System

Abstract

A system is provided.SOLUTION: A system comprising: means for receiving visual and auditory data associated with a memory from a user; means for analyzing the received data to extract features necessary for scent generation; means for generating a scent profile using a generated AI model based on the features; means for transmitting the scent profile; means for operating a scent generation device based on the transmitted scent profile to generate a scent; means for receiving user feedback regarding the generated scent; and means for optimizing the scent profile based on the feedback.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] Current media experiences rely primarily on sight and hearing, and new ways for users to feel emotions are needed. Conventional media content is one-way and therefore unable to fully provide the emotional experiences users desire. This leads to problems of users becoming bored and decreasing satisfaction. Furthermore, current media can only provide experiences that evoke memories and emotions to a limited extent, and richer experiences are needed. Therefore, this invention proposes a new system that evokes emotions and memories through the sense of smell, providing personalized emotional experiences. [Means for solving the problem]

[0005] The present invention solves the above-mentioned problems with a system that includes the following elements. Specifically, the system includes a means for receiving visual and auditory data from a user, a means for analyzing the data and extracting features necessary for scent generation, and a means for creating a scent profile using a generative AI model based on the extracted features. The system also includes a means for transmitting the created scent profile and a means for operating a scent generation device based on the scent profile to generate a scent. The system also includes a means for receiving user feedback regarding the generated scent and a means for optimizing the scent profile based on the feedback. This allows for constantly improving the user's experience and providing a personalized scent experience. Furthermore, the system includes a means for receiving supplemental information from the user, such as weather, season, time, location, and vegetation, allowing for more accurate scent generation. Furthermore, the system includes a means for generating a scent that evokes specific emotions or memories based on features extracted from the analyzed data, thereby further deepening the user's emotional experience.

[0006] "Visual and Audio Data" refers to visual and audio information provided by users, such as photographs, video, audio, and text.

[0007] "Features" refer to elements and attributes extracted from visual and auditory data that are necessary for scent generation. Examples include season, weather, location, and emotion.

[0008] "Generative AI model" refers to an artificial intelligence model that creates scent profiles using predictive and generative algorithms.

[0009] "Fragrance profile" refers to information specifying the combination and ratio of fragrances to produce a particular fragrance.

[0010] "Scent generating device" refers to a device for actually generating and emitting a scent based on a scent profile.

[0011] "Feedback" refers to the user's impressions of the scent they experienced and their opinions on areas for improvement.

[0012] "Optimization" refers to the process of refining the scent profile based on user feedback to improve the user experience.

[0013] "Supplementary information" refers to additional information such as weather, season, time, location, vegetation, etc. provided by the user in addition to visual and auditory data to improve the accuracy of scent generation. [Brief explanation of the drawings]

[0014] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0015] 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.

[0016] First, the terms used in the following description will be explained.

[0017] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0018] 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.

[0019] 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.

[0020] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0021] 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."

[0022] [First embodiment]

[0023] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0024] 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.

[0025] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0026] 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.

[0027] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0028] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0029] 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.

[0030] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] 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.

[0032] 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.

[0033] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0034] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0035] The present invention provides a system that receives visual and auditory data, analyzes the data, and generates a scent to provide a user with a personalized scent experience. This system is mainly composed of a server, a terminal, and a user.

[0036] 1. Data collection

[0037] User

[0038] Users upload visual and audio data related to their memories to the system. This data includes photos, videos, text, and voice data. Supplementary information, such as weather, season, time, location, and vegetation, can also be added. For example, a user can upload photos taken during a trip and enter text about their impressions.

[0039] 2. Data Analysis

[0040] server

[0041] The server analyzes the visual and auditory data received from the user and extracts the features necessary for scent generation. This analysis uses image recognition algorithms and natural language processing (NLP) technology. For example, it can identify the landscape, season, weather, and location from photo data, and analyze emotions from text data.

[0042] 3. Generation of aroma profiles

[0043] server

[0044] Based on the analyzed features, a generative AI model is used to create a scent profile. This profile includes the combination and ratio of fragrances used within the system. For example, to recreate the scent of a spring day, specific fragrance ratios could be set, such as 50% cherry blossom scent, 30% fresh green scent, and 20% warm air scent.

[0045] server

[0046] The generated scent profile is sent to the device.

[0047] 4. Fragrance production

[0048] Terminal

[0049] The device operates the built-in scent generating device based on the scent profile received from the server. It mixes the required scents from the scent cartridges in the specified ratio to generate the scent. This process is performed in real time, and the generated scent is diffused around the user.

[0050] 5. Gathering Feedback

[0051] User

[0052] Users experience the scents generated and provide feedback on their impressions and suggestions for improvement via a dedicated mobile or web app.

[0053] 6. Profile Optimization

[0054] server

[0055] The server receives feedback from users and stores it as evaluation data for the scent profile. Based on this evaluation data, the generative AI model is retrained and the scent profile is optimized. For example, if a user provides feedback saying, "I would like a more refreshing scent," the server creates a new scent profile and reflects it the next time the user uses the device.

[0056] Specific examples

[0057] Data collection and analysis

[0058] A user uploads photos from a trip to France and an episode titled "Spring in Paris: A walk along the Eiffel Tower and the Seine River" to the system. The server uses image recognition to identify the Eiffel Tower, the Seine River, and spring scenery from the photo, and extracts the keywords "Paris," "spring," and "walk" from the text.

[0059] Scent profile generation and scent generation

[0060] Based on the extracted features, the server creates a scent profile, mainly consisting of "floral spring flower scent" and "fresh riverside scent," and sends it to the device. The device then uses a scent generation device to generate a scent based on the profile and diffuse it around the user.

[0061] Gathering feedback and optimizing your profile

[0062] The user gives feedback such as, "The floral scent was good, but I'd like it to be a little fresher." The server analyzes the feedback and reflects in the generative AI model a suggestion to increase the ratio of fresh fragrances in the next scent profile. The new profile is reflected the next time the user uses the product.

[0063] In this way, the present invention is a system that continuously provides users with personalized moving experiences.

[0064] The processing flow will be explained below.

[0065] Step 1:

[0066] User

[0067] Users upload visual and audio data to the system, providing, for example, photos of their trip and textual anecdotes about it, and can also provide supplemental information such as weather, season, time, location, and vegetation.

[0068] Step 2:

[0069] server

[0070] The server stores the visual and auditory data and supplementary information received from the user, and securely stores the data using a database or cloud storage.

[0071] Step 3:

[0072] server

[0073] The server analyzes the stored data, using image recognition algorithms to identify key objects and scenes from the photos, and natural language processing (NLP) techniques to extract emotions and keywords from the text data, thereby obtaining the features necessary for generating the scent.

[0074] Step 4:

[0075] server

[0076] The server uses a generative AI model based on the extracted features to create a scent profile, which includes the combinations and proportions of fragrances used within the system.

[0077] Step 5:

[0078] server

[0079] The server then sends the created scent profile to the device, securely transferring the data using a communication protocol.

[0080] Step 6:

[0081] Terminal

[0082] The device operates the built-in scent generating device based on the scent profile received from the server, mixing the required scents from the scent cartridges in a specific ratio to create the specified scent.

[0083] Step 7:

[0084] Terminal

[0085] The device then uses a diffuser to diffuse the generated scent around the user, adjusting it to spread throughout the room so that the user can experience the scent.

[0086] Step 8:

[0087] User

[0088] Users can experience the generated scent and provide feedback on their impressions and suggestions for improvement through a dedicated application, such as "I'd like a more floral scent."

[0089] Step 9:

[0090] server

[0091] The server receives feedback from users and stores it as evaluation data, which is used to retrain the generative AI model and optimize the scent profile.

[0092] Step 10:

[0093] server

[0094] The server then sends the new optimized scent profile to the device to be reflected in the next scent generation, thereby continuously improving the user experience.

[0095] Through the above steps, the present invention provides a personalized scent experience to the user.

[0096] Example 1

[0097] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0098] Conventional scent generation systems were unable to utilize visual and auditory data to generate scents, making it difficult to provide a personalized scent experience based on a user's specific memories and emotions. Furthermore, mechanisms for effectively incorporating user feedback on the generated scent and optimizing the scent profile were not adequately developed. Furthermore, supplementary information such as weather, season, time, location, and vegetation could not be reflected, resulting in a lower quality of overall experience.

[0099] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0100] In this invention, the server includes means for receiving visual and auditory data from a user, means for analyzing the received data using an image recognition algorithm and natural language processing technology to extract features necessary for scent generation, means for creating a scent profile using a generative AI model based on the features, means for transmitting the scent profile to a terminal, means for operating a scent generation device on the terminal based on the transmitted scent profile to generate a scent, means for receiving user feedback regarding the generated scent through a dedicated application, means for optimizing the scent profile based on the feedback, and means for receiving supplementary information such as weather, season, time, location, and vegetation from the user. This makes it possible to provide a personalized scent experience based on the user's memories and emotions and to optimize the scent profile to reflect that feedback.

[0101] "Visual data" refers to visual information such as photos and videos provided by users.

[0102] "Auditory data" refers to aural information such as text or voice data provided by a user.

[0103] "Features" refer to the information elements necessary for scent generation, extracted using image recognition algorithms and natural language processing technology.

[0104] "Generative AI model" refers to an artificial intelligence system that uses machine learning algorithms to generate scent profiles.

[0105] "Fragrance profile" means data that indicates the combination and proportions of fragrance ingredients required to produce a particular fragrance.

[0106] "Scent generation device" refers to a mechanical device for generating an actual scent based on a scent profile.

[0107] "Feedback" refers to the user's impressions of the scent they experienced and their opinions on areas for improvement.

[0108] "Supplemental information" refers to user-provided background information for the visual and audio data, such as weather, season, time, location, vegetation, etc.

[0109] "Image recognition algorithm" refers to an algorithm that automatically identifies specific objects, scenes, and situations from photos and videos.

[0110] "Natural language processing technology" refers to algorithms for analyzing meaning and emotions from text and voice data.

[0111] "User" refers to an individual who utilizes the system to provide visual and auditory data and receive a personalized scent experience.

[0112] "Terminal" refers to a device, such as a computer or smartphone, that a user uses to upload visual and auditory data and experience the generated scent.

[0113] MODE FOR CARRYING OUT THE INVENTION

[0114] This invention is a system that receives visual and auditory data, analyzes it, and generates a scent to provide a user with a personalized scent experience. This system is mainly composed of a server, a terminal, and a user. The specific roles and operation procedures of each component are described below.

[0115] 1. Data collection

[0116] User

[0117] Users log in to the system using a device such as a smartphone or PC via a dedicated application or browser. They upload visual and auditory data related to their memories, including photos, videos, text, and voice data. They also enter supplementary information, such as weather, season, time, location, and vegetation, as needed. For example, a user might upload an episode of "Spring in Paris, a walk with the Eiffel Tower and along the Seine River," along with photos.

[0118] 2. Data Analysis

[0119] server

[0120] The server analyzes the visual and auditory data received from the user and extracts the features necessary for scent generation. Visual data (photos and videos) is analyzed using image recognition algorithms (e.g., TensorFlow and OpenCV). For example, the Eiffel Tower, the Seine River, and spring scenery in a photo are identified. Next, auditory data (text and voice data) is analyzed using natural language processing (NLP) techniques (e.g., Google BERT and spaCy) to extract emotions and keywords. For example, keywords such as "Paris," "spring," and "walk" and emotions such as "fun" and "relaxed" are analyzed.

[0121] 3. Generation of aroma profiles

[0122] server

[0123] The server uses a generative AI model (e.g., GPT-4) to create a scent profile based on the analyzed features. The user inputs a prompt to the generative AI model: "Please create a scent profile inspired by spring in Paris, the Eiffel Tower, and a walk along the Seine." As a result, the server calculates specific fragrance ratios and combinations. For example, it generates a profile that is "40% floral spring flower scent, 30% fresh riverside scent, and 30% warm sunshine scent."

[0124] server

[0125] The generated scent profile is sent to the terminal in data format.

[0126] 4. Fragrance production

[0127] Terminal

[0128] The device operates its built-in scent-generating device based on the scent profile received from the server. The control software inside the device extracts and mixes the scent from the scent cartridge in the specified ratio to create the scent. This process is carried out in real time, and the generated scent is diffused around the user, providing them with a personalized scent experience.

[0129] 5. Gathering Feedback

[0130] User

[0131] Users experience the generated scent and provide feedback on their impressions and suggestions for improvement via a dedicated mobile or web app. For example, they can enter specific opinions such as, "I liked the floral scent, but I'd like it to be a little fresher." The feedback is then sent to the server.

[0132] 6. Profile Optimization

[0133] server

[0134] The server receives feedback from users and stores it as scent profile evaluation data. Based on this, it retrains the generative AI model and optimizes the scent profile. For example, based on feedback that a "slightly fresher scent" is desired, a new profile is created that incorporates more fresh scent components. The next time the user uses the service, a new scent profile that reflects these improvements will be provided.

[0135] In this way, the present invention is a system that provides a user with a personalized and inspiring scent experience.

[0136] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0137] Step 1: Collect data

[0138] User

[0139] Users use devices such as smartphones or PCs to log in via a dedicated application or browser on the system and upload visual and auditory data related to their memories.

[0140] input

[0141] Photos, videos, text, voice data, and supplementary information (weather, season, time, location, vegetation) provided by users.

[0142] output

[0143] The visual and audio data and accompanying supplemental information are transmitted to the server side.

[0144] Specific actions

[0145] The user presses the "Upload" button within the application, selects photos, videos, text, or voice data from the file selection screen, and presses the send button.

[0146] Step 2: Analyze the data

[0147] server

[0148] The server analyzes the received visual and auditory data and extracts the features necessary for scent generation.

[0149] input

[0150] Visual data (photos, videos), auditory data (text, voice data), and supplemental information sent by the user.

[0151] output

[0152] Analyzed feature information (e.g., location, season, scenery, emotion, keywords).

[0153] Data processing / data calculation

[0154] The server analyzes the visual data using image recognition algorithms (e.g., TensorFlow and OpenCV) to identify scenery, locations, etc. It also analyzes the auditory data using natural language processing techniques (e.g., Google BERT and spaCy) to extract emotions and keywords.

[0155] Specific actions

[0156] Image recognition algorithms run on the server, identifying the Eiffel Tower, the Seine River, and spring scenery from the photo, while natural language processing technology analyzes the text to extract keywords like "Paris," "spring," and "walk," as well as emotions like "fun" and "relaxing."

[0157] Step 3: Generate an aroma profile

[0158] server

[0159] The server creates a scent profile using a generative AI model based on the analyzed features.

[0160] input

[0161] Analyzed feature information (e.g., location, season, scenery, emotion, keywords).

[0162] output

[0163] Fragrance profile data (specific fragrance combinations and ratios).

[0164] Data processing / data calculation

[0165] The server inputs a prompt into a generative AI model (e.g., GPT-4) to generate a scent profile. The prompt is, "Please create a scent profile inspired by spring in Paris, the Eiffel Tower, and a walk along the Seine."

[0166] Specific actions

[0167] The generative AI model calculates based on the prompt and creates a specific scent profile, such as "40% floral spring flower scent, 30% fresh riverside scent, and 30% warm sunshine scent."

[0168] Step 4: Submit your scent profile

[0169] server

[0170] The server sends the generated scent profile to the terminal.

[0171] input

[0172] Generated scent profile data.

[0173] output

[0174] Scent profile data sent to the device.

[0175] Specific actions

[0176] The scent profile data is transferred from the server to the terminal via the network.

[0177] Step 5: Scent generation

[0178] Terminal

[0179] The terminal operates the built-in scent generating device based on the scent profile received from the server.

[0180] input

[0181] Scent profile data sent from the server.

[0182] output

[0183] The scent produced is diffused into the surrounding area.

[0184] Data processing / data calculation

[0185] The control software in the device extracts and mixes the fragrance from the fragrance cartridge in the specified ratio.

[0186] Specific actions

[0187] The fragrance generating device extracts the required fragrances from the fragrance cartridges and mixes and diffuses them based on the profile.

[0188] Step 6: Gather feedback

[0189] User

[0190] Users experience the generated scent and provide feedback on their impressions and suggestions for improvement.

[0191] input

[0192] User feedback information (impressions, areas for improvement).

[0193] output

[0194] Feedback data sent to the server.

[0195] Specific actions

[0196] Users fill out and submit a feedback form in a dedicated mobile or web app.

[0197] Step 7: Optimize your profile

[0198] server

[0199] The server receives user feedback and stores it as evaluation data for the scent profile. Based on this evaluation data, the generative AI model is retrained and the scent profile is optimized.

[0200] input

[0201] User feedback data.

[0202] output

[0203] Optimized scent profile data.

[0204] Data processing / data calculation

[0205] Based on the feedback, the generative AI model is retrained to generate new scent profiles.

[0206] Specific actions

[0207] The server analyzes the feedback data, updates the generative AI model, and reflects it in the next scent profile.

[0208] (Application example 1)

[0209] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0210] The present invention relates to a technology for generating scents by analyzing visual and auditory data, and aims to realize a more interactive and immediate experience in a system that provides users with a personalized scent experience. Another object of the present invention is to provide customers with a special shopping experience by analyzing customers' visual and auditory data in a store in real time and generating a customized scent on the spot.

[0211] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0212] In this invention, the server includes means for receiving visual and auditory data related to memories from a user, means for analyzing the received data and extracting features necessary for scent generation, means for creating a scent profile using a generative AI model based on the features, means for transmitting the scent profile, means for operating a scent generation device based on the transmitted scent profile to generate a scent, means for receiving user feedback on the generated scent, means for optimizing the scent profile based on the feedback, and means for collecting customer visual and auditory data in a store and generating a personalized scent on the spot, thereby providing users with a personalized and moving experience and improving the customer's purchasing experience.

[0213] "Visual data" refers to visually related data such as images and videos provided by users.

[0214] "Auditory data" refers to data related to the sense of hearing, such as voice or music, provided by the user.

[0215] "Scent generation" refers to the process of physically creating a specific scent based on extracted data.

[0216] "Features" refers to information necessary for scent generation extracted from visual and auditory data.

[0217] "Generative AI model" refers to an artificial intelligence model for creating scent profiles based on data analysis.

[0218] A "fragrance profile" is a setting that defines which specific fragrances are used and in what proportions.

[0219] An "aroma generating device" is a device that generates an actual aroma based on an aroma profile.

[0220] "Feedback" refers to opinions and information such as impressions from users after use and suggestions for improvement.

[0221] "In-store" refers to places where customers visit, such as commercial facilities, sales areas, and exhibition spaces.

[0222] "Personalization" refers to customizing something to suit the preferences and experiences of individual users.

[0223] The present invention relates to a system for providing a personalized scent experience to customers in a physical store. Hereinafter, an embodiment of the system will be described.

[0224] 1. Data Collection

[0225] User:

[0226] Users use smartphones or smart glasses to collect visual and auditory data, including the sights and sounds of the store.

[0227] 2. Data Analysis

[0228] server:

[0229] The server analyzes the visual and auditory data received from the user. It uses OpenCV for camera control to analyze the visual data, and the SpeechRecognition library to analyze the audio data. Scenery, objects, and seasonal sensations are extracted from the visual data, while the sound atmosphere and text information are extracted from the audio data.

[0230] 3. Generation of Aroma Profiles

[0231] server:

[0232] Based on the analyzed features, the server creates a scent profile using a generative AI model that determines which fragrances to mix and in what proportions based on information obtained from the user's visual and auditory data.

[0233] 4. Fragrance production

[0234] Device:

[0235] The terminal operates a scent-generating device based on the scent profile received from the server. This device mixes the required fragrances in the specified ratio and diffuses them around the customer. The AromaBlender device generates the scent in real time.

[0236] 5. Collect feedback

[0237] User:

[0238] Users provide feedback about the scents generated through a dedicated mobile or web app.

[0239] 6. Profile Optimization

[0240] server:

[0241] The server analyzes user feedback and stores the scent profile as evaluation data, which is then used to retrain the generative AI model and optimize the scent profile.

[0242] Specific examples

[0243] A customer wears smart glasses and walks around the store. The scenery of the flower section is captured as visual data, and the music in the store is recorded as auditory data. The server analyzes this data and extracts the characteristics of "floral scents" and "refreshing scents." A scent profile is created based on the generative AI model, and the AromaBlender device generates the scent and provides it to the customer. If the customer gives feedback such as "I liked the floral scent, but I would like it to be a little more refreshing," the scent profile that reflects this feedback will be used the next time.

[0244] Example prompt sentence:

[0245] Smart glasses installed in stores collect the sights and sounds customers see and hear in real time. When a customer is in a specific area of ​​the store, the data is analyzed to generate the optimal scent for that area, and that scent is then delivered through a scent-generating device. For example, if a customer is in the flower section, a scent generated based on "floral scents" and "forest scents" will be distributed.

[0246] In this way, the present invention realizes a system that continuously provides users with personalized and moving experiences, and can also improve customer engagement through special shopping experiences in physical stores.

[0247] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0248] Step 1:

[0249] Data collection

[0250] Users use smartphones or smart glasses to collect visual and auditory data within a store. Specifically, when a user is in a specific area, the camera captures the scenery and the microphone records the surrounding sounds. The input is the scenery seen by the customer (image data) and the sounds heard (audio data), and this data is saved on the device.

[0251] Step 2:

[0252] Sending data

[0253] The device transmits the collected visual and auditory data to a server. Specifically, if the device is connected to the Internet, the collected image data and audio data are uploaded to the server. The input is the data stored on the device, and the output is the data transmitted to the server.

[0254] Step 3:

[0255] Data analysis

[0256] The server analyzes the received visual and auditory data. Specifically, it runs an image recognition algorithm using OpenCV on the visual data to identify scenery and objects. For the auditory data, it uses the SpeechRecognition library to convert the audio data into text and extract specific keywords and emotions. The input is the received visual and auditory data, and the output is feature data as the analysis result.

[0257] Step 4:

[0258] Generation of aroma profiles

[0259] The server uses a generative AI model to create a scent profile based on the analyzed features. Specifically, the AI ​​model calculates the optimal scent combination and its ratio based on the input feature data. The input is feature data, and the output is a scent profile.

[0260] Step 5:

[0261] Sending your scent profile

[0262] The server sends the generated scent profile to the terminal. Specifically, scent profile data is sent from the server to the terminal. The input is the scent profile, and the output is the scent profile sent to the terminal.

