System

Abstract

A system is provided.SOLUTION: A system comprising: means for acquiring video data; means for acquiring audio data; means for analyzing the video data and the audio data to specify an attribute and a situation of an object; means for generating an optimum advertisement from advertisement generation information prepared in advance based on a result of the analysis; and means for distributing the generated advertisement via a display device.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] Conventional video advertisements not only require time and money to produce, but are also difficult to tailor to viewers' interests and attributes. Furthermore, they are unable to generate and distribute advertisements appropriate for specific locations and times in real time, reducing advertising effectiveness. The objective of this invention is to solve these problems by reducing the time and cost required for advertisement production and distribution, while providing original advertisements that capture viewers' attention and maximize advertising effectiveness. [Means for solving the problem]

[0005] The present invention relates to a system including a means for acquiring video data, a means for acquiring audio data, a means for analyzing the video data and the audio data to identify target attributes and circumstances, a means for generating an optimal advertisement from pre-prepared advertisement generation information based on the analysis results, and a means for delivering the generated advertisement via a display device. The system further includes a means for selecting an optimal advertisement from multiple advertising materials based on the analysis results, and a means for linking the selected advertising materials to generate a customized advertisement. The system also includes a means for applying a pay-per-view model based on the number of impressions and playback time based on the analysis results of the video data and audio data, and a means for calculating fees based on the pay-per-view model. This enables the system to generate and deliver advertisements optimized for the current situation in real time, maximizing advertising effectiveness.

[0006] "Video data" is digital data that includes visual information captured by a photographing device such as a camera.

[0007] "Audio data" is digital data containing auditory information captured by a recording device such as a microphone.

[0008] "Analysis" is the process of extracting specific information from acquired video and audio data and understanding its content.

[0009] "Attributes" refer to the characteristics or features of a target that are identified from the analysis results, such as age or gender.

[0010] The "situation" includes the target's behavior and the environment at the time, which are identified from the analysis results, such as the content of the conversation and the intention of the behavior.

[0011] "Information for advertisement generation" refers to information used to generate advertisements, such as pre-prepared advertisement materials and data, for example, demo videos and product information.

[0012] The "means for generating advertisements" refers to a method or mechanism for creating optimal advertisements based on the analysis results.

[0013] "Display device" refers to a device for physically displaying the generated advertisement, such as a digital signage or a monitor.

[0014] The "pay-as-you-go model" is a billing method in which the fee varies depending on the number of times an advertisement is played and the duration of the playback. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

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

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

[0018] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

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

[0020] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

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

[0022] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0023] [First embodiment]

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

[0025] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0026] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0027] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0028] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0029] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

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

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

[0032] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0033] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

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

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

[0036] The present invention relates to a system that generates and distributes advertisements in real time that are optimized for the current situation based on video and audio data acquired from a camera and a microphone. Specific embodiments for implementing the present invention will be described below.

[0037] Data collection

[0038] Subject: Server

[0039] The server collects real-time video and audio data from the installed cameras and microphones. The video data consists of successive image frames sent from the cameras, and the audio data is digitized sound waves sent from the microphones. The collected data is temporarily stored in a database for analysis.

[0040] Data analysis

[0041] Subject: Server

[0042] The server analyzes the collected video data and identifies people using a facial recognition algorithm. From this identified person, the server infers their attributes (e.g., age and gender). It also uses a voice recognition engine to convert the voice data into text and analyzes the conversation content and situation. This makes it possible to understand the interests and behavior of the target.

[0043] Ad Generation

[0044] Subject: Server

[0045] The server generates optimal advertisements based on the information obtained through the analysis and compares it with pre-prepared advertisement generation information. Specifically, it selects advertising materials based on the analysis results and creates customized advertising videos using a generative AI model. This advertisement generation process makes it possible to create personalized advertisements for each user in a short amount of time.

