIs based on the mode specified by the user and instructs each to perform a task. At this time, the server provides the specified prompt sentence to each generative AI.

[0382] The generative AIs execute tasks based on prompts provided by the server and generate results. For example, for the task of "writing a poem," generative AI 1 generates a poem about an autumn evening, with trees shining golden, while generative AI 2 generates a poem about a calm lake surface, with the wind rippling the water. These results are then sent back to the server.

[0383] In cooperative mode, the server runs an algorithm to combine the generated results and generate a single combined result, such as "Autumn evening, the trees shine golden, the surface of the lake is calm, the wind ripples the surface of the water."

[0384] In competitive mode, the server evaluates each generated result and determines the winner. Evaluation criteria include the creativity and length of the result. For example, the server evaluates "Power to Create the Future" and "Key to Open a New Era" and determines the winner as "Key to Open a New Era" based on the length and creativity of the result.

[0385] Finally, the server provides the integrated results or the results selected by the evaluation to the user, who can receive the results through the terminal and decide on the next action.

[0386] The following specific hardware and software is used to implement this system:

[0387] Hardware used:

[0388] Server: AWS EC2 instance

[0389] Software used:

[0390] Generative AI: Generative AI models like GPT-3 and BERT

[0391] Integration and evaluation algorithms: Unique integration algorithms and evaluation logic implemented in Python

[0392] In this way, the present invention promotes new ideas and expressions through cooperation and competition between different generative artificial intelligences, making it possible to provide more advanced output to users.

[0393] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0394] Step 1:

[0395] The user registers the generative artificial intelligence on the server.

[0396] Specific operation: The user accesses the server from their device and enters information such as the name and characteristics of the generative AI, the generative AI model to be used (e.g., GPT-3), etc. The information entered by the user is sent to the server, which then stores it in a database.

[0397] Input: Name of generative artificial intelligence, characteristics, information of the generative AI model to be used

[0398] Output: Generative artificial intelligence information stored in a database

[0399] Step 2:

[0400] A user requests a specific task from the server.

[0401] Specific operation: The user uses the terminal to enter the details of the task, select cooperative or competitive mode, and send a prompt to the server to be used for execution.

[0402] Input: Task details, cooperative or competitive mode selection, prompt

[0403] Output: Task request information saved on the server

[0404] Step 3:

[0405] The server selects a generative AI and gives it instructions for the task.

[0406] Specific operation: The server selects two generative AIs from the database and provides each with a designated prompt sentence, which the server then sends to the generative AIs via an API call.

[0407] Input: Task request information, prompt text

[0408] Output: Task instructions for generative AI

[0409] Step 4:

[0410] Generative artificial intelligence performs tasks and generates results.

[0411] Specific operation: The generative artificial intelligence performs tasks based on prompts provided by the server and generates results, which are then sent back to the server.

[0412] Input: prompt statement

[0413] Output: The results generated by generative artificial intelligence

[0414] Step 5:

[0415] The server aggregates or evaluates the results.

[0416] Specific operation: In cooperative mode, the server runs an algorithm that combines multiple generated results to produce a single combined result. In competitive mode, the server evaluates the generated results and determines a winner. Evaluation criteria include the creativity and length of the results.

[0417] Input: Results received from the generative artificial intelligence

[0418] Output: Consolidated results or winners by rating

[0419] Step 6:

[0420] The results are presented to the user.

[0421] Specific operation: The server sends the integrated results or the results selected by the evaluation to the user and displays them on the user's device, so that the user can decide on the next action based on them.

[0422] Input: Winners by combined results or ratings

[0423] Output: The final result provided to the user

[0424] (Application example 1)

[0425] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0426] The modern advertising industry requires the rapid generation of more effective and creative slogans and advertising materials. However, conventional methods require a large amount of human resources and time, making efficient generation difficult. Furthermore, limited means for generating new ideas and expressions often result in a lack of quality and variety in advertisements. Therefore, there is a need for a system that utilizes generative artificial intelligence to generate efficient and creative advertising slogans.

[0427] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0428] In this invention, the server includes means for registering different generative AIs, cooperation means for the generative AIs to cooperate with each other to perform tasks, competition means for the generative AIs to compete with each other to perform tasks, means for integrating the results of the tasks, means for evaluating the results of the competition, means for inputting a prompt sentence from the user, and means for generating catchy slogans generated by the generative AIs for advertising media. This enables the rapid generation of high-quality and diverse advertising copy by having different generative AIs generate advertising catchy slogans through cooperation or competition and integrating or evaluating the results.

[0429] "Generative artificial intelligence" is an artificial intelligence system that generates information based on user prompts.

[0430] The "registration means" is a means having a function for registering different generative artificial intelligences in the server.

[0431] A "cooperative means" is a means having the function of allowing generative artificial intelligences to cooperate with each other to carry out tasks.

[0432] A "competition means" is a means that has the function of allowing generative artificial intelligences to compete with each other to perform tasks.

[0433] An "integration means" is a means having the function of integrating multiple results generated by a generative artificial intelligence into one.

[0434] The "evaluation means" is a means having a function for evaluating the results generated by the competition and selecting the superior results.

[0435] The "input means" is a means having a function for a user to input a prompt sentence to the server.

[0436] The "generation means" is a means having the function of generating a catchy slogan generated by the generative artificial intelligence for an advertising medium.

[0437] This invention relates to a system for generating effective advertising slogans through cooperation and competition between different generative artificial intelligences. This system can improve the quality and efficiency of advertising campaigns by having generative artificial intelligences cooperate to generate advertising materials or compete to select the best results.

[0438] System Program

[0439] The server contains the following main facilities:

[0440] 1. A means of registering different generative AIs

[0441] 2. A means of cooperation for generative AIs to work together to accomplish tasks

[0442] 3. A means of competition for generative AIs to perform tasks in competition with each other

[0443] 4. Means of integrating task results

[0444] 5. Means of assessing the results of the competition

[0445] 6. A means of inputting a prompt from the user

[0446] 7. A method for generating catchphrases generated by generative AI for advertising media

[0447] Program processing

[0448] The user inputs a prompt sentence, for example, a prompt sentence that generates an "advertising catchphrase for a new smartphone." This prompt sentence is sent to the server via the user terminal.

[0449] The server uses different generative AI models (such as GPT-2) to generate catchphrases based on the input prompt. Each generative AI model generates a unique catchphrase, resulting in a wide variety of ideas.

[0450] In the collaborative mode, a means for integrating these catchphrases is activated, combining or arranging multiple catchphrases to generate a single integrated result. For example, the integrated result "Experience the next generation of smart life. The technology of the future is here" may be generated as an "advertising catchphrase for a new smartphone."

[0451] In the competitive mode, the server evaluates multiple generated slogans and selects the most effective one. For example, between the slogans "Experience the next generation of smart life" and "The technology of the future is here," the server selects the best one based on the evaluation criteria.

[0452] This allows users to obtain high-quality advertising copy generated through the cooperation and competition of different generative AIs. The main software used is a generative AI model (e.g., GPT-2), TensorFlow, and the Hugging Face Transformers library. For hardware, a server equipped with a high-performance CPU or GPU is used.

[0453] As an example, suppose the user enters the following prompt:

[0454] Example prompt: "What beautiful hair can you achieve with your new hair care products?"

[0455] In this way, generative AI models can be used cooperatively and competitively to generate sophisticated advertising slogans.

[0456] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0457] Step 1:

[0458] The user inputs a prompt sentence. For example, a prompt sentence that generates an "advertising catchphrase for a new smartphone" is input from the user terminal and sent to the server. In this step, the input data is the prompt sentence, and the output is the prompt sentence sent to the server.

[0459] Step 2:

[0460] The server analyzes the received prompt sentence and prepares it to be passed to the generative AI model. Specifically, it tokenizes the prompt sentence and converts it into a format that the generative AI model can understand. In this step, the input data is the prompt sentence sent by the user, and the output is the tokenized data.

[0461] Step 3:

[0462] The server inputs the tokenized data into different generative artificial intelligence models and generates catchphrases based on each model. For example, Model1 and Model2 are used, each of which generates its own catchphrase. In this step, the input data is the tokenized prompt sentence, and the output is the generated multiple catchphrases.

[0463] Step 4:

[0464] In the collaborative mode, the server integrates the generated catchphrases. Specifically, it combines or organizes multiple catchphrases to generate a single integrated result. For example, the integrated result is "Experience the next generation of smart life. The technology of the future is here." In this step, the input data are multiple catchphrases, and the output is the integrated catchphrase.

[0465] Step 5:

[0466] In the competitive mode, the server evaluates the generated copy and selects the most effective one. It compares each copy based on evaluation criteria, such as copy length or creativity, and selects the best copy. In this step, the input data are multiple copy titles, and the output is the copy that is evaluated as the best.

[0467] Step 6:

[0468] The server sends the final tagline to the user terminal, so that the user can receive the generated advertising tagline. In this step, the input data is the synthesized tagline or the evaluated tagline, and the output is the tagline sent to the user terminal.

[0469] As a specific example of operation, when the prompt sentence "What kind of beautiful hair can you get by using new hair care products?" is input and processing is performed in collaborative mode, the server receives the output from both generative artificial intelligence models, combines them, and generates an integrated result "Use new hair care products to get vibrant, shiny, beautiful hair," which is provided to the user.

[0470] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0471] The present invention provides a more advanced experience by combining a system in which generative artificial intelligences execute tasks through cooperation and competition with each other and integrate or evaluate the generated results with an emotion engine that recognizes the user's emotions. This system promotes diversity and creativity in generative artificial intelligences and is capable of generating output in response to the user's emotions. Specific embodiments of the system are described below.

[0472] First, the user creates a generative AI and registers it on the server. Each generative AI has a different name and characteristics and is designed to perform various tasks. Based on instructions from the user, the server initiates procedures to make the generative AI cooperate or compete. Furthermore, the emotion engine recognizes the user's emotions and reflects this information in the generative AI's task execution.

[0473] In the case of collaboration, the server selects two GAIs to perform a specific task. Each GAI generates its own unique result for the task. The server receives these results and uses a means to integrate them to produce a single integrated result. This result is a rich work that leverages the knowledge and creativity of the different GAIs.

[0474] In the case of a competition, the server similarly selects two generative AIs to perform a specific task. Each generative AI generates a result for the competition. The server receives these results and uses a means to evaluate them to determine the winner. This evaluation process encourages competition between generative AIs, resulting in more creative and efficient results.

[0475] The emotion engine recognizes the user's emotions and adjusts the generative AI's output accordingly. This process generates content appropriate to the user's current emotional state. For example, if the user is recognized as sad, the emotion engine can instruct the generative AI to generate a comforting poem or message.

[0476] As a concrete example, consider the case where a user requests a generative AI to write a poem, while at the same time the emotion engine recognizes the emotion as "sad." The user registers two generative AIs on a server and requests the task of "writing a poem" in cooperative mode. The emotion engine sends the emotion "sad" to the server, and the server takes this into consideration and has the generative AIs execute the task. If generative AI 1 generates "a poem like tears falling on a quiet lakeside" and generative AI 2 generates "a poem that conveys the melancholy of the stars disappearing in the night sky," the server will integrate the results of both and generate a combined result: "a poem like tears falling on a quiet lakeside and a poem that conveys the melancholy of the stars disappearing in the night sky." This integrated result takes into account the user's emotional state.

[0477] Let's also consider the case where a user requests the task of generating a slogan from a generative AI in competitive mode, and at the same time, the emotion engine recognizes "joy." In this case, the user assigns the task of "creating a slogan" to two generative AIs, and the emotion engines send the emotion "joy" to the server. If generative AI 1 generates the slogan "The power to spread happiness" and generative AI 2 generates the slogan "The key to spreading smiles," the server evaluates these results and declares "The key to spreading smiles" as the winner. In this way, the emotion engines provide the optimal result according to the user's emotions.

[0478] By combining generative artificial intelligence with an emotion engine, this system can provide a more personalized experience for users and further enhance the diversity and creativity of generative artificial intelligence.

[0479] The processing flow will be explained below.

[0480] Step 1:

[0481] Users create generative AI. Users create instances of generative AI and set their names and characteristics, which allows the AI ​​to perform specific tasks.

[0482] Step 2:

[0483] The user registers the generative AI on the server. The user calls the register_ai_agent method on the server to add the generative AI to the server's database. This allows the server to manage and track the generative AI.

[0484] Step 3:

[0485] The user activates the emotion engine and recognizes emotions. The emotion engine connected to the device analyzes the user's facial expressions and voice to determine their current emotional state. This information is then sent to the server.