[0263] Step 6:

[0264] Fragrance production

[0265] The terminal operates the scent generation device based on the scent profile received from the server. Specifically, it uses the AromaBlender device to mix the fragrances defined in the scent profile in the specified ratio and diffuse the scent around the customer. The input is the scent profile and the output is the generated scent.

[0266] Step 7:

[0267] Collecting feedback

[0268] Users provide feedback on the generated scent, including their impressions and suggestions for improvement. Specifically, they use a dedicated mobile or web app to enter the scent characteristics they perceived and desired changes in text format. The input is the user's impressions and suggestions for improvement, and the output is feedback data.

[0269] Step 8:

[0270] Profile Optimization

[0271] The server analyzes the feedback from users and stores it as evaluation data for the scent profile. Specifically, it retrains the generative AI model based on the feedback content and optimizes the scent profile. The input is the feedback data, and the output is the optimized scent profile.

[0272] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0273] The present invention is a system that analyzes visual and auditory data received from a user and generates a scent to provide the user with a personalized scent experience. By combining this system with an emotion engine, it also has the function of recognizing the user's emotional state and reflecting this in the scent generation. A specific embodiment of the system is described below.

[0274] 1. Data collection

[0275] User

[0276] Users can provide visual and audio data related to their memories, such as photos and videos of their trips, as well as text and audio descriptions of their experiences. They can also input supplementary information such as weather, season, time, location, and vegetation.

[0277] 2. Data Analysis

[0278] server

[0279] The server stores and analyzes the received data, using image recognition algorithms to identify key objects and scenes from photos and videos, and natural language processing (NLP) techniques to extract emotions and keywords from text and audio data.

[0280] 3. Utilizing the Emotion Engine

[0281] server

[0282] The emotion engine installed on the server recognizes the user's emotions based on information extracted from visual and auditory data. For example, if image analysis provides photos or videos showing many smiling faces, the emotion engine will recognize the emotion of "joy."

[0283] 4. Generation of aroma profiles

[0284] server

[0285] Based on the extracted features and the emotions recognized by the emotion engine, a generative AI model is used to create a scent profile. This scent profile includes the combination and proportion of fragrances used within the system. For example, if the emotion of "joy" is recognized, a profile containing a lot of floral fragrances is generated.

[0286] server

[0287] The generated scent profile is sent to the device.

[0288] 5. Aroma production

[0289] Terminal

[0290] The device operates the built-in scent generating device based on the received scent profile, mixing the required scents from the scent cartridges in the specified ratio to create the specified scent.

[0291] Terminal

[0292] The generated scent is diffused around the user using a diffuser, allowing the user to experience a scent that matches the predicted emotion.

[0293] 6. Gather feedback and optimize your profile

[0294] User

[0295] Users can experience the generated scent and provide feedback on their impressions and suggestions for improvement through a dedicated application, such as "I wish the scent was a little stronger."

[0296] server

[0297] The server receives user feedback and stores it as evaluation data. It then retrains the generative AI model based on the feedback and optimizes the scent profile. The new profile is reflected in the next scent generation.

[0298] Specific examples

[0299] Data collection and analysis

[0300] A user uploads a photo taken at the beach in the summer along with the text "A fun summer day" to the system. The server uses image recognition to identify the sea, beach, and summer features in the photo, and extracts the positive emotion "It was fun" from the text.

[0301] Utilizing the Emotion Engine

[0302] The emotion engine installed on the server recognizes the emotion of "fun" from this data.

[0303] Scent profile generation and scent generation

[0304] Based on the extracted features and the recognized emotions, the server creates a scent profile mainly consisting of "fruity scents" and "sea breeze scents" and sends it to the device. The device then uses a scent generation device to generate a scent based on the profile and diffuses it around the user.

[0305] Gathering feedback and optimizing your profile

[0306] The user gives feedback such as "I want a more fruity scent." The server analyzes this feedback, reflects it in the generative AI model, and suggests increasing the ratio of fruity fragrances in the next scent profile, then sends the new profile to the device.

[0307] In this way, the present invention is a system that utilizes an emotion engine to provide a personalized scent experience that is tailored to the user's emotional state.

[0308] The processing flow will be explained below.

[0309] Step 1:

[0310] User

[0311] Users upload visual and audio data related to their memories (e.g., photos and videos of their trips, as well as text and audio descriptions of their experiences) to the system, and can also enter supplemental information such as weather, season, time, location, and vegetation.

[0312] Step 2:

[0313] server

[0314] The server securely stores the visual and auditory data received from the user, as well as any supplementary information, using a database or cloud storage.

[0315] Step 3:

[0316] server

[0317] The server analyzes the stored data. First, it uses image recognition algorithms to identify key objects and scenes (e.g., landscapes, buildings, seasons, and weather) from photos and videos. Next, it uses natural language processing (NLP) techniques to analyze the text and audio data and extract emotions and keywords (e.g., happy, sad, and place names).

[0318] Step 4:

[0319] server

[0320] The server uses an emotion engine to recognize the user's emotions based on the analyzed features and extracted emotions. For example, if the emotion "fun" is extracted from image and text analysis, the emotion engine will recognize this as "joy."

[0321] Step 5:

[0322] server

[0323] A generative AI model is used based on the emotion recognized by the emotion engine and the feature values ​​to create a scent profile. This scent profile includes the appropriate combination of fragrances and their proportions. For example, if the emotion of "joy" is recognized, a profile containing a lot of floral fragrances is generated.

[0324] Step 6:

[0325] server

[0326] The server then sends the generated scent profile to the device, transferring the data using a secure communication protocol.

[0327] Step 7:

[0328] Terminal

[0329] The device operates the built-in scent generating device based on the received scent profile, mixing the required scents from the scent cartridges in the specified ratio to create the specified scent.

[0330] Step 8:

[0331] Terminal

[0332] The generated scent is diffused around the user using a diffuser, allowing the user to experience a scent that matches the recognized emotion.

[0333] Step 9:

[0334] User

[0335] Users can experience the generated scent and provide feedback on their impressions and suggestions for improvement through a dedicated application. For example, they could input their opinion, such as, "I want a more fruity scent."

[0336] Step 10:

[0337] server

[0338] The server receives feedback from users and stores it as evaluation data. Based on this feedback, the generative AI model is retrained and the scent profile is optimized. The new profile is reflected the next time a scent is generated.

[0339] Specific examples

[0340] Data collection and analysis

[0341] Step 1-3: User and Server

[0342] A user uploads a photo they took at the beach in the summer along with the text "It was a fun summer day" to the system. The server uses an image recognition algorithm to identify the sea, beach, and summer scenery from the received photo, and then uses NLP technology to extract the positive emotion of "It was fun" from the text.

[0343] Utilizing the Emotion Engine

[0344] Step 4: Server

[0345] The emotion engine installed on the server recognizes the emotion of "fun" from this data.

[0346] Scent profile generation and scent generation

[0347] Step 5-8: Server and Terminal

[0348] Based on the extracted features and the recognized emotions, the server creates a scent profile mainly consisting of "fruity scents" and "sea breeze scents" and sends it to the device. The device then uses a scent generation device to generate a scent based on the profile and diffuse it around the user.

[0349] Gathering feedback and optimizing your profile

[0350] Steps 9-10: Users and Servers

[0351] The user gives feedback such as "I want a more fruity scent." The server analyzes this feedback, reflects it in the generative AI model, and suggests increasing the ratio of fruity fragrances in the next scent profile, then sends the new profile to the device.

[0352] In this way, the present invention is a system that utilizes an emotion engine to provide a personalized scent experience that is tailored to the user's emotional state.

[0353] Example 2

[0354] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0355] Conventional technologies have struggled to generate personalized scents based on a user's emotional state and memories. Furthermore, the process of efficiently collecting user feedback and optimizing scent profiles has been ineffective. To address these challenges, the present invention aims to provide a system that analyzes a user's visual and auditory data and generates scents based on their emotions.

[0356] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving visual and auditory data from a user, means for saving the received data, means for analyzing the saved data and identifying key objects and scenes using an image recognition algorithm, means for extracting emotions and keywords from text and voice data using natural language processing technology, means for recognizing emotions based on features extracted from the visual and auditory data, means for creating a scent profile using a generative AI model based on the recognized emotions, means for transmitting the created scent profile to a terminal, means for operating a scent generation device based on the received scent profile to generate a scent, means for diffusing the generated scent around the user, means for receiving user feedback regarding the generated scent, and means for optimizing the scent profile based on the received feedback. This makes it possible to generate a scent personalized to the user's emotional state and optimize the scent profile based on the feedback.

[0357] "Visual data" refers to visually related information, such as images and videos provided by users.

[0358] "Auditory data" refers to information related to the sense of hearing, such as voice or music provided by the user.

[0359] "Supplementary information" refers to additional information related to the visual and auditory data, such as weather, season, time, location, vegetation, etc.

[0360] "Database" means a digital information management system for storing visual data, auditory data, and supplemental information.

[0361] An "image recognition algorithm" is a technology that analyzes images and videos and recognizes specific objects and scenes.

[0362] "Natural language processing technology" is a technology for analyzing text and voice data and extracting emotions and keywords.

[0363] The "emotion engine" is a technology for recognizing a user's emotions based on features extracted from visual and auditory data.

[0364] A "generative AI model" is an artificial intelligence model for creating scent profiles based on recognized emotions.

[0365] A "fragrance profile" is information that defines a combination of multiple fragrances and their proportions.

[0366] A "terminal" is a device that generates a scent based on a scent profile and diffuses it around the user.

[0367] An "aroma generating device" is a device that extracts and mixes fragrances from a fragrance cartridge to generate a fragrance.

[0368] A "diffuser" is a device that diffuses the generated scent around the user.

[0369] "Feedback" refers to the user's thoughts and suggestions for improvement regarding the generated fragrance.

[0370] "Optimization" refers to the process of improving the scent profile based on user feedback.

[0371] The present invention is a system that analyzes visual and auditory data provided by a user and generates a scent based on the user's emotions. This system operates in cooperation with a server, a terminal, and the user.

[0372] 1. Data collection

[0373] User

[0374] Through a dedicated application, users provide visual and audio data related to their memories, such as photos, videos, text, and audio, and can also input supplemental information such as weather, season, time, location, and vegetation.

[0375] 2. Data storage

[0376] server

[0377] The server stores the received data in a database such as MongoDB or MySQL. The data is classified as visual data, auditory data, and supplementary information.

[0378] 3. Image Recognition and Natural Language Processing

[0379] server

[0380] The server uses image recognition algorithms (e.g., OpenCV or TensorFlow) to identify key objects and scenes from photos and videos. It also uses natural language processing (NLP) techniques (e.g., BERT) to extract emotions and keywords from text and audio data. It converts audio data into text using the Google Speech-to-Text API for analysis.

[0381] 4. Utilizing the Emotion Engine

[0382] server

[0383] The emotion engine installed on the server recognizes the user's emotions based on features extracted from visual and auditory data. For example, if there are many photos of smiling faces or positive keywords such as "it was fun," it will recognize the emotions of "joy" and "enjoyment."

[0384] 5. Generation of aroma profiles

[0385] server

[0386] The server uses a generative AI model (e.g., GPT-3) to create a scent profile based on the recognized emotion. This scent profile includes a combination of fragrances and their proportions. For example, if the emotion of "joy" is recognized, a profile containing a lot of floral fragrances is generated.

[0387] 6. Submit your scent profile

[0388] server

[0389] The server sends the created scent profile to the device in JSON format.

[0390] 7. Scent production and diffusion

[0391] Terminal

[0392] The device operates a built-in scent generating device (e.g., an Arduino-controlled diffuser) based on the received scent profile. It mixes the required fragrances from the fragrance cartridges in the specified ratio to generate the specified scent. It also diffuses the generated scent around the user using the diffuser.

[0393] 8. Gather feedback and optimize

[0394] User

[0395] Users can experience the generated scent and provide feedback on their impressions and suggestions for improvement through a dedicated application. For example, they could say, "I'd like a more fruity scent."

[0396] server

[0397] The server analyzes the received feedback and retrains the generative AI model to optimize the scent profile. The new profile is reflected in the next scent generation.

[0398] Specific examples

[0399] Data collection and analysis

[0400] A user uploads a photo taken at the beach in the summer along with the text "A fun summer day" to the system. The server uses an image recognition algorithm to identify the characteristics of the sea, beach, and summer from the photo, and then uses natural language processing technology to extract the positive emotion of "It was fun" from the text.

[0401] Utilizing the Emotion Engine

[0402] The emotion engine installed on the server recognizes the emotion of "fun" from this data.

[0403] Scent profile generation and scent generation

[0404] Based on the extracted features and the recognized emotions, the server uses a generative AI model to create a scent profile mainly consisting of "fruity scents" and "sea breeze scents" and sends it to the device. The device then uses a scent generation device to generate a scent based on the profile and diffuses it around the user.

[0405] Gathering feedback and optimizing your profile

[0406] The user gives feedback such as "I want a more fruity scent." The server analyzes this feedback, reflects it in the generative AI model, and suggests increasing the ratio of fruity fragrances in the next scent profile, then sends the new profile to the device.

[0407] Prompt Sentence Examples

[0408] "Analyze a user-provided summer beach photo and the text 'A fun summer day' and generate a corresponding scent profile. If the user provides feedback that they would like a more fruity scent, explain how you would optimize the scent profile."

[0409] As a result, the present invention provides a system that provides a personalized scent experience tailored to the user's emotional state and optimizes the scent profile based on user feedback.

[0410] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0411] Step 1: Collect data

[0412] User

[0413] Using a dedicated application, users upload photos and videos of their trip, text and audio impressions, and supplementary information such as weather, season, time, location, and vegetation. The inputs are visual data (photos and videos), auditory data (audio), and supplementary information. The output is this data sent to the server.

[0414] Step 2: Save your data

[0415] server

[0416] The server stores the received visual data, auditory data, and supplementary information in a database. The input is the data sent by the user. The output is the data stored in the database. Database systems such as MongoDB and MySQL are used at this stage.

[0417] Step 3: Perform image recognition

[0418] server

[0419] The server uses image recognition algorithms (e.g., OpenCV or TensorFlow) to analyze the stored photo and video data. It identifies key objects and scenes and stores them as metadata. The input is visual data stored in a database. The output is metadata for the identified objects and scenes.

[0420] Step 4: Performing Natural Language Processing

[0421] server

[0422] The server uses natural language processing (NLP) techniques (e.g., BERT) to extract sentiment and keywords from text and audio data. The audio data is converted to text using the Google Speech-to-Text API and then analyzed. The input is the text and audio data stored in the database. The output is the extracted sentiment and keywords.

[0423] Step 5: Leverage the Emotion Engine

[0424] server

[0425] The emotion engine installed on the server recognizes the user's emotions based on features extracted from visual and auditory data. The input is features extracted through image recognition and natural language processing. The output is the recognized user emotion. For example, emotions of "joy" and "fun" can be recognized from a photo of a smiling face and the keyword "fun."

[0426] Step 6: Generate an aroma profile

[0427] server

[0428] The server creates a scent profile using a generative AI model (e.g., GPT-3) based on the recognized emotion. The scent profile includes a combination of fragrances and their proportions. The input is the emotion recognized by the emotion engine. The output is the generated scent profile. For example, a floral scent profile is generated based on the emotion of "joy."

[0429] Step 7: Submit your scent profile

[0430] server

[0431] The server sends the generated scent profile to the device. The profile is sent in JSON format, etc. The input is the generated scent profile. The output is the profile being transmitted to the destination device.

[0432] Step 8: Scent generation

[0433] Terminal

[0434] The device operates the built-in scent generation device (e.g., an Arduino-controlled diffuser) based on the received scent profile. It mixes the required scents from the scent cartridges in the specified ratio to generate the specified scent. The input is the received scent profile. The output is the generated scent.

[0435] Step 9: Diffuse the scent

[0436] Terminal

[0437] The device diffuses the generated scent around the user using a diffuser or the like. The input is the generated scent. The output is the scent diffused in the user's environment.

[0438] Step 10: Gather feedback

[0439] User

[0440] The user experiences the generated scent and provides feedback and suggestions for improvement through a dedicated application. For example, the user may input feedback such as "I wish the scent was a little stronger." The input is the user's feedback. The output is evaluation data sent to the server.

[0441] Step 11: Analyze feedback

[0442] server

[0443] The server receives feedback from users and stores it in a database. The input is the user feedback. The output is the stored rating data.

[0444] Step 12: Optimizing the aroma profile

[0445] server

[0446] The server retrains the generative AI model based on the received feedback and optimizes the scent profile. The new profile is reflected in the next scent generation. The input is the feedback and the existing scent profile. The output is the new optimized scent profile.

[0447] (Application example 2)

[0448] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0449] Conventional scent generation systems have struggled to efficiently analyze visual and auditory data from users and generate personalized scents tailored to individual emotions and situations. Furthermore, the management and blending of fragrances used in manufacturing sites relies on manual labor, resulting in problems of inefficiency and inaccuracy. Large-scale manufacturing facilities, in particular, lack systems that automatically generate scents based on product and environmental information. This creates a demand for efficient, emotion-driven scent experiences.

[0450] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving visual and auditory data related to memories, means for analyzing the received data and extracting features, means for creating a scent profile using a generative AI model, means for transmitting the scent profile, means for generating a scent by operating a scent generation device based on the transmitted scent profile, means for receiving user feedback on the generated scent, means for optimizing the scent profile based on the feedback, means for managing fragrances used in a manufacturing facility and automatically generating a scent based on specified conditions, means for analyzing product and environmental information of the manufacturing facility using image processing and natural language processing techniques, and means for recognizing an emotional state using an emotion engine based on the analyzed information. This enables automation of fragrance management and blending in a manufacturing facility, thereby realizing the provision of personalized scents according to the user's emotional state.

[0451] "Visual data" refers to image information such as photos and videos provided by users.

[0452] "Auditory data" refers to acoustic information such as voice or music provided by the user.

[0453] "Features" refer to the information and attributes necessary for scent generation extracted from visual and auditory data.

[0454] "Generative AI model" refers to an artificial intelligence model for generating a scent profile based on extracted features.

[0455] "Scent profile" refers to data that defines the composition and characteristics of the scent generated by a scent generating device.

[0456] "Scent generating device" refers to a device that physically generates a scent based on a received scent profile.

[0457] "Feedback" refers to the user providing their thoughts and opinions about the generated scent.

[0458] "Manufacturing facility" means an industrial facility or location for producing products.

[0459] "Fragrance control" refers to the process of effectively controlling and maintaining the types and amounts of fragrances used within a manufacturing facility.

[0460] "Image processing" refers to the technology of analyzing visual data to identify key objects and scenes.

[0461] "Natural language processing" refers to the technology of extracting emotions and keywords from text and voice data.

[0462] "Emotion engine" refers to a system for recognizing a user's emotions from visual and auditory data.

[0463] The present invention is a system that provides a user with a personalized scent experience based on visual and auditory data, and will be described in a form that can be particularly applied to factory robots.

[0464] Hardware and Software

[0465] At the heart of the system is a robot powered by NVIDIA Jetson Nano. The robot has a built-in device that manages fragrances and automatically generates scents within the manufacturing facility. The robot runs the following software:

[0466] 1. Emotion recognition using TensorFlow.

[0467] 2. Generate scent profiles using PyTorch.

[0468] 3. Perform image processing using OpenCV.

[0469] Data collection

[0470] The user (operator of the manufacturing facility) provides the robot with images and videos of the product, and inputs their emotions and supplementary information (weather, season, etc.) via text and voice. The robot collects this visual and auditory data and sends it to the server.

[0471] Data analysis

[0472] The server uses OpenCV to identify key objects and patterns in the collected visual data, TensorFlow to extract emotions and keywords from the auditory data, and natural language processing to analyze text and audio data and recognize user emotions with an emotion engine.

[0473] Generation of aroma profiles

[0474] The server then uses a generative AI model trained with PyTorch to create a scent profile based on the extracted features and the recognized emotion. For example, if the emotion of "joy" is recognized, a scent profile rich in floral fragrances will be generated.

[0475] Fragrance production

[0476] The scent profile is sent from the server to the robot, which then uses a scent generator to mix the required fragrances from the fragrance cartridges in the specified ratio to create the desired scent, which is then diffused throughout the manufacturing facility using a diffuser.

[0477] Gathering feedback and optimizing your profile

[0478] The user provides feedback on the generated scent through the operator's application. This feedback is sent to the server and stored as evaluation data. The server then uses the feedback to retrain the generative AI model and optimize the scent profile for future uses.

[0479] Examples and prompts

[0480] Specific examples

[0481] During testing of a new product line in a factory, a robot takes photos and videos of the product, and an operator inputs text such as, "This product is very good and gives off a spring-like feeling." Image processing algorithms identify the new product and its context, and extract positive emotions and seasonal information such as "good" and "spring" from the text data. Based on this, a generative AI model generates a scent profile dominated by floral and fresh spring scents.

[0482] Prompt Sentence Examples

[0483] Example prompt for emotion recognition engine on text data:

[0484] Emotion Recognition Pipeline:

[0485] Input: "This product is very well made and gives a spring-like feeling."

[0486] Output: {'label': 'Positive', 'score': 0.95}

[0487] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0488] Step 1:

[0489] The user provides the robot with product images and videos, and inputs their emotions and supplementary information (such as the weather and season) via text and voice. The input data consists of image files, video files, text data, and voice files. The robot collects this data and sends it to a server. Data collection is performed using devices such as cameras and microphones.

[0490] Step 2:

[0491] The server analyzes the collected visual data using OpenCV. The input is image and video data. The server identifies key objects and patterns from the images and extracts this information as features. Specific operations include image preprocessing, edge detection, and object recognition. The output is a list of key objects and patterns.

[0492] Step 3:

[0493] The server analyzes the auditory data using TensorFlow. The inputs are audio files and text data. The server uses natural language processing techniques to extract emotions and keywords from the audio data. Specific operations include transcribing the audio, sentiment analysis, and keyword extraction. The output is a list of emotion labels and keywords.

[0494] Step 4:

[0495] The server generates a scent profile using a generative AI model trained with PyTorch based on the analyzed features and recognized emotions. The inputs are the features and emotion labels. The server inputs this data into the generative AI model to generate a scent profile that defines the scent composition and characteristics. Specific operations include the inference process of the generative model. The output is a scent profile.

[0496] Step 5:

[0497] The scent profile is sent from the server to the robot. The robot operates the scent generation device based on the received scent profile. The input is the scent profile. The robot mixes the required fragrances from the fragrance cartridges in the specified ratio to generate the specified scent. Specific operations include extracting the fragrances, mixing them, and diffusing them using a diffuser. The output is the generated scent.