[0046] Ad serving

[0047] Subject: Device

[0048] The terminal (digital signage) receives the advertising video generated from the server and displays it instantly. The terminal then plays this advertising video on the screen for passersby and users to view. This allows for the delivery of advertising that is appropriate for the situation in real time, which can be expected to be more effective.

[0049] Pay-as-you-go model

[0050] Subject: Server

[0051] The server collects ad playback data, recording the number of plays and the playback time. Based on this data, a pay-per-view model is applied to calculate the fees for advertisers, enabling them to deliver ads in a cost-effective manner.

[0052] Specific examples

[0053] Example 1: Supermarket

[0054] Cameras and microphones installed at the entrance of supermarkets capture video and audio of families entering the store in real time. The server identifies parents and children from the video and analyzes the audio for conversations about "children's toys." Based on this, the server generates promotional advertisements for children's toys and instantly distributes them to digital signage in the store.

[0055] Example 2: Inside a station

[0056] Cameras and microphones installed inside the station capture images of passersby and audio of them talking about "travel." The server analyzes the data, identifies age groups and genders, and generates advertisements for travel agencies. These advertisements are displayed on digital signage, promoting travel products to passersby.

[0057] As explained above, the present invention makes it possible to generate and distribute advertisements that are optimal for each situation in real time, thereby maximizing the effectiveness of advertisements.

[0058] The processing flow will be explained below.

[0059] Step 1:

[0060] Subject: Server

[0061] The server starts streaming video and audio data from the installed cameras and microphones. The video data is captured frame by frame, and the audio data is collected in real time. These data are temporarily stored in a database for analysis.

[0062] Step 2:

[0063] Subject: Server

[0064] The server analyzes the video data and identifies people using a facial recognition algorithm. It uses a face detection library such as OpenCV to detect faces in the frame and estimate attributes such as age and gender for the detected faces.

[0065] Step 3:

[0066] Subject: Server

[0067] The server sends the voice data to a speech recognition engine, which converts it into text in real time. It then transcribes the conversation using speech recognition services such as Google Cloud Speech-to-Text API or Watson Speech to Text, and uses text analysis to identify interests and behaviors.

[0068] Step 4:

[0069] Subject: Server

[0070] The server combines the results of facial and voice recognition analysis to identify target attributes and context, which serves as the basis for generating customized advertisements that reflect the user's interests and current behavior.

[0071] Step 5:

[0072] Subject: Server

[0073] The server compares the analysis results with pre-prepared information for generating advertisements and selects the most suitable advertising materials. For example, if the customer is a family, it will select advertising materials for products aimed at children. This selected material is then input into the generative AI model.

[0074] Step 6:

[0075] Subject: Server

[0076] The generative AI model generates customized ad videos based on the selected ad materials and analysis results, and uses facial recognition results to quickly create personalized ads relevant to users.

[0077] Step 7:

[0078] Subject: Device

[0079] The terminal (digital signage) receives the generated advertising video from the server and immediately displays it. The video file is downloaded using a file transfer protocol and played on the display via a video player application.

[0080] Step 8:

[0081] Subject: Server

[0082] The server collects ad playback data, records the number of plays and the playback time, and periodically transmits ad playback event data from the digital signage terminal.

[0083] Step 9:

[0084] Subject: Server

[0085] The server calculates the fee for the advertiser based on the recorded playback data using a pay-per-view model, refers to a fee schedule based on the number of playbacks and duration, calculates the total amount, and provides a fee report to the advertiser.

[0086] The above steps enable the generation and distribution of optimal advertisements in real time according to the situation, maximizing the effectiveness of advertising.

[0087] Example 1

[0088] 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."

[0089] Conventional ad delivery systems have had difficulty generating and delivering optimal ads in real time based on target attributes and circumstances. In particular, they have been unable to effectively utilize video and audio data to instantly provide personalized ads based on analysis results. Furthermore, they lacked an efficient method for accurately calculating fees when applying a pay-per-use model to advertisers. As a result, maximizing advertising effectiveness and improving cost-effectiveness have not been fully achieved.

