System

Abstract

An object of a system according to an embodiment is to propose an optimal cocktail based on a customer's desire.SOLUTION: A system includes a reception part, a generation part, an output part, a feedback collection part, and a fine tuning part. The reception unit inputs a desire of a customer. The generation unit analyzes the information input by the reception unit and generates a recipe of an appropriate cocktail from a large-scale database. The output unit outputs the recipe of the cocktail generated by the generation unit in a moving image format. The output unit outputs the appearance of the cocktail generated by the generation unit in an image format. The output unit outputs the name of the cocktail generated by the generation unit in a text format. The feedback collection unit collects feedback on the recipe of the cocktail output by the output unit. The fine-tuning unit fine-tunes the model of the generative AI based on the feedback collected by the feedback collection unit.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including 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 technology makes it difficult to suggest the best cocktail based on a customer's preferences, and there is room for improvement to meet individual needs.

[0005] The system according to the embodiment aims to propose the best cocktail based on the customer's wishes. [Means for solving the problem]

[0006] The system according to the embodiment includes a reception unit, a generation unit, an output unit, a feedback collection unit, and a fine-tuning unit. The reception unit inputs a customer's wishes. The generation unit analyzes the information input by the reception unit and generates an appropriate cocktail recipe from a large database. The output unit outputs the cocktail recipe generated by the generation unit in video format. The output unit outputs the appearance of the cocktail generated by the generation unit in image format. The output unit outputs the name of the cocktail generated by the generation unit in text format. The feedback collection unit collects feedback on the cocktail recipe output by the output unit. The fine-tuning unit fine-tunes the generation AI model based on the feedback collected by the feedback collection unit. [Effects of the Invention]

[0007] The system according to the embodiment can suggest the best cocktail based on the customer's wishes. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [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. DETAILED DESCRIPTION OF THE INVENTION

[0009] 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.

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

[0011] 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, the 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), an APU (Accelerated Processing Unit), or a TPU (Tensor Processing Unit).

[0012] 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.

[0013] 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.

[0014] 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), and Bluetooth (registered trademark).

[0015] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. 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."

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

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

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

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

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

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

[0022] 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.

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

[0024] 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.

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

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

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

[0028] (Example 1) A cocktail generation system according to an embodiment of the present invention takes a customer's preferences as input and generates an optimal cocktail from a vast database. The cocktail generation system analyzes the customer's preferences using a generation AI to generate an optimal cocktail recipe. The generated cocktail recipe is output in video format, the cocktail's appearance is output in image format, and the cocktail's name is output in text format. Furthermore, by collecting feedback and fine-tuning the generation AI model, it is possible to provide personalized suggestions that reflect the bartender's characteristics and the customer's preferences. For example, a cocktail generation system inputs a customer's preferences, such as "I want a fruity and sweet cocktail." This information is input into the generation AI. The generation AI then analyzes the input information and generates an optimal cocktail recipe from a vast database. The generation AI generates an optimal cocktail recipe based on past cocktail recipes and customer feedback. For example, it generates a recipe for a fruity and sweet cocktail. The generated cocktail recipe is output in video format. For example, a step-by-step video showing how to make the cocktail is generated. The cocktail's appearance is output in image format. For example, an image of the finished cocktail is generated. The cocktail's name is output in text format. For example, a name such as "Fruity Sweet Cocktail" is generated. Finally, feedback from bartenders and customers is collected to fine-tune the generative AI model. This allows the cocktail generation system to make personalized suggestions that reflect the bartender's characteristics and the customer's preferences. This allows the cocktail generation system to generate the optimal cocktail based on the customer's wishes and improve its performance by incorporating feedback. For example, bartenders can easily create original cocktails and provide cocktails that satisfy customers.

[0029] A cocktail creation system according to an embodiment includes a reception unit, a generation unit, an output unit, a feedback collection unit, and a fine-tuning unit. The reception unit inputs customer preferences. The customer preferences include, but are not limited to, taste preferences, alcohol strength, and the presence or absence of specific ingredients. The reception unit can receive the customer preferences via, for example, text input, voice input, or image input. The generation unit uses a generation AI to analyze the information input by the reception unit and generate an optimal cocktail recipe from a vast database. The generation AI generates the optimal cocktail recipe based on, for example, past cocktail recipes and customer feedback. The generation unit generates the cocktail recipe using, for example, a text generation AI (e.g., LLM). The generation unit can also generate the cocktail recipe using a multimodal generation AI. The generation unit can also generate the optimal cocktail recipe based on, for example, past cocktail recipes and customer feedback. The output unit outputs the cocktail recipe generated by the generation unit in video format. The video format can include, for example, the length, resolution, and format of the video, but is not limited to, these examples. The output unit generates, for example, a video showing step-by-step instructions for making a cocktail. The output unit also outputs the appearance of the generated cocktail in image format. Examples of image formats include, but are not limited to, image resolution, file format, and display method. The output unit also generates, for example, an image of the finished cocktail. The output unit also outputs the name of the generated cocktail in text format. Examples of text formats include, but are not limited to, the number of characters, font, and display method. For example, the output unit generates a name such as "Fruity Sweet Cocktail." The feedback collection unit collects feedback from bartenders and customers. Examples of feedback include, but are not limited to, questionnaires, free-form comments, and evaluation criteria. For example, the feedback collection unit collects feedback such as "I wish it was a little more sour."The fine-tuning unit fine-tunes the generative AI model based on the feedback collected by the feedback collection unit. Fine-tuning includes, but is not limited to, the method of adjusting the model parameters and the type of data used. For example, the fine-tuning unit adjusts the generative AI model based on the collected feedback and reflects the feedback in the next proposal. As a result, the cocktail generation system according to the embodiment can generate optimal cocktails based on customer preferences and improve its performance by reflecting the feedback. For example, a bartender can easily create original cocktails and provide cocktails that satisfy customers.

[0030] The reception unit can input the customer's preferences using a generation AI. Examples of generation AI include, but are not limited to, text generation AI (e.g., LLM) and multimodal generation AI. The reception unit efficiently inputs the customer's preferences using, for example, a generation AI. For example, the generation AI inputs a preference such as "I would like a fruity and sweet cocktail." The reception unit can also use the generation AI to analyze the customer's preferences and suggest the optimal input method. For example, the generation AI displays a prompt such as "Please tell me your preference for the flavor of this cocktail" and inputs the customer's preferences. In this way, the generation AI can efficiently input the customer's preferences. Some or all of the above-described processing in the reception unit may be performed using, for example, an AI, or may be performed without using an AI. For example, the reception unit inputs the customer's preferences using a generation AI and transmits the information to the generation unit.

[0031] The generation unit can use a generation AI to generate an appropriate cocktail recipe based on past cocktail recipes and customer feedback. Examples of generation AI include, but are not limited to, text generation AI (e.g., LLM) and multimodal generation AI. The generation unit can use a generation AI to generate an optimal cocktail recipe based on past cocktail recipes and customer feedback. For example, the generation AI receives a prompt such as "Please generate a recipe for a fruity and sweet cocktail" and generates an optimal cocktail recipe. The generation unit can also use a generation AI to analyze past cocktail recipes and customer feedback and generate an optimal cocktail recipe. For example, the generation AI receives a prompt such as "Please generate an optimal cocktail recipe based on past cocktail recipes and customer feedback" and generates an optimal cocktail recipe. In this way, the generation AI can generate an optimal cocktail recipe using past data. Some or all of the above-described processing in the generation unit may be performed using, for example, an AI, or may be performed without using an AI. For example, the generation unit uses a generation AI to generate an optimal cocktail recipe based on past cocktail recipes and customer feedback, and transmits the information to the output unit.

[0032] The output unit can output the generated cocktail recipe in video format. Examples of the video format include, but are not limited to, the length of the video, the resolution, and the format used. For example, the output unit generates a video showing step-by-step instructions for making a cocktail. For example, the output unit generates a video showing how to make a cocktail based on a cocktail recipe generated by the generation AI. Furthermore, by outputting the generated cocktail recipe in video format, the output unit can provide visually easy-to-understand information. For example, the output unit generates a video showing how to make a cocktail and provides the video to a customer. By outputting the cocktail recipe in video format, visually easy-to-understand information can be provided. Some or all of the above-described processing in the output unit may be performed using, for example, AI, or may be performed without using AI. For example, the output unit generates a video showing how to make a cocktail based on a cocktail recipe generated by the generation AI, and provides the video to a customer.

[0033] The output unit can output the appearance of the generated cocktail in image format. Image formats include, but are not limited to, image resolution, file format, and display method. The output unit, for example, generates an image of the finished cocktail. For example, the output unit generates an image of the finished cocktail based on the cocktail recipe generated by the generation AI. Furthermore, the output unit can provide visually appealing information by outputting the appearance of the generated cocktail in image format. For example, the output unit generates an image of the finished cocktail and provides the image to a customer. In this way, visually appealing information can be provided by outputting the appearance of the cocktail in image format. Some or all of the above-described processing in the output unit may be performed using AI, or may be performed without using AI. For example, the output unit generates an image of the finished cocktail based on the cocktail recipe generated by the generation AI and provides the image to a customer.

[0034] The output unit can output the generated cocktail name in text format. Examples of text format include, but are not limited to, the number of characters, font, and display method. The output unit generates a name such as "Fruity Sweet Cocktail." For example, the output unit generates a cocktail name based on the cocktail recipe generated by the generation AI. The output unit can also provide concise information by outputting the generated cocktail name in text format. For example, the output unit generates a cocktail name and provides the name to the customer. By outputting the cocktail name in text format, information can be provided concisely. Some or all of the above-described processing in the output unit may be performed using AI, or may be performed without AI. For example, the output unit generates a cocktail name based on the cocktail recipe generated by the generation AI and provides the name to the customer.

[0035] The feedback collection unit can collect feedback from bartenders or customers. Examples of feedback include, but are not limited to, questionnaires, free-form responses, and evaluation criteria. For example, the feedback collection unit collects feedback such as "I'd like it a little more sour." The feedback collection unit can also collect feedback such as "This cocktail is too sweet" from customers. Furthermore, the feedback collection unit can analyze the collected feedback and provide information for improving the generative AI model. For example, the feedback collection unit analyzes the collected feedback and provides data for adjusting the generative AI model. Thus, by collecting feedback, information for improving the generative AI model can be obtained. Some or all of the above-described processing in the feedback collection unit may be performed using AI, or may be performed without using AI. For example, the feedback collection unit collects feedback from bartenders and customers and sends the information to the fine-tuning unit.