[0498] Step 6:

[0499] The user inputs feedback about the generated scent through the operator's application. The input is text data of the feedback. The server receives this feedback and stores it as evaluation data. Specific operations include receiving and storing the feedback data.

[0500] Step 7:

[0501] The server retrains the generative AI model based on the feedback and optimizes the scent profile. The input is the feedback data. The server analyzes the feedback and uses it as learning data for the generative AI model. Specific operations include the model retraining process. The output is an optimized scent profile.

[0502] 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.

[0503] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0504] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0505] [Second embodiment]

[0506] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0507] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0508] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0509] 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.

[0510] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0511] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0512] 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.

[0513] 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.

[0514] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0515] 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.

[0516] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0517] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0518] The present invention provides a system that receives visual and auditory data, analyzes the data, and generates a scent to provide a user with a personalized scent experience. This system is mainly composed of a server, a terminal, and a user.

[0519] 1. Data collection

[0520] User

[0521] Users upload visual and audio data related to their memories to the system. This data includes photos, videos, text, and voice data. Supplementary information, such as weather, season, time, location, and vegetation, can also be added. For example, a user can upload photos taken during a trip and enter text about their impressions.

[0522] 2. Data Analysis

[0523] server

[0524] The server analyzes the visual and auditory data received from the user and extracts the features necessary for scent generation. This analysis uses image recognition algorithms and natural language processing (NLP) technology. For example, it can identify the landscape, season, weather, and location from photo data, and analyze emotions from text data.

[0525] 3. Generation of aroma profiles

[0526] server

[0527] Based on the analyzed features, a generative AI model is used to create a scent profile. This profile includes the combination and ratio of fragrances used within the system. For example, to recreate the scent of a spring day, specific fragrance ratios could be set, such as 50% cherry blossom scent, 30% fresh green scent, and 20% warm air scent.

[0528] server

[0529] The generated scent profile is sent to the device.

[0530] 4. Fragrance production

[0531] Terminal

[0532] The device operates the built-in scent generating device based on the scent profile received from the server. It mixes the required scents from the scent cartridges in the specified ratio to generate the scent. This process is performed in real time, and the generated scent is diffused around the user.

[0533] 5. Gathering Feedback

[0534] User

[0535] Users experience the scents generated and provide feedback on their impressions and suggestions for improvement via a dedicated mobile or web app.

[0536] 6. Profile Optimization

[0537] server

[0538] The server receives feedback from users and stores it as evaluation data for the scent profile. Based on this evaluation data, the generative AI model is retrained and the scent profile is optimized. For example, if a user provides feedback saying, "I would like a more refreshing scent," the server creates a new scent profile and reflects it the next time the user uses the device.

[0539] Specific examples

[0540] Data collection and analysis

[0541] A user uploads photos from a trip to France and an episode titled "Spring in Paris: A walk along the Eiffel Tower and the Seine River" to the system. The server uses image recognition to identify the Eiffel Tower, the Seine River, and spring scenery from the photo, and extracts the keywords "Paris," "spring," and "walk" from the text.

[0542] Scent profile generation and scent generation

[0543] Based on the extracted features, the server creates a scent profile, mainly consisting of "floral spring flower scent" and "fresh riverside scent," and sends it to the device. The device then uses a scent generation device to generate a scent based on the profile and diffuse it around the user.

[0544] Gathering feedback and optimizing your profile

[0545] The user gives feedback such as, "The floral scent was good, but I'd like it to be a little fresher." The server analyzes the feedback and reflects in the generative AI model a suggestion to increase the ratio of fresh fragrances in the next scent profile. The new profile is reflected the next time the user uses the product.

[0546] In this way, the present invention is a system that continuously provides users with personalized moving experiences.

[0547] The processing flow will be explained below.

[0548] Step 1:

[0549] User

[0550] Users upload visual and audio data to the system, providing, for example, photos of their trip and textual anecdotes about it, and can also provide supplemental information such as weather, season, time, location, and vegetation.

[0551] Step 2:

[0552] server

[0553] The server stores the visual and auditory data and supplementary information received from the user, and securely stores the data using a database or cloud storage.

[0554] Step 3:

[0555] server

[0556] The server analyzes the stored data, using image recognition algorithms to identify key objects and scenes from the photos, and natural language processing (NLP) techniques to extract emotions and keywords from the text data, thereby obtaining the features necessary for generating the scent.

[0557] Step 4:

[0558] server

[0559] The server uses a generative AI model based on the extracted features to create a scent profile, which includes the combinations and proportions of fragrances used within the system.

[0560] Step 5:

[0561] server

[0562] The server then sends the created scent profile to the device, securely transferring the data using a communication protocol.

[0563] Step 6:

[0564] Terminal

[0565] The device operates the built-in scent generating device based on the scent profile received from the server, mixing the required scents from the scent cartridges in a specific ratio to create the specified scent.

[0566] Step 7:

[0567] Terminal

[0568] The device then uses a diffuser to diffuse the generated scent around the user, adjusting it to spread throughout the room so that the user can experience the scent.

[0569] Step 8:

[0570] User

[0571] Users can experience the generated scent and provide feedback on their impressions and suggestions for improvement through a dedicated application, such as "I'd like a more floral scent."

[0572] Step 9:

[0573] server

[0574] The server receives feedback from users and stores it as evaluation data, which is used to retrain the generative AI model and optimize the scent profile.

[0575] Step 10:

[0576] server

[0577] The server then sends the new optimized scent profile to the device to be reflected in the next scent generation, thereby continuously improving the user experience.

[0578] Through the above steps, the present invention provides a personalized scent experience to the user.

[0579] Example 1

[0580] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0581] Conventional scent generation systems were unable to utilize visual and auditory data to generate scents, making it difficult to provide a personalized scent experience based on a user's specific memories and emotions. Furthermore, mechanisms for effectively incorporating user feedback on the generated scent and optimizing the scent profile were not adequately developed. Furthermore, supplementary information such as weather, season, time, location, and vegetation could not be reflected, resulting in a lower quality of overall experience.

[0582] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0583] In this invention, the server includes means for receiving visual and auditory data from a user, means for analyzing the received data using an image recognition algorithm and natural language processing technology to extract features necessary for scent generation, means for creating a scent profile using a generative AI model based on the features, means for transmitting the scent profile to a terminal, means for operating a scent generation device on the terminal based on the transmitted scent profile to generate a scent, means for receiving user feedback regarding the generated scent through a dedicated application, means for optimizing the scent profile based on the feedback, and means for receiving supplementary information such as weather, season, time, location, and vegetation from the user. This makes it possible to provide a personalized scent experience based on the user's memories and emotions and to optimize the scent profile to reflect that feedback.

[0584] "Visual data" refers to visual information such as photos and videos provided by users.

[0585] "Auditory data" refers to aural information such as text or voice data provided by a user.

[0586] "Features" refer to the information elements necessary for scent generation, extracted using image recognition algorithms and natural language processing technology.

[0587] "Generative AI model" refers to an artificial intelligence system that uses machine learning algorithms to generate scent profiles.

[0588] "Fragrance profile" means data that indicates the combination and proportions of fragrance ingredients required to produce a particular fragrance.

[0589] "Scent generation device" refers to a mechanical device for generating an actual scent based on a scent profile.

[0590] "Feedback" refers to the user's impressions of the scent they experienced and their opinions on areas for improvement.

[0591] "Supplemental information" refers to user-provided background information for the visual and audio data, such as weather, season, time, location, vegetation, etc.

[0592] "Image recognition algorithm" refers to an algorithm that automatically identifies specific objects, scenes, and situations from photos and videos.

[0593] "Natural language processing technology" refers to algorithms for analyzing meaning and emotions from text and voice data.

[0594] "User" refers to an individual who utilizes the system to provide visual and auditory data and receive a personalized scent experience.

[0595] "Terminal" refers to a device, such as a computer or smartphone, that a user uses to upload visual and auditory data and experience the generated scent.

[0596] MODE FOR CARRYING OUT THE INVENTION

[0597] This invention is a system that receives visual and auditory data, analyzes it, and generates a scent to provide a user with a personalized scent experience. This system is mainly composed of a server, a terminal, and a user. The specific roles and operation procedures of each component are described below.

[0598] 1. Data collection

[0599] User

[0600] Users log in to the system using a device such as a smartphone or PC via a dedicated application or browser. They upload visual and auditory data related to their memories, including photos, videos, text, and voice data. They also enter supplementary information, such as weather, season, time, location, and vegetation, as needed. For example, a user might upload an episode of "Spring in Paris, a walk with the Eiffel Tower and along the Seine River," along with photos.

[0601] 2. Data Analysis

[0602] server

[0603] The server analyzes the visual and auditory data received from the user and extracts the features necessary for scent generation. Visual data (photos and videos) is analyzed using image recognition algorithms (e.g., TensorFlow and OpenCV). For example, the Eiffel Tower, the Seine River, and spring scenery in a photo are identified. Next, auditory data (text and voice data) is analyzed using natural language processing (NLP) techniques (e.g., Google BERT and spaCy) to extract emotions and keywords. For example, keywords such as "Paris," "spring," and "walk" and emotions such as "fun" and "relaxed" are analyzed.

[0604] 3. Generation of aroma profiles

[0605] server

[0606] The server uses a generative AI model (e.g., GPT-4) to create a scent profile based on the analyzed features. The user inputs a prompt to the generative AI model: "Please create a scent profile inspired by spring in Paris, the Eiffel Tower, and a walk along the Seine." As a result, the server calculates specific fragrance ratios and combinations. For example, it generates a profile that is "40% floral spring flower scent, 30% fresh riverside scent, and 30% warm sunshine scent."

[0607] server

[0608] The generated scent profile is sent to the terminal in data format.

[0609] 4. Fragrance production

[0610] Terminal

[0611] The device operates its built-in scent-generating device based on the scent profile received from the server. The control software inside the device extracts and mixes the scent from the scent cartridge in the specified ratio to create the scent. This process is carried out in real time, and the generated scent is diffused around the user, providing them with a personalized scent experience.

[0612] 5. Gathering Feedback

[0613] User

[0614] Users experience the generated scent and provide feedback on their impressions and suggestions for improvement via a dedicated mobile or web app. For example, they can enter specific opinions such as, "I liked the floral scent, but I'd like it to be a little fresher." The feedback is then sent to the server.

[0615] 6. Profile Optimization

[0616] server

[0617] The server receives feedback from users and stores it as scent profile evaluation data. Based on this, it retrains the generative AI model and optimizes the scent profile. For example, based on feedback that a "slightly fresher scent" is desired, a new profile is created that incorporates more fresh scent components. The next time the user uses the service, a new scent profile that reflects these improvements will be provided.

[0618] In this way, the present invention is a system that provides a user with a personalized and inspiring scent experience.

[0619] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0620] Step 1: Collect data

[0621] User

[0622] Users use devices such as smartphones or PCs to log in via a dedicated application or browser on the system and upload visual and auditory data related to their memories.

[0623] input

[0624] Photos, videos, text, voice data, and supplementary information (weather, season, time, location, vegetation) provided by users.

[0625] output

[0626] The visual and audio data and accompanying supplemental information are transmitted to the server side.

[0627] Specific actions

[0628] The user presses the "Upload" button within the application, selects photos, videos, text, or voice data from the file selection screen, and presses the send button.

[0629] Step 2: Analyze the data

[0630] server

[0631] The server analyzes the received visual and auditory data and extracts the features necessary for scent generation.

[0632] input

[0633] Visual data (photos, videos), auditory data (text, voice data), and supplemental information sent by the user.

[0634] output

[0635] Analyzed feature information (e.g., location, season, scenery, emotion, keywords).

[0636] Data processing / data calculation

[0637] The server analyzes the visual data using image recognition algorithms (e.g., TensorFlow and OpenCV) to identify scenery, locations, etc. It also analyzes the auditory data using natural language processing techniques (e.g., Google BERT and spaCy) to extract emotions and keywords.

[0638] Specific actions

[0639] Image recognition algorithms run on the server, identifying the Eiffel Tower, the Seine River, and spring scenery from the photo, while natural language processing technology analyzes the text to extract keywords like "Paris," "spring," and "walk," as well as emotions like "fun" and "relaxing."

[0640] Step 3: Generate an aroma profile

[0641] server

[0642] The server creates a scent profile using a generative AI model based on the analyzed features.

[0643] input

[0644] Analyzed feature information (e.g., location, season, scenery, emotion, keywords).

[0645] output

[0646] Fragrance profile data (specific fragrance combinations and ratios).

[0647] Data processing / data calculation

[0648] The server inputs a prompt into a generative AI model (e.g., GPT-4) to generate a scent profile. The prompt is, "Please create a scent profile inspired by spring in Paris, the Eiffel Tower, and a walk along the Seine."

[0649] Specific actions

[0650] The generative AI model calculates based on the prompt and creates a specific scent profile, such as "40% floral spring flower scent, 30% fresh riverside scent, and 30% warm sunshine scent."

[0651] Step 4: Submit your scent profile

[0652] server

[0653] The server sends the generated scent profile to the terminal.

[0654] input

[0655] Generated scent profile data.

[0656] output

[0657] Scent profile data sent to the device.

[0658] Specific actions

[0659] The scent profile data is transferred from the server to the terminal via the network.

[0660] Step 5: Scent generation

[0661] Terminal

[0662] The terminal operates the built-in scent generating device based on the scent profile received from the server.

[0663] input

[0664] Scent profile data sent from the server.

[0665] output

[0666] The scent produced is diffused into the surrounding area.

[0667] Data processing / data calculation

[0668] The control software in the device extracts and mixes the fragrance from the fragrance cartridge in the specified ratio.

[0669] Specific actions

[0670] The fragrance generating device extracts the required fragrances from the fragrance cartridges and mixes and diffuses them based on the profile.

[0671] Step 6: Gather feedback

[0672] User

[0673] Users experience the generated scent and provide feedback on their impressions and suggestions for improvement.

[0674] input

[0675] User feedback information (impressions, areas for improvement).

[0676] output

[0677] Feedback data sent to the server.

[0678] Specific actions

[0679] Users fill out and submit a feedback form in a dedicated mobile or web app.

[0680] Step 7: Optimize your profile

[0681] server

[0682] The server receives user feedback and stores it as evaluation data for the scent profile. Based on this evaluation data, the generative AI model is retrained and the scent profile is optimized.

[0683] input

[0684] User feedback data.

[0685] output

[0686] Optimized scent profile data.

[0687] Data processing / data calculation

[0688] Based on the feedback, the generative AI model is retrained to generate new scent profiles.

[0689] Specific actions

[0690] The server analyzes the feedback data, updates the generative AI model, and reflects it in the next scent profile.

[0691] (Application example 1)

[0692] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0693] The present invention relates to a technology for generating scents by analyzing visual and auditory data, and aims to realize a more interactive and immediate experience in a system that provides users with a personalized scent experience. Another object of the present invention is to provide customers with a special shopping experience by analyzing customers' visual and auditory data in a store in real time and generating a customized scent on the spot.

[0694] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0695] In this invention, the server includes means for receiving visual and auditory data related to memories from a user, means for analyzing the received data and extracting features necessary for scent generation, means for creating a scent profile using a generative AI model based on the features, means for transmitting the scent profile, means for operating a scent generation device based on the transmitted scent profile to generate a scent, means for receiving user feedback on the generated scent, means for optimizing the scent profile based on the feedback, and means for collecting customer visual and auditory data in a store and generating a personalized scent on the spot, thereby providing users with a personalized and moving experience and improving the customer's purchasing experience.

[0696] "Visual data" refers to visually related data such as images and videos provided by users.

[0697] "Auditory data" refers to data related to the sense of hearing, such as voice or music, provided by the user.

[0698] "Scent generation" refers to the process of physically creating a specific scent based on extracted data.

[0699] "Features" refers to information necessary for scent generation extracted from visual and auditory data.

[0700] "Generative AI model" refers to an artificial intelligence model for creating scent profiles based on data analysis.

[0701] A "fragrance profile" is a setting that defines which specific fragrances are used and in what proportions.

[0702] An "aroma generating device" is a device that generates an actual aroma based on an aroma profile.

[0703] "Feedback" refers to opinions and information such as impressions from users after use and suggestions for improvement.

[0704] "In-store" refers to places where customers visit, such as commercial facilities, sales areas, and exhibition spaces.

[0705] "Personalization" refers to customizing something to suit the preferences and experiences of individual users.

[0706] The present invention relates to a system for providing a personalized scent experience to customers in a physical store. Hereinafter, an embodiment of the system will be described.

[0707] 1. Data Collection

[0708] User:

[0709] Users use smartphones or smart glasses to collect visual and auditory data, including the sights and sounds of the store.

[0710] 2. Data Analysis

[0711] server:

[0712] The server analyzes the visual and auditory data received from the user. It uses OpenCV for camera control to analyze the visual data, and the SpeechRecognition library to analyze the audio data. Scenery, objects, and seasonal sensations are extracted from the visual data, while the sound atmosphere and text information are extracted from the audio data.

[0713] 3. Generation of Aroma Profiles

[0714] server:

[0715] Based on the analyzed features, the server creates a scent profile using a generative AI model that determines which fragrances to mix and in what proportions based on information obtained from the user's visual and auditory data.

[0716] 4. Fragrance production

[0717] Device:

[0718] The terminal operates a scent-generating device based on the scent profile received from the server. This device mixes the required fragrances in the specified ratio and diffuses them around the customer. The AromaBlender device generates the scent in real time.

[0719] 5. Collect feedback

[0720] User:

[0721] Users provide feedback about the scents generated through a dedicated mobile or web app.

[0722] 6. Profile Optimization

[0723] server:

[0724] The server analyzes user feedback and stores the scent profile as evaluation data, which is then used to retrain the generative AI model and optimize the scent profile.

[0725] Specific examples

[0726] A customer wears smart glasses and walks around the store. The scenery of the flower section is captured as visual data, and the music in the store is recorded as auditory data. The server analyzes this data and extracts the characteristics of "floral scents" and "refreshing scents." A scent profile is created based on the generative AI model, and the AromaBlender device generates the scent and provides it to the customer. If the customer gives feedback such as "I liked the floral scent, but I would like it to be a little more refreshing," the scent profile that reflects this feedback will be used the next time.

[0727] Example prompt sentence:

[0728] Smart glasses installed in stores collect the sights and sounds customers see and hear in real time. When a customer is in a specific area of ​​the store, the data is analyzed to generate the optimal scent for that area, and that scent is then delivered through a scent-generating device. For example, if a customer is in the flower section, a scent generated based on "floral scents" and "forest scents" will be distributed.

[0729] In this way, the present invention realizes a system that continuously provides users with personalized and moving experiences, and can also improve customer engagement through special shopping experiences in physical stores.

[0730] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0731] Step 1:

[0732] Data collection

[0733] Users use smartphones or smart glasses to collect visual and auditory data within a store. Specifically, when a user is in a specific area, the camera captures the scenery and the microphone records the surrounding sounds. The input is the scenery seen by the customer (image data) and the sounds heard (audio data), and this data is saved on the device.

[0734] Step 2:

[0735] Sending data

[0736] The device transmits the collected visual and auditory data to a server. Specifically, if the device is connected to the Internet, the collected image data and audio data are uploaded to the server. The input is the data stored on the device, and the output is the data transmitted to the server.

[0737] Step 3:

[0738] Data analysis

[0739] The server analyzes the received visual and auditory data. Specifically, it runs an image recognition algorithm using OpenCV on the visual data to identify scenery and objects. For the auditory data, it uses the SpeechRecognition library to convert the audio data into text and extract specific keywords and emotions. The input is the received visual and auditory data, and the output is feature data as the analysis result.

[0740] Step 4:

[0741] Generation of aroma profiles

[0742] The server uses a generative AI model to create a scent profile based on the analyzed features. Specifically, the AI ​​model calculates the optimal scent combination and its ratio based on the input feature data. The input is feature data, and the output is a scent profile.

[0743] Step 5:

[0744] Sending your scent profile

[0745] The server sends the generated scent profile to the terminal. Specifically, scent profile data is sent from the server to the terminal. The input is the scent profile, and the output is the scent profile sent to the terminal.

[0746] Step 6:

[0747] Fragrance production

[0748] The terminal operates the scent generation device based on the scent profile received from the server. Specifically, it uses the AromaBlender device to mix the fragrances defined in the scent profile in the specified ratio and diffuse the scent around the customer. The input is the scent profile and the output is the generated scent.

[0749] Step 7:

[0750] Collecting feedback

[0751] Users provide feedback on the generated scent, including their impressions and suggestions for improvement. Specifically, they use a dedicated mobile or web app to enter the scent characteristics they perceived and desired changes in text format. The input is the user's impressions and suggestions for improvement, and the output is feedback data.

[0752] Step 8:

[0753] Profile Optimization

[0754] The server analyzes the feedback from users and stores it as evaluation data for the scent profile. Specifically, it retrains the generative AI model based on the feedback content and optimizes the scent profile. The input is the feedback data, and the output is the optimized scent profile.

[0755] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0756] The present invention is a system that analyzes visual and auditory data received from a user and generates a scent to provide the user with a personalized scent experience. By combining this system with an emotion engine, it also has the function of recognizing the user's emotional state and reflecting this in the scent generation. A specific embodiment of the system is described below.

[0757] 1. Data collection

[0758] User

[0759] Users can provide visual and audio data related to their memories, such as photos and videos of their trips, as well as text and audio descriptions of their experiences. They can also input supplementary information such as weather, season, time, location, and vegetation.

[0760] 2. Data Analysis

[0761] server

[0762] The server stores and analyzes the received data, using image recognition algorithms to identify key objects and scenes from photos and videos, and natural language processing (NLP) techniques to extract emotions and keywords from text and audio data.

[0763] 3. Utilizing the Emotion Engine

[0764] server

[0765] The emotion engine installed on the server recognizes the user's emotions based on information extracted from visual and auditory data. For example, if image analysis provides photos or videos showing many smiling faces, the emotion engine will recognize the emotion of "joy."

[0766] 4. Generation of aroma profiles

[0767] server

[0768] Based on the extracted features and the emotions recognized by the emotion engine, a generative AI model is used to create a scent profile. This scent profile includes the combination and proportion of fragrances used within the system. For example, if the emotion of "joy" is recognized, a profile containing a lot of floral fragrances is generated.

[0769] server

[0770] The generated scent profile is sent to the device.