[0486] Step 4:

[0487] The user submits a task to the server. The user selects either collaboration or competition mode and specifies the task content, along with the emotional information recognized by the emotion engine. For example, to submit the task of "writing a poem," the user calls the create_collaboration or create_competition method.

[0488] Step 5:

[0489] The server receives the emotion information and assigns a task to the generative AI. The server takes into account the emotion information received from the emotion engine and instructs the generative AI to perform the task. The generative AI uses the generate method to generate an output that matches the emotion.

[0490] Step 6:

[0491] The server receives the output of the generative AI. Each generative AI sends the results of its task to the server. The server collects and stores these results.

[0492] Step 7:

[0493] In the collaborative case, the server combines the outputs of the generative AI. The server uses the combine_results method to combine multiple results into a single combined result, for example by adding "AND" to the beginning of the sentence.

[0494] Step 8:

[0495] In the case of a competition, the server evaluates the outputs of the generative AI. The server uses the evaluate_results method to compare the results of the competition between the generative AIs and determine the winner. Evaluation criteria can include the length of the output, creativity, etc.

[0496] Step 9:

[0497] The server provides the integrated or evaluated results to the user, and the server returns the final results to the user and displays them to the user, allowing the user to enjoy personalized content generated through the cooperation of generative artificial intelligence and the emotion engine.

[0498] For example, if a user feels sad, the emotion engine sends that information to the server. When the user requests the task of "writing a poem" in collaborative mode, the server uses this emotion information to have the generative AI execute the task. The generative AI outputs results such as "a poem like tears falling on a quiet lakeside" or "a poem that conveys the melancholy of stars disappearing in the night sky," and the server integrates these and provides them to the user. In this way, appropriate content is generated according to the user's emotions.

[0499] Example 2

[0500] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0501] Conventional generative AI systems provide fixed outputs without considering the user's emotional state, making it difficult to obtain results that are fully satisfying to the user. In addition, there is a lack of a mechanism for enabling generative AIs to cooperate or compete to complete tasks, which makes it difficult to maximize diversity and creativity.

[0502] The identification processing by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a means for registering different generative AIs, a cooperation means for the generative AIs to cooperate with each other to perform tasks, a competition means for the generative AIs to compete with each other to perform tasks, a means for integrating the results of the tasks, a means for evaluating the results of the competition, an emotion recognition means for recognizing the emotional state of the user, and a means for adjusting the output of the generative AI based on the emotional state. This allows the results of task execution to be output in a form appropriate to the user's emotional state, making it possible to provide an advanced experience that makes use of diversity and creativity.

[0503] "Generative AI" refers to an AI system that generates creative output based on user input and instructions.

[0504] "Cooperative means" refers to a means by which two or more generative artificial intelligences work together to accomplish a task.

[0505] "Competitive means" refers to a means by which two or more generative artificial intelligences independently perform the same task and compete with each other over the results they generate.

[0506] "Means for integrating results" refers to a means for combining outputs generated by multiple generative artificial intelligences into a single integrated output.

[0507] "Means for evaluating results" refers to a means for comparing outputs generated by multiple generative AIs and determining their relative merits.

[0508] "Emotion recognition means" refers to a means for recognizing the user's emotional state in real time and transmitting that information to a server.

[0509] "Means for adjusting output" refers to means for appropriately adjusting the output of the generative artificial intelligence based on the recognized emotional state of the user.

[0510] A "task" refers to a specific task or instruction that a generative artificial intelligence performs.

[0511] A "prompt sentence" refers to text input that allows a user to specify the task content and conditions in detail to a generative artificial intelligence.

[0512] MODE FOR CARRYING OUT THE INVENTION

[0513] The present invention combines an emotion recognition means for recognizing the emotions of a user with a system in which different generative artificial intelligence (hereinafter referred to as generative AI models) perform tasks through cooperation or competition and integrate or evaluate the results. Specific embodiments are described below.

[0514] 1. Creating and registering a generative AI model

[0515] Users create generative AI models to perform specific tasks and register them on a server. Generative AI models are developed using dedicated software (e.g., GPT-4 API or similar AI model development tools). The created generative AI models are assigned metadata that describes their name, characteristics, and task suitability.

[0516] 2. Recognizing user emotions using emotion recognition means

[0517] The device used by the user has a built-in camera and microphone, and emotion recognition means (e.g., Emotion API) analyzes the user's facial expressions and tone of voice through these devices. This analysis recognizes the user's emotional state in real time and transmits it to the server.

[0518] 3. Assigning tasks and specifying prompts

[0519] The user requests a task from the generative AI model through the server's web interface. The details of the task and the desired conditions are specified in a prompt. For example, the prompt could be set in the form, "When the user is feeling sad, please generate a poem to comfort them."

[0520] 4. Server selects generative AI model and assigns tasks

[0521] The server selects the most suitable model to execute the specified task from the registered generative AI models. In cooperative mode, the server assigns the same task to two generative AI models, and in competitive mode, the server assigns the task to two generative AI models in the same way.

[0522] 5. Generative AI model executes tasks and generates results

[0523] The selected generative AI model performs the task and sends its results to the server, which adjusts the output based on the user's emotional state. For example, if the user is recognized as sad, the generative AI model can be instructed to generate a comforting poem or message.

[0524] 6. Synthesis or evaluation of results

[0525] In collaborative mode, the server integrates the results received from the generative AI models, for example by generating different poems and combining them into a single integrated poem using NLP (Natural Language Processing) tools. In competitive mode, the server compares the generated results using an evaluation algorithm (e.g., a ranking model) and selects the best result.

[0526] 7. Providing Results

[0527] The final result is sent from the server to the user's device and displayed to the user. For example, the integrated result may be "A poem like tears falling on a quiet lakeside and a poem that evokes the melancholy of stars disappearing in the night sky."

[0528] Specific examples

[0529] When creating a poem (cooperative mode)

[0530] A specific example will be given in which a user requests the task of "writing a poem" and the emotion recognition means recognizes the emotion as "sad."

[0531] Example prompt sentence:

[0532] "When the user is feeling sad, generate a comforting poem."

[0533] In this case, the user registers two generative AI models on the server and requests them to create a poem in collaborative mode. The emotion recognition means detects "sadness," and the server has the two generative AI models generate a poem. Generative AI model 1 generates a poem like "tears falling on a quiet lake," while generative AI model 2 generates a poem that conveys the melancholy of stars disappearing in the night sky. The server then combines these results and displays them on the user's device.

[0534] When generating a slogan (competitive mode)

[0535] A specific example is given in which a task of creating a slogan is requested and the emotion recognition means recognizes "joy."

[0536] Example prompt sentence:

[0537] "When the user's emotion is joy, generate a slogan that emphasizes joy."

[0538] In this case, the user requests two generative AI models to create a slogan. The emotion recognition means detects "joy," and the server assigns the task to the two generative AI models. If generative AI model 1 generates the slogan "The power to spread happiness" and generative AI model 2 generates the slogan "The key to spreading smiles," the server uses an evaluation algorithm to select the best result and displays it on the user's device.

[0539] In this way, the user can obtain personalized results according to his / her emotional state.

[0540] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0541] Step 1:

[0542] Input: A user creates a generative AI model and sends its metadata (name, features, task suitability) to the server.

[0543] How it works: Users develop generative AI models using dedicated software tools (e.g., GPT-4 API) and upload them to a server through a web interface.

[0544] Output: The generative AI model registered on the server is stored in the database and displayed in the user's generative AI model list.

[0545] Step 2:

[0546] Input: Data on the user's facial expressions and voice obtained from the camera and microphone built into the user's device.

[0547] How it works: An emotion recognizer (e.g., Emotion API) analyzes this data and recognizes the user's emotional state in real time.

[0548] Output: The recognized emotional state (e.g. sad, happy) is sent to the server.

[0549] Step 3:

[0550] Input: A user-specified prompt (e.g., "When the user feels sad, please generate a comforting poem.") and task request information.

[0551] How it works: The user enters a prompt through the server's web interface to request a specific task from the generative AI model.

[0552] Output: The prompt statement and task request information received by the server are added to the processing queue.

[0553] Step 4:

[0554] Input: A list of generative AI models registered by the user and the specified task information.

[0555] How it works: The server selects the best generative AI model to perform the task. In cooperative mode, it selects two generative AI models, and in competitive mode, it selects two generative AI models.

[0556] Output: The selected generative AI model generates allocation data for a specific task.

[0557] Step 5:

[0558] Input: Assigned task information and selected generative AI model.

[0559] How it works: The server assigns tasks to each generative AI model. For example, the server assigns the same task of "writing a poem" to two generative AI models in cooperative mode.

[0560] Output: The generative AI model receives task information to be executed.

[0561] Step 6:

[0562] Input: Task results generated by the generative AI model and the user's emotional state data.

[0563] How it works: The generative AI model performs a given task and produces a result, adjusting for the user's emotional state.

[0564] Output: The generated results are sent to the server.

[0565] Step 7:

[0566] Input: The resulting data sent from the generative AI model.

[0567] How it works: The server combines the results to produce a single combined output in collaborative mode, or selects the best results using an evaluation algorithm (e.g., a ranking model) in competitive mode.

[0568] Output: Final results data that are integrated or evaluated are produced.

[0569] Step 8:

[0570] Input: Final result data.

[0571] How it works: The server sends the final result to the user's device and displays it to the user, for example, the synthesized poem or the selected slogan.

[0572] Output: The final result is displayed on the user's terminal and the user views the result.

[0573] (Application example 2)

[0574] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0575] In advertising creation, there is a demand for providing personalized content that reflects the user's emotions. Conventional advertising generation systems have difficulty taking user emotions into account, and as a result, they often generate advertisements that fail to attract the user's interest. To solve this problem, a system is needed that can properly recognize user emotions and adjust the output of generative artificial intelligence based on those emotions.

[0576] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for registering different generative AIs, a cooperation means for the generative AIs to cooperate with each other to perform tasks, a competition means for the generative AIs to compete with each other to perform tasks, a means for integrating the results of the tasks, a means for evaluating the results of the competition, a means including an emotion engine that recognizes the user's emotions, and a means for adjusting the output of the generative AI based on the user's emotions. This makes it possible to generate personalized advertising content that is optimized for the user's emotions.

[0577] ---

[0578] "Generative AI" is an AI system that can generate useful information or results based on user input.

[0579] A "collaborative means" is a method by which multiple generative artificial intelligences work together to accomplish tasks and generate results.

[0580] A "competitive method" is a method by which multiple generative artificial intelligences each carry out a task in their own unique way and compete for the results.

[0581] "Means for integrating task results" refers to a method for combining the results output by multiple generative artificial intelligences into one to generate an integrated result.

[0582] The "means for evaluating the results of the competition" is a method for comparing the results output by multiple generative artificial intelligences and selecting the optimal result.

[0583] An "emotion engine" is a system that recognizes the user's emotions from their facial expressions, voice, etc.

[0584] "Means for adjusting the output of generative artificial intelligence based on user emotions" refers to a method for adjusting the results output by generative artificial intelligence based on the user's emotional information recognized by the emotion engine.

[0585] The present invention relates to a system for generating optimal advertisements for users by combining generative artificial intelligence and an emotion engine. Specific embodiments for carrying out the present invention will be described below.

[0586] The system includes a server, a user terminal (such as a smartphone), and multiple generative AI models for registering generative AI. Users can use their own terminals to operate the generative AI and perform tasks such as creating advertisements.

[0587] First, the user enters an overview of the product or service for which they want to create an ad and related keywords into the device. Next, the emotion engine recognizes the user's emotions through the camera and microphone. The emotion engine uses, for example, AWS Rekognition or Microsoft Azure Emotion API.

[0588] The server sends a prompt message to a generative AI based on the user's emotional data and the keywords entered, and begins generating ad copy. For example, a generative AI model such as OpenAI's GPT-3 is used to generate ad copy that corresponds to the user's emotions.

[0589] The generated ad copy is then evaluated and integrated by the server, and finally output in a form that best suits the user's emotional state. This process provides personalized ad content to the user.

[0590] As a concrete example, consider a user who wants to create an ad for a new camera. If the user has the emotion "happy" along with the keywords "new camera," "high quality," and "professional," the following prompt sentence will be sent to the generative AI model:

[0591] Prompt Sentence Examples

[0592] Keywords: new camera, high quality, professional

[0593] Emotion: Joy

[0594] Generate compelling ad copy.

[0595] Based on this prompt, the generative AI model generates the following ad copy:

[0596] "A high-quality professional camera! Capture special moments from a new perspective. Why not use this camera to take a photo that will double your joy?"

[0597] The advertising copy generated in this way is optimized to the user's emotions, which is expected to increase the effectiveness of the advertisement. This system allows users to create optimal advertisements very easily.