[0036] The fine-tuning unit can fine-tune the generative AI model based on the collected feedback. Fine-tuning includes, but is not limited to, the method of adjusting the model parameters and the type of data used. The fine-tuning unit, for example, adjusts the generative AI model based on the collected feedback and reflects the results in the next proposal. For example, the fine-tuning unit adjusts the generative AI model based on feedback such as "I'd like it a little more sour." The fine-tuning unit can also adjust the generative AI model based on feedback such as "This cocktail is too sweet." By fine-tuning the generative AI model based on feedback, the accuracy of the proposal can be improved. Some or all of the above-mentioned processing in the fine-tuning unit may be performed using AI, for example, or without AI. For example, the fine-tuning unit adjusts the generative AI model based on the collected feedback and transmits the information to the generation unit.

[0037] The reception unit can analyze the customer's past preference input history and suggest the optimal input method. The past preference input history includes, but is not limited to, the type of preference entered, the frequency, and the storage period for the history. For example, the reception unit can automatically display preferences that the customer has frequently entered in the past as candidates. The reception unit can also prioritize and suggest input methods (voice, text, etc.) that the customer has used in the past. Furthermore, the reception unit can predict and suggest preferences to be used during specific time periods based on the customer's past input history. By analyzing the past input history, a more efficient input method can be suggested. Some or all of the above-described processing in the reception unit may be performed using, for example, AI, or may be performed without AI. For example, the reception unit analyzes the customer's past preference input history and suggests the optimal input method based on that information.

[0038] When inputting the desired drink, the reception unit can filter the desired drink based on the customer's current mood and physical condition. Methods for inputting the current mood and physical condition include, but are not limited to, self-reporting, sensor measurement, and questionnaires. For example, if the customer is tired, the reception unit can suggest a refreshing cocktail. Furthermore, if the customer is energetic, the reception unit can suggest a cocktail that will replenish energy. Furthermore, if the customer wants to relax, the reception unit can suggest a relaxing cocktail. This allows for suggestions based on the customer's mood and physical condition, thereby providing a more appropriate cocktail. Some or all of the above-described processing in the reception unit may be performed using, for example, AI, or may be performed without using AI. For example, the reception unit filters the desired input method based on the customer's current mood and physical condition, and transmits the information to the generation unit.

[0039] When a customer inputs their preferences, the reception unit can select the optimal input means according to their input method. Examples of input methods include, but are not limited to, voice input, text input, and touch input. For example, when a customer inputs their preferences by voice, the reception unit can support the input using voice recognition technology. Furthermore, when a customer inputs their preferences by text, the reception unit can also support the input using text analysis technology. Furthermore, when a customer inputs their preferences using an image, the reception unit can also support the input using image recognition technology. This improves input efficiency by selecting the optimal means according to the customer's input method. Some or all of the above-described processing in the reception unit may be performed using, for example, AI, or may be performed without AI. For example, the reception unit selects the optimal input means according to the customer's input method and transmits the information to the generation unit.

[0040] When inputting preferences, the reception unit can prioritize inputting highly relevant preferences by taking into account the customer's geographical location information. Examples of geographical location information include, but are not limited to, methods such as GPS data, IP address, and location information services. For example, if the customer is at the beach, the reception unit can suggest cocktails that go well with the beach. Furthermore, if the customer is in a cold region, the reception unit can suggest hot cocktails. Furthermore, if the customer is in an urban area, the reception unit can suggest cocktails that match the city's trends. By taking geographical location information into consideration, more relevant suggestions can be made. Some or all of the above-described processing in the reception unit may be performed using, for example, AI, or may be performed without using AI. For example, the reception unit adjusts the method of inputting preferences based on the geographical location information and transmits the information to the generation unit.

[0041] When inputting preferences, the reception unit can analyze the customer's social media activity and input related preferences. Social media activity includes, but is not limited to, for example, the content of posts, the number of likes, and the number of followers. The reception unit can, for example, suggest cocktails related to places where the customer has checked in on social media. The reception unit can also analyze the content of the customer's social media posts to suggest related cocktails. Furthermore, the reception unit can also suggest related cocktails by referring to the activities of the customer's friends on social media. In this way, analyzing social media activity can make more relevant suggestions. Some or all of the above-mentioned processing by the reception unit may be performed using, for example, AI, or may be performed without using AI. For example, the reception unit adjusts the method for inputting preferences based on the social media activity and sends the information to the generation unit.

[0042] The reception unit can customize the input method by reflecting the customer's past feedback when the customer inputs their desired input. Examples of past feedback include, but are not limited to, questionnaires, free-form text, and evaluation criteria. The reception unit, for example, suggests the optimal input method based on the customer's past feedback. The reception unit can also preferentially suggest a specific input method based on the customer's past feedback. Furthermore, the reception unit can analyze the customer's past feedback and customize the input method. This allows the customer to provide a more appropriate input method by reflecting the past feedback. Some or all of the above-described processing in the reception unit may be performed using, for example, AI, or may be performed without using AI. For example, the reception unit customizes the customer's desired input method based on the past feedback and transmits the information to the generation unit.

[0043] When generating a recipe, the generation unit can adjust the level of detail of the recipe based on the importance of the cocktail. Examples of the importance of a cocktail include, but are not limited to, customer ratings, sales performance, and the importance of the event. For example, the generation unit generates a detailed recipe for a cocktail for a special event. The generation unit can also generate a simplified recipe for an everyday cocktail. Furthermore, the generation unit can adjust the level of detail of the recipe based on factors that the customer considers particularly important (e.g., taste, appearance). This allows for adjusting the level of detail of the recipe according to the importance of the cocktail, thereby providing a more appropriate recipe. Some or all of the above-described processing in the generation unit may be performed using, for example, AI, or may be performed without using AI. For example, the generation unit adjusts the level of detail of the recipe based on the importance of the cocktail, and transmits the information to the output unit.

[0044] When generating a recipe, the generation unit can apply different generation algorithms depending on the cocktail category. Cocktail categories include, but are not limited to, for example, the type of alcohol, flavor characteristics, and serving occasion. For example, in the case of a fruity cocktail, the generation unit applies a generation algorithm that emphasizes the fruit combination. In addition, in the case of a classic cocktail, the generation unit can also apply a generation algorithm based on a traditional recipe. Furthermore, in the case of a non-alcoholic cocktail, the generation unit can also apply a generation algorithm that emphasizes ingredients that do not contain alcohol. In this way, by applying a generation algorithm depending on the cocktail category, a more appropriate recipe can be provided. Some or all of the above-mentioned processing in the generation unit may be performed, for example, using AI or without AI. For example, the generation unit applies a generation algorithm based on the cocktail category and transmits the information to the output unit.

[0045] When generating a recipe, the generation unit can improve the accuracy of the generation by referring to the customer's past recipe generation results. Past recipe generation results include, but are not limited to, the number of generated recipes, evaluation results, and feedback. For example, the generation unit generates a new recipe by referring to cocktail recipes that the customer has previously liked. The generation unit can also improve the accuracy of the generation based on feedback provided by the customer in the past. Furthermore, the generation unit can analyze the customer's past recipe generation results and generate an optimal recipe. In this way, the accuracy of the generation can be improved by referring to the past recipe generation results. Some or all of the above-described processing in the generation unit may be performed, for example, using AI or without AI. For example, the generation unit improves the accuracy of the generation based on the past recipe generation results and sends the information to the output unit.

[0046] When generating a recipe, the generation unit can determine the generation priority based on the time of submission of the cocktail. Examples of the submission time include, but are not limited to, the season, the timing of an event, and the customer's schedule. For example, if a customer is in a hurry, the generation unit can prioritize generating a cocktail that can be generated quickly. Furthermore, if a customer needs a cocktail for a specific event, the generation unit can prioritize generating a cocktail that suits the event. Furthermore, if a customer needs a cocktail during a specific time period, the generation unit can prioritize generating a cocktail that suits the time period. Thus, by determining the generation priority based on the submission time, cocktails can be served at a more appropriate time. Some or all of the above-described processing in the generation unit may be performed using, for example, AI, or may be performed without using AI. For example, the generation unit determines the generation priority based on the submission time and transmits the information to the output unit.

[0047] When generating a recipe, the generation unit can adjust the order of cocktail creation based on the relevance of the cocktails. Examples of cocktail relevance include, but are not limited to, common ingredients, similar flavors, and common serving occasions. For example, if a customer desires a cocktail based on a specific theme, the generation unit can prioritize creating cocktails related to that theme. Furthermore, if a customer desires a cocktail using specific ingredients, the generation unit can also prioritize creating cocktails using those ingredients. Furthermore, if a customer desires a specific flavor, the generation unit can also prioritize creating cocktails related to that flavor. By adjusting the order of creation based on the relevance, a more appropriate cocktail can be provided. Some or all of the above-described processing in the generation unit may be performed, for example, using AI, or may be performed without using AI. For example, the generation unit adjusts the order of creation based on the relevance of the cocktails and transmits the information to the output unit.

[0048] When generating a recipe, the generation unit can adjust the use of technical terminology in the recipe generation according to the customer's level of expertise. Examples of technical terminology include, but are not limited to, the bartender's years of experience, the customer's level of knowledge, and past feedback. For example, if the customer is a beginner, the generation unit can generate a recipe using simple technical terminology. For example, if the customer is an intermediate bartender, the generation unit can generate a recipe using appropriate technical terminology. Furthermore, if the customer is an advanced bartender, the generation unit can generate a recipe using detailed technical terminology. This allows for the provision of a more appropriate recipe by adjusting the use of technical terminology according to the level of expertise. Some or all of the above-described processing in the generation unit may be performed using, for example, AI, or may be performed without AI. For example, the generation unit adjusts the use of technical terminology in the recipe generation based on the customer's level of expertise and sends the information to the output unit.

[0049] The output unit can adjust the level of detail of the output based on the importance of the cocktail when outputting. Examples of the importance of a cocktail include, but are not limited to, customer ratings, sales performance, and the importance of the event. For example, the output unit provides detailed output for a cocktail for a special event. The output unit can also provide simplified output for an everyday cocktail. Furthermore, the output unit can adjust the level of detail of the output based on factors that the customer considers particularly important (e.g., taste, appearance). This allows for more appropriate information to be provided by adjusting the level of detail of the output according to the importance of the cocktail. Some or all of the above-described processing in the output unit may be performed using, for example, AI, or may be performed without using AI. For example, the output unit adjusts the level of detail of the output based on the importance of the cocktail and provides the information to the customer.

[0050] The output unit can apply different output formats depending on the cocktail category at the time of output. Cocktail categories include, but are not limited to, for example, the type of alcohol, flavor characteristics, and serving occasion. For example, in the case of a fruity cocktail, the output unit can apply an output format that emphasizes the vividness of the fruit. In addition, in the case of a classic cocktail, the output unit can also apply an output format that emphasizes a traditional atmosphere. Furthermore, in the case of a non-alcoholic cocktail, the output unit can apply an output format that emphasizes a healthy image. In this way, by applying an output format according to the cocktail category, more appropriate information can be provided. Some or all of the above-mentioned processing in the output unit may be performed, for example, using AI or without using AI. For example, the output unit applies an output format based on the cocktail category and provides the information to a customer.