[0771] 5. Aroma production

[0772] Terminal

[0773] The device operates the built-in scent generating device based on the received scent profile, mixing the required scents from the scent cartridges in the specified ratio to create the specified scent.

[0774] Terminal

[0775] The generated scent is diffused around the user using a diffuser, allowing the user to experience a scent that matches the predicted emotion.

[0776] 6. Gather feedback and optimize your profile

[0777] User

[0778] Users can experience the generated scent and provide feedback on their impressions and suggestions for improvement through a dedicated application, such as "I wish the scent was a little stronger."

[0779] server

[0780] The server receives user feedback and stores it as evaluation data. It then retrains the generative AI model based on the feedback and optimizes the scent profile. The new profile is reflected in the next scent generation.

[0781] Specific examples

[0782] Data collection and analysis

[0783] A user uploads a photo taken at the beach in the summer along with the text "A fun summer day" to the system. The server uses image recognition to identify the sea, beach, and summer features in the photo, and extracts the positive emotion "It was fun" from the text.

[0784] Utilizing the Emotion Engine

[0785] The emotion engine installed on the server recognizes the emotion of "fun" from this data.

[0786] Scent profile generation and scent generation

[0787] Based on the extracted features and the recognized emotions, the server creates a scent profile mainly consisting of "fruity scents" and "sea breeze scents" and sends it to the device. The device then uses a scent generation device to generate a scent based on the profile and diffuses it around the user.

[0788] Gathering feedback and optimizing your profile

[0789] The user gives feedback such as "I want a more fruity scent." The server analyzes this feedback, reflects it in the generative AI model, and suggests increasing the ratio of fruity fragrances in the next scent profile, then sends the new profile to the device.

[0790] In this way, the present invention is a system that utilizes an emotion engine to provide a personalized scent experience that is tailored to the user's emotional state.

[0791] The processing flow will be explained below.

[0792] Step 1:

[0793] User

[0794] Users upload visual and audio data related to their memories (e.g., photos and videos of their trips, as well as text and audio descriptions of their experiences) to the system, and can also enter supplemental information such as weather, season, time, location, and vegetation.

[0795] Step 2:

[0796] server

[0797] The server securely stores the visual and auditory data received from the user, as well as any supplementary information, using a database or cloud storage.

[0798] Step 3:

[0799] server

[0800] The server analyzes the stored data. First, it uses image recognition algorithms to identify key objects and scenes (e.g., landscapes, buildings, seasons, and weather) from photos and videos. Next, it uses natural language processing (NLP) techniques to analyze the text and audio data and extract emotions and keywords (e.g., happy, sad, and place names).

[0801] Step 4:

[0802] server

[0803] The server uses an emotion engine to recognize the user's emotions based on the analyzed features and extracted emotions. For example, if the emotion "fun" is extracted from image and text analysis, the emotion engine will recognize this as "joy."

[0804] Step 5:

[0805] server

[0806] A generative AI model is used based on the emotion recognized by the emotion engine and the feature values ​​to create a scent profile. This scent profile includes the appropriate combination of fragrances and their proportions. For example, if the emotion of "joy" is recognized, a profile containing a lot of floral fragrances is generated.

[0807] Step 6:

[0808] server

[0809] The server then sends the generated scent profile to the device, transferring the data using a secure communication protocol.

[0810] Step 7:

[0811] Terminal

[0812] The device operates the built-in scent generating device based on the received scent profile, mixing the required scents from the scent cartridges in the specified ratio to create the specified scent.

[0813] Step 8:

[0814] Terminal

[0815] The generated scent is diffused around the user using a diffuser, allowing the user to experience a scent that matches the recognized emotion.

[0816] Step 9:

[0817] User

[0818] Users can experience the generated scent and provide feedback on their impressions and suggestions for improvement through a dedicated application. For example, they could input their opinion, such as, "I want a more fruity scent."

[0819] Step 10:

[0820] server

[0821] The server receives feedback from users and stores it as evaluation data. Based on this feedback, the generative AI model is retrained and the scent profile is optimized. The new profile is reflected the next time a scent is generated.

[0822] Specific examples

[0823] Data collection and analysis

[0824] Step 1-3: User and Server

[0825] A user uploads a photo they took at the beach in the summer along with the text "It was a fun summer day" to the system. The server uses an image recognition algorithm to identify the sea, beach, and summer scenery from the received photo, and then uses NLP technology to extract the positive emotion of "It was fun" from the text.

[0826] Utilizing the Emotion Engine

[0827] Step 4: Server

[0828] The emotion engine installed on the server recognizes the emotion of "fun" from this data.

[0829] Scent profile generation and scent generation

[0830] Step 5-8: Server and Terminal

[0831] Based on the extracted features and the recognized emotions, the server creates a scent profile mainly consisting of "fruity scents" and "sea breeze scents" and sends it to the device. The device then uses a scent generation device to generate a scent based on the profile and diffuse it around the user.

[0832] Gathering feedback and optimizing your profile

[0833] Steps 9-10: Users and Servers

[0834] The user gives feedback such as "I want a more fruity scent." The server analyzes this feedback, reflects it in the generative AI model, and suggests increasing the ratio of fruity fragrances in the next scent profile, then sends the new profile to the device.

[0835] In this way, the present invention is a system that utilizes an emotion engine to provide a personalized scent experience that is tailored to the user's emotional state.

[0836] Example 2

[0837] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0838] Conventional technologies have struggled to generate personalized scents based on a user's emotional state and memories. Furthermore, the process of efficiently collecting user feedback and optimizing scent profiles has been ineffective. To address these challenges, the present invention aims to provide a system that analyzes a user's visual and auditory data and generates scents based on their emotions.

[0839] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving visual and auditory data from a user, means for saving the received data, means for analyzing the saved data and identifying key objects and scenes using an image recognition algorithm, means for extracting emotions and keywords from text and voice data using natural language processing technology, means for recognizing emotions based on features extracted from the visual and auditory data, means for creating a scent profile using a generative AI model based on the recognized emotions, means for transmitting the created scent profile to a terminal, means for operating a scent generation device based on the received scent profile to generate a scent, means for diffusing the generated scent around the user, means for receiving user feedback regarding the generated scent, and means for optimizing the scent profile based on the received feedback. This makes it possible to generate a scent personalized to the user's emotional state and optimize the scent profile based on the feedback.

[0840] "Visual data" refers to visually related information, such as images and videos provided by users.

[0841] "Auditory data" refers to information related to the sense of hearing, such as voice or music provided by the user.

[0842] "Supplementary information" refers to additional information related to the visual and auditory data, such as weather, season, time, location, vegetation, etc.

[0843] "Database" means a digital information management system for storing visual data, auditory data, and supplemental information.

[0844] An "image recognition algorithm" is a technology that analyzes images and videos and recognizes specific objects and scenes.

[0845] "Natural language processing technology" is a technology for analyzing text and voice data and extracting emotions and keywords.

[0846] The "emotion engine" is a technology for recognizing a user's emotions based on features extracted from visual and auditory data.

[0847] A "generative AI model" is an artificial intelligence model for creating scent profiles based on recognized emotions.

[0848] A "fragrance profile" is information that defines a combination of multiple fragrances and their proportions.

[0849] A "terminal" is a device that generates a scent based on a scent profile and diffuses it around the user.

[0850] An "aroma generating device" is a device that extracts and mixes fragrances from a fragrance cartridge to generate a fragrance.

[0851] A "diffuser" is a device that diffuses the generated scent around the user.

[0852] "Feedback" refers to the user's thoughts and suggestions for improvement regarding the generated fragrance.

[0853] "Optimization" refers to the process of improving the scent profile based on user feedback.

[0854] The present invention is a system that analyzes visual and auditory data provided by a user and generates a scent based on the user's emotions. This system operates in cooperation with a server, a terminal, and the user.

[0855] 1. Data collection

[0856] User

[0857] Through a dedicated application, users provide visual and audio data related to their memories, such as photos, videos, text, and audio, and can also input supplemental information such as weather, season, time, location, and vegetation.

[0858] 2. Data storage

[0859] server

[0860] The server stores the received data in a database such as MongoDB or MySQL. The data is classified as visual data, auditory data, and supplementary information.

[0861] 3. Image Recognition and Natural Language Processing

[0862] server

[0863] The server uses image recognition algorithms (e.g., OpenCV or TensorFlow) to identify key objects and scenes from photos and videos. It also uses natural language processing (NLP) techniques (e.g., BERT) to extract emotions and keywords from text and audio data. It converts audio data into text using the Google Speech-to-Text API for analysis.

[0864] 4. Utilizing the Emotion Engine

[0865] server

[0866] The emotion engine installed on the server recognizes the user's emotions based on features extracted from visual and auditory data. For example, if there are many photos of smiling faces or positive keywords such as "it was fun," it will recognize the emotions of "joy" and "enjoyment."

[0867] 5. Generation of aroma profiles

[0868] server

[0869] The server uses a generative AI model (e.g., GPT-3) to create a scent profile based on the recognized emotion. This scent profile includes a combination of fragrances and their proportions. For example, if the emotion of "joy" is recognized, a profile containing a lot of floral fragrances is generated.

[0870] 6. Submit your scent profile

[0871] server

[0872] The server sends the created scent profile to the device in JSON format.

[0873] 7. Scent production and diffusion

[0874] Terminal

[0875] The device operates a built-in scent generating device (e.g., an Arduino-controlled diffuser) based on the received scent profile. It mixes the required fragrances from the fragrance cartridges in the specified ratio to generate the specified scent. It also diffuses the generated scent around the user using the diffuser.

[0876] 8. Gather feedback and optimize

[0877] User

[0878] Users can experience the generated scent and provide feedback on their impressions and suggestions for improvement through a dedicated application. For example, they could say, "I'd like a more fruity scent."

[0879] server

[0880] The server analyzes the received feedback and retrains the generative AI model to optimize the scent profile. The new profile is reflected in the next scent generation.

[0881] Specific examples

[0882] Data collection and analysis

[0883] A user uploads a photo taken at the beach in the summer along with the text "A fun summer day" to the system. The server uses an image recognition algorithm to identify the characteristics of the sea, beach, and summer from the photo, and then uses natural language processing technology to extract the positive emotion of "It was fun" from the text.

[0884] Utilizing the Emotion Engine

[0885] The emotion engine installed on the server recognizes the emotion of "fun" from this data.

[0886] Scent profile generation and scent generation

[0887] Based on the extracted features and the recognized emotions, the server uses a generative AI model to create a scent profile mainly consisting of "fruity scents" and "sea breeze scents" and sends it to the device. The device then uses a scent generation device to generate a scent based on the profile and diffuses it around the user.

[0888] Gathering feedback and optimizing your profile

[0889] The user gives feedback such as "I want a more fruity scent." The server analyzes this feedback, reflects it in the generative AI model, and suggests increasing the ratio of fruity fragrances in the next scent profile, then sends the new profile to the device.

[0890] Prompt Sentence Examples

[0891] "Analyze a user-provided summer beach photo and the text 'A fun summer day' and generate a corresponding scent profile. If the user provides feedback that they would like a more fruity scent, explain how you would optimize the scent profile."

[0892] As a result, the present invention provides a system that provides a personalized scent experience tailored to the user's emotional state and optimizes the scent profile based on user feedback.

[0893] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0894] Step 1: Collect data

[0895] User

[0896] Using a dedicated application, users upload photos and videos of their trip, text and audio impressions, and supplementary information such as weather, season, time, location, and vegetation. The inputs are visual data (photos and videos), auditory data (audio), and supplementary information. The output is this data sent to the server.

[0897] Step 2: Save your data

[0898] server

[0899] The server stores the received visual data, auditory data, and supplementary information in a database. The input is the data sent by the user. The output is the data stored in the database. Database systems such as MongoDB and MySQL are used at this stage.

[0900] Step 3: Perform image recognition

[0901] server

[0902] The server uses image recognition algorithms (e.g., OpenCV or TensorFlow) to analyze the stored photo and video data. It identifies key objects and scenes and stores them as metadata. The input is visual data stored in a database. The output is metadata for the identified objects and scenes.

[0903] Step 4: Performing Natural Language Processing

[0904] server

[0905] The server uses natural language processing (NLP) techniques (e.g., BERT) to extract sentiment and keywords from text and audio data. The audio data is converted to text using the Google Speech-to-Text API and then analyzed. The input is the text and audio data stored in the database. The output is the extracted sentiment and keywords.

[0906] Step 5: Leverage the Emotion Engine

[0907] server

[0908] The emotion engine installed on the server recognizes the user's emotions based on features extracted from visual and auditory data. The input is features extracted through image recognition and natural language processing. The output is the recognized user emotion. For example, emotions of "joy" and "fun" can be recognized from a photo of a smiling face and the keyword "fun."

[0909] Step 6: Generate an aroma profile

[0910] server

[0911] The server creates a scent profile using a generative AI model (e.g., GPT-3) based on the recognized emotion. The scent profile includes a combination of fragrances and their proportions. The input is the emotion recognized by the emotion engine. The output is the generated scent profile. For example, a floral scent profile is generated based on the emotion of "joy."

[0912] Step 7: Submit your scent profile

[0913] server

[0914] The server sends the generated scent profile to the device. The profile is sent in JSON format, etc. The input is the generated scent profile. The output is the profile being transmitted to the destination device.

[0915] Step 8: Scent generation

[0916] Terminal

[0917] The device operates the built-in scent generation device (e.g., an Arduino-controlled diffuser) based on the received scent profile. It mixes the required scents from the scent cartridges in the specified ratio to generate the specified scent. The input is the received scent profile. The output is the generated scent.

[0918] Step 9: Diffuse the scent

[0919] Terminal

[0920] The device diffuses the generated scent around the user using a diffuser or the like. The input is the generated scent. The output is the scent diffused in the user's environment.

[0921] Step 10: Gather feedback

[0922] User

[0923] The user experiences the generated scent and provides feedback and suggestions for improvement through a dedicated application. For example, the user may input feedback such as "I wish the scent was a little stronger." The input is the user's feedback. The output is evaluation data sent to the server.

[0924] Step 11: Analyze feedback

[0925] server

[0926] The server receives feedback from users and stores it in a database. The input is the user feedback. The output is the stored rating data.

[0927] Step 12: Optimizing the aroma profile

[0928] server

[0929] The server retrains the generative AI model based on the received feedback and optimizes the scent profile. The new profile is reflected in the next scent generation. The input is the feedback and the existing scent profile. The output is the new optimized scent profile.

[0930] (Application example 2)

[0931] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0932] Conventional scent generation systems have struggled to efficiently analyze visual and auditory data from users and generate personalized scents tailored to individual emotions and situations. Furthermore, the management and blending of fragrances used in manufacturing sites relies on manual labor, resulting in problems of inefficiency and inaccuracy. Large-scale manufacturing facilities, in particular, lack systems that automatically generate scents based on product and environmental information. This creates a demand for efficient, emotion-driven scent experiences.

[0933] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving visual and auditory data related to memories, means for analyzing the received data and extracting features, means for creating a scent profile using a generative AI model, means for transmitting the scent profile, means for generating a scent by operating a scent generation device based on the transmitted scent profile, means for receiving user feedback on the generated scent, means for optimizing the scent profile based on the feedback, means for managing fragrances used in a manufacturing facility and automatically generating a scent based on specified conditions, means for analyzing product and environmental information of the manufacturing facility using image processing and natural language processing techniques, and means for recognizing an emotional state using an emotion engine based on the analyzed information. This enables automation of fragrance management and blending in a manufacturing facility, thereby realizing the provision of personalized scents according to the user's emotional state.

[0934] "Visual data" refers to image information such as photos and videos provided by users.

[0935] "Auditory data" refers to acoustic information such as voice or music provided by the user.

[0936] "Features" refer to the information and attributes necessary for scent generation extracted from visual and auditory data.

[0937] "Generative AI model" refers to an artificial intelligence model for generating a scent profile based on extracted features.

[0938] "Scent profile" refers to data that defines the composition and characteristics of the scent generated by a scent generating device.

[0939] "Scent generating device" refers to a device that physically generates a scent based on a received scent profile.

[0940] "Feedback" refers to the user providing their thoughts and opinions about the generated scent.

[0941] "Manufacturing facility" means an industrial facility or location for producing products.

[0942] "Fragrance control" refers to the process of effectively controlling and maintaining the types and amounts of fragrances used within a manufacturing facility.

[0943] "Image processing" refers to the technology of analyzing visual data to identify key objects and scenes.

[0944] "Natural language processing" refers to the technology of extracting emotions and keywords from text and voice data.

[0945] "Emotion engine" refers to a system for recognizing a user's emotions from visual and auditory data.

[0946] The present invention is a system that provides a user with a personalized scent experience based on visual and auditory data, and will be described in a form that can be particularly applied to factory robots.

[0947] Hardware and Software

[0948] At the heart of the system is a robot powered by NVIDIA Jetson Nano. The robot has a built-in device that manages fragrances and automatically generates scents within the manufacturing facility. The robot runs the following software:

[0949] 1. Emotion recognition using TensorFlow.

[0950] 2. Generate scent profiles using PyTorch.

[0951] 3. Perform image processing using OpenCV.

[0952] Data collection

[0953] The user (operator of the manufacturing facility) provides the robot with images and videos of the product, and inputs their emotions and supplementary information (weather, season, etc.) via text and voice. The robot collects this visual and auditory data and sends it to the server.

[0954] Data analysis

[0955] The server uses OpenCV to identify key objects and patterns in the collected visual data, TensorFlow to extract emotions and keywords from the auditory data, and natural language processing to analyze text and audio data and recognize user emotions with an emotion engine.

[0956] Generation of aroma profiles

[0957] The server then uses a generative AI model trained with PyTorch to create a scent profile based on the extracted features and the recognized emotion. For example, if the emotion of "joy" is recognized, a scent profile rich in floral fragrances will be generated.

[0958] Fragrance production

[0959] The scent profile is sent from the server to the robot, which then uses a scent generator to mix the required fragrances from the fragrance cartridges in the specified ratio to create the desired scent, which is then diffused throughout the manufacturing facility using a diffuser.

[0960] Gathering feedback and optimizing your profile

[0961] The user provides feedback on the generated scent through the operator's application. This feedback is sent to the server and stored as evaluation data. The server then uses the feedback to retrain the generative AI model and optimize the scent profile for future uses.

[0962] Examples and prompts

[0963] Specific examples

[0964] During testing of a new product line in a factory, a robot takes photos and videos of the product, and an operator inputs text such as, "This product is very good and gives off a spring-like feeling." Image processing algorithms identify the new product and its context, and extract positive emotions and seasonal information such as "good" and "spring" from the text data. Based on this, a generative AI model generates a scent profile dominated by floral and fresh spring scents.

[0965] Prompt Sentence Examples

[0966] Example prompt for emotion recognition engine on text data:

[0967] Emotion Recognition Pipeline:

[0968] Input: "This product is very well made and gives a spring-like feeling."

[0969] Output: {'label': 'Positive', 'score': 0.95}

[0970] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0971] Step 1:

[0972] The user provides the robot with product images and videos, and inputs their emotions and supplementary information (such as the weather and season) via text and voice. The input data consists of image files, video files, text data, and voice files. The robot collects this data and sends it to a server. Data collection is performed using devices such as cameras and microphones.

[0973] Step 2:

[0974] The server analyzes the collected visual data using OpenCV. The input is image and video data. The server identifies key objects and patterns from the images and extracts this information as features. Specific operations include image preprocessing, edge detection, and object recognition. The output is a list of key objects and patterns.

[0975] Step 3:

[0976] The server analyzes the auditory data using TensorFlow. The inputs are audio files and text data. The server uses natural language processing techniques to extract emotions and keywords from the audio data. Specific operations include transcribing the audio, sentiment analysis, and keyword extraction. The output is a list of emotion labels and keywords.

[0977] Step 4:

[0978] The server generates a scent profile using a generative AI model trained with PyTorch based on the analyzed features and recognized emotions. The inputs are the features and emotion labels. The server inputs this data into the generative AI model to generate a scent profile that defines the scent composition and characteristics. Specific operations include the inference process of the generative model. The output is a scent profile.

[0979] Step 5:

[0980] The scent profile is sent from the server to the robot. The robot operates the scent generation device based on the received scent profile. The input is the scent profile. The robot mixes the required fragrances from the fragrance cartridges in the specified ratio to generate the specified scent. Specific operations include extracting the fragrances, mixing them, and diffusing them using a diffuser. The output is the generated scent.

[0981] Step 6:

[0982] The user inputs feedback about the generated scent through the operator's application. The input is text data of the feedback. The server receives this feedback and stores it as evaluation data. Specific operations include receiving and storing the feedback data.

[0983] Step 7:

[0984] The server retrains the generative AI model based on the feedback and optimizes the scent profile. The input is the feedback data. The server analyzes the feedback and uses it as learning data for the generative AI model. Specific operations include the model retraining process. The output is an optimized scent profile.

[0985] 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.

[0986] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0987] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0988] [Third embodiment]

[0989] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0990] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0991] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0992] 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.

[0993] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0994] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0995] 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.

[0996] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0997] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0998] 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.

[0999] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1000] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[1001] The present invention provides a system that receives visual and auditory data, analyzes the data, and generates a scent to provide a user with a personalized scent experience. This system is mainly composed of a server, a terminal, and a user.

[1002] 1. Data collection

[1003] User

[1004] Users upload visual and audio data related to their memories to the system. This data includes photos, videos, text, and voice data. Supplementary information, such as weather, season, time, location, and vegetation, can also be added. For example, a user can upload photos taken during a trip and enter text about their impressions.

[1005] 2. Data Analysis

[1006] server

[1007] The server analyzes the visual and auditory data received from the user and extracts the features necessary for scent generation. This analysis uses image recognition algorithms and natural language processing (NLP) technology. For example, it can identify the landscape, season, weather, and location from photo data, and analyze emotions from text data.

[1008] 3. Generation of aroma profiles

[1009] server

[1010] Based on the analyzed features, a generative AI model is used to create a scent profile. This profile includes the combination and ratio of fragrances used within the system. For example, to recreate the scent of a spring day, specific fragrance ratios could be set, such as 50% cherry blossom scent, 30% fresh green scent, and 20% warm air scent.

[1011] server

[1012] The generated scent profile is sent to the device.

[1013] 4. Fragrance production

[1014] Terminal

[1015] The device operates the built-in scent generating device based on the scent profile received from the server. It mixes the required scents from the scent cartridges in the specified ratio to generate the scent. This process is performed in real time, and the generated scent is diffused around the user.