[0598] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0599] ---

[0600] Step 1:

[0601] The user inputs into the device the summary and keywords of the product or service for which they want to create an ad. The input data collects keywords such as "new camera," "high image quality," and "professional" as well as emotional information. Terminal input: product summary, keywords, emotional information. Output: collected product summary, keywords, emotional information.

[0602] Step 2:

[0603] The emotion engine recognizes the user's current emotional state using the device's camera and microphone. The emotion engine analyzes the user's facial expressions and tone of voice to extract emotion data, for example, using AWS Rekognition or Microsoft Azure Emotion API. Device input: User's facial photo or voice data. Output: Recognized emotion data (e.g., joy).

[0604] Step 3:

[0605] The device sends the operation details and emotional data to the server. Based on the received data, the server prepares information to send a prompt to the generative AI. Server input: product summary, keywords, emotional data. Output: prompt to the generative AI model.

[0606] Step 4:

[0607] Send a prompt to a generative AI model (e.g., OpenAI GPT-3) to start generating ad copy. Example prompt: "Keywords: new camera, high quality, professional.\nEmotion: joy.\nPlease generate a compelling ad copy." Server input: prompt. Output: generated ad copy.

[0608] Step 5:

[0609] The server receives the ad copy output from the generative AI model. The server evaluates and combines the results to determine the optimal ad copy. Server input: Generated ad copy (multiple variations). Output: Optimal ad copy.

[0610] Step 6:

[0611] The optimized ad copy is sent back to the device and displayed to the user. Device input: Optimized ad copy received from the server. Output: Ad copy displayed to the user.

[0612] Step 7:

[0613] The user can review the generated ad copy and modify or approve it as needed. Terminal input: Displayed ad copy. Output: User modified or approved ad copy.

[0614] ---

[0615] This allows users to easily generate and view personalized, emotion-based ads.

[0616] 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 audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio 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 audio data.

[0617] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) 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 voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0618] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0619] [Third embodiment]

[0620] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0621] 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.

[0622] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. 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).

[0623] 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.

[0624] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0625] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and 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 surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0626] 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 carried out in a secure state.

[0627] 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.

[0628] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. 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.

[0629] 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.

[0630] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0631] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0632] The present invention provides a system for enabling different generative artificial intelligences to perform tasks through cooperation or competition, and for integrating or evaluating the generated results. This system promotes diversity and creativity in generative artificial intelligences and enables efficient learning. Specific embodiments of the system are described below.

[0633] First, a user creates a generative AI and registers it on the server. Each generative AI has a different name and characteristics and is designed to perform various tasks. Based on instructions from the user, the server initiates a procedure to have the generative AI cooperate or compete.

[0634] In the case of collaboration, the server selects two GAIs to perform a specific task. Each GAI generates its own unique result for the task. The server receives these results and uses a means to integrate them to produce a single integrated result. This result is a rich work that leverages the knowledge and creativity of the different GAIs.

[0635] In the case of a competition, the server similarly selects two generative AIs to perform a specific task. Each generative AI generates a result for the competition. The server receives these results and uses a means to evaluate them to determine the winner. This evaluation process encourages competition between generative AIs, resulting in more creative and efficient results.

[0636] As a concrete example, consider the case where a user requests a generative artificial intelligence to create a poem. The user registers two generative artificial intelligences on the server and requests the task of "writing a poem" in cooperative mode. The server instructs generative artificial intelligence 1 and generative artificial intelligence 2 to perform the task of "writing a poem." If generative artificial intelligence 1 generates a poem "Autumn evening, trees shining golden," and generative artificial intelligence 2 generates a poem "A calm lake surface, the wind rippling the water," the server integrates the results of both and generates the integrated result "Autumn evening, trees shining golden, a calm lake surface, the wind rippling the water." This integrated result is provided to the user.

[0637] Consider also the case where a user requests the task of generating a slogan from a generative AI in a competitive mode. The user assigns the task of "creating a slogan" to two generative AIs. If generative AI 1 generates the "power to create the future" and generative AI 2 generates the "key to open a new era," the server evaluates these results and determines the winner based on, for example, the length or creativity of the result. In this case, the "key to open a new era" is declared the winner based on an evaluation criterion that, for example, the longer the result, the better.

[0638] In this way, the system of the present invention provides a platform for generating new ideas and expressions through cooperation and competition between generative AIs, promoting interaction among generative AIs and the generation of new ideas, and achieving more advanced output.

[0639] The processing flow will be explained below.

[0640] Step 1:

[0641] The user creates a generative AI. The user creates an instance of the generative AI and sets its name and characteristics. For example, the user creates instances ai_agent1 and ai_agent2 using the AIAgent class.

[0642] Step 2:

[0643] The user registers the generative AI on the server. The user calls the register_ai_agent method on the server to add the generative AI to the server's database. This allows the server to manage and track the generative AI.

[0644] Step 3:

[0645] The user submits a task to the server. The user selects either collaboration or competition mode and specifies the task content. For example, to submit the task "write a poem," the user calls the create_collaboration or create_competition method.

[0646] Step 4:

[0647] The server assigns a task to the specified generative AI. The server passes the task to the specified generative AI, and each executes the task. The generative AI generates output for the task using the generate method.

[0648] Step 5:

[0649] The server receives the output of the generative AI. Each generative AI sends the results of its task to the server. The server collects and stores these results.

[0650] Step 6:

[0651] In the collaborative case, the server combines the outputs of the generative AI. The server uses the combine_results method to combine multiple results into a single combined result, for example by adding "AND" to the beginning of the sentence.

[0652] Step 7:

[0653] In the case of a competition, the server evaluates the outputs of the generative AI. The server uses the evaluate_results method to compare the results of the competition between the generative AIs and determine the winner. Evaluation criteria can include the length of the output, creativity, etc.

[0654] Step 8:

[0655] The server provides the synthesized or evaluated results to the user. The server returns the final result to the user and displays it to the user, which may be a synthesized poem or a slogan of the competition winner.

[0656] Through the above processing steps, users can take advantage of cooperation and competition between generative AIs to obtain a variety of ideas and methods of expression.

[0657] Example 1

[0658] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0659] In conventional generative AI systems, different generative AIs perform tasks through cooperation or competition, but there is a lack of means to efficiently integrate or evaluate the results. It is also difficult to maximize the diversity and creativity of generative AI. As a result, task results tend to fall into a uniform pattern, and more advanced and effective output is required.

[0660] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0661] In this invention, the server includes means for registering generative AIs, cooperation means for generative AIs to cooperate with each other to perform tasks, competition means for generative AIs to compete with each other to perform tasks, means for issuing instructions to the generative AIs based on a specified task, means for receiving the results of the tasks generated by the generative AIs, means for integrating the results of the tasks, means for evaluating the results of the competition, and means for providing the generated results to a user. This makes it possible for different generative AIs to perform tasks through cooperation or competition, and for the results to be efficiently integrated or evaluated, thereby maximizing the diversity and creativity of the generative AIs.

[0662] "Generative AI" is an AI model that has the ability to generate and answer questions for given tasks.

[0663] A "server" is a computer system that registers, manages, and instructs generative artificial intelligences, integrates or evaluates results, and provides results to users.

[0664] "Registration means" refers to a method or process for registering a generative artificial intelligence on a server, and includes the function of storing necessary data.

[0665] A "collaboration means" is a method or process by which multiple generative artificial intelligences cooperate to perform a task and integrate the results.

[0666] A "competition method" is a method or process by which multiple generative artificial intelligences compete to perform tasks and evaluate the results.

[0667] "Instruction means" refers to a method or process by which the server instructs the generative artificial intelligence on tasks and prompt sentences.

[0668] "Result receiving means" refers to a method or process for returning the results generated by the generative artificial intelligence to the server.

[0669] An "integration means" is a method or process for combining the results obtained from multiple generative artificial intelligences.

[0670] An "evaluation means" is a method or process for comparing and evaluating multiple results generated by a generative artificial intelligence and determining a winner.

[0671] A "results delivery means" is a method or process for delivering the aggregated results or selected results of an evaluation to a user.

[0672] This invention provides a system for different generative artificial intelligences to perform tasks through cooperation or competition, and then integrate or evaluate the generated results. This system promotes diversity and creativity in generative artificial intelligences and enables efficient learning.

[0673] First, the user creates a generative AI and registers it on the server. The generative AI is built using a generative AI model such as GPT-3 or BERT. The user uses a device to input the name and characteristics of the generative AI, as well as information about the generative AI model to be used, and sends this information to the server. The server then stores this information in a database.

[0674] The user then submits a specific task to the server, such as writing a poem or generating a slogan. Using the terminal, the user enters the details of the task, selects cooperative or competitive mode, and specifies a prompt to be used for execution. Examples of prompts include:

[0675] Co-op prompt:

[0676] "Write a poem on the theme of autumn evenings."

[0677] Competitive mode prompt:

[0678] "Create a slogan for a new product."

[0679] The server selects two of the registered generative AIs based on the mode specified by the user and instructs each to perform a task. At this time, the server provides the specified prompt sentence to each generative AI.

[0680] The generative AIs execute tasks based on prompts provided by the server and generate results. For example, for the task of "writing a poem," generative AI 1 generates a poem about an autumn evening, with trees shining golden, while generative AI 2 generates a poem about a calm lake surface, with the wind rippling the water. These results are then sent back to the server.

[0681] In cooperative mode, the server runs an algorithm to combine the generated results and generate a single combined result, such as "Autumn evening, the trees shine golden, the surface of the lake is calm, the wind ripples the surface of the water."

[0682] In competitive mode, the server evaluates each generated result and determines the winner. Evaluation criteria include the creativity and length of the result. For example, the server evaluates "Power to Create the Future" and "Key to Open a New Era" and determines the winner as "Key to Open a New Era" based on the length and creativity of the result.

[0683] Finally, the server provides the integrated results or the results selected by the evaluation to the user, who can receive the results through the terminal and decide on the next action.

[0684] The following specific hardware and software is used to implement this system:

[0685] Hardware used:

[0686] Server: AWS EC2 instance

[0687] Software used:

[0688] Generative AI: Generative AI models like GPT-3 and BERT

[0689] Integration and evaluation algorithms: Unique integration algorithms and evaluation logic implemented in Python

[0690] In this way, the present invention promotes new ideas and expressions through cooperation and competition between different generative artificial intelligences, making it possible to provide more advanced output to users.

[0691] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0692] Step 1:

[0693] The user registers the generative artificial intelligence on the server.

[0694] Specific operation: The user accesses the server from their device and enters information such as the name and characteristics of the generative AI, the generative AI model to be used (e.g., GPT-3), etc. The information entered by the user is sent to the server, which then stores it in a database.

[0695] Input: Name of generative artificial intelligence, characteristics, information of the generative AI model to be used

[0696] Output: Generative artificial intelligence information stored in a database

[0697] Step 2:

[0698] A user requests a specific task from the server.

[0699] Specific operation: The user uses the terminal to enter the details of the task, select cooperative or competitive mode, and send a prompt to the server to be used for execution.

[0700] Input: Task details, cooperative or competitive mode selection, prompt

[0701] Output: Task request information saved on the server

[0702] Step 3:

[0703] The server selects a generative AI and gives it instructions for the task.

[0704] Specific operation: The server selects two generative AIs from the database and provides each with a designated prompt sentence, which the server then sends to the generative AIs via an API call.

[0705] Input: Task request information, prompt text

[0706] Output: Task instructions for generative AI

[0707] Step 4:

[0708] Generative artificial intelligence performs tasks and generates results.

[0709] Specific operation: The generative artificial intelligence performs tasks based on prompts provided by the server and generates results, which are then sent back to the server.

[0710] Input: prompt statement

[0711] Output: The results generated by generative artificial intelligence

[0712] Step 5:

[0713] The server aggregates or evaluates the results.

[0714] Specific operation: In cooperative mode, the server runs an algorithm that combines multiple generated results to produce a single combined result. In competitive mode, the server evaluates the generated results and determines a winner. Evaluation criteria include the creativity and length of the results.

[0715] Input: Results received from the generative artificial intelligence

[0716] Output: Consolidated results or winners by rating

[0717] Step 6:

[0718] The results are presented to the user.

[0719] Specific operation: The server sends the integrated results or the results selected by the evaluation to the user and displays them on the user's device, so that the user can decide on the next action based on them.

[0720] Input: Winners by combined results or ratings

[0721] Output: The final result provided to the user

[0722] (Application example 1)

[0723] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0724] The modern advertising industry requires the rapid generation of more effective and creative slogans and advertising materials. However, conventional methods require a large amount of human resources and time, making efficient generation difficult. Furthermore, limited means for generating new ideas and expressions often result in a lack of quality and variety in advertisements. Therefore, there is a need for a system that utilizes generative artificial intelligence to generate efficient and creative advertising slogans.