[0051] The output unit can improve the accuracy of output by referring to the customer's past output results. Past output results include, but are not limited to, the number of output information, evaluation results, and feedback. For example, the output unit generates a new output by referring to the customer's previously preferred output format. The output unit can also improve the accuracy of the output based on feedback provided by the customer in the past. Furthermore, the output unit can analyze the customer's past output results and generate optimal output. This allows the accuracy of the output to be improved by referring to the past output results. Some or all of the above-described processing in the output unit may be performed using, for example, AI, or may be performed without using AI. For example, the output unit improves the accuracy of the output based on the past output results and provides the information to the customer.

[0052] At the time of output, the output unit can determine the output priority based on the time of submission of the cocktail. Examples of the submission time include, but are not limited to, the season, the timing of an event, and the customer's schedule. For example, if a customer is in a hurry, the output unit can prioritize cocktails that can be output quickly. Furthermore, if a customer needs a cocktail for a specific event, the output unit can prioritize cocktails that are suited to the event. Furthermore, if a customer needs a cocktail for a specific time period, the output unit can prioritize cocktails that are suited to that time period. Thus, by determining the output priority based on the time of submission, information can be provided at a more appropriate time. Some or all of the above-described processing in the output unit may be performed, for example, using AI, or may be performed without using AI. For example, the output unit can determine the output priority based on the time of submission and provide the information to the customer.

[0053] The output unit can adjust the order of output based on the relevance of the cocktails when outputting. Examples of the relevance of the cocktails include, but are not limited to, common ingredients, similar flavors, and common serving occasions. For example, if a customer desires a cocktail based on a specific theme, the output unit can prioritize outputting cocktails related to that theme. Also, if a customer desires a cocktail using specific ingredients, the output unit can prioritize outputting cocktails using those ingredients. Furthermore, if a customer desires a specific flavor, the output unit can prioritize outputting cocktails related to that flavor. In this way, adjusting the order of output based on relevance allows for providing more appropriate cocktails. Some or all of the above-described processing in the output unit may be performed, for example, using AI, or may be performed without using AI. For example, the output unit adjusts the order of output based on the relevance of the cocktails and provides the information to the customer.

[0054] The output unit can adjust the use of technical terminology in the output depending on the customer's level of expertise. Examples of the level of expertise include, but are not limited to, the bartender's years of experience, the customer's level of knowledge, and past feedback. For example, if the customer is a beginner, the output unit can use simple technical terminology to output. If the customer is an intermediate bartender, the output unit can also use appropriate technical terminology to output. Furthermore, if the customer is an advanced bartender, the output unit can also use detailed technical terminology to output. This allows for more appropriate information to be provided by adjusting the use of technical terminology depending on the level of expertise. Some or all of the above-described processing in the output unit may be performed using, for example, AI, or may be performed without AI. For example, the output unit adjusts the use of technical terminology in the output based on the customer's level of expertise and provides the information to the customer.

[0055] When collecting feedback, the feedback collection unit can optimize the collection algorithm by referring to past feedback data. Past feedback data includes, but is not limited to, the number of feedbacks, evaluation results, and collection methods. For example, the feedback collection unit proposes an optimal feedback collection method based on the past feedback data. The feedback collection unit can also analyze the past feedback data and optimize the collection algorithm. Furthermore, the feedback collection unit can also improve the accuracy of feedback collection by referring to the past feedback data. By referring to the past feedback data, the collection algorithm can be optimized and more appropriate feedback can be obtained. Some or all of the above-described processing in the feedback collection unit may be performed using, for example, AI, or may be performed without using AI. For example, the feedback collection unit optimizes the collection algorithm based on the past feedback data and transmits the information to the fine-tuning unit.

[0056] When collecting feedback, the feedback collection unit can customize the collection method by reflecting the customer's feedback history. The feedback history includes, but is not limited to, for example, the number of feedbacks, evaluation results, and collection methods. The feedback collection unit, for example, suggests an optimal collection method based on feedback provided by the customer in the past. The feedback collection unit can also preferentially suggest a specific collection method based on the customer's past feedback history. Furthermore, the feedback collection unit can analyze the customer's past feedback history and customize the collection method. In this way, by reflecting the feedback history, a more appropriate feedback collection method can be provided. Some or all of the above-mentioned processing in the feedback collection unit may be performed using, for example, AI, or may be performed without using AI. For example, the feedback collection unit customizes the collection method based on the past feedback history and sends the information to the fine-tuning unit.

[0057] When collecting feedback, the feedback collection unit can prioritize collecting more relevant feedback by taking into account the customer's geographical location information. Examples of geographical location information include, but are not limited to, GPS data, IP address, and location information services. For example, if the customer is in a specific area, the feedback collection unit can prioritize collecting feedback related to that area. Also, if the customer is in a specific location, the feedback collection unit can prioritize collecting feedback related to that location. Furthermore, if the customer is participating in a specific event, the feedback collection unit can prioritize collecting feedback related to the event. In this way, more relevant feedback can be obtained by taking the geographical location information into account. Some or all of the above-described processing in the feedback collection unit may be performed using, for example, AI, or may be performed without using AI. For example, the feedback collection unit adjusts the feedback collection method based on the geographical location information and sends the information to the fine-tuning unit.

[0058] When collecting feedback, the feedback collection unit can analyze the customer's social media activity and collect relevant feedback. Social media activity includes, but is not limited to, for example, the content of posts, the number of likes, and the number of followers. For example, the feedback collection unit collects feedback regarding places where the customer has checked in on social media. The feedback collection unit can also analyze the content of the customer's social media posts to collect relevant feedback. Furthermore, the feedback collection unit can also collect relevant feedback by referring to the activities of the customer's friends on social media. In this way, more relevant feedback can be obtained by analyzing social media activity. Some or all of the above-mentioned processing in the feedback collection unit may be performed using, for example, AI, or may be performed without using AI. For example, the feedback collection unit adjusts the feedback collection method based on the social media activity and sends the information to the fine-tuning unit.

[0059] During fine tuning, the fine tuning unit can optimize the fine tuning algorithm by referring to past training data. Examples of past training data include, but are not limited to, the number of data, evaluation results, and collection methods. For example, the fine tuning unit proposes an optimal fine tuning method based on the past training data. The fine tuning unit can also analyze the past training data and optimize the fine tuning algorithm. Furthermore, the fine tuning unit can improve the accuracy of fine tuning by referring to the past training data. This allows the fine tuning algorithm to be optimized and more appropriate model adjustments to be performed by referring to the past training data. Some or all of the above-described processing in the fine tuning unit may be performed using, for example, AI, or without AI. For example, the fine tuning unit optimizes the fine tuning algorithm based on the past training data and transmits the information to the generation unit.

[0060] During fine tuning, the fine-tuning unit can update the fine-tuning data to reflect customer feedback. Examples of feedback include, but are not limited to, questionnaires, free-form descriptions, and evaluation criteria. The fine-tuning unit updates the fine-tuning data based on, for example, customer feedback. The fine-tuning unit can also analyze customer feedback and optimize the fine-tuning data. Furthermore, the fine-tuning unit can improve the accuracy of fine-tuning by referring to customer feedback. This allows for more appropriate fine-tuning data to be provided by reflecting the feedback. Some or all of the above-described processing in the fine-tuning unit may be performed using, for example, AI, or may be performed without using AI. For example, the fine-tuning unit updates the fine-tuning data based on feedback and transmits the information to the generation unit.

[0061] During fine-tuning, the fine-tuning unit can weight the fine-tuning data based on the time of submission of the cocktail. Examples of the submission time include, but are not limited to, the season, the timing of an event, and the customer's schedule. For example, if a customer is in a hurry, the fine-tuning unit adjusts the weighting of the data to perform fine-tuning quickly. Furthermore, if a customer needs a cocktail for a specific event, the fine-tuning unit can weight the data according to the event. Furthermore, if a customer needs a cocktail during a specific time period, the fine-tuning unit can weight the data according to the time period. Thus, weighting the fine-tuning data based on the time of submission enables more appropriate model adjustment. Some or all of the above-described processing in the fine-tuning unit may be performed using, for example, AI, or may be performed without AI. For example, the fine-tuning unit weights the fine-tuning data based on the time of submission and transmits the information to the generation unit.

[0062] During fine-tuning, the fine-tuning unit can integrate information from different data sources to enrich the fine-tuning data. Examples of different data sources include, but are not limited to, external APIs, internal databases, and third-party data sources. For example, the fine-tuning unit can integrate customer feedback data and past learning data to perform fine-tuning. The fine-tuning unit can also integrate customer social media activity data to perform fine-tuning. Furthermore, the fine-tuning unit can also integrate customer geographic location information to perform fine-tuning. This allows for the integration of information from different data sources to provide richer fine-tuning data. Some or all of the above-described processing in the fine-tuning unit may be performed using, for example, AI, or may be performed without AI. For example, the fine-tuning unit integrates information from different data sources and performs fine-tuning based on that information.

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

[0064] The reception unit can analyze the user's past cocktail order history and suggest cocktails based on the user's preferences. For example, it can suggest similar cocktails based on the taste and alcohol content of cocktails the user has ordered in the past. Also, if the user has a preference for a particular ingredient, it can preferentially suggest cocktails that use that ingredient. Furthermore, it can suggest cocktails that are suitable for a season or event based on cocktails the user has ordered for a particular season or event. This makes it possible to make more personalized cocktail suggestions by utilizing the user's past order history.

[0065] When generating a cocktail recipe, the generation unit can adjust the recipe taking into account the user's health condition. For example, if the user has an allergy, the generation unit can generate a cocktail using ingredients that do not contain the allergen. Also, if the user has a specific health goal (e.g., calorie restriction or carbohydrate restriction), the generation unit can generate a cocktail that suits that goal. Furthermore, if the user wants to consume a specific nutrient, the generation unit can generate a cocktail using ingredients that contain that nutrient. This makes it possible to provide cocktail recipes that are suited to the user's health condition.

[0066] The output unit can output the generated cocktail recipe in multiple languages. For example, the recipe can be output in English, Japanese, Spanish, or other languages ​​according to the user's language setting. The output unit can also appropriately translate the names of the cocktails and ingredients based on the language selected by the user. Furthermore, if the user is learning a different language, the output unit can support language learning by providing the recipe in that language. This allows the cocktail recipe to be provided according to the user's language setting.

[0067] The reception unit can analyze the user's social media activity and suggest related cocktails. For example, it can suggest cocktails related to places where the user has checked in on social media. It can also analyze the content of the user's social media posts and suggest cocktails based on that content. It can also suggest related cocktails based on the activity of the user's friends on social media. This makes it possible to make personalized cocktail suggestions that utilize social media activity.