[1016] 5. Gathering Feedback

[1017] User

[1018] Users experience the scents generated and provide feedback on their impressions and suggestions for improvement via a dedicated mobile or web app.

[1019] 6. Profile Optimization

[1020] server

[1021] The server receives feedback from users and stores it as evaluation data for the scent profile. Based on this evaluation data, the generative AI model is retrained and the scent profile is optimized. For example, if a user provides feedback saying, "I would like a more refreshing scent," the server creates a new scent profile and reflects it the next time the user uses the device.

[1022] Specific examples

[1023] Data collection and analysis

[1024] A user uploads photos from a trip to France and an episode titled "Spring in Paris: A walk along the Eiffel Tower and the Seine River" to the system. The server uses image recognition to identify the Eiffel Tower, the Seine River, and spring scenery from the photo, and extracts the keywords "Paris," "spring," and "walk" from the text.

[1025] Scent profile generation and scent generation

[1026] Based on the extracted features, the server creates a scent profile, mainly consisting of "floral spring flower scent" and "fresh riverside scent," and sends it to the device. The device then uses a scent generation device to generate a scent based on the profile and diffuse it around the user.

[1027] Gathering feedback and optimizing your profile

[1028] The user gives feedback such as, "The floral scent was good, but I'd like it to be a little fresher." The server analyzes the feedback and reflects in the generative AI model a suggestion to increase the ratio of fresh fragrances in the next scent profile. The new profile is reflected the next time the user uses the product.

[1029] In this way, the present invention is a system that continuously provides users with personalized moving experiences.

[1030] The processing flow will be explained below.

[1031] Step 1:

[1032] User

[1033] Users upload visual and audio data to the system, providing, for example, photos of their trip and textual anecdotes about it, and can also provide supplemental information such as weather, season, time, location, and vegetation.

[1034] Step 2:

[1035] server

[1036] The server stores the visual and auditory data and supplementary information received from the user, and securely stores the data using a database or cloud storage.

[1037] Step 3:

[1038] server

[1039] The server analyzes the stored data, using image recognition algorithms to identify key objects and scenes from the photos, and natural language processing (NLP) techniques to extract emotions and keywords from the text data, thereby obtaining the features necessary for generating the scent.

[1040] Step 4:

[1041] server

[1042] The server uses a generative AI model based on the extracted features to create a scent profile, which includes the combinations and proportions of fragrances used within the system.

[1043] Step 5:

[1044] server

[1045] The server then sends the created scent profile to the device, securely transferring the data using a communication protocol.

[1046] Step 6:

[1047] Terminal

[1048] The device operates the built-in scent generating device based on the scent profile received from the server, mixing the required scents from the scent cartridges in a specific ratio to create the specified scent.

[1049] Step 7:

[1050] Terminal

[1051] The device then uses a diffuser to diffuse the generated scent around the user, adjusting it to spread throughout the room so that the user can experience the scent.

[1052] Step 8:

[1053] User

[1054] Users can experience the generated scent and provide feedback on their impressions and suggestions for improvement through a dedicated application, such as "I'd like a more floral scent."

[1055] Step 9:

[1056] server

[1057] The server receives feedback from users and stores it as evaluation data, which is used to retrain the generative AI model and optimize the scent profile.

[1058] Step 10:

[1059] server

[1060] The server then sends the new optimized scent profile to the device to be reflected in the next scent generation, thereby continuously improving the user experience.

[1061] Through the above steps, the present invention provides a personalized scent experience to the user.

[1062] Example 1

[1063] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1064] Conventional scent generation systems were unable to utilize visual and auditory data to generate scents, making it difficult to provide a personalized scent experience based on a user's specific memories and emotions. Furthermore, mechanisms for effectively incorporating user feedback on the generated scent and optimizing the scent profile were not adequately developed. Furthermore, supplementary information such as weather, season, time, location, and vegetation could not be reflected, resulting in a lower quality of overall experience.

[1065] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1066] In this invention, the server includes means for receiving visual and auditory data from a user, means for analyzing the received data using an image recognition algorithm and natural language processing technology to extract features necessary for scent generation, means for creating a scent profile using a generative AI model based on the features, means for transmitting the scent profile to a terminal, means for operating a scent generation device on the terminal based on the transmitted scent profile to generate a scent, means for receiving user feedback regarding the generated scent through a dedicated application, means for optimizing the scent profile based on the feedback, and means for receiving supplementary information such as weather, season, time, location, and vegetation from the user. This makes it possible to provide a personalized scent experience based on the user's memories and emotions and to optimize the scent profile to reflect that feedback.

[1067] "Visual data" refers to visual information such as photos and videos provided by users.

[1068] "Auditory data" refers to aural information such as text or voice data provided by a user.

[1069] "Features" refer to the information elements necessary for scent generation, extracted using image recognition algorithms and natural language processing technology.

[1070] "Generative AI model" refers to an artificial intelligence system that uses machine learning algorithms to generate scent profiles.

[1071] "Fragrance profile" means data that indicates the combination and proportions of fragrance ingredients required to produce a particular fragrance.

[1072] "Scent generation device" refers to a mechanical device for generating an actual scent based on a scent profile.

[1073] "Feedback" refers to the user's impressions of the scent they experienced and their opinions on areas for improvement.

[1074] "Supplemental information" refers to user-provided background information for the visual and audio data, such as weather, season, time, location, vegetation, etc.

[1075] "Image recognition algorithm" refers to an algorithm that automatically identifies specific objects, scenes, and situations from photos and videos.

[1076] "Natural language processing technology" refers to algorithms for analyzing meaning and emotions from text and voice data.

[1077] "User" refers to an individual who utilizes the system to provide visual and auditory data and receive a personalized scent experience.

[1078] "Terminal" refers to a device, such as a computer or smartphone, that a user uses to upload visual and auditory data and experience the generated scent.

[1079] MODE FOR CARRYING OUT THE INVENTION

[1080] This invention is a system that receives visual and auditory data, analyzes it, and generates a scent to provide a user with a personalized scent experience. This system is mainly composed of a server, a terminal, and a user. The specific roles and operation procedures of each component are described below.

[1081] 1. Data collection

[1082] User

[1083] Users log in to the system using a device such as a smartphone or PC via a dedicated application or browser. They upload visual and auditory data related to their memories, including photos, videos, text, and voice data. They also enter supplementary information, such as weather, season, time, location, and vegetation, as needed. For example, a user might upload an episode of "Spring in Paris, a walk with the Eiffel Tower and along the Seine River," along with photos.

[1084] 2. Data Analysis

[1085] server

[1086] The server analyzes the visual and auditory data received from the user and extracts the features necessary for scent generation. Visual data (photos and videos) is analyzed using image recognition algorithms (e.g., TensorFlow and OpenCV). For example, the Eiffel Tower, the Seine River, and spring scenery in a photo are identified. Next, auditory data (text and voice data) is analyzed using natural language processing (NLP) techniques (e.g., Google BERT and spaCy) to extract emotions and keywords. For example, keywords such as "Paris," "spring," and "walk" and emotions such as "fun" and "relaxed" are analyzed.

[1087] 3. Generation of aroma profiles

[1088] server

[1089] The server uses a generative AI model (e.g., GPT-4) to create a scent profile based on the analyzed features. The user inputs a prompt to the generative AI model: "Please create a scent profile inspired by spring in Paris, the Eiffel Tower, and a walk along the Seine." As a result, the server calculates specific fragrance ratios and combinations. For example, it generates a profile that is "40% floral spring flower scent, 30% fresh riverside scent, and 30% warm sunshine scent."

[1090] server

[1091] The generated scent profile is sent to the terminal in data format.

[1092] 4. Fragrance production

[1093] Terminal

[1094] The device operates its built-in scent-generating device based on the scent profile received from the server. The control software inside the device extracts and mixes the scent from the scent cartridge in the specified ratio to create the scent. This process is carried out in real time, and the generated scent is diffused around the user, providing them with a personalized scent experience.

[1095] 5. Gathering Feedback

[1096] User

[1097] Users experience the generated scent and provide feedback on their impressions and suggestions for improvement via a dedicated mobile or web app. For example, they can enter specific opinions such as, "I liked the floral scent, but I'd like it to be a little fresher." The feedback is then sent to the server.

[1098] 6. Profile Optimization

[1099] server

[1100] The server receives feedback from users and stores it as scent profile evaluation data. Based on this, it retrains the generative AI model and optimizes the scent profile. For example, based on feedback that a "slightly fresher scent" is desired, a new profile is created that incorporates more fresh scent components. The next time the user uses the service, a new scent profile that reflects these improvements will be provided.

[1101] In this way, the present invention is a system that provides a user with a personalized and inspiring scent experience.

[1102] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1103] Step 1: Collect data

[1104] User

[1105] Users use devices such as smartphones or PCs to log in via a dedicated application or browser on the system and upload visual and auditory data related to their memories.

[1106] input

[1107] Photos, videos, text, voice data, and supplementary information (weather, season, time, location, vegetation) provided by users.

[1108] output

[1109] The visual and audio data and accompanying supplemental information are transmitted to the server side.

[1110] Specific actions

[1111] The user presses the "Upload" button within the application, selects photos, videos, text, or voice data from the file selection screen, and presses the send button.

[1112] Step 2: Analyze the data

[1113] server

[1114] The server analyzes the received visual and auditory data and extracts the features necessary for scent generation.

[1115] input

[1116] Visual data (photos, videos), auditory data (text, voice data), and supplemental information sent by the user.

[1117] output

[1118] Analyzed feature information (e.g., location, season, scenery, emotion, keywords).

[1119] Data processing / data calculation

[1120] The server analyzes the visual data using image recognition algorithms (e.g., TensorFlow and OpenCV) to identify scenery, locations, etc. It also analyzes the auditory data using natural language processing techniques (e.g., Google BERT and spaCy) to extract emotions and keywords.

[1121] Specific actions

[1122] Image recognition algorithms run on the server, identifying the Eiffel Tower, the Seine River, and spring scenery from the photo, while natural language processing technology analyzes the text to extract keywords like "Paris," "spring," and "walk," as well as emotions like "fun" and "relaxing."

[1123] Step 3: Generate an aroma profile

[1124] server

[1125] The server creates a scent profile using a generative AI model based on the analyzed features.

[1126] input

[1127] Analyzed feature information (e.g., location, season, scenery, emotion, keywords).

[1128] output

[1129] Fragrance profile data (specific fragrance combinations and ratios).

[1130] Data processing / data calculation

[1131] The server inputs a prompt into a generative AI model (e.g., GPT-4) to generate a scent profile. The prompt is, "Please create a scent profile inspired by spring in Paris, the Eiffel Tower, and a walk along the Seine."

[1132] Specific actions

[1133] The generative AI model calculates based on the prompt and creates a specific scent profile, such as "40% floral spring flower scent, 30% fresh riverside scent, and 30% warm sunshine scent."

[1134] Step 4: Submit your scent profile

[1135] server

[1136] The server sends the generated scent profile to the terminal.

[1137] input

[1138] Generated scent profile data.

[1139] output

[1140] Scent profile data sent to the device.

[1141] Specific actions

[1142] The scent profile data is transferred from the server to the terminal via the network.

[1143] Step 5: Scent generation

[1144] Terminal

[1145] The terminal operates the built-in scent generating device based on the scent profile received from the server.

[1146] input

[1147] Scent profile data sent from the server.

[1148] output

[1149] The scent produced is diffused into the surrounding area.

[1150] Data processing / data calculation

[1151] The control software in the device extracts and mixes the fragrance from the fragrance cartridge in the specified ratio.

[1152] Specific actions

[1153] The fragrance generating device extracts the required fragrances from the fragrance cartridges and mixes and diffuses them based on the profile.

[1154] Step 6: Gather feedback

[1155] User

[1156] Users experience the generated scent and provide feedback on their impressions and suggestions for improvement.

[1157] input

[1158] User feedback information (impressions, areas for improvement).

[1159] output

[1160] Feedback data sent to the server.

[1161] Specific actions

[1162] Users fill out and submit a feedback form in a dedicated mobile or web app.

[1163] Step 7: Optimize your profile

[1164] server

[1165] The server receives user feedback and stores it as evaluation data for the scent profile. Based on this evaluation data, the generative AI model is retrained and the scent profile is optimized.

[1166] input

[1167] User feedback data.

[1168] output

[1169] Optimized scent profile data.

[1170] Data processing / data calculation

[1171] Based on the feedback, the generative AI model is retrained to generate new scent profiles.

[1172] Specific actions

[1173] The server analyzes the feedback data, updates the generative AI model, and reflects it in the next scent profile.

[1174] (Application example 1)

[1175] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1176] The present invention relates to a technology for generating scents by analyzing visual and auditory data, and aims to realize a more interactive and immediate experience in a system that provides users with a personalized scent experience. Another object of the present invention is to provide customers with a special shopping experience by analyzing customers' visual and auditory data in a store in real time and generating a customized scent on the spot.

[1177] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1178] In this invention, the server includes means for receiving visual and auditory data related to memories from a user, means for analyzing the received data and extracting features necessary for scent generation, means for creating a scent profile using a generative AI model based on the features, means for transmitting the scent profile, means for operating a scent generation device based on the transmitted scent profile to generate a scent, means for receiving user feedback on the generated scent, means for optimizing the scent profile based on the feedback, and means for collecting customer visual and auditory data in a store and generating a personalized scent on the spot, thereby providing users with a personalized and moving experience and improving the customer's purchasing experience.

[1179] "Visual data" refers to visually related data such as images and videos provided by users.

[1180] "Auditory data" refers to data related to the sense of hearing, such as voice or music, provided by the user.

[1181] "Scent generation" refers to the process of physically creating a specific scent based on extracted data.

[1182] "Features" refers to information necessary for scent generation extracted from visual and auditory data.

[1183] "Generative AI model" refers to an artificial intelligence model for creating scent profiles based on data analysis.

[1184] A "fragrance profile" is a setting that defines which specific fragrances are used and in what proportions.

[1185] An "aroma generating device" is a device that generates an actual aroma based on an aroma profile.

[1186] "Feedback" refers to opinions and information such as impressions from users after use and suggestions for improvement.

[1187] "In-store" refers to places where customers visit, such as commercial facilities, sales areas, and exhibition spaces.

[1188] "Personalization" refers to customizing something to suit the preferences and experiences of individual users.

[1189] The present invention relates to a system for providing a personalized scent experience to customers in a physical store. Hereinafter, an embodiment of the system will be described.

[1190] 1. Data Collection

[1191] User:

[1192] Users use smartphones or smart glasses to collect visual and auditory data, including the sights and sounds of the store.

[1193] 2. Data Analysis

[1194] server:

[1195] The server analyzes the visual and auditory data received from the user. It uses OpenCV for camera control to analyze the visual data, and the SpeechRecognition library to analyze the audio data. Scenery, objects, and seasonal sensations are extracted from the visual data, while the sound atmosphere and text information are extracted from the audio data.

[1196] 3. Generation of Aroma Profiles

[1197] server:

[1198] Based on the analyzed features, the server creates a scent profile using a generative AI model that determines which fragrances to mix and in what proportions based on information obtained from the user's visual and auditory data.

[1199] 4. Fragrance production

[1200] Device:

[1201] The terminal operates a scent-generating device based on the scent profile received from the server. This device mixes the required fragrances in the specified ratio and diffuses them around the customer. The AromaBlender device generates the scent in real time.

[1202] 5. Collect feedback

[1203] User:

[1204] Users provide feedback about the scents generated through a dedicated mobile or web app.

[1205] 6. Profile Optimization

[1206] server:

[1207] The server analyzes user feedback and stores the scent profile as evaluation data, which is then used to retrain the generative AI model and optimize the scent profile.

[1208] Specific examples

[1209] A customer wears smart glasses and walks around the store. The scenery of the flower section is captured as visual data, and the music in the store is recorded as auditory data. The server analyzes this data and extracts the characteristics of "floral scents" and "refreshing scents." A scent profile is created based on the generative AI model, and the AromaBlender device generates the scent and provides it to the customer. If the customer gives feedback such as "I liked the floral scent, but I would like it to be a little more refreshing," the scent profile that reflects this feedback will be used the next time.

[1210] Example prompt sentence:

[1211] Smart glasses installed in stores collect the sights and sounds customers see and hear in real time. When a customer is in a specific area of ​​the store, the data is analyzed to generate the optimal scent for that area, and that scent is then delivered through a scent-generating device. For example, if a customer is in the flower section, a scent generated based on "floral scents" and "forest scents" will be distributed.

[1212] In this way, the present invention realizes a system that continuously provides users with personalized and moving experiences, and can also improve customer engagement through special shopping experiences in physical stores.

[1213] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1214] Step 1:

[1215] Data collection

[1216] Users use smartphones or smart glasses to collect visual and auditory data within a store. Specifically, when a user is in a specific area, the camera captures the scenery and the microphone records the surrounding sounds. The input is the scenery seen by the customer (image data) and the sounds heard (audio data), and this data is saved on the device.

[1217] Step 2:

[1218] Sending data

[1219] The device transmits the collected visual and auditory data to a server. Specifically, if the device is connected to the Internet, the collected image data and audio data are uploaded to the server. The input is the data stored on the device, and the output is the data transmitted to the server.

[1220] Step 3:

[1221] Data analysis

[1222] The server analyzes the received visual and auditory data. Specifically, it runs an image recognition algorithm using OpenCV on the visual data to identify scenery and objects. For the auditory data, it uses the SpeechRecognition library to convert the audio data into text and extract specific keywords and emotions. The input is the received visual and auditory data, and the output is feature data as the analysis result.

[1223] Step 4:

[1224] Generation of aroma profiles

[1225] The server uses a generative AI model to create a scent profile based on the analyzed features. Specifically, the AI ​​model calculates the optimal scent combination and its ratio based on the input feature data. The input is feature data, and the output is a scent profile.

[1226] Step 5:

[1227] Sending your scent profile

[1228] The server sends the generated scent profile to the terminal. Specifically, scent profile data is sent from the server to the terminal. The input is the scent profile, and the output is the scent profile sent to the terminal.

[1229] Step 6:

[1230] Fragrance production

[1231] The terminal operates the scent generation device based on the scent profile received from the server. Specifically, it uses the AromaBlender device to mix the fragrances defined in the scent profile in the specified ratio and diffuse the scent around the customer. The input is the scent profile and the output is the generated scent.

[1232] Step 7:

[1233] Collecting feedback

[1234] Users provide feedback on the generated scent, including their impressions and suggestions for improvement. Specifically, they use a dedicated mobile or web app to enter the scent characteristics they perceived and desired changes in text format. The input is the user's impressions and suggestions for improvement, and the output is feedback data.

[1235] Step 8:

[1236] Profile Optimization

[1237] The server analyzes the feedback from users and stores it as evaluation data for the scent profile. Specifically, it retrains the generative AI model based on the feedback content and optimizes the scent profile. The input is the feedback data, and the output is the optimized scent profile.

[1238] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1239] The present invention is a system that analyzes visual and auditory data received from a user and generates a scent to provide the user with a personalized scent experience. By combining this system with an emotion engine, it also has the function of recognizing the user's emotional state and reflecting this in the scent generation. A specific embodiment of the system is described below.

[1240] 1. Data collection

[1241] User

[1242] Users can provide visual and audio data related to their memories, such as photos and videos of their trips, as well as text and audio descriptions of their experiences. They can also input supplementary information such as weather, season, time, location, and vegetation.

[1243] 2. Data Analysis

[1244] server

[1245] The server stores and analyzes the received data, using image recognition algorithms to identify key objects and scenes from photos and videos, and natural language processing (NLP) techniques to extract emotions and keywords from text and audio data.

[1246] 3. Utilizing the Emotion Engine

[1247] server

[1248] The emotion engine installed on the server recognizes the user's emotions based on information extracted from visual and auditory data. For example, if image analysis provides photos or videos showing many smiling faces, the emotion engine will recognize the emotion of "joy."

[1249] 4. Generation of aroma profiles

[1250] server

[1251] Based on the extracted features and the emotions recognized by the emotion engine, a generative AI model is used to create a scent profile. This scent profile includes the combination and proportion of fragrances used within the system. For example, if the emotion of "joy" is recognized, a profile containing a lot of floral fragrances is generated.

[1252] server

[1253] The generated scent profile is sent to the device.

[1254] 5. Aroma production

[1255] Terminal

[1256] The device operates the built-in scent generating device based on the received scent profile, mixing the required scents from the scent cartridges in the specified ratio to create the specified scent.

[1257] Terminal

[1258] The generated scent is diffused around the user using a diffuser, allowing the user to experience a scent that matches the predicted emotion.

[1259] 6. Gather feedback and optimize your profile

[1260] User

[1261] Users can experience the generated scent and provide feedback on their impressions and suggestions for improvement through a dedicated application, such as "I wish the scent was a little stronger."

[1262] server

[1263] The server receives user feedback and stores it as evaluation data. It then retrains the generative AI model based on the feedback and optimizes the scent profile. The new profile is reflected in the next scent generation.

[1264] Specific examples

[1265] Data collection and analysis

[1266] A user uploads a photo taken at the beach in the summer along with the text "A fun summer day" to the system. The server uses image recognition to identify the sea, beach, and summer features in the photo, and extracts the positive emotion "It was fun" from the text.

[1267] Utilizing the Emotion Engine

[1268] The emotion engine installed on the server recognizes the emotion of "fun" from this data.

[1269] Scent profile generation and scent generation

[1270] Based on the extracted features and the recognized emotions, the server creates a scent profile mainly consisting of "fruity scents" and "sea breeze scents" and sends it to the device. The device then uses a scent generation device to generate a scent based on the profile and diffuses it around the user.

[1271] Gathering feedback and optimizing your profile

[1272] The user gives feedback such as "I want a more fruity scent." The server analyzes this feedback, reflects it in the generative AI model, and suggests increasing the ratio of fruity fragrances in the next scent profile, then sends the new profile to the device.

[1273] In this way, the present invention is a system that utilizes an emotion engine to provide a personalized scent experience that is tailored to the user's emotional state.

[1274] The processing flow will be explained below.

[1275] Step 1:

[1276] User

[1277] Users upload visual and audio data related to their memories (e.g., photos and videos of their trips, as well as text and audio descriptions of their experiences) to the system, and can also enter supplemental information such as weather, season, time, location, and vegetation.

[1278] Step 2:

[1279] server

[1280] The server securely stores the visual and auditory data received from the user, as well as any supplementary information, using a database or cloud storage.