[0725] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0726] In this invention, the server includes means for registering different generative AIs, cooperation means for the generative AIs to cooperate with each other to perform tasks, competition means for the generative AIs to compete with each other to perform tasks, means for integrating the results of the tasks, means for evaluating the results of the competition, means for inputting a prompt sentence from the user, and means for generating catchy slogans generated by the generative AIs for advertising media. This enables the rapid generation of high-quality and diverse advertising copy by having different generative AIs generate advertising catchy slogans through cooperation or competition and integrating or evaluating the results.

[0727] "Generative artificial intelligence" is an artificial intelligence system that generates information based on user prompts.

[0728] The "registration means" is a means having a function for registering different generative artificial intelligences in the server.

[0729] A "cooperative means" is a means having the function of allowing generative artificial intelligences to cooperate with each other to carry out tasks.

[0730] A "competition means" is a means that has the function of allowing generative artificial intelligences to compete with each other to perform tasks.

[0731] An "integration means" is a means having the function of integrating multiple results generated by a generative artificial intelligence into one.

[0732] The "evaluation means" is a means having a function for evaluating the results generated by the competition and selecting the superior results.

[0733] The "input means" is a means having a function for a user to input a prompt sentence to the server.

[0734] The "generation means" is a means having the function of generating a catchy slogan generated by the generative artificial intelligence for an advertising medium.

[0735] This invention relates to a system for generating effective advertising slogans through cooperation and competition between different generative artificial intelligences. This system can improve the quality and efficiency of advertising campaigns by having generative artificial intelligences cooperate to generate advertising materials or compete to select the best results.

[0736] System Program

[0737] The server contains the following main facilities:

[0738] 1. A means of registering different generative AIs

[0739] 2. A means of cooperation for generative AIs to work together to accomplish tasks

[0740] 3. A means of competition for generative AIs to perform tasks in competition with each other

[0741] 4. Means of integrating task results

[0742] 5. Means of assessing the results of the competition

[0743] 6. A means of inputting a prompt from the user

[0744] 7. A method for generating catchphrases generated by generative AI for advertising media

[0745] Program processing

[0746] The user inputs a prompt sentence, for example, a prompt sentence that generates an "advertising catchphrase for a new smartphone." This prompt sentence is sent to the server via the user terminal.

[0747] The server uses different generative AI models (such as GPT-2) to generate catchphrases based on the input prompt. Each generative AI model generates a unique catchphrase, resulting in a wide variety of ideas.

[0748] In the collaborative mode, a means for integrating these catchphrases is activated, combining or arranging multiple catchphrases to generate a single integrated result. For example, the integrated result "Experience the next generation of smart life. The technology of the future is here" may be generated as an "advertising catchphrase for a new smartphone."

[0749] In the competitive mode, the server evaluates multiple generated slogans and selects the most effective one. For example, between the slogans "Experience the next generation of smart life" and "The technology of the future is here," the server selects the best one based on the evaluation criteria.

[0750] This allows users to obtain high-quality advertising copy generated through the cooperation and competition of different generative AIs. The main software used is a generative AI model (e.g., GPT-2), TensorFlow, and the Hugging Face Transformers library. For hardware, a server equipped with a high-performance CPU or GPU is used.

[0751] As an example, suppose the user enters the following prompt:

[0752] Example prompt: "What beautiful hair can you achieve with your new hair care products?"

[0753] In this way, generative AI models can be used cooperatively and competitively to generate sophisticated advertising slogans.

[0754] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0755] Step 1:

[0756] The user inputs a prompt sentence. For example, a prompt sentence that generates an "advertising catchphrase for a new smartphone" is input from the user terminal and sent to the server. In this step, the input data is the prompt sentence, and the output is the prompt sentence sent to the server.

[0757] Step 2:

[0758] The server analyzes the received prompt sentence and prepares it to be passed to the generative AI model. Specifically, it tokenizes the prompt sentence and converts it into a format that the generative AI model can understand. In this step, the input data is the prompt sentence sent by the user, and the output is the tokenized data.

[0759] Step 3:

[0760] The server inputs the tokenized data into different generative artificial intelligence models and generates catchphrases based on each model. For example, Model1 and Model2 are used, each of which generates its own catchphrase. In this step, the input data is the tokenized prompt sentence, and the output is the generated multiple catchphrases.

[0761] Step 4:

[0762] In the collaborative mode, the server integrates the generated catchphrases. Specifically, it combines or organizes multiple catchphrases to generate a single integrated result. For example, the integrated result is "Experience the next generation of smart life. The technology of the future is here." In this step, the input data are multiple catchphrases, and the output is the integrated catchphrase.

[0763] Step 5:

[0764] In the competitive mode, the server evaluates the generated copy and selects the most effective one. It compares each copy based on evaluation criteria, such as copy length or creativity, and selects the best copy. In this step, the input data are multiple copy titles, and the output is the copy that is evaluated as the best.

[0765] Step 6:

[0766] The server sends the final tagline to the user terminal, so that the user can receive the generated advertising tagline. In this step, the input data is the synthesized tagline or the evaluated tagline, and the output is the tagline sent to the user terminal.

[0767] As a specific example of operation, when the prompt sentence "What kind of beautiful hair can you get by using new hair care products?" is input and processing is performed in collaborative mode, the server receives the output from both generative artificial intelligence models, combines them, and generates an integrated result "Use new hair care products to get vibrant, shiny, beautiful hair," which is provided to the user.

[0768] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0769] The present invention provides a more advanced experience by combining a system in which generative artificial intelligences execute tasks through cooperation and competition with each other and integrate or evaluate the generated results with an emotion engine that recognizes the user's emotions. This system promotes diversity and creativity in generative artificial intelligences and is capable of generating output in response to the user's emotions. Specific embodiments of the system are described below.

[0770] First, the user creates a generative AI and registers it on the server. Each generative AI has a different name and characteristics and is designed to perform various tasks. Based on instructions from the user, the server initiates procedures to make the generative AI cooperate or compete. Furthermore, the emotion engine recognizes the user's emotions and reflects this information in the generative AI's task execution.

[0771] In the case of collaboration, the server selects two GAIs to perform a specific task. Each GAI generates its own unique result for the task. The server receives these results and uses a means to integrate them to produce a single integrated result. This result is a rich work that leverages the knowledge and creativity of the different GAIs.

[0772] In the case of a competition, the server similarly selects two generative AIs to perform a specific task. Each generative AI generates a result for the competition. The server receives these results and uses a means to evaluate them to determine the winner. This evaluation process encourages competition between generative AIs, resulting in more creative and efficient results.

[0773] The emotion engine recognizes the user's emotions and adjusts the generative AI's output accordingly. This process generates content appropriate to the user's current emotional state. For example, if the user is recognized as sad, the emotion engine can instruct the generative AI to generate a comforting poem or message.

[0774] As a concrete example, consider the case where a user requests a generative AI to write a poem, while at the same time the emotion engine recognizes the emotion as "sad." The user registers two generative AIs on a server and requests the task of "writing a poem" in cooperative mode. The emotion engine sends the emotion "sad" to the server, and the server takes this into consideration and has the generative AIs execute the task. If generative AI 1 generates "a poem like tears falling on a quiet lakeside" and generative AI 2 generates "a poem that conveys the melancholy of the stars disappearing in the night sky," the server will integrate the results of both and generate a combined result: "a poem like tears falling on a quiet lakeside and a poem that conveys the melancholy of the stars disappearing in the night sky." This integrated result takes into account the user's emotional state.

[0775] Let's also consider the case where a user requests the task of generating a slogan from a generative AI in competitive mode, and at the same time, the emotion engine recognizes "joy." In this case, the user assigns the task of "creating a slogan" to two generative AIs, and the emotion engines send the emotion "joy" to the server. If generative AI 1 generates the slogan "The power to spread happiness" and generative AI 2 generates the slogan "The key to spreading smiles," the server evaluates these results and declares "The key to spreading smiles" as the winner. In this way, the emotion engines provide the optimal result according to the user's emotions.

[0776] By combining generative artificial intelligence with an emotion engine, this system can provide a more personalized experience for users and further enhance the diversity and creativity of generative artificial intelligence.

[0777] The processing flow will be explained below.

[0778] Step 1:

[0779] Users create generative AI. Users create instances of generative AI and set their names and characteristics, which allows the AI ​​to perform specific tasks.

[0780] Step 2:

[0781] The user registers the generative AI on the server. The user calls the register_ai_agent method on the server to add the generative AI to the server's database. This allows the server to manage and track the generative AI.

[0782] Step 3:

[0783] The user activates the emotion engine and recognizes emotions. The emotion engine connected to the device analyzes the user's facial expressions and voice to determine their current emotional state. This information is then sent to the server.

[0784] Step 4:

[0785] The user submits a task to the server. The user selects either collaboration or competition mode and specifies the task content, along with the emotional information recognized by the emotion engine. For example, to submit the task of "writing a poem," the user calls the create_collaboration or create_competition method.

[0786] Step 5:

[0787] The server receives the emotion information and assigns a task to the generative AI. The server takes into account the emotion information received from the emotion engine and instructs the generative AI to perform the task. The generative AI uses the generate method to generate an output that matches the emotion.

[0788] Step 6:

[0789] The server receives the output of the generative AI. Each generative AI sends the results of its task to the server. The server collects and stores these results.

[0790] Step 7:

[0791] In the collaborative case, the server combines the outputs of the generative AI. The server uses the combine_results method to combine multiple results into a single combined result, for example by adding "AND" to the beginning of the sentence.

[0792] Step 8:

[0793] In the case of a competition, the server evaluates the outputs of the generative AI. The server uses the evaluate_results method to compare the results of the competition between the generative AIs and determine the winner. Evaluation criteria can include the length of the output, creativity, etc.

[0794] Step 9:

[0795] The server provides the integrated or evaluated results to the user, and the server returns the final results to the user and displays them to the user, allowing the user to enjoy personalized content generated through the cooperation of generative artificial intelligence and the emotion engine.

[0796] For example, if a user feels sad, the emotion engine sends that information to the server. When the user requests the task of "writing a poem" in collaborative mode, the server uses this emotion information to have the generative AI execute the task. The generative AI outputs results such as "a poem like tears falling on a quiet lakeside" or "a poem that conveys the melancholy of stars disappearing in the night sky," and the server integrates these and provides them to the user. In this way, appropriate content is generated according to the user's emotions.

[0797] Example 2

[0798] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0799] Conventional generative AI systems provide fixed outputs without considering the user's emotional state, making it difficult to obtain results that are fully satisfying to the user. In addition, there is a lack of a mechanism for enabling generative AIs to cooperate or compete to complete tasks, which makes it difficult to maximize diversity and creativity.

[0800] The identification processing by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a means for registering different generative AIs, a cooperation means for the generative AIs to cooperate with each other to perform tasks, a competition means for the generative AIs to compete with each other to perform tasks, a means for integrating the results of the tasks, a means for evaluating the results of the competition, an emotion recognition means for recognizing the emotional state of the user, and a means for adjusting the output of the generative AI based on the emotional state. This allows the results of task execution to be output in a form appropriate to the user's emotional state, making it possible to provide an advanced experience that makes use of diversity and creativity.

[0801] "Generative AI" refers to an AI system that generates creative output based on user input and instructions.

[0802] "Cooperative means" refers to a means by which two or more generative artificial intelligences work together to accomplish a task.

[0803] "Competitive means" refers to a means by which two or more generative artificial intelligences independently perform the same task and compete with each other over the results they generate.

[0804] "Means for integrating results" refers to a means for combining outputs generated by multiple generative artificial intelligences into a single integrated output.

[0805] "Means for evaluating results" refers to a means for comparing outputs generated by multiple generative AIs and determining their relative merits.

[0806] "Emotion recognition means" refers to a means for recognizing the user's emotional state in real time and transmitting that information to a server.

[0807] "Means for adjusting output" refers to means for appropriately adjusting the output of the generative artificial intelligence based on the recognized emotional state of the user.

[0808] A "task" refers to a specific task or instruction that a generative artificial intelligence performs.

[0809] A "prompt sentence" refers to text input that allows a user to specify the task content and conditions in detail to a generative artificial intelligence.

[0810] MODE FOR CARRYING OUT THE INVENTION

[0811] The present invention combines an emotion recognition means for recognizing the emotions of a user with a system in which different generative artificial intelligence (hereinafter referred to as generative AI models) perform tasks through cooperation or competition and integrate or evaluate the results. Specific embodiments are described below.