[0068] The output unit can output the generated cocktail recipe in a visually appealing format. For example, it can generate a video showing how to make the cocktail in an animated format. It can also display the finished cocktail as a 3D model. Furthermore, it can provide visually appealing information by displaying the name of the cocktail and the names of its ingredients in colorful fonts. This makes it possible to provide a visually appealing cocktail recipe.

[0069] The fine-tuning unit can integrate information from different data sources to enrich the fine-tuning data. For example, it can integrate customer feedback data and past learning data to perform fine-tuning. The fine-tuning unit can also integrate customer social media activity data to perform fine-tuning. Furthermore, the fine-tuning unit can also integrate customer geographic location information to perform fine-tuning. In this way, by integrating information from different data sources, it can provide richer fine-tuning data.

[0070] The processing flow of the first embodiment will be briefly explained below.

[0071] Step 1: The reception unit inputs the customer's preferences. The customer's preferences may include, for example, taste preferences, alcohol strength, whether or not a particular ingredient is included, etc. The reception unit can accept the customer's preferences by text input, voice input, image input, or other methods. Step 2: The generation unit uses a generation AI to analyze the information entered by the reception unit and generate the optimal cocktail recipe from a huge database. The generation AI generates the optimal cocktail recipe based on past cocktail recipes and customer feedback. The generation unit generates the cocktail recipe using a text generation AI (e.g., LLM) or a multimodal generation AI. Step 3: The output unit outputs the cocktail recipe generated by the generation unit in a video format. The video format includes the video length, resolution, format used, etc. The output unit generates a video that shows how to make the cocktail step by step. Step 4: The output unit outputs the appearance of the generated cocktail in an image format. The image format includes the image resolution, file format, display method, etc. The output unit generates an image of the finished cocktail. Step 5: The output unit outputs the generated cocktail name in text format. The text format includes the number of characters, font, display method, etc. The output unit generates a name such as "Fruity Sweet Cocktail." Step 6: The feedback collection department collects feedback from bartenders and customers. Feedback can be in the form of a survey, free-form comments, or evaluation criteria. The feedback collection department collects feedback such as "I would like it to be a little more sour" from a bartender. Step 7: The fine-tuning unit fine-tunes the generative AI model based on the feedback collected by the feedback collection unit. Fine-tuning includes how to adjust the model parameters and the type of data to use. The fine-tuning unit adjusts the generative AI model based on the collected feedback and reflects it in the next proposal.

[0072] (Example 2) A cocktail generation system according to an embodiment of the present invention takes a customer's preferences as input and generates an optimal cocktail from a vast database. The cocktail generation system analyzes the customer's preferences using a generation AI to generate an optimal cocktail recipe. The generated cocktail recipe is output in video format, the cocktail's appearance is output in image format, and the cocktail's name is output in text format. Furthermore, by collecting feedback and fine-tuning the generation AI model, it is possible to provide personalized suggestions that reflect the bartender's characteristics and the customer's preferences. For example, a cocktail generation system inputs a customer's preferences, such as "I want a fruity and sweet cocktail." This information is input into the generation AI. The generation AI then analyzes the input information and generates an optimal cocktail recipe from a vast database. The generation AI generates an optimal cocktail recipe based on past cocktail recipes and customer feedback. For example, it generates a recipe for a fruity and sweet cocktail. The generated cocktail recipe is output in video format. For example, a step-by-step video showing how to make the cocktail is generated. The cocktail's appearance is output in image format. For example, an image of the finished cocktail is generated. The cocktail's name is output in text format. For example, a name such as "Fruity Sweet Cocktail" is generated. Finally, feedback from bartenders and customers is collected to fine-tune the generative AI model. This allows the cocktail generation system to make personalized suggestions that reflect the bartender's characteristics and the customer's preferences. This allows the cocktail generation system to generate the optimal cocktail based on the customer's wishes and improve its performance by incorporating feedback. For example, bartenders can easily create original cocktails and provide cocktails that satisfy customers.

[0073] A cocktail creation system according to an embodiment includes a reception unit, a generation unit, an output unit, a feedback collection unit, and a fine-tuning unit. The reception unit inputs customer preferences. The customer preferences include, but are not limited to, taste preferences, alcohol strength, and the presence or absence of specific ingredients. The reception unit can receive the customer preferences via, for example, text input, voice input, or image input. The generation unit uses a generation AI to analyze the information input by the reception unit and generate an optimal cocktail recipe from a vast database. The generation AI generates the optimal cocktail recipe based on, for example, past cocktail recipes and customer feedback. The generation unit generates the cocktail recipe using, for example, a text generation AI (e.g., LLM). The generation unit can also generate the cocktail recipe using a multimodal generation AI. The generation unit can also generate the optimal cocktail recipe based on, for example, past cocktail recipes and customer feedback. The output unit outputs the cocktail recipe generated by the generation unit in video format. The video format can include, for example, the length, resolution, and format of the video, but is not limited to, these examples. The output unit generates, for example, a video showing step-by-step instructions for making a cocktail. The output unit also outputs the appearance of the generated cocktail in image format. Examples of image formats include, but are not limited to, image resolution, file format, and display method. The output unit also generates, for example, an image of the finished cocktail. The output unit also outputs the name of the generated cocktail in text format. Examples of text formats include, but are not limited to, the number of characters, font, and display method. For example, the output unit generates a name such as "Fruity Sweet Cocktail." The feedback collection unit collects feedback from bartenders and customers. Examples of feedback include, but are not limited to, questionnaires, free-form comments, and evaluation criteria. For example, the feedback collection unit collects feedback such as "I wish it was a little more sour."The fine-tuning unit fine-tunes the generative AI model based on the feedback collected by the feedback collection unit. Fine-tuning includes, but is not limited to, the method of adjusting the model parameters and the type of data used. For example, the fine-tuning unit adjusts the generative AI model based on the collected feedback and reflects the feedback in the next proposal. As a result, the cocktail generation system according to the embodiment can generate optimal cocktails based on customer preferences and improve its performance by reflecting the feedback. For example, a bartender can easily create original cocktails and provide cocktails that satisfy customers.

[0074] The reception unit can input the customer's preferences using a generation AI. Examples of generation AI include, but are not limited to, text generation AI (e.g., LLM) and multimodal generation AI. The reception unit efficiently inputs the customer's preferences using, for example, a generation AI. For example, the generation AI inputs a preference such as "I would like a fruity and sweet cocktail." The reception unit can also use the generation AI to analyze the customer's preferences and suggest the optimal input method. For example, the generation AI displays a prompt such as "Please tell me your preference for the flavor of this cocktail" and inputs the customer's preferences. In this way, the generation AI can efficiently input the customer's preferences. Some or all of the above-described processing in the reception unit may be performed using, for example, an AI, or may be performed without using an AI. For example, the reception unit inputs the customer's preferences using a generation AI and transmits the information to the generation unit.

[0075] The generation unit can use a generation AI to generate an appropriate cocktail recipe based on past cocktail recipes and customer feedback. Examples of generation AI include, but are not limited to, text generation AI (e.g., LLM) and multimodal generation AI. The generation unit can use a generation AI to generate an optimal cocktail recipe based on past cocktail recipes and customer feedback. For example, the generation AI receives a prompt such as "Please generate a recipe for a fruity and sweet cocktail" and generates an optimal cocktail recipe. The generation unit can also use a generation AI to analyze past cocktail recipes and customer feedback and generate an optimal cocktail recipe. For example, the generation AI receives a prompt such as "Please generate an optimal cocktail recipe based on past cocktail recipes and customer feedback" and generates an optimal cocktail recipe. In this way, the generation AI can generate an optimal cocktail recipe using past data. Some or all of the above-described processing in the generation unit may be performed using, for example, an AI, or may be performed without using an AI. For example, the generation unit uses a generation AI to generate an optimal cocktail recipe based on past cocktail recipes and customer feedback, and transmits the information to the output unit.

[0076] The output unit can output the generated cocktail recipe in video format. Examples of the video format include, but are not limited to, the length of the video, the resolution, and the format used. For example, the output unit generates a video showing step-by-step instructions for making a cocktail. For example, the output unit generates a video showing how to make a cocktail based on a cocktail recipe generated by the generation AI. Furthermore, by outputting the generated cocktail recipe in video format, the output unit can provide visually easy-to-understand information. For example, the output unit generates a video showing how to make a cocktail and provides the video to a customer. By outputting the cocktail recipe in video format, visually easy-to-understand information can be provided. Some or all of the above-described processing in the output unit may be performed using, for example, AI, or may be performed without using AI. For example, the output unit generates a video showing how to make a cocktail based on a cocktail recipe generated by the generation AI, and provides the video to a customer.

[0077] The output unit can output the appearance of the generated cocktail in image format. Image formats include, but are not limited to, image resolution, file format, and display method. The output unit, for example, generates an image of the finished cocktail. For example, the output unit generates an image of the finished cocktail based on the cocktail recipe generated by the generation AI. Furthermore, the output unit can provide visually appealing information by outputting the appearance of the generated cocktail in image format. For example, the output unit generates an image of the finished cocktail and provides the image to a customer. In this way, visually appealing information can be provided by outputting the appearance of the cocktail in image format. Some or all of the above-described processing in the output unit may be performed using AI, or may be performed without using AI. For example, the output unit generates an image of the finished cocktail based on the cocktail recipe generated by the generation AI and provides the image to a customer.

[0078] The output unit can output the generated cocktail name in text format. Examples of text format include, but are not limited to, the number of characters, font, and display method. The output unit generates a name such as "Fruity Sweet Cocktail." For example, the output unit generates a cocktail name based on the cocktail recipe generated by the generation AI. The output unit can also provide concise information by outputting the generated cocktail name in text format. For example, the output unit generates a cocktail name and provides the name to the customer. By outputting the cocktail name in text format, information can be provided concisely. Some or all of the above-described processing in the output unit may be performed using AI, or may be performed without AI. For example, the output unit generates a cocktail name based on the cocktail recipe generated by the generation AI and provides the name to the customer.

[0079] The feedback collection unit can collect feedback from bartenders or customers. Examples of feedback include, but are not limited to, questionnaires, free-form responses, and evaluation criteria. For example, the feedback collection unit collects feedback such as "I'd like it a little more sour." The feedback collection unit can also collect feedback such as "This cocktail is too sweet" from customers. Furthermore, the feedback collection unit can analyze the collected feedback and provide information for improving the generative AI model. For example, the feedback collection unit analyzes the collected feedback and provides data for adjusting the generative AI model. Thus, by collecting feedback, information for improving the generative AI model can be obtained. Some or all of the above-described processing in the feedback collection unit may be performed using AI, or may be performed without using AI. For example, the feedback collection unit collects feedback from bartenders and customers and sends the information to the fine-tuning unit.