[1281] Step 3:

[1282] server

[1283] The server analyzes the stored data. First, it uses image recognition algorithms to identify key objects and scenes (e.g., landscapes, buildings, seasons, and weather) from photos and videos. Next, it uses natural language processing (NLP) techniques to analyze the text and audio data and extract emotions and keywords (e.g., happy, sad, and place names).

[1284] Step 4:

[1285] server

[1286] The server uses an emotion engine to recognize the user's emotions based on the analyzed features and extracted emotions. For example, if the emotion "fun" is extracted from image and text analysis, the emotion engine will recognize this as "joy."

[1287] Step 5:

[1288] server

[1289] A generative AI model is used based on the emotion recognized by the emotion engine and the feature values ​​to create a scent profile. This scent profile includes the appropriate combination of fragrances and their proportions. For example, if the emotion of "joy" is recognized, a profile containing a lot of floral fragrances is generated.

[1290] Step 6:

[1291] server

[1292] The server then sends the generated scent profile to the device, transferring the data using a secure communication protocol.

[1293] Step 7:

[1294] Terminal

[1295] The device operates the built-in scent generating device based on the received scent profile, mixing the required scents from the scent cartridges in the specified ratio to create the specified scent.

[1296] Step 8:

[1297] Terminal

[1298] The generated scent is diffused around the user using a diffuser, allowing the user to experience a scent that matches the recognized emotion.

[1299] Step 9:

[1300] User

[1301] Users can experience the generated scent and provide feedback on their impressions and suggestions for improvement through a dedicated application. For example, they could input their opinion, such as, "I want a more fruity scent."

[1302] Step 10:

[1303] server

[1304] The server receives feedback from users and stores it as evaluation data. Based on this feedback, the generative AI model is retrained and the scent profile is optimized. The new profile is reflected the next time a scent is generated.

[1305] Specific examples

[1306] Data collection and analysis

[1307] Step 1-3: User and Server

[1308] A user uploads a photo they took at the beach in the summer along with the text "It was a fun summer day" to the system. The server uses an image recognition algorithm to identify the sea, beach, and summer scenery from the received photo, and then uses NLP technology to extract the positive emotion of "It was fun" from the text.

[1309] Utilizing the Emotion Engine

[1310] Step 4: Server

[1311] The emotion engine installed on the server recognizes the emotion of "fun" from this data.

[1312] Scent profile generation and scent generation

[1313] Step 5-8: Server and Terminal

[1314] Based on the extracted features and the recognized emotions, the server creates a scent profile mainly consisting of "fruity scents" and "sea breeze scents" and sends it to the device. The device then uses a scent generation device to generate a scent based on the profile and diffuse it around the user.

[1315] Gathering feedback and optimizing your profile

[1316] Steps 9-10: Users and Servers

[1317] The user gives feedback such as "I want a more fruity scent." The server analyzes this feedback, reflects it in the generative AI model, and suggests increasing the ratio of fruity fragrances in the next scent profile, then sends the new profile to the device.

[1318] In this way, the present invention is a system that utilizes an emotion engine to provide a personalized scent experience that is tailored to the user's emotional state.

[1319] Example 2

[1320] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1321] Conventional technologies have struggled to generate personalized scents based on a user's emotional state and memories. Furthermore, the process of efficiently collecting user feedback and optimizing scent profiles has been ineffective. To address these challenges, the present invention aims to provide a system that analyzes a user's visual and auditory data and generates scents based on their emotions.

[1322] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving visual and auditory data from a user, means for saving the received data, means for analyzing the saved data and identifying key objects and scenes using an image recognition algorithm, means for extracting emotions and keywords from text and voice data using natural language processing technology, means for recognizing emotions based on features extracted from the visual and auditory data, means for creating a scent profile using a generative AI model based on the recognized emotions, means for transmitting the created scent profile to a terminal, means for operating a scent generation device based on the received scent profile to generate a scent, means for diffusing the generated scent around the user, means for receiving user feedback regarding the generated scent, and means for optimizing the scent profile based on the received feedback. This makes it possible to generate a scent personalized to the user's emotional state and optimize the scent profile based on the feedback.

[1323] "Visual data" refers to visually related information, such as images and videos provided by users.

[1324] "Auditory data" refers to information related to the sense of hearing, such as voice or music provided by the user.

[1325] "Supplementary information" refers to additional information related to the visual and auditory data, such as weather, season, time, location, vegetation, etc.

[1326] "Database" means a digital information management system for storing visual data, auditory data, and supplemental information.

[1327] An "image recognition algorithm" is a technology that analyzes images and videos and recognizes specific objects and scenes.

[1328] "Natural language processing technology" is a technology for analyzing text and voice data and extracting emotions and keywords.

[1329] The "emotion engine" is a technology for recognizing a user's emotions based on features extracted from visual and auditory data.

[1330] A "generative AI model" is an artificial intelligence model for creating scent profiles based on recognized emotions.

[1331] A "fragrance profile" is information that defines a combination of multiple fragrances and their proportions.

[1332] A "terminal" is a device that generates a scent based on a scent profile and diffuses it around the user.

[1333] An "aroma generating device" is a device that extracts and mixes fragrances from a fragrance cartridge to generate a fragrance.

[1334] A "diffuser" is a device that diffuses the generated scent around the user.

[1335] "Feedback" refers to the user's thoughts and suggestions for improvement regarding the generated fragrance.

[1336] "Optimization" refers to the process of improving the scent profile based on user feedback.

[1337] The present invention is a system that analyzes visual and auditory data provided by a user and generates a scent based on the user's emotions. This system operates in cooperation with a server, a terminal, and the user.

[1338] 1. Data collection

[1339] User

[1340] Through a dedicated application, users provide visual and audio data related to their memories, such as photos, videos, text, and audio, and can also input supplemental information such as weather, season, time, location, and vegetation.

[1341] 2. Data storage

[1342] server

[1343] The server stores the received data in a database such as MongoDB or MySQL. The data is classified as visual data, auditory data, and supplementary information.

[1344] 3. Image Recognition and Natural Language Processing

[1345] server

[1346] The server uses image recognition algorithms (e.g., OpenCV or TensorFlow) to identify key objects and scenes from photos and videos. It also uses natural language processing (NLP) techniques (e.g., BERT) to extract emotions and keywords from text and audio data. It converts audio data into text using the Google Speech-to-Text API for analysis.

[1347] 4. Utilizing the Emotion Engine

[1348] server

[1349] The emotion engine installed on the server recognizes the user's emotions based on features extracted from visual and auditory data. For example, if there are many photos of smiling faces or positive keywords such as "it was fun," it will recognize the emotions of "joy" and "enjoyment."

[1350] 5. Generation of aroma profiles

[1351] server

[1352] The server uses a generative AI model (e.g., GPT-3) to create a scent profile based on the recognized emotion. This scent profile includes a combination of fragrances and their proportions. For example, if the emotion of "joy" is recognized, a profile containing a lot of floral fragrances is generated.

[1353] 6. Submit your scent profile

[1354] server

[1355] The server sends the created scent profile to the device in JSON format.

[1356] 7. Scent production and diffusion

[1357] Terminal

[1358] The device operates a built-in scent generating device (e.g., an Arduino-controlled diffuser) based on the received scent profile. It mixes the required fragrances from the fragrance cartridges in the specified ratio to generate the specified scent. It also diffuses the generated scent around the user using the diffuser.

[1359] 8. Gather feedback and optimize

[1360] User

[1361] Users can experience the generated scent and provide feedback on their impressions and suggestions for improvement through a dedicated application. For example, they could say, "I'd like a more fruity scent."

[1362] server

[1363] The server analyzes the received feedback and retrains the generative AI model to optimize the scent profile. The new profile is reflected in the next scent generation.

[1364] Specific examples

[1365] Data collection and analysis

[1366] A user uploads a photo taken at the beach in the summer along with the text "A fun summer day" to the system. The server uses an image recognition algorithm to identify the characteristics of the sea, beach, and summer from the photo, and then uses natural language processing technology to extract the positive emotion of "It was fun" from the text.

[1367] Utilizing the Emotion Engine

[1368] The emotion engine installed on the server recognizes the emotion of "fun" from this data.

[1369] Scent profile generation and scent generation

[1370] Based on the extracted features and the recognized emotions, the server uses a generative AI model to create a scent profile mainly consisting of "fruity scents" and "sea breeze scents" and sends it to the device. The device then uses a scent generation device to generate a scent based on the profile and diffuses it around the user.

[1371] Gathering feedback and optimizing your profile

[1372] The user gives feedback such as "I want a more fruity scent." The server analyzes this feedback, reflects it in the generative AI model, and suggests increasing the ratio of fruity fragrances in the next scent profile, then sends the new profile to the device.

[1373] Prompt Sentence Examples

[1374] "Analyze a user-provided summer beach photo and the text 'A fun summer day' and generate a corresponding scent profile. If the user provides feedback that they would like a more fruity scent, explain how you would optimize the scent profile."

[1375] As a result, the present invention provides a system that provides a personalized scent experience tailored to the user's emotional state and optimizes the scent profile based on user feedback.

[1376] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1377] Step 1: Collect data

[1378] User

[1379] Using a dedicated application, users upload photos and videos of their trip, text and audio impressions, and supplementary information such as weather, season, time, location, and vegetation. The inputs are visual data (photos and videos), auditory data (audio), and supplementary information. The output is this data sent to the server.

[1380] Step 2: Save your data

[1381] server

[1382] The server stores the received visual data, auditory data, and supplementary information in a database. The input is the data sent by the user. The output is the data stored in the database. Database systems such as MongoDB and MySQL are used at this stage.

[1383] Step 3: Perform image recognition

[1384] server

[1385] The server uses image recognition algorithms (e.g., OpenCV or TensorFlow) to analyze the stored photo and video data. It identifies key objects and scenes and stores them as metadata. The input is visual data stored in a database. The output is metadata for the identified objects and scenes.

[1386] Step 4: Performing Natural Language Processing

[1387] server

[1388] The server uses natural language processing (NLP) techniques (e.g., BERT) to extract sentiment and keywords from text and audio data. The audio data is converted to text using the Google Speech-to-Text API and then analyzed. The input is the text and audio data stored in the database. The output is the extracted sentiment and keywords.

[1389] Step 5: Leverage the Emotion Engine

[1390] server

[1391] The emotion engine installed on the server recognizes the user's emotions based on features extracted from visual and auditory data. The input is features extracted through image recognition and natural language processing. The output is the recognized user emotion. For example, emotions of "joy" and "fun" can be recognized from a photo of a smiling face and the keyword "fun."

[1392] Step 6: Generate an aroma profile

[1393] server

[1394] The server creates a scent profile using a generative AI model (e.g., GPT-3) based on the recognized emotion. The scent profile includes a combination of fragrances and their proportions. The input is the emotion recognized by the emotion engine. The output is the generated scent profile. For example, a floral scent profile is generated based on the emotion of "joy."

[1395] Step 7: Submit your scent profile

[1396] server

[1397] The server sends the generated scent profile to the device. The profile is sent in JSON format, etc. The input is the generated scent profile. The output is the profile being transmitted to the destination device.

[1398] Step 8: Scent generation

[1399] Terminal

[1400] The device operates the built-in scent generation device (e.g., an Arduino-controlled diffuser) based on the received scent profile. It mixes the required scents from the scent cartridges in the specified ratio to generate the specified scent. The input is the received scent profile. The output is the generated scent.

[1401] Step 9: Diffuse the scent

[1402] Terminal

[1403] The device diffuses the generated scent around the user using a diffuser or the like. The input is the generated scent. The output is the scent diffused in the user's environment.

[1404] Step 10: Gather feedback

[1405] User

[1406] The user experiences the generated scent and provides feedback and suggestions for improvement through a dedicated application. For example, the user may input feedback such as "I wish the scent was a little stronger." The input is the user's feedback. The output is evaluation data sent to the server.

[1407] Step 11: Analyze feedback

[1408] server

[1409] The server receives feedback from users and stores it in a database. The input is the user feedback. The output is the stored rating data.

[1410] Step 12: Optimizing the aroma profile

[1411] server

[1412] The server retrains the generative AI model based on the received feedback and optimizes the scent profile. The new profile is reflected in the next scent generation. The input is the feedback and the existing scent profile. The output is the new optimized scent profile.

[1413] (Application example 2)

[1414] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1415] Conventional scent generation systems have struggled to efficiently analyze visual and auditory data from users and generate personalized scents tailored to individual emotions and situations. Furthermore, the management and blending of fragrances used in manufacturing sites relies on manual labor, resulting in problems of inefficiency and inaccuracy. Large-scale manufacturing facilities, in particular, lack systems that automatically generate scents based on product and environmental information. This creates a demand for efficient, emotion-driven scent experiences.

[1416] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving visual and auditory data related to memories, means for analyzing the received data and extracting features, means for creating a scent profile using a generative AI model, means for transmitting the scent profile, means for generating a scent by operating a scent generation device based on the transmitted scent profile, means for receiving user feedback on the generated scent, means for optimizing the scent profile based on the feedback, means for managing fragrances used in a manufacturing facility and automatically generating a scent based on specified conditions, means for analyzing product and environmental information of the manufacturing facility using image processing and natural language processing techniques, and means for recognizing an emotional state using an emotion engine based on the analyzed information. This enables automation of fragrance management and blending in a manufacturing facility, thereby realizing the provision of personalized scents according to the user's emotional state.

[1417] "Visual data" refers to image information such as photos and videos provided by users.

[1418] "Auditory data" refers to acoustic information such as voice or music provided by the user.

[1419] "Features" refer to the information and attributes necessary for scent generation extracted from visual and auditory data.

[1420] "Generative AI model" refers to an artificial intelligence model for generating a scent profile based on extracted features.

[1421] "Scent profile" refers to data that defines the composition and characteristics of the scent generated by a scent generating device.

[1422] "Scent generating device" refers to a device that physically generates a scent based on a received scent profile.

[1423] "Feedback" refers to the user providing their thoughts and opinions about the generated scent.

[1424] "Manufacturing facility" means an industrial facility or location for producing products.

[1425] "Fragrance control" refers to the process of effectively controlling and maintaining the types and amounts of fragrances used within a manufacturing facility.

[1426] "Image processing" refers to the technology of analyzing visual data to identify key objects and scenes.

[1427] "Natural language processing" refers to the technology of extracting emotions and keywords from text and voice data.

[1428] "Emotion engine" refers to a system for recognizing a user's emotions from visual and auditory data.

[1429] The present invention is a system that provides a user with a personalized scent experience based on visual and auditory data, and will be described in a form that can be particularly applied to factory robots.

[1430] Hardware and Software

[1431] At the heart of the system is a robot powered by NVIDIA Jetson Nano. The robot has a built-in device that manages fragrances and automatically generates scents within the manufacturing facility. The robot runs the following software:

[1432] 1. Emotion recognition using TensorFlow.

[1433] 2. Generate scent profiles using PyTorch.

[1434] 3. Perform image processing using OpenCV.

[1435] Data collection

[1436] The user (operator of the manufacturing facility) provides the robot with images and videos of the product, and inputs their emotions and supplementary information (weather, season, etc.) via text and voice. The robot collects this visual and auditory data and sends it to the server.

[1437] Data analysis

[1438] The server uses OpenCV to identify key objects and patterns in the collected visual data, TensorFlow to extract emotions and keywords from the auditory data, and natural language processing to analyze text and audio data and recognize user emotions with an emotion engine.

[1439] Generation of aroma profiles

[1440] The server then uses a generative AI model trained with PyTorch to create a scent profile based on the extracted features and the recognized emotion. For example, if the emotion of "joy" is recognized, a scent profile rich in floral fragrances will be generated.

[1441] Fragrance production

[1442] The scent profile is sent from the server to the robot, which then uses a scent generator to mix the required fragrances from the fragrance cartridges in the specified ratio to create the desired scent, which is then diffused throughout the manufacturing facility using a diffuser.

[1443] Gathering feedback and optimizing your profile

[1444] The user provides feedback on the generated scent through the operator's application. This feedback is sent to the server and stored as evaluation data. The server then uses the feedback to retrain the generative AI model and optimize the scent profile for future uses.

[1445] Examples and prompts

[1446] Specific examples

[1447] During testing of a new product line in a factory, a robot takes photos and videos of the product, and an operator inputs text such as, "This product is very good and gives off a spring-like feeling." Image processing algorithms identify the new product and its context, and extract positive emotions and seasonal information such as "good" and "spring" from the text data. Based on this, a generative AI model generates a scent profile dominated by floral and fresh spring scents.

[1448] Prompt Sentence Examples

[1449] Example prompt for emotion recognition engine on text data:

[1450] Emotion Recognition Pipeline:

[1451] Input: "This product is very well made and gives a spring-like feeling."

[1452] Output: {'label': 'Positive', 'score': 0.95}

[1453] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1454] Step 1:

[1455] The user provides the robot with product images and videos, and inputs their emotions and supplementary information (such as the weather and season) via text and voice. The input data consists of image files, video files, text data, and voice files. The robot collects this data and sends it to a server. Data collection is performed using devices such as cameras and microphones.

[1456] Step 2:

[1457] The server analyzes the collected visual data using OpenCV. The input is image and video data. The server identifies key objects and patterns from the images and extracts this information as features. Specific operations include image preprocessing, edge detection, and object recognition. The output is a list of key objects and patterns.

[1458] Step 3:

[1459] The server analyzes the auditory data using TensorFlow. The inputs are audio files and text data. The server uses natural language processing techniques to extract emotions and keywords from the audio data. Specific operations include transcribing the audio, sentiment analysis, and keyword extraction. The output is a list of emotion labels and keywords.

[1460] Step 4:

[1461] The server generates a scent profile using a generative AI model trained with PyTorch based on the analyzed features and recognized emotions. The inputs are the features and emotion labels. The server inputs this data into the generative AI model to generate a scent profile that defines the scent composition and characteristics. Specific operations include the inference process of the generative model. The output is a scent profile.

[1462] Step 5:

[1463] The scent profile is sent from the server to the robot. The robot operates the scent generation device based on the received scent profile. The input is the scent profile. The robot mixes the required fragrances from the fragrance cartridges in the specified ratio to generate the specified scent. Specific operations include extracting the fragrances, mixing them, and diffusing them using a diffuser. The output is the generated scent.

[1464] Step 6:

[1465] The user inputs feedback about the generated scent through the operator's application. The input is text data of the feedback. The server receives this feedback and stores it as evaluation data. Specific operations include receiving and storing the feedback data.

[1466] Step 7:

[1467] The server retrains the generative AI model based on the feedback and optimizes the scent profile. The input is the feedback data. The server analyzes the feedback and uses it as learning data for the generative AI model. Specific operations include the model retraining process. The output is an optimized scent profile.

[1468] 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.

[1469] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1470] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1471] [Fourth embodiment]

[1472] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1473] 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.

[1474] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1475] 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.

[1476] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1477] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1478] 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.

[1479] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1480] 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.

[1481] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1482] 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.

[1483] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1484] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1485] The present invention provides a system that receives visual and auditory data, analyzes the data, and generates a scent to provide a user with a personalized scent experience. This system is mainly composed of a server, a terminal, and a user.

[1486] 1. Data collection

[1487] User

[1488] Users upload visual and audio data related to their memories to the system. This data includes photos, videos, text, and voice data. Supplementary information, such as weather, season, time, location, and vegetation, can also be added. For example, a user can upload photos taken during a trip and enter text about their impressions.

[1489] 2. Data Analysis

[1490] server

[1491] The server analyzes the visual and auditory data received from the user and extracts the features necessary for scent generation. This analysis uses image recognition algorithms and natural language processing (NLP) technology. For example, it can identify the landscape, season, weather, and location from photo data, and analyze emotions from text data.

[1492] 3. Generation of aroma profiles

[1493] server

[1494] Based on the analyzed features, a generative AI model is used to create a scent profile. This profile includes the combination and ratio of fragrances used within the system. For example, to recreate the scent of a spring day, specific fragrance ratios could be set, such as 50% cherry blossom scent, 30% fresh green scent, and 20% warm air scent.

[1495] server

[1496] The generated scent profile is sent to the device.

[1497] 4. Fragrance production

[1498] Terminal

[1499] The device operates the built-in scent generating device based on the scent profile received from the server. It mixes the required scents from the scent cartridges in the specified ratio to generate the scent. This process is performed in real time, and the generated scent is diffused around the user.

[1500] 5. Gathering Feedback

[1501] User

[1502] Users experience the scents generated and provide feedback on their impressions and suggestions for improvement via a dedicated mobile or web app.

[1503] 6. Profile Optimization

[1504] server

[1505] The server receives feedback from users and stores it as evaluation data for the scent profile. Based on this evaluation data, the generative AI model is retrained and the scent profile is optimized. For example, if a user provides feedback saying, "I would like a more refreshing scent," the server creates a new scent profile and reflects it the next time the user uses the device.

[1506] Specific examples

[1507] Data collection and analysis

[1508] A user uploads photos from a trip to France and an episode titled "Spring in Paris: A walk along the Eiffel Tower and the Seine River" to the system. The server uses image recognition to identify the Eiffel Tower, the Seine River, and spring scenery from the photo, and extracts the keywords "Paris," "spring," and "walk" from the text.

[1509] Scent profile generation and scent generation

[1510] Based on the extracted features, the server creates a scent profile, mainly consisting of "floral spring flower scent" and "fresh riverside scent," and sends it to the device. The device then uses a scent generation device to generate a scent based on the profile and diffuse it around the user.

[1511] Gathering feedback and optimizing your profile

[1512] The user gives feedback such as, "The floral scent was good, but I'd like it to be a little fresher." The server analyzes the feedback and reflects in the generative AI model a suggestion to increase the ratio of fresh fragrances in the next scent profile. The new profile is reflected the next time the user uses the product.

[1513] In this way, the present invention is a system that continuously provides users with personalized moving experiences.

[1514] The processing flow will be explained below.

[1515] Step 1:

[1516] User

[1517] Users upload visual and audio data to the system, providing, for example, photos of their trip and textual anecdotes about it, and can also provide supplemental information such as weather, season, time, location, and vegetation.

[1518] Step 2:

[1519] server

[1520] The server stores the visual and auditory data and supplementary information received from the user, and securely stores the data using a database or cloud storage.

[1521] Step 3:

[1522] server

[1523] The server analyzes the stored data, using image recognition algorithms to identify key objects and scenes from the photos, and natural language processing (NLP) techniques to extract emotions and keywords from the text data, thereby obtaining the features necessary for generating the scent.