[0090] 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.

[0091] In this invention, the server includes a means for acquiring video data, a means for acquiring audio data, and an analysis means, which allows the server to generate and distribute optimal advertisements in real time according to the attributes and circumstances of the target audience, thereby maximizing the effectiveness of the advertisements.

[0092] "Video data" refers to digital data consisting of successive image frames captured in real time from an image capture device such as a camera.

[0093] "Audio data" refers to data obtained by digitizing sound waves acquired from a sound acquisition device such as a microphone.

[0094] The "analysis results" are the results of calculations performed using the acquired video and audio data to identify the attributes and circumstances of the target.

[0095] "Information for advertisement generation" refers to materials and data prepared in advance that will be the basis for the advertisement to be generated.

[0096] A "generative AI model" is an artificial intelligence model that uses machine learning algorithms to generate content based on input prompts.

[0097] A "prompt" is an instruction or question input to a generative AI model that provides the requirements for generating a specific output.

[0098] A "display device" is a screen or display for visually displaying the generated advertisement.

[0099] The "pay-as-you-go model" is a billing method in which fees are calculated based on the number of times an advertisement is displayed and the duration of its playback.

[0100] The present invention is a system that generates and distributes advertisements in real time that are optimized for the current situation based on video and audio data acquired from a camera and microphone. Specific embodiments for implementing the present invention will be described below.

[0101] Data collection

[0102] server

[0103] The server acquires video and audio data in real time from the installed cameras and microphones. This video data consists of successive image frames sent from the cameras, and the audio data is digitized sound waves sent from the microphones. Specifically, the server acquires the video stream from the cameras using RTSP (Real-Time Streaming Protocol), and the audio data is received via the Internet. This data is temporarily stored in a database for analysis.

[0104] Data analysis

[0105] server

[0106] The server analyzes the collected video and audio data. Using a facial recognition algorithm, it detects the faces of people in the video frames and infers their attributes (age, gender). Specifically, it uses OpenCV (an open-source computer vision library). It also uses the Google Cloud Speech-to-Text API as a speech recognition engine to convert the audio data into text. The text data can then be analyzed to understand the content and context of the conversation.

[0107] Ad Generation

[0108] server

[0109] Based on the results of the data analysis, the server uses a generative AI model (e.g., GPT-4) to generate the optimal advertisement from pre-prepared information for advertisement generation. The server inputs the prompt text into the generative AI model, which generates the advertisement script as a result. The generated script is then used in video editing software such as Adobe Premiere Pro to edit and complete the advertisement video. An example of a specific prompt text is, "Based on the camera footage and microphone audio data, generate an advertisement that is optimal for the current situation. As a specific scenario, please create an advertisement for a family entering a supermarket and talking about children's toys."

[0110] Ad serving

[0111] Terminal

[0112] Devices such as digital signage receive advertising videos generated from the server and instantly display them on the screen. The device uses a basic HTTP request to download the latest advertising video from the server and then plays the video file for passersby or users to view, enabling real-time advertising.

[0113] Pay-as-you-go model

[0114] server

[0115] The server collects ad playback data, recording the number of plays and the playback time. It receives access logs and playback logs sent from the device and uses this data to apply a pay-per-view model to calculate fees for advertisers. This allows advertisers to deliver ads in a cost-effective manner.

[0116] Specific examples

[0117] Example 1: Supermarket

[0118] Cameras and microphones installed at the entrance of the supermarket capture video and audio of families entering the store in real time. The server identifies parents and children from the video and analyzes the conversation about "children's toys" from the audio. Based on the analysis results, prompts are input into a generative AI model to generate promotional advertisements for children's toys. The completed advertising video is distributed to and displayed on digital signage terminals in the store.