[0080] The fine-tuning unit can fine-tune the generative AI model based on the collected feedback. Fine-tuning includes, but is not limited to, the method of adjusting the model parameters and the type of data used. The fine-tuning unit, for example, adjusts the generative AI model based on the collected feedback and reflects the results in the next proposal. For example, the fine-tuning unit adjusts the generative AI model based on feedback such as "I'd like it a little more sour." The fine-tuning unit can also adjust the generative AI model based on feedback such as "This cocktail is too sweet." By fine-tuning the generative AI model based on feedback, the accuracy of the proposal can be improved. Some or all of the above-mentioned processing in the fine-tuning unit may be performed using AI, for example, or without AI. For example, the fine-tuning unit adjusts the generative AI model based on the collected feedback and transmits the information to the generation unit.

[0081] The reception unit can estimate the customer's emotions and adjust the desired input method based on the estimated customer emotions. Emotion estimation includes, but is not limited to, technologies such as facial expression recognition, voice analysis, and text analysis. For example, if the customer is relaxed, the reception unit can provide detailed input options and suggest a customizable input method. Furthermore, if the customer is stressed, the reception unit can provide a simple interface and minimize input steps. Furthermore, if the customer is in a hurry, the reception unit can prioritize voice input to enable quick input of the customer's preferences. This allows for a more appropriate input experience by adjusting the input method according to the customer's emotions. Emotion estimation is achieved using an emotion estimation function, for example, an emotion engine or a generative AI. The generative AI can be, for example, a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to these examples. Some or all of the above-described processing in the reception unit may be performed using AI, or without AI. For example, the reception unit can estimate the customer's emotions using an emotion estimation function and adjust the desired input method based on that information.

[0082] The reception unit can analyze the customer's past preference input history and suggest the optimal input method. The past preference input history includes, but is not limited to, the type of preference entered, the frequency, and the storage period for the history. For example, the reception unit can automatically display preferences that the customer has frequently entered in the past as candidates. The reception unit can also prioritize and suggest input methods (voice, text, etc.) that the customer has used in the past. Furthermore, the reception unit can predict and suggest preferences to be used during specific time periods based on the customer's past input history. By analyzing the past input history, a more efficient input method can be suggested. Some or all of the above-described processing in the reception unit may be performed using, for example, AI, or may be performed without AI. For example, the reception unit analyzes the customer's past preference input history and suggests the optimal input method based on that information.

[0083] When inputting the desired drink, the reception unit can filter the desired drink based on the customer's current mood and physical condition. Methods for inputting the current mood and physical condition include, but are not limited to, self-reporting, sensor measurement, and questionnaires. For example, if the customer is tired, the reception unit can suggest a refreshing cocktail. Furthermore, if the customer is energetic, the reception unit can suggest a cocktail that will replenish energy. Furthermore, if the customer wants to relax, the reception unit can suggest a relaxing cocktail. This allows for suggestions based on the customer's mood and physical condition, thereby providing a more appropriate cocktail. Some or all of the above-described processing in the reception unit may be performed using, for example, AI, or may be performed without using AI. For example, the reception unit filters the desired input method based on the customer's current mood and physical condition, and transmits the information to the generation unit.

[0084] When a customer inputs their preferences, the reception unit can select the optimal input means according to their input method. Examples of input methods include, but are not limited to, voice input, text input, and touch input. For example, when a customer inputs their preferences by voice, the reception unit can support the input using voice recognition technology. Furthermore, when a customer inputs their preferences by text, the reception unit can also support the input using text analysis technology. Furthermore, when a customer inputs their preferences using an image, the reception unit can also support the input using image recognition technology. This improves input efficiency by selecting the optimal means according to the customer's input method. Some or all of the above-described processing in the reception unit may be performed using, for example, AI, or may be performed without AI. For example, the reception unit selects the optimal input means according to the customer's input method and transmits the information to the generation unit.

[0085] The reception unit can estimate the customer's emotions and prioritize the input preferences based on the estimated emotions. Emotion estimation includes, but is not limited to, techniques such as facial expression recognition, voice analysis, and text analysis. For example, if the customer is relaxed, the reception unit can prioritize cocktails with a relaxing effect. Furthermore, if the customer is stressed, the reception unit can prioritize cocktails with a stress-relieving effect. Furthermore, if the customer is energetic, the reception unit can prioritize cocktails that replenish energy. This allows for more appropriate recommendations by determining priorities based on the customer's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generative AI. The generative AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the reception unit may be performed using, for example, AI, or without AI. For example, the reception unit can estimate the customer's emotions using an emotion estimation function and prioritize preferences based on the information.

[0086] When inputting preferences, the reception unit can prioritize inputting highly relevant preferences by taking into account the customer's geographical location information. Examples of geographical location information include, but are not limited to, methods such as GPS data, IP address, and location information services. For example, if the customer is at the beach, the reception unit can suggest cocktails that go well with the beach. Furthermore, if the customer is in a cold region, the reception unit can suggest hot cocktails. Furthermore, if the customer is in an urban area, the reception unit can suggest cocktails that match the city's trends. By taking geographical location information into consideration, more relevant suggestions can be made. Some or all of the above-described processing in the reception unit may be performed using, for example, AI, or may be performed without using AI. For example, the reception unit adjusts the method of inputting preferences based on the geographical location information and transmits the information to the generation unit.

[0087] When inputting preferences, the reception unit can analyze the customer's social media activity and input related preferences. Social media activity includes, but is not limited to, for example, the content of posts, the number of likes, and the number of followers. The reception unit can, for example, suggest cocktails related to places where the customer has checked in on social media. The reception unit can also analyze the content of the customer's social media posts to suggest related cocktails. Furthermore, the reception unit can also suggest related cocktails by referring to the activities of the customer's friends on social media. In this way, analyzing social media activity can make more relevant suggestions. Some or all of the above-mentioned processing by the reception unit may be performed using, for example, AI, or may be performed without using AI. For example, the reception unit adjusts the method for inputting preferences based on the social media activity and sends the information to the generation unit.

[0088] The reception unit can customize the input method by reflecting the customer's past feedback when the customer inputs their desired input. Examples of past feedback include, but are not limited to, questionnaires, free-form text, and evaluation criteria. The reception unit, for example, suggests the optimal input method based on the customer's past feedback. The reception unit can also preferentially suggest a specific input method based on the customer's past feedback. Furthermore, the reception unit can analyze the customer's past feedback and customize the input method. This allows the customer to provide a more appropriate input method by reflecting the past feedback. Some or all of the above-described processing in the reception unit may be performed using, for example, AI, or may be performed without using AI. For example, the reception unit customizes the customer's desired input method based on the past feedback and transmits the information to the generation unit.

[0089] The generation unit can estimate the customer's emotions and adjust the cocktail recipe generation method based on the estimated customer's emotions. Emotion estimation includes, but is not limited to, technologies such as facial expression recognition, voice analysis, and text analysis. For example, if the customer is relaxed, the generation unit can generate a cocktail recipe with a relaxing effect. Furthermore, if the customer is stressed, the generation unit can generate a cocktail recipe with a stress-relieving effect. Furthermore, if the customer is energetic, the generation unit can generate a cocktail recipe that replenishes energy. This allows for adjusting the recipe generation method according to the customer's emotions to provide a more appropriate cocktail. Emotion estimation is achieved using an emotion estimation function, for example, an emotion engine or generation AI. The generation AI can be, for example, a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to these examples. Some or all of the above-described processing in the generation unit may be performed using AI, or without AI. For example, the generation unit can estimate the customer's emotions using the emotion estimation function and adjust the cocktail recipe generation method based on the information.

[0090] When generating a recipe, the generation unit can adjust the level of detail of the recipe based on the importance of the cocktail. Examples of the importance of a cocktail include, but are not limited to, customer ratings, sales performance, and the importance of the event. For example, the generation unit generates a detailed recipe for a cocktail for a special event. The generation unit can also generate a simplified recipe for an everyday cocktail. Furthermore, the generation unit can adjust the level of detail of the recipe based on factors that the customer considers particularly important (e.g., taste, appearance). This allows for adjusting the level of detail of the recipe according to the importance of the cocktail, thereby providing a more appropriate recipe. Some or all of the above-described processing in the generation unit may be performed using, for example, AI, or may be performed without using AI. For example, the generation unit adjusts the level of detail of the recipe based on the importance of the cocktail, and transmits the information to the output unit.

[0091] When generating a recipe, the generation unit can apply different generation algorithms depending on the cocktail category. Cocktail categories include, but are not limited to, for example, the type of alcohol, flavor characteristics, and serving occasion. For example, in the case of a fruity cocktail, the generation unit applies a generation algorithm that emphasizes the fruit combination. In addition, in the case of a classic cocktail, the generation unit can also apply a generation algorithm based on a traditional recipe. Furthermore, in the case of a non-alcoholic cocktail, the generation unit can also apply a generation algorithm that emphasizes ingredients that do not contain alcohol. In this way, by applying a generation algorithm depending on the cocktail category, a more appropriate recipe can be provided. Some or all of the above-mentioned processing in the generation unit may be performed, for example, using AI or without AI. For example, the generation unit applies a generation algorithm based on the cocktail category and transmits the information to the output unit.

[0092] When generating a recipe, the generation unit can improve the accuracy of the generation by referring to the customer's past recipe generation results. Past recipe generation results include, but are not limited to, the number of generated recipes, evaluation results, and feedback. For example, the generation unit generates a new recipe by referring to cocktail recipes that the customer has previously liked. The generation unit can also improve the accuracy of the generation based on feedback provided by the customer in the past. Furthermore, the generation unit can analyze the customer's past recipe generation results and generate an optimal recipe. In this way, the accuracy of the generation can be improved by referring to the past recipe generation results. Some or all of the above-described processing in the generation unit may be performed, for example, using AI or without AI. For example, the generation unit improves the accuracy of the generation based on the past recipe generation results and sends the information to the output unit.

[0093] The generation unit can estimate the customer's emotions and adjust the type of cocktail to be generated based on the estimated customer's emotions. Emotion estimation includes, but is not limited to, technologies such as facial expression recognition, voice analysis, and text analysis. For example, if the customer is relaxed, the generation unit can generate a cocktail with a relaxing effect. Furthermore, if the customer is stressed, the generation unit can generate a cocktail with a stress-relieving effect. Furthermore, if the customer is energetic, the generation unit can generate a cocktail that replenishes energy. This allows for adjusting the type of cocktail according to the customer's emotions to provide a more appropriate cocktail. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the generation unit may be performed using AI, or without AI. For example, the generation unit can estimate the customer's emotions using an emotion estimation function and adjust the type of cocktail based on the information.