[1524] Step 4:

[1525] server

[1526] The server uses a generative AI model based on the extracted features to create a scent profile, which includes the combinations and proportions of fragrances used within the system.

[1527] Step 5:

[1528] server

[1529] The server then sends the created scent profile to the device, securely transferring the data using a communication protocol.

[1530] Step 6:

[1531] Terminal

[1532] The device operates the built-in scent generating device based on the scent profile received from the server, mixing the required scents from the scent cartridges in a specific ratio to create the specified scent.

[1533] Step 7:

[1534] Terminal

[1535] The device then uses a diffuser to diffuse the generated scent around the user, adjusting it to spread throughout the room so that the user can experience the scent.

[1536] Step 8:

[1537] User

[1538] Users can experience the generated scent and provide feedback on their impressions and suggestions for improvement through a dedicated application, such as "I'd like a more floral scent."

[1539] Step 9:

[1540] server

[1541] The server receives feedback from users and stores it as evaluation data, which is used to retrain the generative AI model and optimize the scent profile.

[1542] Step 10:

[1543] server

[1544] The server then sends the new optimized scent profile to the device to be reflected in the next scent generation, thereby continuously improving the user experience.

[1545] Through the above steps, the present invention provides a personalized scent experience to the user.

[1546] Example 1

[1547] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1548] Conventional scent generation systems were unable to utilize visual and auditory data to generate scents, making it difficult to provide a personalized scent experience based on a user's specific memories and emotions. Furthermore, mechanisms for effectively incorporating user feedback on the generated scent and optimizing the scent profile were not adequately developed. Furthermore, supplementary information such as weather, season, time, location, and vegetation could not be reflected, resulting in a lower quality of overall experience.

[1549] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1550] In this invention, the server includes means for receiving visual and auditory data from a user, means for analyzing the received data using an image recognition algorithm and natural language processing technology to extract features necessary for scent generation, means for creating a scent profile using a generative AI model based on the features, means for transmitting the scent profile to a terminal, means for operating a scent generation device on the terminal based on the transmitted scent profile to generate a scent, means for receiving user feedback regarding the generated scent through a dedicated application, means for optimizing the scent profile based on the feedback, and means for receiving supplementary information such as weather, season, time, location, and vegetation from the user. This makes it possible to provide a personalized scent experience based on the user's memories and emotions and to optimize the scent profile to reflect that feedback.

[1551] "Visual data" refers to visual information such as photos and videos provided by users.

[1552] "Auditory data" refers to aural information such as text or voice data provided by a user.

[1553] "Features" refer to the information elements necessary for scent generation, extracted using image recognition algorithms and natural language processing technology.

[1554] "Generative AI model" refers to an artificial intelligence system that uses machine learning algorithms to generate scent profiles.

[1555] "Fragrance profile" means data that indicates the combination and proportions of fragrance ingredients required to produce a particular fragrance.

[1556] "Scent generation device" refers to a mechanical device for generating an actual scent based on a scent profile.

[1557] "Feedback" refers to the user's impressions of the scent they experienced and their opinions on areas for improvement.

[1558] "Supplemental information" refers to user-provided background information for the visual and audio data, such as weather, season, time, location, vegetation, etc.

[1559] "Image recognition algorithm" refers to an algorithm that automatically identifies specific objects, scenes, and situations from photos and videos.

[1560] "Natural language processing technology" refers to algorithms for analyzing meaning and emotions from text and voice data.

[1561] "User" refers to an individual who utilizes the system to provide visual and auditory data and receive a personalized scent experience.

[1562] "Terminal" refers to a device, such as a computer or smartphone, that a user uses to upload visual and auditory data and experience the generated scent.

[1563] MODE FOR CARRYING OUT THE INVENTION

[1564] This invention is a system that receives visual and auditory data, analyzes it, and generates a scent to provide a user with a personalized scent experience. This system is mainly composed of a server, a terminal, and a user. The specific roles and operation procedures of each component are described below.

[1565] 1. Data collection

[1566] User

[1567] Users log in to the system using a device such as a smartphone or PC via a dedicated application or browser. They upload visual and auditory data related to their memories, including photos, videos, text, and voice data. They also enter supplementary information, such as weather, season, time, location, and vegetation, as needed. For example, a user might upload an episode of "Spring in Paris, a walk with the Eiffel Tower and along the Seine River," along with photos.

[1568] 2. Data Analysis

[1569] server

[1570] The server analyzes the visual and auditory data received from the user and extracts the features necessary for scent generation. Visual data (photos and videos) is analyzed using image recognition algorithms (e.g., TensorFlow and OpenCV). For example, the Eiffel Tower, the Seine River, and spring scenery in a photo are identified. Next, auditory data (text and voice data) is analyzed using natural language processing (NLP) techniques (e.g., Google BERT and spaCy) to extract emotions and keywords. For example, keywords such as "Paris," "spring," and "walk" and emotions such as "fun" and "relaxed" are analyzed.

[1571] 3. Generation of aroma profiles

[1572] server

[1573] The server uses a generative AI model (e.g., GPT-4) to create a scent profile based on the analyzed features. The user inputs a prompt to the generative AI model: "Please create a scent profile inspired by spring in Paris, the Eiffel Tower, and a walk along the Seine." As a result, the server calculates specific fragrance ratios and combinations. For example, it generates a profile that is "40% floral spring flower scent, 30% fresh riverside scent, and 30% warm sunshine scent."

[1574] server

[1575] The generated scent profile is sent to the terminal in data format.

[1576] 4. Fragrance production

[1577] Terminal

[1578] The device operates its built-in scent-generating device based on the scent profile received from the server. The control software inside the device extracts and mixes the scent from the scent cartridge in the specified ratio to create the scent. This process is carried out in real time, and the generated scent is diffused around the user, providing them with a personalized scent experience.

[1579] 5. Gathering Feedback

[1580] User

[1581] Users experience the generated scent and provide feedback on their impressions and suggestions for improvement via a dedicated mobile or web app. For example, they can enter specific opinions such as, "I liked the floral scent, but I'd like it to be a little fresher." The feedback is then sent to the server.

[1582] 6. Profile Optimization

[1583] server

[1584] The server receives feedback from users and stores it as scent profile evaluation data. Based on this, it retrains the generative AI model and optimizes the scent profile. For example, based on feedback that a "slightly fresher scent" is desired, a new profile is created that incorporates more fresh scent components. The next time the user uses the service, a new scent profile that reflects these improvements will be provided.

[1585] In this way, the present invention is a system that provides a user with a personalized and inspiring scent experience.

[1586] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1587] Step 1: Collect data

[1588] User

[1589] Users use devices such as smartphones or PCs to log in via a dedicated application or browser on the system and upload visual and auditory data related to their memories.

[1590] input

[1591] Photos, videos, text, voice data, and supplementary information (weather, season, time, location, vegetation) provided by users.

[1592] output

[1593] The visual and audio data and accompanying supplemental information are transmitted to the server side.

[1594] Specific actions

[1595] The user presses the "Upload" button within the application, selects photos, videos, text, or voice data from the file selection screen, and presses the send button.

[1596] Step 2: Analyze the data

[1597] server

[1598] The server analyzes the received visual and auditory data and extracts the features necessary for scent generation.

[1599] input

[1600] Visual data (photos, videos), auditory data (text, voice data), and supplemental information sent by the user.

[1601] output

[1602] Analyzed feature information (e.g., location, season, scenery, emotion, keywords).

[1603] Data processing / data calculation

[1604] The server analyzes the visual data using image recognition algorithms (e.g., TensorFlow and OpenCV) to identify scenery, locations, etc. It also analyzes the auditory data using natural language processing techniques (e.g., Google BERT and spaCy) to extract emotions and keywords.

[1605] Specific actions

[1606] Image recognition algorithms run on the server, identifying the Eiffel Tower, the Seine River, and spring scenery from the photo, while natural language processing technology analyzes the text to extract keywords like "Paris," "spring," and "walk," as well as emotions like "fun" and "relaxing."

[1607] Step 3: Generate an aroma profile

[1608] server

[1609] The server creates a scent profile using a generative AI model based on the analyzed features.

[1610] input

[1611] Analyzed feature information (e.g., location, season, scenery, emotion, keywords).

[1612] output

[1613] Fragrance profile data (specific fragrance combinations and ratios).

[1614] Data processing / data calculation

[1615] The server inputs a prompt into a generative AI model (e.g., GPT-4) to generate a scent profile. The prompt is, "Please create a scent profile inspired by spring in Paris, the Eiffel Tower, and a walk along the Seine."

[1616] Specific actions

[1617] The generative AI model calculates based on the prompt and creates a specific scent profile, such as "40% floral spring flower scent, 30% fresh riverside scent, and 30% warm sunshine scent."

[1618] Step 4: Submit your scent profile

[1619] server

[1620] The server sends the generated scent profile to the terminal.

[1621] input

[1622] Generated scent profile data.

[1623] output

[1624] Scent profile data sent to the device.

[1625] Specific actions

[1626] The scent profile data is transferred from the server to the terminal via the network.

[1627] Step 5: Scent generation

[1628] Terminal

[1629] The terminal operates the built-in scent generating device based on the scent profile received from the server.

[1630] input

[1631] Scent profile data sent from the server.

[1632] output

[1633] The scent produced is diffused into the surrounding area.

[1634] Data processing / data calculation

[1635] The control software in the device extracts and mixes the fragrance from the fragrance cartridge in the specified ratio.

[1636] Specific actions

[1637] The fragrance generating device extracts the required fragrances from the fragrance cartridges and mixes and diffuses them based on the profile.

[1638] Step 6: Gather feedback

[1639] User

[1640] Users experience the generated scent and provide feedback on their impressions and suggestions for improvement.

[1641] input

[1642] User feedback information (impressions, areas for improvement).

[1643] output

[1644] Feedback data sent to the server.

[1645] Specific actions

[1646] Users fill out and submit a feedback form in a dedicated mobile or web app.

[1647] Step 7: Optimize your profile

[1648] server

[1649] The server receives user feedback and stores it as evaluation data for the scent profile. Based on this evaluation data, the generative AI model is retrained and the scent profile is optimized.

[1650] input

[1651] User feedback data.

[1652] output

[1653] Optimized scent profile data.

[1654] Data processing / data calculation

[1655] Based on the feedback, the generative AI model is retrained to generate new scent profiles.

[1656] Specific actions

[1657] The server analyzes the feedback data, updates the generative AI model, and reflects it in the next scent profile.

[1658] (Application example 1)

[1659] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1660] The present invention relates to a technology for generating scents by analyzing visual and auditory data, and aims to realize a more interactive and immediate experience in a system that provides users with a personalized scent experience. Another object of the present invention is to provide customers with a special shopping experience by analyzing customers' visual and auditory data in a store in real time and generating a customized scent on the spot.

[1661] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1662] In this invention, the server includes means for receiving visual and auditory data related to memories from a user, means for analyzing the received data and extracting features necessary for scent generation, means for creating a scent profile using a generative AI model based on the features, means for transmitting the scent profile, means for operating a scent generation device based on the transmitted scent profile to generate a scent, means for receiving user feedback on the generated scent, means for optimizing the scent profile based on the feedback, and means for collecting customer visual and auditory data in a store and generating a personalized scent on the spot, thereby providing users with a personalized and moving experience and improving the customer's purchasing experience.

[1663] "Visual data" refers to visually related data such as images and videos provided by users.

[1664] "Auditory data" refers to data related to the sense of hearing, such as voice or music, provided by the user.

[1665] "Scent generation" refers to the process of physically creating a specific scent based on extracted data.

[1666] "Features" refers to information necessary for scent generation extracted from visual and auditory data.

[1667] "Generative AI model" refers to an artificial intelligence model for creating scent profiles based on data analysis.

[1668] A "fragrance profile" is a setting that defines which specific fragrances are used and in what proportions.

[1669] An "aroma generating device" is a device that generates an actual aroma based on an aroma profile.

[1670] "Feedback" refers to opinions and information such as impressions from users after use and suggestions for improvement.

[1671] "In-store" refers to places where customers visit, such as commercial facilities, sales areas, and exhibition spaces.

[1672] "Personalization" refers to customizing something to suit the preferences and experiences of individual users.

[1673] The present invention relates to a system for providing a personalized scent experience to customers in a physical store. Hereinafter, an embodiment of the system will be described.

[1674] 1. Data Collection

[1675] User:

[1676] Users use smartphones or smart glasses to collect visual and auditory data, including the sights and sounds of the store.

[1677] 2. Data Analysis

[1678] server:

[1679] The server analyzes the visual and auditory data received from the user. It uses OpenCV for camera control to analyze the visual data, and the SpeechRecognition library to analyze the audio data. Scenery, objects, and seasonal sensations are extracted from the visual data, while the sound atmosphere and text information are extracted from the audio data.

[1680] 3. Generation of Aroma Profiles

[1681] server:

[1682] Based on the analyzed features, the server creates a scent profile using a generative AI model that determines which fragrances to mix and in what proportions based on information obtained from the user's visual and auditory data.

[1683] 4. Fragrance production

[1684] Device:

[1685] The terminal operates a scent-generating device based on the scent profile received from the server. This device mixes the required fragrances in the specified ratio and diffuses them around the customer. The AromaBlender device generates the scent in real time.

[1686] 5. Collect feedback

[1687] User:

[1688] Users provide feedback about the scents generated through a dedicated mobile or web app.

[1689] 6. Profile Optimization

[1690] server:

[1691] The server analyzes user feedback and stores the scent profile as evaluation data, which is then used to retrain the generative AI model and optimize the scent profile.

[1692] Specific examples

[1693] A customer wears smart glasses and walks around the store. The scenery of the flower section is captured as visual data, and the music in the store is recorded as auditory data. The server analyzes this data and extracts the characteristics of "floral scents" and "refreshing scents." A scent profile is created based on the generative AI model, and the AromaBlender device generates the scent and provides it to the customer. If the customer gives feedback such as "I liked the floral scent, but I would like it to be a little more refreshing," the scent profile that reflects this feedback will be used the next time.

[1694] Example prompt sentence:

[1695] Smart glasses installed in stores collect the sights and sounds customers see and hear in real time. When a customer is in a specific area of ​​the store, the data is analyzed to generate the optimal scent for that area, and that scent is then delivered through a scent-generating device. For example, if a customer is in the flower section, a scent generated based on "floral scents" and "forest scents" will be distributed.

[1696] In this way, the present invention realizes a system that continuously provides users with personalized and moving experiences, and can also improve customer engagement through special shopping experiences in physical stores.

[1697] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1698] Step 1:

[1699] Data collection

[1700] Users use smartphones or smart glasses to collect visual and auditory data within a store. Specifically, when a user is in a specific area, the camera captures the scenery and the microphone records the surrounding sounds. The input is the scenery seen by the customer (image data) and the sounds heard (audio data), and this data is saved on the device.

[1701] Step 2:

[1702] Sending data

[1703] The device transmits the collected visual and auditory data to a server. Specifically, if the device is connected to the Internet, the collected image data and audio data are uploaded to the server. The input is the data stored on the device, and the output is the data transmitted to the server.

[1704] Step 3:

[1705] Data analysis

[1706] The server analyzes the received visual and auditory data. Specifically, it runs an image recognition algorithm using OpenCV on the visual data to identify scenery and objects. For the auditory data, it uses the SpeechRecognition library to convert the audio data into text and extract specific keywords and emotions. The input is the received visual and auditory data, and the output is feature data as the analysis result.

[1707] Step 4:

[1708] Generation of aroma profiles

[1709] The server uses a generative AI model to create a scent profile based on the analyzed features. Specifically, the AI ​​model calculates the optimal scent combination and its ratio based on the input feature data. The input is feature data, and the output is a scent profile.

[1710] Step 5:

[1711] Sending your scent profile

[1712] The server sends the generated scent profile to the terminal. Specifically, scent profile data is sent from the server to the terminal. The input is the scent profile, and the output is the scent profile sent to the terminal.

[1713] Step 6:

[1714] Fragrance production

[1715] The terminal operates the scent generation device based on the scent profile received from the server. Specifically, it uses the AromaBlender device to mix the fragrances defined in the scent profile in the specified ratio and diffuse the scent around the customer. The input is the scent profile and the output is the generated scent.

[1716] Step 7:

[1717] Collecting feedback

[1718] Users provide feedback on the generated scent, including their impressions and suggestions for improvement. Specifically, they use a dedicated mobile or web app to enter the scent characteristics they perceived and desired changes in text format. The input is the user's impressions and suggestions for improvement, and the output is feedback data.

[1719] Step 8:

[1720] Profile Optimization

[1721] The server analyzes the feedback from users and stores it as evaluation data for the scent profile. Specifically, it retrains the generative AI model based on the feedback content and optimizes the scent profile. The input is the feedback data, and the output is the optimized scent profile.

[1722] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1723] The present invention is a system that analyzes visual and auditory data received from a user and generates a scent to provide the user with a personalized scent experience. By combining this system with an emotion engine, it also has the function of recognizing the user's emotional state and reflecting this in the scent generation. A specific embodiment of the system is described below.

[1724] 1. Data collection

[1725] User

[1726] Users can provide visual and audio data related to their memories, such as photos and videos of their trips, as well as text and audio descriptions of their experiences. They can also input supplementary information such as weather, season, time, location, and vegetation.

[1727] 2. Data Analysis

[1728] server

[1729] The server stores and analyzes the received data, using image recognition algorithms to identify key objects and scenes from photos and videos, and natural language processing (NLP) techniques to extract emotions and keywords from text and audio data.

[1730] 3. Utilizing the Emotion Engine

[1731] server

[1732] The emotion engine installed on the server recognizes the user's emotions based on information extracted from visual and auditory data. For example, if image analysis provides photos or videos showing many smiling faces, the emotion engine will recognize the emotion of "joy."

[1733] 4. Generation of aroma profiles

[1734] server

[1735] Based on the extracted features and the emotions recognized by the emotion engine, a generative AI model is used to create a scent profile. This scent profile includes the combination and proportion of fragrances used within the system. For example, if the emotion of "joy" is recognized, a profile containing a lot of floral fragrances is generated.

[1736] server

[1737] The generated scent profile is sent to the device.

[1738] 5. Aroma production

[1739] Terminal

[1740] The device operates the built-in scent generating device based on the received scent profile, mixing the required scents from the scent cartridges in the specified ratio to create the specified scent.

[1741] Terminal

[1742] The generated scent is diffused around the user using a diffuser, allowing the user to experience a scent that matches the predicted emotion.

[1743] 6. Gather feedback and optimize your profile

[1744] User

[1745] Users can experience the generated scent and provide feedback on their impressions and suggestions for improvement through a dedicated application, such as "I wish the scent was a little stronger."

[1746] server

[1747] The server receives user feedback and stores it as evaluation data. It then retrains the generative AI model based on the feedback and optimizes the scent profile. The new profile is reflected in the next scent generation.

[1748] Specific examples

[1749] Data collection and analysis

[1750] A user uploads a photo taken at the beach in the summer along with the text "A fun summer day" to the system. The server uses image recognition to identify the sea, beach, and summer features in the photo, and extracts the positive emotion "It was fun" from the text.

[1751] Utilizing the Emotion Engine

[1752] The emotion engine installed on the server recognizes the emotion of "fun" from this data.

[1753] Scent profile generation and scent generation

[1754] Based on the extracted features and the recognized emotions, the server creates a scent profile mainly consisting of "fruity scents" and "sea breeze scents" and sends it to the device. The device then uses a scent generation device to generate a scent based on the profile and diffuses it around the user.

[1755] Gathering feedback and optimizing your profile

[1756] The user gives feedback such as "I want a more fruity scent." The server analyzes this feedback, reflects it in the generative AI model, and suggests increasing the ratio of fruity fragrances in the next scent profile, then sends the new profile to the device.

[1757] In this way, the present invention is a system that utilizes an emotion engine to provide a personalized scent experience that is tailored to the user's emotional state.

[1758] The processing flow will be explained below.

[1759] Step 1:

[1760] User

[1761] Users upload visual and audio data related to their memories (e.g., photos and videos of their trips, as well as text and audio descriptions of their experiences) to the system, and can also enter supplemental information such as weather, season, time, location, and vegetation.

[1762] Step 2:

[1763] server

[1764] The server securely stores the visual and auditory data received from the user, as well as any supplementary information, using a database or cloud storage.

[1765] Step 3:

[1766] server

[1767] The server analyzes the stored data. First, it uses image recognition algorithms to identify key objects and scenes (e.g., landscapes, buildings, seasons, and weather) from photos and videos. Next, it uses natural language processing (NLP) techniques to analyze the text and audio data and extract emotions and keywords (e.g., happy, sad, and place names).

[1768] Step 4:

[1769] server

[1770] The server uses an emotion engine to recognize the user's emotions based on the analyzed features and extracted emotions. For example, if the emotion "fun" is extracted from image and text analysis, the emotion engine will recognize this as "joy."

[1771] Step 5:

[1772] server

[1773] A generative AI model is used based on the emotion recognized by the emotion engine and the feature values ​​to create a scent profile. This scent profile includes the appropriate combination of fragrances and their proportions. For example, if the emotion of "joy" is recognized, a profile containing a lot of floral fragrances is generated.

[1774] Step 6:

[1775] server

[1776] The server then sends the generated scent profile to the device, transferring the data using a secure communication protocol.

[1777] Step 7:

[1778] Terminal

[1779] The device operates the built-in scent generating device based on the received scent profile, mixing the required scents from the scent cartridges in the specified ratio to create the specified scent.

[1780] Step 8:

[1781] Terminal

[1782] The generated scent is diffused around the user using a diffuser, allowing the user to experience a scent that matches the recognized emotion.

[1783] Step 9:

[1784] User

[1785] Users can experience the generated scent and provide feedback on their impressions and suggestions for improvement through a dedicated application. For example, they could input their opinion, such as, "I want a more fruity scent."

[1786] Step 10:

[1787] server

[1788] The server receives feedback from users and stores it as evaluation data. Based on this feedback, the generative AI model is retrained and the scent profile is optimized. The new profile is reflected the next time a scent is generated.

[1789] Specific examples

[1790] Data collection and analysis

[1791] Step 1-3: User and Server

[1792] A user uploads a photo they took at the beach in the summer along with the text "It was a fun summer day" to the system. The server uses an image recognition algorithm to identify the sea, beach, and summer scenery from the received photo, and then uses NLP technology to extract the positive emotion of "It was fun" from the text.

[1793] Utilizing the Emotion Engine

[1794] Step 4: Server

[1795] The emotion engine installed on the server recognizes the emotion of "fun" from this data.