[0812] 1. Creating and registering a generative AI model

[0813] Users create generative AI models to perform specific tasks and register them on a server. Generative AI models are developed using dedicated software (e.g., GPT-4 API or similar AI model development tools). The created generative AI models are assigned metadata that describes their name, characteristics, and task suitability.

[0814] 2. Recognizing user emotions using emotion recognition means

[0815] The device used by the user has a built-in camera and microphone, and emotion recognition means (e.g., Emotion API) analyzes the user's facial expressions and tone of voice through these devices. This analysis recognizes the user's emotional state in real time and transmits it to the server.

[0816] 3. Assigning tasks and specifying prompts

[0817] The user requests a task from the generative AI model through the server's web interface. The details of the task and the desired conditions are specified in a prompt. For example, the prompt could be set in the form, "When the user is feeling sad, please generate a poem to comfort them."

[0818] 4. Server selects generative AI model and assigns tasks

[0819] The server selects the most suitable model to execute the specified task from the registered generative AI models. In cooperative mode, the server assigns the same task to two generative AI models, and in competitive mode, the server assigns the task to two generative AI models in the same way.

[0820] 5. Generative AI model executes tasks and generates results

[0821] The selected generative AI model performs the task and sends its results to the server, which adjusts the output based on the user's emotional state. For example, if the user is recognized as sad, the generative AI model can be instructed to generate a comforting poem or message.

[0822] 6. Synthesis or evaluation of results

[0823] In collaborative mode, the server integrates the results received from the generative AI models, for example by generating different poems and combining them into a single integrated poem using NLP (Natural Language Processing) tools. In competitive mode, the server compares the generated results using an evaluation algorithm (e.g., a ranking model) and selects the best result.

[0824] 7. Providing Results

[0825] The final result is sent from the server to the user's device and displayed to the user. For example, the integrated result may be "A poem like tears falling on a quiet lakeside and a poem that evokes the melancholy of stars disappearing in the night sky."

[0826] Specific examples

[0827] When creating a poem (cooperative mode)

[0828] A specific example will be given in which a user requests the task of "writing a poem" and the emotion recognition means recognizes the emotion as "sad."

[0829] Example prompt sentence:

[0830] "When the user is feeling sad, generate a comforting poem."

[0831] In this case, the user registers two generative AI models on the server and requests them to create a poem in collaborative mode. The emotion recognition means detects "sadness," and the server has the two generative AI models generate a poem. Generative AI model 1 generates a poem like "tears falling on a quiet lake," while generative AI model 2 generates a poem that conveys the melancholy of stars disappearing in the night sky. The server then combines these results and displays them on the user's device.

[0832] When generating a slogan (competitive mode)

[0833] A specific example is given in which a task of creating a slogan is requested and the emotion recognition means recognizes "joy."

[0834] Example prompt sentence:

[0835] "When the user's emotion is joy, generate a slogan that emphasizes joy."

[0836] In this case, the user requests two generative AI models to create a slogan. The emotion recognition means detects "joy," and the server assigns the task to the two generative AI models. If generative AI model 1 generates the slogan "The power to spread happiness" and generative AI model 2 generates the slogan "The key to spreading smiles," the server uses an evaluation algorithm to select the best result and displays it on the user's device.

[0837] In this way, the user can obtain personalized results according to his / her emotional state.

[0838] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0839] Step 1:

[0840] Input: A user creates a generative AI model and sends its metadata (name, features, task suitability) to the server.

[0841] How it works: Users develop generative AI models using dedicated software tools (e.g., GPT-4 API) and upload them to a server through a web interface.

[0842] Output: The generative AI model registered on the server is stored in the database and displayed in the user's generative AI model list.

[0843] Step 2:

[0844] Input: Data on the user's facial expressions and voice obtained from the camera and microphone built into the user's device.

[0845] How it works: An emotion recognizer (e.g., Emotion API) analyzes this data and recognizes the user's emotional state in real time.

[0846] Output: The recognized emotional state (e.g. sad, happy) is sent to the server.

[0847] Step 3:

[0848] Input: A user-specified prompt (e.g., "When the user feels sad, please generate a comforting poem.") and task request information.

[0849] How it works: The user enters a prompt through the server's web interface to request a specific task from the generative AI model.

[0850] Output: The prompt statement and task request information received by the server are added to the processing queue.

[0851] Step 4:

[0852] Input: A list of generative AI models registered by the user and the specified task information.

[0853] How it works: The server selects the best generative AI model to perform the task. In cooperative mode, it selects two generative AI models, and in competitive mode, it selects two generative AI models.

[0854] Output: The selected generative AI model generates allocation data for a specific task.

[0855] Step 5:

[0856] Input: Assigned task information and selected generative AI model.

[0857] How it works: The server assigns tasks to each generative AI model. For example, the server assigns the same task of "writing a poem" to two generative AI models in cooperative mode.

[0858] Output: The generative AI model receives task information to be executed.

[0859] Step 6:

[0860] Input: Task results generated by the generative AI model and the user's emotional state data.

[0861] How it works: The generative AI model performs a given task and produces a result, adjusting for the user's emotional state.

[0862] Output: The generated results are sent to the server.

[0863] Step 7:

[0864] Input: The resulting data sent from the generative AI model.

[0865] How it works: The server combines the results to produce a single combined output in collaborative mode, or selects the best results using an evaluation algorithm (e.g., a ranking model) in competitive mode.

[0866] Output: Final results data that are integrated or evaluated are produced.

[0867] Step 8:

[0868] Input: Final result data.

[0869] How it works: The server sends the final result to the user's device and displays it to the user, for example, the synthesized poem or the selected slogan.

[0870] Output: The final result is displayed on the user's terminal and the user views the result.

[0871] (Application example 2)

[0872] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0873] In advertising creation, there is a demand for providing personalized content that reflects the user's emotions. Conventional advertising generation systems have difficulty taking user emotions into account, and as a result, they often generate advertisements that fail to attract the user's interest. To solve this problem, a system is needed that can properly recognize user emotions and adjust the output of generative artificial intelligence based on those emotions.

[0874] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for registering different generative AIs, a cooperation means for the generative AIs to cooperate with each other to perform tasks, a competition means for the generative AIs to compete with each other to perform tasks, a means for integrating the results of the tasks, a means for evaluating the results of the competition, a means including an emotion engine that recognizes the user's emotions, and a means for adjusting the output of the generative AI based on the user's emotions. This makes it possible to generate personalized advertising content that is optimized for the user's emotions.

[0875] ---

[0876] "Generative AI" is an AI system that can generate useful information or results based on user input.

[0877] A "collaborative means" is a method by which multiple generative artificial intelligences work together to accomplish tasks and generate results.

[0878] A "competitive method" is a method by which multiple generative artificial intelligences each carry out a task in their own unique way and compete for the results.

[0879] "Means for integrating task results" refers to a method for combining the results output by multiple generative artificial intelligences into one to generate an integrated result.

[0880] The "means for evaluating the results of the competition" is a method for comparing the results output by multiple generative artificial intelligences and selecting the optimal result.

[0881] An "emotion engine" is a system that recognizes the user's emotions from their facial expressions, voice, etc.

[0882] "Means for adjusting the output of generative artificial intelligence based on user emotions" refers to a method for adjusting the results output by generative artificial intelligence based on the user's emotional information recognized by the emotion engine.

[0883] The present invention relates to a system for generating optimal advertisements for users by combining generative artificial intelligence and an emotion engine. Specific embodiments for carrying out the present invention will be described below.

[0884] The system includes a server, a user terminal (such as a smartphone), and multiple generative AI models for registering generative AI. Users can use their own terminals to operate the generative AI and perform tasks such as creating advertisements.

[0885] First, the user enters an overview of the product or service for which they want to create an ad and related keywords into the device. Next, the emotion engine recognizes the user's emotions through the camera and microphone. The emotion engine uses, for example, AWS Rekognition or Microsoft Azure Emotion API.

[0886] The server sends a prompt message to a generative AI based on the user's emotional data and the keywords entered, and begins generating ad copy. For example, a generative AI model such as OpenAI's GPT-3 is used to generate ad copy that corresponds to the user's emotions.

[0887] The generated ad copy is then evaluated and integrated by the server, and finally output in a form that best suits the user's emotional state. This process provides personalized ad content to the user.

[0888] As a concrete example, consider a user who wants to create an ad for a new camera. If the user has the emotion "happy" along with the keywords "new camera," "high quality," and "professional," the following prompt sentence will be sent to the generative AI model:

[0889] Prompt Sentence Examples

[0890] Keywords: new camera, high quality, professional

[0891] Emotion: Joy

[0892] Generate compelling ad copy.

[0893] Based on this prompt, the generative AI model generates the following ad copy:

[0894] "A high-quality professional camera! Capture special moments from a new perspective. Why not use this camera to take a photo that will double your joy?"

[0895] The advertising copy generated in this way is optimized to the user's emotions, which is expected to increase the effectiveness of the advertisement. This system allows users to create optimal advertisements very easily.

[0896] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0897] ---

[0898] Step 1:

[0899] The user inputs into the device the summary and keywords of the product or service for which they want to create an ad. The input data collects keywords such as "new camera," "high image quality," and "professional" as well as emotional information. Terminal input: product summary, keywords, emotional information. Output: collected product summary, keywords, emotional information.

[0900] Step 2:

[0901] The emotion engine recognizes the user's current emotional state using the device's camera and microphone. The emotion engine analyzes the user's facial expressions and tone of voice to extract emotion data, for example, using AWS Rekognition or Microsoft Azure Emotion API. Device input: User's facial photo or voice data. Output: Recognized emotion data (e.g., joy).

[0902] Step 3:

[0903] The device sends the operation details and emotional data to the server. Based on the received data, the server prepares information to send a prompt to the generative AI. Server input: product summary, keywords, emotional data. Output: prompt to the generative AI model.

[0904] Step 4:

[0905] Send a prompt to a generative AI model (e.g., OpenAI GPT-3) to start generating ad copy. Example prompt: "Keywords: new camera, high quality, professional.\nEmotion: joy.\nPlease generate a compelling ad copy." Server input: prompt. Output: generated ad copy.

[0906] Step 5:

[0907] The server receives the ad copy output from the generative AI model. The server evaluates and combines the results to determine the optimal ad copy. Server input: Generated ad copy (multiple variations). Output: Optimal ad copy.

[0908] Step 6:

[0909] The optimized ad copy is sent back to the device and displayed to the user. Device input: Optimized ad copy received from the server. Output: Ad copy displayed to the user.

[0910] Step 7:

[0911] The user can review the generated ad copy and modify or approve it as needed. Terminal input: Displayed ad copy. Output: User modified or approved ad copy.

[0912] ---

[0913] This allows users to easily generate and view personalized, emotion-based ads.

[0914] 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 audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio 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 audio data.

[0915] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) 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 voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0916] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[0917] [Fourth embodiment]

[0918] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0919] 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.

[0920] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. 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).

[0921] 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.

[0922] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0923] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and 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 surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0924] 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 carried out in a secure state.

[0925] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. 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.

[0926] 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.

[0927] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. 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.

[0928] 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.

[0929] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0930] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[0931] The present invention provides a system for enabling different generative artificial intelligences to perform tasks through cooperation or competition, and for integrating or evaluating the generated results. This system promotes diversity and creativity in generative artificial intelligences and enables efficient learning. Specific embodiments of the system are described below.

[0932] First, a user creates a generative AI and registers it on the server. Each generative AI has a different name and characteristics and is designed to perform various tasks. Based on instructions from the user, the server initiates a procedure to have the generative AI cooperate or compete.

[0933] In the case of collaboration, the server selects two GAIs to perform a specific task. Each GAI generates its own unique result for the task. The server receives these results and uses a means to integrate them to produce a single integrated result. This result is a rich work that leverages the knowledge and creativity of the different GAIs.

[0934] In the case of a competition, the server similarly selects two generative AIs to perform a specific task. Each generative AI generates a result for the competition. The server receives these results and uses a means to evaluate them to determine the winner. This evaluation process encourages competition between generative AIs, resulting in more creative and efficient results.

[0935] As a concrete example, consider the case where a user requests a generative artificial intelligence to create a poem. The user registers two generative artificial intelligences on the server and requests the task of "writing a poem" in cooperative mode. The server instructs generative artificial intelligence 1 and generative artificial intelligence 2 to perform the task of "writing a poem." If generative artificial intelligence 1 generates a poem "Autumn evening, trees shining golden," and generative artificial intelligence 2 generates a poem "A calm lake surface, the wind rippling the water," the server integrates the results of both and generates the integrated result "Autumn evening, trees shining golden, a calm lake surface, the wind rippling the water." This integrated result is provided to the user.