[0094] When generating a recipe, the generation unit can determine the generation priority based on the time of submission of the cocktail. Examples of the submission time include, but are not limited to, the season, the timing of an event, and the customer's schedule. For example, if a customer is in a hurry, the generation unit can prioritize generating a cocktail that can be generated quickly. Furthermore, if a customer needs a cocktail for a specific event, the generation unit can prioritize generating a cocktail that suits the event. Furthermore, if a customer needs a cocktail during a specific time period, the generation unit can prioritize generating a cocktail that suits the time period. Thus, by determining the generation priority based on the submission time, cocktails can be served at a more appropriate time. Some or all of the above-described processing in the generation unit may be performed using, for example, AI, or may be performed without using AI. For example, the generation unit determines the generation priority based on the submission time and transmits the information to the output unit.

[0095] When generating a recipe, the generation unit can adjust the order of cocktail creation based on the relevance of the cocktails. Examples of cocktail relevance include, but are not limited to, common ingredients, similar flavors, and common serving occasions. For example, if a customer desires a cocktail based on a specific theme, the generation unit can prioritize creating cocktails related to that theme. Furthermore, if a customer desires a cocktail using specific ingredients, the generation unit can also prioritize creating cocktails using those ingredients. Furthermore, if a customer desires a specific flavor, the generation unit can also prioritize creating cocktails related to that flavor. By adjusting the order of creation based on the relevance, a more appropriate cocktail can be provided. Some or all of the above-described processing in the generation unit may be performed, for example, using AI, or may be performed without using AI. For example, the generation unit adjusts the order of creation based on the relevance of the cocktails and transmits the information to the output unit.

[0096] When generating a recipe, the generation unit can adjust the use of technical terminology in the recipe generation according to the customer's level of expertise. Examples of technical terminology include, but are not limited to, the bartender's years of experience, the customer's level of knowledge, and past feedback. For example, if the customer is a beginner, the generation unit can generate a recipe using simple technical terminology. For example, if the customer is an intermediate bartender, the generation unit can generate a recipe using appropriate technical terminology. Furthermore, if the customer is an advanced bartender, the generation unit can generate a recipe using detailed technical terminology. This allows for the provision of a more appropriate recipe by adjusting the use of technical terminology according to the level of expertise. Some or all of the above-described processing in the generation unit may be performed using, for example, AI, or may be performed without AI. For example, the generation unit adjusts the use of technical terminology in the recipe generation based on the customer's level of expertise and sends the information to the output unit.

[0097] The output unit can estimate the customer's emotion and adjust the output expression method based on the estimated customer's emotion. Emotion estimation includes, but is not limited to, techniques such as facial expression recognition, voice analysis, and text analysis. For example, if the customer is relaxed, the output unit outputs an expression that has a relaxing effect. Furthermore, if the customer is stressed, the output unit can output an expression that has a stress-relieving effect. Furthermore, if the customer is energetic, the output unit can output an expression that replenishes energy. This allows for more appropriate output by adjusting the expression method according to the customer's emotion. Emotion estimation is achieved using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the output unit may be performed using, for example, AI, or without AI. For example, the output unit estimates the customer's emotion using an emotion estimation function and adjusts the output expression method based on that information.

[0098] The output unit can adjust the level of detail of the output based on the importance of the cocktail when outputting. Examples of the importance of a cocktail include, but are not limited to, customer ratings, sales performance, and the importance of the event. For example, the output unit provides detailed output for a cocktail for a special event. The output unit can also provide simplified output for an everyday cocktail. Furthermore, the output unit can adjust the level of detail of the output based on factors that the customer considers particularly important (e.g., taste, appearance). This allows for more appropriate information to be provided by adjusting the level of detail of the output according to the importance of the cocktail. Some or all of the above-described processing in the output unit may be performed using, for example, AI, or may be performed without using AI. For example, the output unit adjusts the level of detail of the output based on the importance of the cocktail and provides the information to the customer.

[0099] The output unit can apply different output formats depending on the cocktail category at the time of output. Cocktail categories include, but are not limited to, for example, the type of alcohol, flavor characteristics, and serving occasion. For example, in the case of a fruity cocktail, the output unit can apply an output format that emphasizes the vividness of the fruit. In addition, in the case of a classic cocktail, the output unit can also apply an output format that emphasizes a traditional atmosphere. Furthermore, in the case of a non-alcoholic cocktail, the output unit can apply an output format that emphasizes a healthy image. In this way, by applying an output format according to the cocktail category, more appropriate information can be provided. Some or all of the above-mentioned processing in the output unit may be performed, for example, using AI or without using AI. For example, the output unit applies an output format based on the cocktail category and provides the information to a customer.

[0100] The output unit can improve the accuracy of output by referring to the customer's past output results. Past output results include, but are not limited to, the number of output information, evaluation results, and feedback. For example, the output unit generates a new output by referring to the customer's previously preferred output format. The output unit can also improve the accuracy of the output based on feedback provided by the customer in the past. Furthermore, the output unit can analyze the customer's past output results and generate optimal output. This allows the accuracy of the output to be improved by referring to the past output results. Some or all of the above-described processing in the output unit may be performed using, for example, AI, or may be performed without using AI. For example, the output unit improves the accuracy of the output based on the past output results and provides the information to the customer.

[0101] The output unit can estimate the customer's emotions and adjust the length of the output based on the estimated customer emotions. Emotion estimation includes, but is not limited to, techniques such as facial expression recognition, voice analysis, and text analysis. For example, if the customer is relaxed, the output unit can provide a longer output with detailed explanations. If the customer is in a hurry, the output unit can also provide a shorter output that focuses on the main points. Furthermore, if the customer is excited, the output unit can also provide an output with visually stimulating effects. This allows for more appropriate information to be provided by adjusting the length of the output based on the customer's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the output unit may be performed using, for example, AI, or without AI. For example, the output unit can estimate the customer's emotions using an emotion estimation function and adjust the length of the output based on that information.

[0102] At the time of output, the output unit can determine the output priority based on the time of submission of the cocktail. Examples of the submission time include, but are not limited to, the season, the timing of an event, and the customer's schedule. For example, if a customer is in a hurry, the output unit can prioritize cocktails that can be output quickly. Furthermore, if a customer needs a cocktail for a specific event, the output unit can prioritize cocktails that are suited to the event. Furthermore, if a customer needs a cocktail for a specific time period, the output unit can prioritize cocktails that are suited to that time period. Thus, by determining the output priority based on the time of submission, information can be provided at a more appropriate time. Some or all of the above-described processing in the output unit may be performed, for example, using AI, or may be performed without using AI. For example, the output unit can determine the output priority based on the time of submission and provide the information to the customer.

[0103] The output unit can adjust the order of output based on the relevance of the cocktails when outputting. Examples of the relevance of the cocktails include, but are not limited to, common ingredients, similar flavors, and common serving occasions. For example, if a customer desires a cocktail based on a specific theme, the output unit can prioritize outputting cocktails related to that theme. Also, if a customer desires a cocktail using specific ingredients, the output unit can prioritize outputting cocktails using those ingredients. Furthermore, if a customer desires a specific flavor, the output unit can prioritize outputting cocktails related to that flavor. In this way, adjusting the order of output based on relevance allows for providing more appropriate cocktails. Some or all of the above-described processing in the output unit may be performed, for example, using AI, or may be performed without using AI. For example, the output unit adjusts the order of output based on the relevance of the cocktails and provides the information to the customer.

[0104] The output unit can adjust the use of technical terminology in the output depending on the customer's level of expertise. Examples of the level of expertise include, but are not limited to, the bartender's years of experience, the customer's level of knowledge, and past feedback. For example, if the customer is a beginner, the output unit can use simple technical terminology to output. If the customer is an intermediate bartender, the output unit can also use appropriate technical terminology to output. Furthermore, if the customer is an advanced bartender, the output unit can also use detailed technical terminology to output. This allows for more appropriate information to be provided by adjusting the use of technical terminology depending on the level of expertise. Some or all of the above-described processing in the output unit may be performed using, for example, AI, or may be performed without AI. For example, the output unit adjusts the use of technical terminology in the output based on the customer's level of expertise and provides the information to the customer.

[0105] The feedback collection unit can estimate the customer's emotions and adjust the feedback collection method based on the estimated customer emotions. Emotion estimation includes, but is not limited to, techniques such as facial expression recognition, voice analysis, and text analysis. For example, if the customer is relaxed, the feedback collection unit can request detailed feedback. Furthermore, if the customer is stressed, the feedback collection unit can also request simple feedback. Furthermore, if the customer is in a hurry, the feedback collection unit can quickly collect feedback. This allows for more appropriate feedback by adjusting the feedback collection method according to the customer's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the feedback collection unit may be performed using, for example, AI, or without AI. For example, the feedback collection unit can estimate the customer's emotions using an emotion estimation function and adjust the feedback collection method based on the information.

[0106] When collecting feedback, the feedback collection unit can optimize the collection algorithm by referring to past feedback data. Past feedback data includes, but is not limited to, the number of feedbacks, evaluation results, and collection methods. For example, the feedback collection unit proposes an optimal feedback collection method based on the past feedback data. The feedback collection unit can also analyze the past feedback data and optimize the collection algorithm. Furthermore, the feedback collection unit can also improve the accuracy of feedback collection by referring to the past feedback data. By referring to the past feedback data, the collection algorithm can be optimized and more appropriate feedback can be obtained. Some or all of the above-described processing in the feedback collection unit may be performed using, for example, AI, or may be performed without using AI. For example, the feedback collection unit optimizes the collection algorithm based on the past feedback data and transmits the information to the fine-tuning unit.

[0107] When collecting feedback, the feedback collection unit can customize the collection method by reflecting the customer's feedback history. The feedback history includes, but is not limited to, for example, the number of feedbacks, evaluation results, and collection methods. The feedback collection unit, for example, suggests an optimal collection method based on feedback provided by the customer in the past. The feedback collection unit can also preferentially suggest a specific collection method based on the customer's past feedback history. Furthermore, the feedback collection unit can analyze the customer's past feedback history and customize the collection method. In this way, by reflecting the feedback history, a more appropriate feedback collection method can be provided. Some or all of the above-mentioned processing in the feedback collection unit may be performed using, for example, AI, or may be performed without using AI. For example, the feedback collection unit customizes the collection method based on the past feedback history and sends the information to the fine-tuning unit.

[0108] The feedback collection unit can estimate the customer's emotions and prioritize feedback based on the estimated customer emotions. Emotion estimation includes, but is not limited to, techniques such as facial expression recognition, voice analysis, and text analysis. For example, if the customer is relaxed, the feedback collection unit can prioritize collecting detailed feedback. Furthermore, if the customer is stressed, the feedback collection unit can also prioritize collecting simple feedback. Furthermore, if the customer is in a hurry, the feedback collection unit can quickly collect feedback. This allows for more appropriate feedback to be obtained by prioritizing feedback according to the customer's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the feedback collection unit may be performed using, for example, AI, or without AI. For example, the feedback collection unit can estimate the customer's emotions using an emotion estimation function and prioritize feedback based on the information.