[1796] Scent profile generation and scent generation

[1797] Step 5-8: Server and Terminal

[1798] Based on the extracted features and the recognized emotions, the server creates a scent profile mainly consisting of "fruity scents" and "sea breeze scents" and sends it to the device. The device then uses a scent generation device to generate a scent based on the profile and diffuse it around the user.

[1799] Gathering feedback and optimizing your profile

[1800] Steps 9-10: Users and Servers

[1801] The user gives feedback such as "I want a more fruity scent." The server analyzes this feedback, reflects it in the generative AI model, and suggests increasing the ratio of fruity fragrances in the next scent profile, then sends the new profile to the device.

[1802] In this way, the present invention is a system that utilizes an emotion engine to provide a personalized scent experience that is tailored to the user's emotional state.

[1803] Example 2

[1804] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1805] Conventional technologies have struggled to generate personalized scents based on a user's emotional state and memories. Furthermore, the process of efficiently collecting user feedback and optimizing scent profiles has been ineffective. To address these challenges, the present invention aims to provide a system that analyzes a user's visual and auditory data and generates scents based on their emotions.

[1806] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving visual and auditory data from a user, means for saving the received data, means for analyzing the saved data and identifying key objects and scenes using an image recognition algorithm, means for extracting emotions and keywords from text and voice data using natural language processing technology, means for recognizing emotions based on features extracted from the visual and auditory data, means for creating a scent profile using a generative AI model based on the recognized emotions, means for transmitting the created scent profile to a terminal, means for operating a scent generation device based on the received scent profile to generate a scent, means for diffusing the generated scent around the user, means for receiving user feedback regarding the generated scent, and means for optimizing the scent profile based on the received feedback. This makes it possible to generate a scent personalized to the user's emotional state and optimize the scent profile based on the feedback.

[1807] "Visual data" refers to visually related information, such as images and videos provided by users.

[1808] "Auditory data" refers to information related to the sense of hearing, such as voice or music provided by the user.

[1809] "Supplementary information" refers to additional information related to the visual and auditory data, such as weather, season, time, location, vegetation, etc.

[1810] "Database" means a digital information management system for storing visual data, auditory data, and supplemental information.

[1811] An "image recognition algorithm" is a technology that analyzes images and videos and recognizes specific objects and scenes.

[1812] "Natural language processing technology" is a technology for analyzing text and voice data and extracting emotions and keywords.

[1813] The "emotion engine" is a technology for recognizing a user's emotions based on features extracted from visual and auditory data.

[1814] A "generative AI model" is an artificial intelligence model for creating scent profiles based on recognized emotions.

[1815] A "fragrance profile" is information that defines a combination of multiple fragrances and their proportions.

[1816] A "terminal" is a device that generates a scent based on a scent profile and diffuses it around the user.

[1817] An "aroma generating device" is a device that extracts and mixes fragrances from a fragrance cartridge to generate a fragrance.

[1818] A "diffuser" is a device that diffuses the generated scent around the user.

[1819] "Feedback" refers to the user's thoughts and suggestions for improvement regarding the generated fragrance.

[1820] "Optimization" refers to the process of improving the scent profile based on user feedback.

[1821] The present invention is a system that analyzes visual and auditory data provided by a user and generates a scent based on the user's emotions. This system operates in cooperation with a server, a terminal, and the user.

[1822] 1. Data collection

[1823] User

[1824] Through a dedicated application, users provide visual and audio data related to their memories, such as photos, videos, text, and audio, and can also input supplemental information such as weather, season, time, location, and vegetation.

[1825] 2. Data storage

[1826] server

[1827] The server stores the received data in a database such as MongoDB or MySQL. The data is classified as visual data, auditory data, and supplementary information.

[1828] 3. Image Recognition and Natural Language Processing

[1829] server

[1830] The server uses image recognition algorithms (e.g., OpenCV or TensorFlow) to identify key objects and scenes from photos and videos. It also uses natural language processing (NLP) techniques (e.g., BERT) to extract emotions and keywords from text and audio data. It converts audio data into text using the Google Speech-to-Text API for analysis.

[1831] 4. Utilizing the Emotion Engine

[1832] server

[1833] The emotion engine installed on the server recognizes the user's emotions based on features extracted from visual and auditory data. For example, if there are many photos of smiling faces or positive keywords such as "it was fun," it will recognize the emotions of "joy" and "enjoyment."

[1834] 5. Generation of aroma profiles

[1835] server

[1836] The server uses a generative AI model (e.g., GPT-3) to create a scent profile based on the recognized emotion. This scent profile includes a combination of fragrances and their proportions. For example, if the emotion of "joy" is recognized, a profile containing a lot of floral fragrances is generated.

[1837] 6. Submit your scent profile

[1838] server

[1839] The server sends the created scent profile to the device in JSON format.

[1840] 7. Scent production and diffusion

[1841] Terminal

[1842] The device operates a built-in scent generating device (e.g., an Arduino-controlled diffuser) based on the received scent profile. It mixes the required fragrances from the fragrance cartridges in the specified ratio to generate the specified scent. It also diffuses the generated scent around the user using the diffuser.

[1843] 8. Gather feedback and optimize

[1844] User

[1845] Users can experience the generated scent and provide feedback on their impressions and suggestions for improvement through a dedicated application. For example, they could say, "I'd like a more fruity scent."

[1846] server

[1847] The server analyzes the received feedback and retrains the generative AI model to optimize the scent profile. The new profile is reflected in the next scent generation.

[1848] Specific examples

[1849] Data collection and analysis

[1850] A user uploads a photo taken at the beach in the summer along with the text "A fun summer day" to the system. The server uses an image recognition algorithm to identify the characteristics of the sea, beach, and summer from the photo, and then uses natural language processing technology to extract the positive emotion of "It was fun" from the text.

[1851] Utilizing the Emotion Engine

[1852] The emotion engine installed on the server recognizes the emotion of "fun" from this data.

[1853] Scent profile generation and scent generation

[1854] Based on the extracted features and the recognized emotions, the server uses a generative AI model to create a scent profile mainly consisting of "fruity scents" and "sea breeze scents" and sends it to the device. The device then uses a scent generation device to generate a scent based on the profile and diffuses it around the user.

[1855] Gathering feedback and optimizing your profile

[1856] The user gives feedback such as "I want a more fruity scent." The server analyzes this feedback, reflects it in the generative AI model, and suggests increasing the ratio of fruity fragrances in the next scent profile, then sends the new profile to the device.

[1857] Prompt Sentence Examples

[1858] "Analyze a user-provided summer beach photo and the text 'A fun summer day' and generate a corresponding scent profile. If the user provides feedback that they would like a more fruity scent, explain how you would optimize the scent profile."

[1859] As a result, the present invention provides a system that provides a personalized scent experience tailored to the user's emotional state and optimizes the scent profile based on user feedback.

[1860] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1861] Step 1: Collect data

[1862] User

[1863] Using a dedicated application, users upload photos and videos of their trip, text and audio impressions, and supplementary information such as weather, season, time, location, and vegetation. The inputs are visual data (photos and videos), auditory data (audio), and supplementary information. The output is this data sent to the server.

[1864] Step 2: Save your data

[1865] server

[1866] The server stores the received visual data, auditory data, and supplementary information in a database. The input is the data sent by the user. The output is the data stored in the database. Database systems such as MongoDB and MySQL are used at this stage.

[1867] Step 3: Perform image recognition

[1868] server

[1869] The server uses image recognition algorithms (e.g., OpenCV or TensorFlow) to analyze the stored photo and video data. It identifies key objects and scenes and stores them as metadata. The input is visual data stored in a database. The output is metadata for the identified objects and scenes.

[1870] Step 4: Performing Natural Language Processing

[1871] server

[1872] The server uses natural language processing (NLP) techniques (e.g., BERT) to extract sentiment and keywords from text and audio data. The audio data is converted to text using the Google Speech-to-Text API and then analyzed. The input is the text and audio data stored in the database. The output is the extracted sentiment and keywords.

[1873] Step 5: Leverage the Emotion Engine

[1874] server

[1875] The emotion engine installed on the server recognizes the user's emotions based on features extracted from visual and auditory data. The input is features extracted through image recognition and natural language processing. The output is the recognized user emotion. For example, emotions of "joy" and "fun" can be recognized from a photo of a smiling face and the keyword "fun."

[1876] Step 6: Generate an aroma profile

[1877] server

[1878] The server creates a scent profile using a generative AI model (e.g., GPT-3) based on the recognized emotion. The scent profile includes a combination of fragrances and their proportions. The input is the emotion recognized by the emotion engine. The output is the generated scent profile. For example, a floral scent profile is generated based on the emotion of "joy."

[1879] Step 7: Submit your scent profile

[1880] server

[1881] The server sends the generated scent profile to the device. The profile is sent in JSON format, etc. The input is the generated scent profile. The output is the profile being transmitted to the destination device.

[1882] Step 8: Scent generation

[1883] Terminal

[1884] The device operates the built-in scent generation device (e.g., an Arduino-controlled diffuser) based on the received scent profile. It mixes the required scents from the scent cartridges in the specified ratio to generate the specified scent. The input is the received scent profile. The output is the generated scent.

[1885] Step 9: Diffuse the scent

[1886] Terminal

[1887] The device diffuses the generated scent around the user using a diffuser or the like. The input is the generated scent. The output is the scent diffused in the user's environment.

[1888] Step 10: Gather feedback

[1889] User

[1890] The user experiences the generated scent and provides feedback and suggestions for improvement through a dedicated application. For example, the user may input feedback such as "I wish the scent was a little stronger." The input is the user's feedback. The output is evaluation data sent to the server.

[1891] Step 11: Analyze feedback

[1892] server

[1893] The server receives feedback from users and stores it in a database. The input is the user feedback. The output is the stored rating data.

[1894] Step 12: Optimizing the aroma profile

[1895] server

[1896] The server retrains the generative AI model based on the received feedback and optimizes the scent profile. The new profile is reflected in the next scent generation. The input is the feedback and the existing scent profile. The output is the new optimized scent profile.

[1897] (Application example 2)

[1898] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1899] Conventional scent generation systems have struggled to efficiently analyze visual and auditory data from users and generate personalized scents tailored to individual emotions and situations. Furthermore, the management and blending of fragrances used in manufacturing sites relies on manual labor, resulting in problems of inefficiency and inaccuracy. Large-scale manufacturing facilities, in particular, lack systems that automatically generate scents based on product and environmental information. This creates a demand for efficient, emotion-driven scent experiences.

[1900] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving visual and auditory data related to memories, means for analyzing the received data and extracting features, means for creating a scent profile using a generative AI model, means for transmitting the scent profile, means for generating a scent by operating a scent generation device based on the transmitted scent profile, means for receiving user feedback on the generated scent, means for optimizing the scent profile based on the feedback, means for managing fragrances used in a manufacturing facility and automatically generating a scent based on specified conditions, means for analyzing product and environmental information of the manufacturing facility using image processing and natural language processing techniques, and means for recognizing an emotional state using an emotion engine based on the analyzed information. This enables automation of fragrance management and blending in a manufacturing facility, thereby realizing the provision of personalized scents according to the user's emotional state.

[1901] "Visual data" refers to image information such as photos and videos provided by users.

[1902] "Auditory data" refers to acoustic information such as voice or music provided by the user.

[1903] "Features" refer to the information and attributes necessary for scent generation extracted from visual and auditory data.

[1904] "Generative AI model" refers to an artificial intelligence model for generating a scent profile based on extracted features.

[1905] "Scent profile" refers to data that defines the composition and characteristics of the scent generated by a scent generating device.

[1906] "Scent generating device" refers to a device that physically generates a scent based on a received scent profile.

[1907] "Feedback" refers to the user providing their thoughts and opinions about the generated scent.

[1908] "Manufacturing facility" means an industrial facility or location for producing products.

[1909] "Fragrance control" refers to the process of effectively controlling and maintaining the types and amounts of fragrances used within a manufacturing facility.

[1910] "Image processing" refers to the technology of analyzing visual data to identify key objects and scenes.

[1911] "Natural language processing" refers to the technology of extracting emotions and keywords from text and voice data.

[1912] "Emotion engine" refers to a system for recognizing a user's emotions from visual and auditory data.

[1913] The present invention is a system that provides a user with a personalized scent experience based on visual and auditory data, and will be described in a form that can be particularly applied to factory robots.

[1914] Hardware and Software

[1915] At the heart of the system is a robot powered by NVIDIA Jetson Nano. The robot has a built-in device that manages fragrances and automatically generates scents within the manufacturing facility. The robot runs the following software:

[1916] 1. Emotion recognition using TensorFlow.

[1917] 2. Generate scent profiles using PyTorch.

[1918] 3. Perform image processing using OpenCV.

[1919] Data collection

[1920] The user (operator of the manufacturing facility) provides the robot with images and videos of the product, and inputs their emotions and supplementary information (weather, season, etc.) via text and voice. The robot collects this visual and auditory data and sends it to the server.

[1921] Data analysis

[1922] The server uses OpenCV to identify key objects and patterns in the collected visual data, TensorFlow to extract emotions and keywords from the auditory data, and natural language processing to analyze text and audio data and recognize user emotions with an emotion engine.

[1923] Generation of aroma profiles

[1924] The server then uses a generative AI model trained with PyTorch to create a scent profile based on the extracted features and the recognized emotion. For example, if the emotion of "joy" is recognized, a scent profile rich in floral fragrances will be generated.

[1925] Fragrance production

[1926] The scent profile is sent from the server to the robot, which then uses a scent generator to mix the required fragrances from the fragrance cartridges in the specified ratio to create the desired scent, which is then diffused throughout the manufacturing facility using a diffuser.

[1927] Gathering feedback and optimizing your profile

[1928] The user provides feedback on the generated scent through the operator's application. This feedback is sent to the server and stored as evaluation data. The server then uses the feedback to retrain the generative AI model and optimize the scent profile for future uses.

[1929] Examples and prompts

[1930] Specific examples

[1931] During testing of a new product line in a factory, a robot takes photos and videos of the product, and an operator inputs text such as, "This product is very good and gives off a spring-like feeling." Image processing algorithms identify the new product and its context, and extract positive emotions and seasonal information such as "good" and "spring" from the text data. Based on this, a generative AI model generates a scent profile dominated by floral and fresh spring scents.

[1932] Prompt Sentence Examples

[1933] Example prompt for emotion recognition engine on text data:

[1934] Emotion Recognition Pipeline:

[1935] Input: "This product is very well made and gives a spring-like feeling."

[1936] Output: {'label': 'Positive', 'score': 0.95}

[1937] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1938] Step 1:

[1939] The user provides the robot with product images and videos, and inputs their emotions and supplementary information (such as the weather and season) via text and voice. The input data consists of image files, video files, text data, and voice files. The robot collects this data and sends it to a server. Data collection is performed using devices such as cameras and microphones.

[1940] Step 2:

[1941] The server analyzes the collected visual data using OpenCV. The input is image and video data. The server identifies key objects and patterns from the images and extracts this information as features. Specific operations include image preprocessing, edge detection, and object recognition. The output is a list of key objects and patterns.

[1942] Step 3:

[1943] The server analyzes the auditory data using TensorFlow. The inputs are audio files and text data. The server uses natural language processing techniques to extract emotions and keywords from the audio data. Specific operations include transcribing the audio, sentiment analysis, and keyword extraction. The output is a list of emotion labels and keywords.

[1944] Step 4:

[1945] The server generates a scent profile using a generative AI model trained with PyTorch based on the analyzed features and recognized emotions. The inputs are the features and emotion labels. The server inputs this data into the generative AI model to generate a scent profile that defines the scent composition and characteristics. Specific operations include the inference process of the generative model. The output is a scent profile.

[1946] Step 5:

[1947] The scent profile is sent from the server to the robot. The robot operates the scent generation device based on the received scent profile. The input is the scent profile. The robot mixes the required fragrances from the fragrance cartridges in the specified ratio to generate the specified scent. Specific operations include extracting the fragrances, mixing them, and diffusing them using a diffuser. The output is the generated scent.

[1948] Step 6:

[1949] The user inputs feedback about the generated scent through the operator's application. The input is text data of the feedback. The server receives this feedback and stores it as evaluation data. Specific operations include receiving and storing the feedback data.

[1950] Step 7:

[1951] The server retrains the generative AI model based on the feedback and optimizes the scent profile. The input is the feedback data. The server analyzes the feedback and uses it as learning data for the generative AI model. Specific operations include the model retraining process. The output is an optimized scent profile.

[1952] 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.

[1953] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1954] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1955] 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.

[1956] 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 includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1957] 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.

[1958] 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).

[1959] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, 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 indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1960] 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."

[1961] 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.

[1962] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1963] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1964] 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.

[1965] 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.

[1966] 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.

[1967] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1968] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1969] 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.

[1970] 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.

[1971] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1972] 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.

[1973] The following is further disclosed regarding the above embodiment.

[1974] (Claim 1)

[1975] means for receiving visual and auditory data related to the memory from the user;

[1976] means for analyzing the received data and extracting features necessary for generating a fragrance;

[1977] A means for creating a scent profile using a generative AI model based on the feature amount;

[1978] means for transmitting said scent profile;

[1979] means for operating a scent generation device based on the transmitted scent profile to generate a scent;

[1980] means for receiving user feedback regarding the generated scent;

[1981] means for optimizing the scent profile based on the feedback;

[1982] A system including:

[1983] (Claim 2)

[1984] 10. The system of claim 1, further comprising means for receiving supplemental information from a user, such as weather, season, time, location, vegetation, etc.

[1985] (Claim 3)

[1986] The system of claim 1, further comprising means for generating a scent that evokes a particular emotion or memory based on features extracted from the analyzed data.

[1987] "Example 1"

[1988] (Claim 1)

[1989] means for receiving visual and audio data from a user;

[1990] means for analyzing the received data using an image recognition algorithm and natural language processing technology to extract features necessary for scent generation;

[1991] A means for creating a scent profile using a generative AI model based on the feature amount;

[1992] means for transmitting the scent profile to a terminal;

[1993] a means for operating a scent generation device based on the scent profile transmitted in the terminal and generating a scent;

[1994] means for receiving user feedback regarding the generated scent through a dedicated application;

[1995] means for optimizing the scent profile based on the feedback;

[1996] A system including:

[1997] (Claim 2)

[1998] 10. The system of claim 1, further comprising means for receiving supplemental information from a user, such as weather, season, time, location, vegetation, etc.

[1999] (Claim 3)

[2000] The system of claim 1, further comprising means for generating a scent that evokes a specific emotion or memory based on the features and emotions extracted from the analyzed data.

[2001] "Application Example 1"

[2002] (Claim 1)

[2003] means for receiving visual and auditory data related to the memory from the user;

[2004] means for analyzing the received data and extracting features necessary for generating a fragrance;

[2005] A means for creating a scent profile using a generative AI model based on the feature amount;

[2006] means for transmitting said scent profile;

[2007] means for operating a scent generation device based on the transmitted scent profile to generate a scent;

[2008] means for receiving user feedback regarding the generated scent;

[2009] means for optimizing the scent profile based on the feedback;

[2010] A means of collecting customer visual and auditory data in-store and generating personalized scents on the spot;

[2011] A system including:

[2012] (Claim 2)

[2013] 10. The system of claim 1, further comprising means for receiving supplemental information from a user, such as weather, season, time, location, vegetation, etc.

[2014] (Claim 3)

[2015] The system of claim 1, further comprising means for generating a scent that evokes a particular emotion or memory based on features extracted from the analyzed data.

[2016] "Example 2: Combining Emotion Engines"

[2017] (Claim 1)

[2018] means for receiving visual and audio data from a user;

[2019] means for storing the received data;

[2020] means for analyzing the stored data and identifying key objects and scenes using image recognition algorithms;

[2021] A means for extracting emotions and keywords from text and voice data using natural language processing technology;

[2022] A means for recognizing emotions based on features extracted from visual and auditory data;

[2023] a means for generating a scent profile using a generative AI model based on the recognized emotion;

[2024] A means for transmitting the created scent profile to a terminal;

[2025] means for operating a scent generation device based on the received scent profile to generate a scent;

[2026] A means for diffusing the generated scent around the user;

[2027] means for receiving user feedback regarding the generated scent;

[2028] means for optimizing the scent profile based on the received feedback;

[2029] A system including:

[2030] (Claim 2)

[2031] 10. The system of claim 1, further comprising means for receiving supplemental information from a user, such as weather, season, time, location, vegetation, etc.

[2032] (Claim 3)

[2033] The system of claim 1, further comprising means for generating a scent that evokes a particular emotion or memory based on features extracted from the analyzed data.

[2034] "Application example 2 when combining emotion engines"

[2035] (Claim 1)

[2036] means for receiving visual and auditory data related to the memory from the user;

[2037] means for analyzing the received data and extracting features necessary for generating a fragrance;

[2038] A means for creating a scent profile using a generative AI model based on the feature amount;

[2039] means for transmitting said scent profile;

[2040] means for operating a scent generation device based on the transmitted scent profile to generate a scent;

[2041] means for receiving user feedback regarding the generated scent;

[2042] means for optimizing the scent profile based on the feedback;

[2043] a means for managing fragrances used within the manufacturing facility and automatically generating fragrances based on specified conditions;

[2044] means for analyzing product and environmental information of a manufacturing facility using image processing and natural language processing techniques;

[2045] means for recognizing an emotional state using an emotion engine based on the analyzed information;

[2046] A system including:

[2047] (Claim 2)

[2048] 10. The system of claim 1, further comprising means for receiving supplemental information from a user, such as weather, season, time, location, vegetation, etc.

[2049] (Claim 3)

[2050] The system of claim 1, further comprising means for generating a scent that evokes a particular emotion or memory based on features extracted from the analyzed data. [Explanation of symbols]

[2051] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for receiving visual and auditory data related to the memory from the user; means for analyzing the received data and extracting features necessary for generating a fragrance; A means for creating a scent profile using a generative AI model based on the feature amount; means for transmitting said scent profile; means for operating a scent generation device based on the transmitted scent profile to generate a scent; means for receiving user feedback regarding the generated scent; means for optimizing the scent profile based on the feedback; A system including:

2. 10. The system of claim 1, further comprising means for receiving supplemental information from a user, such as weather, season, time, location, vegetation, etc.

3. The system according to claim 1 , further comprising means for generating a scent that evokes a specific emotion or memory based on the feature quantity extracted from the analyzed data.

Images