[0936] Consider also the case where a user requests the task of generating a slogan from a generative AI in a competitive mode. The user assigns the task of "creating a slogan" to two generative AIs. If generative AI 1 generates the "power to create the future" and generative AI 2 generates the "key to open a new era," the server evaluates these results and determines the winner based on, for example, the length or creativity of the result. In this case, the "key to open a new era" is declared the winner based on an evaluation criterion that, for example, the longer the result, the better.

[0937] In this way, the system of the present invention provides a platform for generating new ideas and expressions through cooperation and competition between generative AIs, promoting interaction among generative AIs and the generation of new ideas, and achieving more advanced output.

[0938] The processing flow will be explained below.

[0939] Step 1:

[0940] The user creates a generative AI. The user creates an instance of the generative AI and sets its name and characteristics. For example, the user creates instances ai_agent1 and ai_agent2 using the AIAgent class.

[0941] Step 2:

[0942] The user registers the generative AI on the server. The user calls the register_ai_agent method on the server to add the generative AI to the server's database. This allows the server to manage and track the generative AI.

[0943] Step 3:

[0944] The user submits a task to the server. The user selects either collaboration or competition mode and specifies the task content. For example, to submit the task "write a poem," the user calls the create_collaboration or create_competition method.

[0945] Step 4:

[0946] The server assigns a task to the specified generative AI. The server passes the task to the specified generative AI, and each executes the task. The generative AI generates output for the task using the generate method.

[0947] Step 5:

[0948] The server receives the output of the generative AI. Each generative AI sends the results of its task to the server. The server collects and stores these results.

[0949] Step 6:

[0950] In the collaborative case, the server combines the outputs of the generative AI. The server uses the combine_results method to combine multiple results into a single combined result, for example by adding "AND" to the beginning of the sentence.

[0951] Step 7:

[0952] In the case of a competition, the server evaluates the outputs of the generative AI. The server uses the evaluate_results method to compare the results of the competition between the generative AIs and determine the winner. Evaluation criteria can include the length of the output, creativity, etc.

[0953] Step 8:

[0954] The server provides the synthesized or evaluated results to the user. The server returns the final result to the user and displays it to the user, which may be a synthesized poem or a slogan of the competition winner.

[0955] Through the above processing steps, users can take advantage of cooperation and competition between generative AIs to obtain a variety of ideas and methods of expression.

[0956] Example 1

[0957] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[0958] In conventional generative AI systems, different generative AIs perform tasks through cooperation or competition, but there is a lack of means to efficiently integrate or evaluate the results. It is also difficult to maximize the diversity and creativity of generative AI. As a result, task results tend to fall into a uniform pattern, and more advanced and effective output is required.

[0959] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0960] In this invention, the server includes means for registering generative AIs, cooperation means for generative AIs to cooperate with each other to perform tasks, competition means for generative AIs to compete with each other to perform tasks, means for issuing instructions to the generative AIs based on a specified task, means for receiving the results of the tasks generated by the generative AIs, means for integrating the results of the tasks, means for evaluating the results of the competition, and means for providing the generated results to a user. This makes it possible for different generative AIs to perform tasks through cooperation or competition, and for the results to be efficiently integrated or evaluated, thereby maximizing the diversity and creativity of the generative AIs.

[0961] "Generative AI" is an AI model that has the ability to generate and answer questions for given tasks.

[0962] A "server" is a computer system that registers, manages, and instructs generative artificial intelligences, integrates or evaluates results, and provides results to users.

[0963] "Registration means" refers to a method or process for registering a generative artificial intelligence on a server, and includes the function of storing necessary data.

[0964] A "collaboration means" is a method or process by which multiple generative artificial intelligences cooperate to perform a task and integrate the results.

[0965] A "competition method" is a method or process by which multiple generative artificial intelligences compete to perform tasks and evaluate the results.

[0966] "Instruction means" refers to a method or process by which the server instructs the generative artificial intelligence on tasks and prompt sentences.

[0967] "Result receiving means" refers to a method or process for returning the results generated by the generative artificial intelligence to the server.

[0968] An "integration means" is a method or process for combining the results obtained from multiple generative artificial intelligences.

[0969] An "evaluation means" is a method or process for comparing and evaluating multiple results generated by a generative artificial intelligence and determining a winner.

[0970] A "results delivery means" is a method or process for delivering the aggregated results or selected results of an evaluation to a user.

[0971] This invention provides a system for different generative artificial intelligences to perform tasks through cooperation or competition, and then integrate or evaluate the generated results. This system promotes diversity and creativity in generative artificial intelligences and enables efficient learning.

[0972] First, the user creates a generative AI and registers it on the server. The generative AI is built using a generative AI model such as GPT-3 or BERT. The user uses a device to input the name and characteristics of the generative AI, as well as information about the generative AI model to be used, and sends this information to the server. The server then stores this information in a database.

[0973] The user then submits a specific task to the server, such as writing a poem or generating a slogan. Using the terminal, the user enters the details of the task, selects cooperative or competitive mode, and specifies a prompt to be used for execution. Examples of prompts include:

[0974] Co-op prompt:

[0975] "Write a poem on the theme of autumn evenings."

[0976] Competitive mode prompt:

[0977] "Create a slogan for a new product."

[0978] The server selects two of the registered generative AIs based on the mode specified by the user and instructs each to perform a task. At this time, the server provides the specified prompt sentence to each generative AI.

[0979] The generative AIs execute tasks based on prompts provided by the server and generate results. For example, for the task of "writing a poem," generative AI 1 generates a poem about an autumn evening, with trees shining golden, while generative AI 2 generates a poem about a calm lake surface, with the wind rippling the water. These results are then sent back to the server.

[0980] In cooperative mode, the server runs an algorithm to combine the generated results and generate a single combined result, such as "Autumn evening, the trees shine golden, the surface of the lake is calm, the wind ripples the surface of the water."

[0981] In competitive mode, the server evaluates each generated result and determines the winner. Evaluation criteria include the creativity and length of the result. For example, the server evaluates "Power to Create the Future" and "Key to Open a New Era" and determines the winner as "Key to Open a New Era" based on the length and creativity of the result.

[0982] Finally, the server provides the integrated results or the results selected by the evaluation to the user, who can receive the results through the terminal and decide on the next action.

[0983] The following specific hardware and software is used to implement this system:

[0984] Hardware used:

[0985] Server: AWS EC2 instance

[0986] Software used:

[0987] Generative AI: Generative AI models like GPT-3 and BERT

[0988] Integration and evaluation algorithms: Unique integration algorithms and evaluation logic implemented in Python

[0989] In this way, the present invention promotes new ideas and expressions through cooperation and competition between different generative artificial intelligences, making it possible to provide more advanced output to users.

[0990] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0991] Step 1:

[0992] The user registers the generative artificial intelligence on the server.

[0993] Specific operation: The user accesses the server from their device and enters information such as the name and characteristics of the generative AI, the generative AI model to be used (e.g., GPT-3), etc. The information entered by the user is sent to the server, which then stores it in a database.

[0994] Input: Name of generative artificial intelligence, characteristics, information of the generative AI model to be used

[0995] Output: Generative artificial intelligence information stored in a database

[0996] Step 2:

[0997] A user requests a specific task from the server.

[0998] Specific operation: The user uses the terminal to enter the details of the task, select cooperative or competitive mode, and send a prompt to the server to be used for execution.

[0999] Input: Task details, cooperative or competitive mode selection, prompt

[1000] Output: Task request information saved on the server

[1001] Step 3:

[1002] The server selects a generative AI and gives it instructions for the task.

[1003] Specific operation: The server selects two generative AIs from the database and provides each with a designated prompt sentence, which the server then sends to the generative AIs via an API call.

[1004] Input: Task request information, prompt text

[1005] Output: Task instructions for generative AI

[1006] Step 4:

[1007] Generative artificial intelligence performs tasks and generates results.

[1008] Specific operation: The generative artificial intelligence performs tasks based on prompts provided by the server and generates results, which are then sent back to the server.

[1009] Input: prompt statement

[1010] Output: The results generated by generative artificial intelligence

[1011] Step 5:

[1012] The server aggregates or evaluates the results.

[1013] Specific operation: In cooperative mode, the server runs an algorithm that combines multiple generated results to produce a single combined result. In competitive mode, the server evaluates the generated results and determines a winner. Evaluation criteria include the creativity and length of the results.

[1014] Input: Results received from the generative artificial intelligence

[1015] Output: Consolidated results or winners by rating

[1016] Step 6:

[1017] The results are presented to the user.

[1018] Specific operation: The server sends the integrated results or the results selected by the evaluation to the user and displays them on the user's device, so that the user can decide on the next action based on them.

[1019] Input: Winners by combined results or ratings

[1020] Output: The final result provided to the user

[1021] (Application example 1)

[1022] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1023] The modern advertising industry requires the rapid generation of more effective and creative slogans and advertising materials. However, conventional methods require a large amount of human resources and time, making efficient generation difficult. Furthermore, limited means for generating new ideas and expressions often result in a lack of quality and variety in advertisements. Therefore, there is a need for a system that utilizes generative artificial intelligence to generate efficient and creative advertising slogans.

[1024] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1025] In this invention, the server includes means for registering different generative AIs, cooperation means for the generative AIs to cooperate with each other to perform tasks, competition means for the generative AIs to compete with each other to perform tasks, means for integrating the results of the tasks, means for evaluating the results of the competition, means for inputting a prompt sentence from the user, and means for generating catchy slogans generated by the generative AIs for advertising media. This enables the rapid generation of high-quality and diverse advertising copy by having different generative AIs generate advertising catchy slogans through cooperation or competition and integrating or evaluating the results.

[1026] "Generative artificial intelligence" is an artificial intelligence system that generates information based on user prompts.

[1027] The "registration means" is a means having a function for registering different generative artificial intelligences in the server.

[1028] A "cooperative means" is a means having the function of allowing generative artificial intelligences to cooperate with each other to carry out tasks.

[1029] A "competition means" is a means that has the function of allowing generative artificial intelligences to compete with each other to perform tasks.

[1030] An "integration means" is a means having the function of integrating multiple results generated by a generative artificial intelligence into one.

[1031] The "evaluation means" is a means having a function for evaluating the results generated by the competition and selecting the superior results.

[1032] The "input means" is a means having a function for a user to input a prompt sentence to the server.

[1033] The "generation means" is a means having the function of generating a catchy slogan generated by the generative artificial intelligence for an advertising medium.

[1034] This invention relates to a system for generating effective advertising slogans through cooperation and competition between different generative artificial intelligences. This system can improve the quality and efficiency of advertising campaigns by having generative artificial intelligences cooperate to generate advertising materials or compete to select the best results.

[1035] System Program

[1036] The server contains the following main facilities:

[1037] 1. A means of registering different generative AIs

[1038] 2. A means of cooperation for generative AIs to work together to accomplish tasks

[1039] 3. A means of competition for generative AIs to perform tasks in competition with each other

[1040] 4. Means of integrating task results

[1041] 5. Means of assessing the results of the competition

[1042] 6. A means of inputting a prompt from the user

[1043] 7. A method for generating catchphrases generated by generative AI for advertising media

[1044] Program processing

[1045] The user inputs a prompt sentence, for example, a prompt sentence that generates an "advertising catchphrase for a new smartphone." This prompt sentence is sent to the server via the user terminal.

[1046] The server uses different generative AI models (such as GPT-2) to generate catchphrases based on the input prompt. Each generative AI model generates a unique catchphrase, resulting in a wide variety of ideas.

[1047] In the collaborative mode, a means for integrating these catchphrases is activated, combining or arranging multiple catchphrases to generate a single integrated result. For example, the integrated result "Experience the next generation of smart life. The technology of the future is here" may be generated as an "advertising catchphrase for a new smartphone."

[1048] In the competitive mode, the server evaluates multiple generated slogans and selects the most effective one. For example, between the slogans "Experience the next generation of smart life" and "The technology of the future is here," the server selects the best one based on the evaluation criteria.

[1049] This allows users to obtain high-quality advertising copy generated through the cooperation and competition of different generative AIs. The main software used is a generative AI model (e.g., GPT-2), TensorFlow, and the Hugging Face Transformers library. For hardware, a server equipped with a high-performance CPU or GPU is used.

[1050] As an example, suppose the user enters the following prompt:

[1051] Example prompt: "What beautiful hair can you achieve with your new hair care products?"

[1052] In this way, generative AI models can be used cooperatively and competitively to generate sophisticated advertising slogans.

[1053] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1054] Step 1:

[1055] The user inputs a prompt sentence. For example, a prompt sentence that generates an "advertising catchphrase for a new smartphone" is input from the user terminal and sent to the server. In this step, the input data is the prompt sentence, and the output is the prompt sentence sent to the server.