[0109] When collecting feedback, the feedback collection unit can prioritize collecting more relevant feedback by taking into account the customer's geographical location information. Examples of geographical location information include, but are not limited to, GPS data, IP address, and location information services. For example, if the customer is in a specific area, the feedback collection unit can prioritize collecting feedback related to that area. Also, if the customer is in a specific location, the feedback collection unit can prioritize collecting feedback related to that location. Furthermore, if the customer is participating in a specific event, the feedback collection unit can prioritize collecting feedback related to the event. In this way, more relevant feedback can be obtained by taking the geographical location information into account. Some or all of the above-described processing in the feedback collection unit may be performed using, for example, AI, or may be performed without using AI. For example, the feedback collection unit adjusts the feedback collection method based on the geographical location information and sends the information to the fine-tuning unit.

[0110] When collecting feedback, the feedback collection unit can analyze the customer's social media activity and collect relevant feedback. Social media activity includes, but is not limited to, for example, the content of posts, the number of likes, and the number of followers. For example, the feedback collection unit collects feedback regarding places where the customer has checked in on social media. The feedback collection unit can also analyze the content of the customer's social media posts to collect relevant feedback. Furthermore, the feedback collection unit can also collect relevant feedback by referring to the activities of the customer's friends on social media. In this way, more relevant feedback can be obtained by analyzing social media activity. Some or all of the above-mentioned processing in the feedback collection unit may be performed using, for example, AI, or may be performed without using AI. For example, the feedback collection unit adjusts the feedback collection method based on the social media activity and sends the information to the fine-tuning unit.

[0111] The fine-tuning unit can estimate a customer's emotion and adjust the fine-tuning method based on the estimated customer's emotion. Emotion estimation includes, but is not limited to, techniques such as facial expression recognition, voice analysis, and text analysis. For example, if the customer is relaxed, the fine-tuning unit can perform fine-tuning based on detailed feedback. For example, if the customer is stressed, the fine-tuning unit can perform fine-tuning based on simple feedback. Furthermore, if the customer is in a hurry, the fine-tuning unit can perform fine-tuning based on quick feedback. This allows for more appropriate model adjustment by adjusting the fine-tuning method according to the customer's emotion. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generative AI. The generative AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the fine-tuning unit may be performed using, for example, AI, or without AI. For example, the fine-tuning unit can estimate a customer's emotion using an emotion estimation function and adjust the fine-tuning method based on the information.

[0112] During fine tuning, the fine tuning unit can optimize the fine tuning algorithm by referring to past training data. Examples of past training data include, but are not limited to, the number of data, evaluation results, and collection methods. For example, the fine tuning unit proposes an optimal fine tuning method based on the past training data. The fine tuning unit can also analyze the past training data and optimize the fine tuning algorithm. Furthermore, the fine tuning unit can improve the accuracy of fine tuning by referring to the past training data. This allows the fine tuning algorithm to be optimized and more appropriate model adjustments to be performed by referring to the past training data. Some or all of the above-described processing in the fine tuning unit may be performed using, for example, AI, or without AI. For example, the fine tuning unit optimizes the fine tuning algorithm based on the past training data and transmits the information to the generation unit.

[0113] During fine tuning, the fine-tuning unit can update the fine-tuning data to reflect customer feedback. Examples of feedback include, but are not limited to, questionnaires, free-form descriptions, and evaluation criteria. The fine-tuning unit updates the fine-tuning data based on, for example, customer feedback. The fine-tuning unit can also analyze customer feedback and optimize the fine-tuning data. Furthermore, the fine-tuning unit can improve the accuracy of fine-tuning by referring to customer feedback. This allows for more appropriate fine-tuning data to be provided by reflecting the feedback. Some or all of the above-described processing in the fine-tuning unit may be performed using, for example, AI, or may be performed without using AI. For example, the fine-tuning unit updates the fine-tuning data based on feedback and transmits the information to the generation unit.

[0114] The fine-tuning unit can estimate a customer's emotion and adjust the frequency of fine-tuning based on the estimated emotion. Emotion estimation includes, but is not limited to, techniques such as facial expression recognition, voice analysis, and text analysis. For example, if a customer is relaxed, the fine-tuning unit can perform fine-tuning frequently based on detailed feedback. Furthermore, if a customer is stressed, the fine-tuning unit can adjust the frequency based on simple feedback. Furthermore, if a customer is in a hurry, the fine-tuning unit can adjust the frequency based on quick feedback. This allows for more appropriate model adjustment by adjusting the frequency of fine-tuning according to the customer's emotion. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generative AI. The generative AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the fine-tuning unit may be performed using AI, or without AI. For example, the fine-tuning unit can estimate a customer's emotion using an emotion estimation function and adjust the frequency of fine-tuning based on the information.

[0115] During fine-tuning, the fine-tuning unit can weight the fine-tuning data based on the time of submission of the cocktail. Examples of the submission time include, but are not limited to, the season, the timing of an event, and the customer's schedule. For example, if a customer is in a hurry, the fine-tuning unit adjusts the weighting of the data to perform fine-tuning quickly. Furthermore, if a customer needs a cocktail for a specific event, the fine-tuning unit can weight the data according to the event. Furthermore, if a customer needs a cocktail during a specific time period, the fine-tuning unit can weight the data according to the time period. Thus, weighting the fine-tuning data based on the time of submission enables more appropriate model adjustment. Some or all of the above-described processing in the fine-tuning unit may be performed using, for example, AI, or may be performed without AI. For example, the fine-tuning unit weights the fine-tuning data based on the time of submission and transmits the information to the generation unit.

[0116] During fine-tuning, the fine-tuning unit can integrate information from different data sources to enrich the fine-tuning data. Examples of different data sources include, but are not limited to, external APIs, internal databases, and third-party data sources. For example, the fine-tuning unit can integrate customer feedback data and past learning data to perform fine-tuning. The fine-tuning unit can also integrate customer social media activity data to perform fine-tuning. Furthermore, the fine-tuning unit can also integrate customer geographic location information to perform fine-tuning. This allows for the integration of information from different data sources to provide richer fine-tuning data. Some or all of the above-described processing in the fine-tuning unit may be performed using, for example, AI, or may be performed without AI. For example, the fine-tuning unit integrates information from different data sources and performs fine-tuning based on that information. === Hard Collateral 1-1 === Each of the multiple elements, including the reception unit, generation unit, output unit, feedback collection unit, and fine-tuning unit, described above, is realized, for example, by at least one of the smart device 14 and the data processing device 12. For example, the reception unit can input a customer's request using the reception device 38 of the smart device 14. For example, the generation unit executes a generation AI using the specific processing unit 290 of the data processing device 12 to generate an optimal cocktail recipe. For example, the output unit outputs the generated cocktail recipe in video format using the output device 40 of the smart device 14. For example, the feedback collection unit collects feedback from the bartender and customers using the camera 42 and microphone 38B of the smart device 14. For example, the fine-tuning unit fine-tunes the generation AI model based on the feedback collected using the specific processing unit 290 of the data processing device 12. === Hard Collateral 1-2 === Each of the multiple elements, including the above-mentioned reception unit, generation unit, output unit, feedback collection unit, and fine-tuning unit, is realized, for example, by at least one of the smart glasses 214 and the data processing device 12. For example, the reception unit can input a customer's preference using the microphone 238 of the smart glasses 214. For example, the generation unit executes a generation AI using the specific processing unit 290 of the data processing device 12 to generate an optimal cocktail recipe. For example, the output unit outputs the generated cocktail recipe in video format using the speaker 240 of the smart glasses 214. For example, the feedback collection unit collects feedback from the bartender and customers using the camera 42 and microphone 238 of the smart glasses 214. For example, the fine-tuning unit fine-tunes the generation AI model based on the feedback collected using the specific processing unit 290 of the data processing device 12. === Hard Collateral 1-3 === Each of the multiple elements including the above-mentioned reception unit, generation unit, output unit, feedback collection unit, and fine-tuning unit is realized, for example, by at least one of the headset-type terminal 314 and the data processing device 12. For example, the reception unit can input a customer's request using the microphone 238 of the headset-type terminal 314. For example, the generation unit executes a generation AI using the specific processing unit 290 of the data processing device 12 to generate an optimal cocktail recipe. For example, the output unit outputs the generated cocktail recipe in video format using the display 343 of the headset-type terminal 314. For example, the feedback collection unit collects feedback from the bartender and customers using the camera 42 and microphone 238 of the headset-type terminal 314. For example, the fine-tuning unit fine-tunes the generation AI model based on the feedback collected using the specific processing unit 290 of the data processing device 12. === Hard Collateral 1-4 === Each of the multiple elements including the above-mentioned reception unit, generation unit, output unit, feedback collection unit, and fine-tuning unit is realized, for example, by at least one of the robot 414 and the data processing device 12. For example, the reception unit can input a customer's request using the microphone 238 of the robot 414. For example, the generation unit executes a generative AI using the specific processing unit 290 of the data processing device 12 to generate an optimal cocktail recipe. For example, the output unit outputs the generated cocktail recipe in video format using the speaker 240 of the robot 414. For example, the feedback collection unit collects feedback from the bartender and customers using the camera 42 and microphone 238 of the robot 414. For example, the fine-tuning unit fine-tunes the generative AI model based on the feedback collected using the specific processing unit 290 of the data processing device 12.

[0117] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0118] The reception unit can analyze the user's past cocktail order history and suggest cocktails based on the user's preferences. For example, it can suggest similar cocktails based on the taste and alcohol content of cocktails the user has ordered in the past. Also, if the user has a preference for a particular ingredient, it can preferentially suggest cocktails that use that ingredient. Furthermore, it can suggest cocktails that are suitable for a season or event based on cocktails the user has ordered for a particular season or event. This makes it possible to make more personalized cocktail suggestions by utilizing the user's past order history.

[0119] When generating a cocktail recipe, the generation unit can adjust the recipe taking into account the user's health condition. For example, if the user has an allergy, the generation unit can generate a cocktail using ingredients that do not contain the allergen. Also, if the user has a specific health goal (e.g., calorie restriction or carbohydrate restriction), the generation unit can generate a cocktail that suits that goal. Furthermore, if the user wants to consume a specific nutrient, the generation unit can generate a cocktail using ingredients that contain that nutrient. This makes it possible to provide cocktail recipes that are suited to the user's health condition.