[1056] Step 2:

[1057] The server analyzes the received prompt sentence and prepares it to be passed to the generative AI model. Specifically, it tokenizes the prompt sentence and converts it into a format that the generative AI model can understand. In this step, the input data is the prompt sentence sent by the user, and the output is the tokenized data.

[1058] Step 3:

[1059] The server inputs the tokenized data into different generative artificial intelligence models and generates catchphrases based on each model. For example, Model1 and Model2 are used, each of which generates its own catchphrase. In this step, the input data is the tokenized prompt sentence, and the output is the generated multiple catchphrases.

[1060] Step 4:

[1061] In the collaborative mode, the server integrates the generated catchphrases. Specifically, it combines or organizes multiple catchphrases to generate a single integrated result. For example, the integrated result is "Experience the next generation of smart life. The technology of the future is here." In this step, the input data are multiple catchphrases, and the output is the integrated catchphrase.

[1062] Step 5:

[1063] In the competitive mode, the server evaluates the generated copy and selects the most effective one. It compares each copy based on evaluation criteria, such as copy length or creativity, and selects the best copy. In this step, the input data are multiple copy titles, and the output is the copy that is evaluated as the best.

[1064] Step 6:

[1065] The server sends the final tagline to the user terminal, so that the user can receive the generated advertising tagline. In this step, the input data is the synthesized tagline or the evaluated tagline, and the output is the tagline sent to the user terminal.

[1066] As a specific example of operation, when the prompt sentence "What kind of beautiful hair can you get by using new hair care products?" is input and processing is performed in collaborative mode, the server receives the output from both generative artificial intelligence models, combines them, and generates an integrated result "Use new hair care products to get vibrant, shiny, beautiful hair," which is provided to the user.

[1067] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1068] The present invention provides a more advanced experience by combining a system in which generative artificial intelligences execute tasks through cooperation and competition with each other and integrate or evaluate the generated results with an emotion engine that recognizes the user's emotions. This system promotes diversity and creativity in generative artificial intelligences and is capable of generating output in response to the user's emotions. Specific embodiments of the system are described below.

[1069] First, the user creates a generative AI and registers it on the server. Each generative AI has a different name and characteristics and is designed to perform various tasks. Based on instructions from the user, the server initiates procedures to make the generative AI cooperate or compete. Furthermore, the emotion engine recognizes the user's emotions and reflects this information in the generative AI's task execution.

[1070] In the case of collaboration, the server selects two GAIs to perform a specific task. Each GAI generates its own unique result for the task. The server receives these results and uses a means to integrate them to produce a single integrated result. This result is a rich work that leverages the knowledge and creativity of the different GAIs.

[1071] In the case of a competition, the server similarly selects two generative AIs to perform a specific task. Each generative AI generates a result for the competition. The server receives these results and uses a means to evaluate them to determine the winner. This evaluation process encourages competition between generative AIs, resulting in more creative and efficient results.

[1072] The emotion engine recognizes the user's emotions and adjusts the generative AI's output accordingly. This process generates content appropriate to the user's current emotional state. For example, if the user is recognized as sad, the emotion engine can instruct the generative AI to generate a comforting poem or message.

[1073] As a concrete example, consider the case where a user requests a generative AI to write a poem, while at the same time the emotion engine recognizes the emotion as "sad." The user registers two generative AIs on a server and requests the task of "writing a poem" in cooperative mode. The emotion engine sends the emotion "sad" to the server, and the server takes this into consideration and has the generative AIs execute the task. If generative AI 1 generates "a poem like tears falling on a quiet lakeside" and generative AI 2 generates "a poem that conveys the melancholy of the stars disappearing in the night sky," the server will integrate the results of both and generate a combined result: "a poem like tears falling on a quiet lakeside and a poem that conveys the melancholy of the stars disappearing in the night sky." This integrated result takes into account the user's emotional state.

[1074] Let's also consider the case where a user requests the task of generating a slogan from a generative AI in competitive mode, and at the same time, the emotion engine recognizes "joy." In this case, the user assigns the task of "creating a slogan" to two generative AIs, and the emotion engines send the emotion "joy" to the server. If generative AI 1 generates the slogan "The power to spread happiness" and generative AI 2 generates the slogan "The key to spreading smiles," the server evaluates these results and declares "The key to spreading smiles" as the winner. In this way, the emotion engines provide the optimal result according to the user's emotions.

[1075] By combining generative artificial intelligence with an emotion engine, this system can provide a more personalized experience for users and further enhance the diversity and creativity of generative artificial intelligence.

[1076] The processing flow will be explained below.

[1077] Step 1:

[1078] Users create generative AI. Users create instances of generative AI and set their names and characteristics, which allows the AI ​​to perform specific tasks.

[1079] Step 2:

[1080] The user registers the generative AI on the server. The user calls the register_ai_agent method on the server to add the generative AI to the server's database. This allows the server to manage and track the generative AI.

[1081] Step 3:

[1082] The user activates the emotion engine and recognizes emotions. The emotion engine connected to the device analyzes the user's facial expressions and voice to determine their current emotional state. This information is then sent to the server.

[1083] Step 4:

[1084] The user submits a task to the server. The user selects either collaboration or competition mode and specifies the task content, along with the emotional information recognized by the emotion engine. For example, to submit the task of "writing a poem," the user calls the create_collaboration or create_competition method.

[1085] Step 5:

[1086] The server receives the emotion information and assigns a task to the generative AI. The server takes into account the emotion information received from the emotion engine and instructs the generative AI to perform the task. The generative AI uses the generate method to generate an output that matches the emotion.

[1087] Step 6:

[1088] The server receives the output of the generative AI. Each generative AI sends the results of its task to the server. The server collects and stores these results.

[1089] Step 7:

[1090] In the collaborative case, the server combines the outputs of the generative AI. The server uses the combine_results method to combine multiple results into a single combined result, for example by adding "AND" to the beginning of the sentence.

[1091] Step 8:

[1092] In the case of a competition, the server evaluates the outputs of the generative AI. The server uses the evaluate_results method to compare the results of the competition between the generative AIs and determine the winner. Evaluation criteria can include the length of the output, creativity, etc.

[1093] Step 9:

[1094] The server provides the integrated or evaluated results to the user, and the server returns the final results to the user and displays them to the user, allowing the user to enjoy personalized content generated through the cooperation of generative artificial intelligence and the emotion engine.

[1095] For example, if a user feels sad, the emotion engine sends that information to the server. When the user requests the task of "writing a poem" in collaborative mode, the server uses this emotion information to have the generative AI execute the task. The generative AI outputs results such as "a poem like tears falling on a quiet lakeside" or "a poem that conveys the melancholy of stars disappearing in the night sky," and the server integrates these and provides them to the user. In this way, appropriate content is generated according to the user's emotions.

[1096] Example 2

[1097] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1098] Conventional generative AI systems provide fixed outputs without considering the user's emotional state, making it difficult to obtain results that are fully satisfying to the user. In addition, there is a lack of a mechanism for enabling generative AIs to cooperate or compete to complete tasks, which makes it difficult to maximize diversity and creativity.

[1099] The identification processing by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a means for registering different generative AIs, a cooperation means for the generative AIs to cooperate with each other to perform tasks, a competition means for the generative AIs to compete with each other to perform tasks, a means for integrating the results of the tasks, a means for evaluating the results of the competition, an emotion recognition means for recognizing the emotional state of the user, and a means for adjusting the output of the generative AI based on the emotional state. This allows the results of task execution to be output in a form appropriate to the user's emotional state, making it possible to provide an advanced experience that makes use of diversity and creativity.

[1100] "Generative AI" refers to an AI system that generates creative output based on user input and instructions.

[1101] "Cooperative means" refers to a means by which two or more generative artificial intelligences work together to accomplish a task.

[1102] "Competitive means" refers to a means by which two or more generative artificial intelligences independently perform the same task and compete with each other over the results they generate.

[1103] "Means for integrating results" refers to a means for combining outputs generated by multiple generative artificial intelligences into a single integrated output.

[1104] "Means for evaluating results" refers to a means for comparing outputs generated by multiple generative AIs and determining their relative merits.

[1105] "Emotion recognition means" refers to a means for recognizing the user's emotional state in real time and transmitting that information to a server.

[1106] "Means for adjusting output" refers to means for appropriately adjusting the output of the generative artificial intelligence based on the recognized emotional state of the user.

[1107] A "task" refers to a specific task or instruction that a generative artificial intelligence performs.

[1108] A "prompt sentence" refers to text input that allows a user to specify the task content and conditions in detail to a generative artificial intelligence.

[1109] MODE FOR CARRYING OUT THE INVENTION

[1110] The present invention combines an emotion recognition means for recognizing the emotions of a user with a system in which different generative artificial intelligence (hereinafter referred to as generative AI models) perform tasks through cooperation or competition and integrate or evaluate the results. Specific embodiments are described below.

[1111] 1. Creating and registering a generative AI model

[1112] Users create generative AI models to perform specific tasks and register them on a server. Generative AI models are developed using dedicated software (e.g., GPT-4 API or similar AI model development tools). The created generative AI models are assigned metadata that describes their name, characteristics, and task suitability.

[1113] 2. Recognizing user emotions using emotion recognition means

[1114] The device used by the user has a built-in camera and microphone, and emotion recognition means (e.g., Emotion API) analyzes the user's facial expressions and tone of voice through these devices. This analysis recognizes the user's emotional state in real time and transmits it to the server.

[1115] 3. Assigning tasks and specifying prompts

[1116] The user requests a task from the generative AI model through the server's web interface. The details of the task and the desired conditions are specified in a prompt. For example, the prompt could be set in the form, "When the user is feeling sad, please generate a poem to comfort them."

[1117] 4. Server selects generative AI model and assigns tasks

[1118] The server selects the most suitable model to execute the specified task from the registered generative AI models. In cooperative mode, the server assigns the same task to two generative AI models, and in competitive mode, the server assigns the task to two generative AI models in the same way.

[1119] 5. Generative AI model executes tasks and generates results

[1120] The selected generative AI model performs the task and sends its results to the server, which adjusts the output based on the user's emotional state. For example, if the user is recognized as sad, the generative AI model can be instructed to generate a comforting poem or message.

[1121] 6. Synthesis or evaluation of results

[1122] In collaborative mode, the server integrates the results received from the generative AI models, for example by generating different poems and combining them into a single integrated poem using NLP (Natural Language Processing) tools. In competitive mode, the server compares the generated results using an evaluation algorithm (e.g., a ranking model) and selects the best result.

[1123] 7. Providing Results

[1124] The final result is sent from the server to the user's device and displayed to the user. For example, the integrated result may be "A poem like tears falling on a quiet lakeside and a poem that evokes the melancholy of stars disappearing in the night sky."

[1125] Specific examples

[1126] When creating a poem (cooperative mode)

[1127] A specific example will be given in which a user requests the task of "writing a poem" and the emotion recognition means recognizes the emotion as "sad."

[1128] Example prompt sentence:

[1129] "When the user is feeling sad, generate a comforting poem."

[1130] In this case, the user registers two generative AI models on the server and requests them to create a poem in collaborative mode. The emotion recognition means detects "sadness," and the server has the two generative AI models generate a poem. Generative AI model 1 generates a poem like "tears falling on a quiet lake," while generative AI model 2 generates a poem that conveys the melancholy of stars disappearing in the night sky. The server then combines these results and displays them on the user's device.

[1131] When generating a slogan (competitive mode)

[1132] A specific example is given in which a task of creating a slogan is requested and the emotion recognition means recognizes "joy."

[1133] Example prompt sentence:

[1134] "When the user's emotion is joy, generate a slogan that emphasizes joy."

[1135] In this case, the user requests two generative AI models to create a slogan. The emotion recognition means detects "joy," and the server assigns the task to the two generative AI models. If generative AI model 1 generates the slogan "The power to spread happiness" and generative AI model 2 generates the slogan "The key to spreading smiles," the server uses an evaluation algorithm to select the best result and displays it on the user's device.

[1136] In this way, the user can obtain personalized results according to his / her emotional state.

[1137] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1138] Step 1:

[1139] Input: A user creates a generative AI model and sends its metadata (name, features, task suitability) to the server.

[1140] How it works: Users develop generative AI models using dedicated software tools (e.g., GPT-4 API) and upload them to a server through a web interface.

[1141] Output: The generative AI model registered on the server is stored in the database and displayed in the user's generative AI model list.

[1142] Step 2:

[1143] Input: Data on the user's facial expressions and voice obtained from the camera and microphone built into the user's device.

[1144] How it works: An emotion recognizer (e.g., Emotion API) analyzes this data and recognizes the user's emotional state in real time.

[1145] Output: The recognized emotional state (e.g. sad, happy) is sent to the server.

[1146] Step 3:

[1147] Input: A user-specified prompt (e.g., "When the user feels sad, please generate a comforting poem.") and task request information.

[1148] How it works: The user enters a prompt through the server's web interface to request a specific task from the generative AI model.

[1149] Output: The prompt statement and task request information received by the server are added to the processing queue.

[1150] Step 4:

[1151] Input: A list of generative AI models registered by the user and the specified task information.

[1152] How it works: The server selects the best generative AI model to perform the task. In cooperative mode, it selects two generative AI models, and in competitive mode, it selects two generative AI models.

[1153] Output: The selected generative AI model generates allocation data for a specific task.

[1154] Step 5:

[1155] Input: Assigned task information and selected generative AI model.

[1156] How it works: The server assigns tasks to each generative AI model. For example, the server assigns the same task of "writing a poem" to two generative AI models in cooperative mode.

[1157] Output: The generative AI model receives task information to be executed.

[1158] Step 6:

[1159] Input: Task results generated by the generative AI model and the user's emotional state data.

[1160] How it works: The generative AI model performs a given task and produces a result, adjusting for the user's emotional state.

[1161] Output: The generated results are sent to the server.

[1162] Step 7:

[1163] Input: The resulting data sent from the generative AI model.

[1164] How it works: The server combines the results to produce a single combined output in collaborative mode, or selects the best results using an evaluation algorithm (e.g., a ranking model) in competitive mode.

[1165] Output: Final results data that are integrated or evaluated are produced.

[1166] Step 8:

[1167] Input: Final result data.

[1168] How it works: The server sends the final result to the user's device and displays it to the user, for example, the synthesized poem or the selected slogan.

[1169] Output: The final result is displayed on the user's terminal and the user views the result.

[1170] (Application example 2)

[1171] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1172] In advertising creation, there is a demand for providing personalized content that reflects the user's emotions. Conventional advertising generation systems have difficulty taking user emotions into account, and as a result, they often generate advertisements that fail to attract the user's interest. To solve this problem, a system is needed that can properly recognize user emotions and adjust the output of generative artificial intelligence based on those emotions.

[1173] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for registering different generative AIs, a cooperation means for the generative AIs to cooperate with each other to perform tasks, a competition means for the generative AIs to compete with each other to perform tasks, a means for integrating the results of the tasks, a means for evaluating the results of the competition, a means including an emotion engine that recognizes the user's emotions, and a means for adjusting the output of the generative AI based on the user's emotions. This makes it possible to generate personalized advertising content that is optimized for the user's emotions.

[1174] ---

[1175] "Generative AI" is an AI system that can generate useful information or results based on user input.

[1176] A "collaborative means" is a method by which multiple generative artificial intelligences work together to accomplish tasks and generate results.

[1177] A "competitive method" is a method by which multiple generative artificial intelligences each carry out a task in their own unique way and compete for the results.

[1178] "Means for integrating task results" refers to a method for combining the results output by multiple generative artificial intelligences into one to generate an integrated result.

[1179] The "means for evaluating the results of the competition" is a method for comparing the results output by multiple generative artificial intelligences and selecting the optimal result.

[1180] An "emotion engine" is a system that recognizes the user's emotions from their facial expressions, voice, etc.

[1181] "Means for adjusting the output of generative artificial intelligence based on user emotions" refers to a method for adjusting the results output by generative artificial intelligence based on the user's emotional information recognized by the emotion engine.

[1182] The present invention relates to a system for generating optimal advertisements for users by combining generative artificial intelligence and an emotion engine. Specific embodiments for carrying out the present invention will be described below.

[1183] The system includes a server, a user terminal (such as a smartphone), and multiple generative AI models for registering generative AI. Users can use their own terminals to operate the generative AI and perform tasks such as creating advertisements.

[1184] First, the user enters an overview of the product or service for which they want to create an ad and related keywords into the device. Next, the emotion engine recognizes the user's emotions through the camera and microphone. The emotion engine uses, for example, AWS Rekognition or Microsoft Azure Emotion API.

[1185] The server sends a prompt message to a generative AI based on the user's emotional data and the keywords entered, and begins generating ad copy. For example, a generative AI model such as OpenAI's GPT-3 is used to generate ad copy that corresponds to the user's emotions.

[1186] The generated ad copy is then evaluated and integrated by the server, and finally output in a form that best suits the user's emotional state. This process provides personalized ad content to the user.

[1187] As a concrete example, consider a user who wants to create an ad for a new camera. If the user has the emotion "happy" along with the keywords "new camera," "high quality," and "professional," the following prompt sentence will be sent to the generative AI model:

[1188] Prompt Sentence Examples

[1189] Keywords: new camera, high quality, professional

[1190] Emotion: Joy

[1191] Generate compelling ad copy.

[1192] Based on this prompt, the generative AI model generates the following ad copy:

[1193] "A high-quality professional camera! Capture special moments from a new perspective. Why not use this camera to take a photo that will double your joy?"

[1194] The advertising copy generated in this way is optimized to the user's emotions, which is expected to increase the effectiveness of the advertisement. This system allows users to create optimal advertisements very easily.

[1195] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1196] ---

[1197] Step 1:

[1198] The user inputs into the device the summary and keywords of the product or service for which they want to create an ad. The input data collects keywords such as "new camera," "high image quality," and "professional" as well as emotional information. Terminal input: product summary, keywords, emotional information. Output: collected product summary, keywords, emotional information.

[1199] Step 2:

[1200] The emotion engine recognizes the user's current emotional state using the device's camera and microphone. The emotion engine analyzes the user's facial expressions and tone of voice to extract emotion data, for example, using AWS Rekognition or Microsoft Azure Emotion API. Device input: User's facial photo or voice data. Output: Recognized emotion data (e.g., joy).

[1201] Step 3:

[1202] The device sends the operation details and emotional data to the server. Based on the received data, the server prepares information to send a prompt to the generative AI. Server input: product summary, keywords, emotional data. Output: prompt to the generative AI model.

[1203] Step 4:

[1204] Send a prompt to a generative AI model (e.g., OpenAI GPT-3) to start generating ad copy. Example prompt: "Keywords: new camera, high quality, professional.\nEmotion: joy.\nPlease generate a compelling ad copy." Server input: prompt. Output: generated ad copy.

[1205] Step 5:

[1206] The server receives the ad copy output from the generative AI model. The server evaluates and combines the results to determine the optimal ad copy. Server input: Generated ad copy (multiple variations). Output: Optimal ad copy.

[1207] Step 6:

[1208] The optimized ad copy is sent back to the device and displayed to the user. Device input: Optimized ad copy received from the server. Output: Ad copy displayed to the user.

[1209] Step 7:

[1210] The user can review the generated ad copy and modify or approve it as needed. Terminal input: Displayed ad copy. Output: User modified or approved ad copy.

[1211] ---

[1212] This allows users to easily generate and view personalized, emotion-based ads.

[1213] 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 voice indicating a user input regarding 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.

[1214] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) 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 voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1215] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1216] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1217] FIG. 9 is a diagram illustrating 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. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1218] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically 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.

[1219] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1220] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, motorcycles, and other devices, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1221] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs 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 a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1222] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of 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, "relieved," "calm," and "reassuring," have similar emotion values.

[1223] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1224] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1225] 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 USB (Universal Serial Bus) 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.

[1226] Alternatively, 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 in response to a request from the data processing device 12.

[1227] 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.

[1228] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1229] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with 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 processing may be a single processor.

[1230] As an example of a system configured with a single 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 system that uses a processor that realizes the functions of an 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.

[1231] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1232] The above-described description and illustrations are a detailed explanation 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, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects 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 replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1233] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1234] The following is further disclosed regarding the above embodiment.

[1235] (Claim 1)

[1236] a means for registering different generative artificial intelligences;

[1237] A means of cooperation for generative artificial intelligences to cooperate with each other to perform tasks;

[1238] A means for generating artificial intelligences to compete with each other to perform tasks;

[1239] a means of integrating the results of the tasks;

[1240] a means of assessing the results of the competition;

[1241] A system including:

[1242] (Claim 2)

[1243] 2. The system of claim 1, wherein the collaboration means includes means for integrating results of tasks generated by the generative artificial intelligence.

[1244] (Claim 3)

[1245] 2. The system of claim 1, wherein the competition means includes means for comparatively evaluating the results of the tasks generated by the generative artificial intelligence.

[1246] "Example 1"

[1247] (Claim 1)

[1248] a means for registering the generative artificial intelligence;

[1249] A means of cooperation for generative artificial intelligences to cooperate with each other to perform tasks;

[1250] A means for generating artificial intelligences to compete with each other to perform tasks;

[1251] a means for instructing the generative artificial intelligence based on a specified task;

[1252] A means for receiving the results of the task generated by the generative artificial intelligence;

[1253] a means of integrating the results of the tasks;

[1254] a means of assessing the results of the competition;

[1255] means for providing the generated results to a user;

[1256] A system including:

[1257] (Claim 2)

[1258] 2. The system of claim 1, wherein the collaboration means includes means for integrating results of tasks generated by the generative artificial intelligence.

[1259] (Claim 3)

[1260] 2. The system of claim 1, wherein the competition means includes means for comparatively evaluating the results of the tasks generated by the generative artificial intelligence.

[1261] "Application Example 1"

[1262] (Claim 1)

[1263] a means for registering different generative artificial intelligences;

[1264] A means of cooperation for generative artificial intelligences to cooperate with each other to perform tasks;

[1265] A means for generating artificial intelligences to compete with each other to perform tasks;

[1266] a means of integrating the results of the tasks;

[1267] a means of assessing the results of the competition;

[1268] a means for inputting a prompt sentence from a user;

[1269] A means for generating a catchphrase generated by the generative artificial intelligence for an advertising medium;

[1270] A system including:

[1271] (Claim 2)

[1272] 2. The system of claim 1, wherein the collaboration means includes means for integrating results of tasks generated by the generative artificial intelligence.

[1273] (Claim 3)

[1274] 2. The system of claim 1, wherein the competition means includes means for comparatively evaluating the results of the tasks generated by the generative artificial intelligence.

[1275] "Example 2: Combining Emotion Engines"

[1276] (Claim 1)

[1277] a means for registering different generative artificial intelligences;

[1278] A means of cooperation for generative artificial intelligences to cooperate with each other to perform tasks;

[1279] A means for generating artificial intelligences to compete with each other to perform tasks;

[1280] a means of integrating the results of the tasks;

[1281] a means of assessing the results of the competition;

[1282] emotion recognition means for recognizing an emotional state of a user;

[1283] A means for adjusting the output of the generative artificial intelligence based on the emotional state; and

[1284] A system including:

[1285] (Claim 2)

[1286] 2. The system of claim 1, wherein the collaboration means includes means for integrating results of tasks generated by the generative artificial intelligence.

[1287] (Claim 3)

[1288] 2. The system of claim 1, wherein the competition means includes means for comparatively evaluating the results of the tasks generated by the generative artificial intelligence.

[1289] (Claim 4)

[1290] 2. The system according to claim 1, wherein the emotion recognition means recognizes the user's emotion using a camera or a microphone of the user's terminal.

[1291] (Claim 5)

[1292] 2. The system according to claim 1, wherein the task request is specified by the user as a prompt sentence.

[1293] "Application example 2 when combining emotion engines"

[1294] (Claim 1)

[1295] a means for registering different generative artificial intelligences;

[1296] A means of cooperation for generative artificial intelligences to cooperate with each other to perform tasks;

[1297] A means for generating artificial intelligences to compete with each other to perform tasks;

[1298] a means of integrating the results of the tasks;

[1299] a means of assessing the results of the competition;

[1300] means including an emotion engine for recognizing an emotion of a user;

[1301] A means for adjusting the output of the generative artificial intelligence based on the user's emotions;

[1302] A system including:

[1303] (Claim 2)

[1304] 2. The system of claim 1, wherein the collaboration means includes means for integrating results of tasks generated by the generative artificial intelligence.

[1305] (Claim 3)

[1306] 2. The system of claim 1, wherein the competition means includes means for comparatively evaluating the results of the tasks generated by the generative artificial intelligence. [Explanation of symbols]

[1307] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means for registering different generative artificial intelligences; A means of cooperation for generative artificial intelligences to cooperate with each other to perform tasks; A means for generating artificial intelligences to compete with each other to perform tasks; a means of integrating the results of the tasks; a means of assessing the results of the competition; A system including:

2. 2. The system of claim 1, wherein the collaboration means includes means for integrating results of tasks generated by the generative artificial intelligence.

3. 2. The system according to claim 1, wherein the competition means includes means for comparatively evaluating the results of the tasks generated by the generative artificial intelligence.

Images