[0120] The output unit can output the generated cocktail recipe in multiple languages. For example, the recipe can be output in English, Japanese, Spanish, or other languages ​​according to the user's language setting. The output unit can also appropriately translate the names of the cocktails and ingredients based on the language selected by the user. Furthermore, if the user is learning a different language, the output unit can support language learning by providing the recipe in that language. This allows the cocktail recipe to be provided according to the user's language setting.

[0121] The feedback collection unit can collect user feedback in real time and immediately reflect it in the generative AI model. For example, if a user provides feedback immediately after tasting a cocktail, that feedback is immediately reflected in the generative AI. The feedback collection unit can also use a simple questionnaire format or voice input when the user provides feedback. Furthermore, when the user provides feedback, the feedback collection unit can estimate the user's emotions using an emotion estimation function and adjust the content of the feedback based on those emotions. This makes it possible to collect and reflect feedback in real time.

[0122] The fine tuning unit can estimate the user's emotions and adjust the fine tuning parameters based on the estimated emotions. For example, if the user is relaxed, fine tuning can be performed based on detailed feedback. If the user is stressed, fine tuning can be performed based on simple feedback. Furthermore, if the user is in a hurry, fine tuning can be performed based on quick feedback. This makes it possible to fine tune according to the user's emotions.

[0123] The reception unit can analyze the user's social media activity and suggest related cocktails. For example, it can suggest cocktails related to places where the user has checked in on social media. It can also analyze the content of the user's social media posts and suggest cocktails based on that content. It can also suggest related cocktails based on the activity of the user's friends on social media. This makes it possible to make personalized cocktail suggestions that utilize social media activity.

[0124] When generating a cocktail recipe, the generation unit can estimate the user's emotions and adjust the ingredients and steps of the recipe based on the estimated emotions. For example, if the user is relaxed, the generation unit can generate a cocktail recipe using ingredients that have a relaxing effect. Also, if the user is stressed, the generation unit can generate a cocktail recipe using ingredients that have a stress-relieving effect. Furthermore, if the user is energetic, the generation unit can generate a cocktail recipe using ingredients that replenish energy. In this way, it is possible to provide cocktail recipes that correspond to the user's emotions.

[0125] The output unit can output the generated cocktail recipe in a visually appealing format. For example, it can generate a video showing how to make the cocktail in an animated format. It can also display the finished cocktail as a 3D model. Furthermore, it can provide visually appealing information by displaying the name of the cocktail and the names of its ingredients in colorful fonts. This makes it possible to provide a visually appealing cocktail recipe.

[0126] The feedback collection unit can estimate the user's emotions and adjust the feedback collection method based on the estimated emotions. For example, if the user is relaxed, detailed feedback can be requested. If the user is stressed, simple feedback can be requested. Furthermore, if the user is in a hurry, feedback can be collected quickly. This makes it possible to collect feedback according to the user's emotions.

[0127] The fine-tuning unit can integrate information from different data sources to enrich the fine-tuning data. For example, it can integrate customer feedback data and past learning data to perform fine-tuning. The fine-tuning unit can also integrate customer social media activity data to perform fine-tuning. Furthermore, the fine-tuning unit can also integrate customer geographic location information to perform fine-tuning. In this way, by integrating information from different data sources, it can provide richer fine-tuning data.

[0128] The processing flow of the second embodiment will be briefly explained below.

[0129] Step 1: The reception unit inputs the customer's preferences. The customer's preferences may include, for example, taste preferences, alcohol strength, whether or not a particular ingredient is included, etc. The reception unit can accept the customer's preferences by text input, voice input, image input, or other methods. Step 2: The generation unit uses a generation AI to analyze the information entered by the reception unit and generate the optimal cocktail recipe from a huge database. The generation AI generates the optimal cocktail recipe based on past cocktail recipes and customer feedback. The generation unit generates the cocktail recipe using a text generation AI (e.g., LLM) or a multimodal generation AI. Step 3: The output unit outputs the cocktail recipe generated by the generation unit in a video format. The video format includes the video length, resolution, format used, etc. The output unit generates a video that shows how to make the cocktail step by step. Step 4: The output unit outputs the appearance of the generated cocktail in an image format. The image format includes the image resolution, file format, display method, etc. The output unit generates an image of the finished cocktail. Step 5: The output unit outputs the generated cocktail name in text format. The text format includes the number of characters, font, display method, etc. The output unit generates a name such as "Fruity Sweet Cocktail." Step 6: The feedback collection department collects feedback from bartenders and customers. Feedback can be in the form of a survey, free-form comments, or evaluation criteria. The feedback collection department collects feedback such as "I would like it to be a little more sour" from a bartender. Step 7: The fine-tuning unit fine-tunes the generative AI model based on the feedback collected by the feedback collection unit. Fine-tuning includes how to adjust the model parameters and the type of data to use. The fine-tuning unit adjusts the generative AI model based on the collected feedback and reflects it in the next proposal.

[0130] 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.

[0131] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AIs include the data generation model 58, such as a neural network model (e.g., a neural network model), and a neural network model (e.g., a neural network model). 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 speech, text data indicating text, and image data indicating an image is also input to the data generation model 58. 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. The specification processing unit 290 performs the above-mentioned specification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AIs other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0132] Furthermore, the processing by the data processing system 10 described above is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart device 14, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart device 14. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart device 14 or an external device, and the smart device 14 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0133] The correspondence between each part and the device or control part is not limited to the above example, and various modifications are possible.

[0134] [Second embodiment] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0135] 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0136] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0137] 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.

[0138] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0139] 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 user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0140] 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.

[0141] 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.

[0142] 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.

[0143] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0144] In the smart glasses 214, the specific processing is performed by the processor 46. A specific processing program 60 is stored in the storage 50. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific processing program 60 executed on the RAM 48. The smart glasses 214 also have a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0145] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0146] 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.

[0147] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt containing an instruction, as well as inference data such as voice data representing speech, text data representing text, and image data representing an image. 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. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AI other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0148] The data processing system 210 according to the second embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 210 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart glasses 214, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart glasses 214. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the smart glasses 214 or an external device, etc., and the smart glasses 214 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0149] The correspondence between each part and the device or control part is not limited to the above example, and various modifications are possible.

[0150] [Third embodiment] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0151] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0152] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0153] 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.

[0154] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0155] 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 user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0156] 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.

[0157] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset 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.

[0158] 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.

[0159] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0160] In the headset type terminal 314, the identification process is performed by the processor 46. A identification program 60 is stored in the storage 50. The processor 46 reads the identification program 60 from the storage 50 and executes the read identification program 60 on the RAM 48. The identification process is realized by the processor 46 operating as a control unit 46A in accordance with the identification program 60 executed on the RAM 48. Note that the headset type terminal 314 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the identification processing unit 290 using these models.

[0161] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0162] 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.

[0163] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt containing an instruction, as well as inference data such as voice data representing speech, text data representing text, and image data representing an image. 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. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AI other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0164] The data processing system 310 according to the third embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 310 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the headset type terminal 314, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the headset type terminal 314. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the headset type terminal 314 or an external device, etc., and the headset type terminal 314 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0165] The correspondence between each part and the device or control part is not limited to the above example, and various modifications are possible.

[0166] [Fourth embodiment] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0167] 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.

[0168] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0169] 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.

[0170] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0171] 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 image sensor or a CCD image sensor, and captures images of the user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0172] 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.

[0173] The control object 443 includes a display device, LEDs in the eyes, and motors that drive 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.

[0174] 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.

[0175] 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.

[0176] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0177] In the robot 414, the processor 46 performs the identification process. The storage 50 stores the identification program 60. The processor 46 reads the identification program 60 from the storage 50 and executes the read identification program 60 on the RAM 48. The identification process is realized by the processor 46 operating as the control unit 46A in accordance with the identification program 60 executed on the RAM 48. The robot 414 also has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can perform the same process as the identification processing unit 290 using these models.

[0178] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0179] 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.

[0180] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt containing an instruction, as well as inference data such as voice data representing speech, text data representing text, and image data representing an image. 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. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AI other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0181] The data processing system 410 according to the fourth embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 410 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the robot 414, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the robot 414. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the robot 414 or an external device, etc., and the robot 414 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0182] The correspondence between each part and the device or control part is not limited to the above example, and various modifications are possible.

[0183] 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.

[0184] FIG. 9 illustrates 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 behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion encompasses both emotions 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.

[0185] 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.

[0186] 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).

[0187] 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 expressed, and when they approach the ideal, a state of pleasure is expressed. Emotions can also be created for robots, cars, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is expressed, and when they approach the ideal, a state of pleasure is expressed. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on speech emotion recognition and brain physiological signal analysis systems for emotions, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the area called "reaction," where sensation is dominant. The right half of the emotion map lists emotions belonging to the area called "situation," where situational awareness is dominant.

[0188] 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."

[0189] 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.

[0190] In the above embodiment, an example was given in which a specific process is performed by one computer 22, but the technology disclosed herein is not limited to this, and distributed processing of the specific process may be performed by multiple computers including computer 22.

[0191] 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.

[0192] 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.

[0193] 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.

[0194] The hardware resource for executing a specific process can be any of the following types of processors: A processor, for example, 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. A processor also includes 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.

[0195] The hardware resource that executes the specific process 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 process may be a single processor.

[0196] 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.

[0197] 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.

[0198] In the above example, the first to fourth embodiments have been described separately, but some or all of these embodiments may be combined. The smart device 14, smart glasses 214, headset terminal 314, and robot 414 are merely examples, and they may be combined, or other devices may be used. In the above example, the first and second embodiments have been described separately, but they may be combined.

[0199] 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, in order to avoid confusion and to 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.

[0200] 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.

[0201] [Explanation of symbols]

[0202] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot

Claims

1. A reception desk where customers can input their requests; a generation unit that analyzes the information input by the reception unit and generates an appropriate cocktail recipe from a large database; an output unit that outputs the cocktail recipes generated by the generation unit in video format; an output unit that outputs the appearance of the cocktail generated by the generation unit in the form of an image; an output unit that outputs the name of the cocktail generated by the generation unit in text format; a feedback collection unit that collects feedback on the cocktail recipe output by the output unit; and a fine-tuning unit that fine-tunes a model of the generative AI based on the feedback collected by the feedback collection unit. A system characterized by:

2. The reception unit Enter your wishes using the generated AI 2. The system of claim 1.

3. The generation unit Generative AI is used to generate appropriate cocktail recipes based on past cocktail recipes and customer feedback.

2. The system of claim 1.

4. The output unit Output the generated cocktail recipe in video format 2. The system of claim 1.

5. The output unit Output the appearance of the generated cocktail in image format 2. The system of claim 1.

6. The output unit Output the generated cocktail name in text format 2. The system of claim 1.

7. The feedback collection unit: Collect feedback from bartenders or customers 2. The system of claim 1.

8. The fine tuning unit Fine-tuning generative AI models based on collected feedback 2. The system of claim 1.

Images