[0119] Example 2: Inside a station

[0120] Cameras and microphones installed inside the station capture images of passersby and audio of them talking about "travel." The server analyzes this data, identifies age groups and genders, and then generates advertisements for travel agencies. These advertisements are displayed on digital signage, promoting travel products to passersby.

[0121] This makes it possible to generate and deliver optimal advertisements in real time according to the situation, maximizing the effectiveness of advertising.

[0122] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0123] Step 1:

[0124] Data collection

[0125] The server acquires video and audio data from the camera and microphone in real time. Specifically, the server receives a video stream of continuous frames from the camera using RTSP (Real-Time Streaming Protocol). It also receives audio data, which is digitized sound waves, from the microphone via the Internet. This data is temporarily stored in a database for analysis.

[0126] Input: Video stream from camera, audio data from microphone

[0127] Output: Video and audio data stored in a database for analysis

[0128] Step 2:

[0129] Data analysis

[0130] The server runs a facial recognition algorithm (e.g., OpenCV) to detect faces from the collected video data. This algorithm detects faces in the video frames and infers their attributes (age, gender). Additionally, the audio data is converted to text using the Google Cloud Speech-to-Text API. This allows the audio to be analyzed for conversation content and context, providing information to understand the subject's interests and behavior.

[0131] Input: Video data, audio data

[0132] Output: Attributes of identified person, audio data converted to text

[0133] Step 3:

[0134] Ad Generation

[0135] The server uses a generative AI model (e.g., GPT-4) to generate optimal advertisements based on the results of data analysis. The server inputs prompt sentences into the generative AI model and obtains the resulting advertisement script. It then uses video editing software such as Adobe Premiere Pro to edit and generate advertisement videos based on the generated script.

[0136] Input: Identified person's attributes, transcribed voice data, prompt

[0137] Output: Ad script, finished ad video

[0138] Step 4:

[0139] Ad serving

[0140] The terminal (digital signage) receives the generated advertising video from the server and plays it on the screen for immediate display. The terminal uses a basic HTTP request to download the latest advertising video from the server and plays the received video file.

[0141] Input: Ad video

[0142] Output: Ad video displayed on the screen

[0143] Step 5:

[0144] Pay-as-you-go model

[0145] The server collects data each time an advertisement is played, recording the number of plays and the play time. Specifically, it receives access logs and play logs sent from the terminal and applies a pay-per-view model based on that data. This is used to calculate the fee for the advertiser.

[0146] Input: Access log and playback log from the device

[0147] Output: Pricing results based on number of plays and duration

[0148] (Application example 1)

[0149] 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."

[0150] Modern commercial facilities and brick-and-mortar stores require real-time advertising that meets customer needs and interests. However, existing systems have difficulty accurately analyzing customer attributes and conversation content, and quickly generating and delivering individually customized advertisements. Another issue is that they are unable to meet the need for direct advertising display using smart devices.

[0151] 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.

[0152] In this invention, the server includes means for acquiring video data, means for acquiring audio data, means for analyzing the video data and the audio data and identifying target attributes and situations, means for generating an optimal advertisement from pre-prepared advertisement generation information based on the analysis results, means for delivering the generated advertisement via a display device, and means for displaying the advertisement on a smart device in real time. This makes it possible to quickly deliver an optimized advertisement to a smart device based on the real-time attributes and conversation content of a customer.

[0153] "Video data" refers to data made up of successive image frames captured using a photographic device such as a camera.

[0154] "Audio data" refers to data obtained by digitizing sound waves acquired using an acoustic device such as a microphone.

[0155] The "analyzing means" is a means for executing a process to determine a person's attributes and situation based on the acquired video data and audio data.

[0156] "Information for advertisement generation" is a collection of materials and information prepared in advance and used to generate optimal advertisements based on the analysis results.

[0157] A "display device" is a hardware device for providing the generated advertisement to the viewer, and includes digital signage, smartphones, etc.

[0158] "Smart devices" is a general term for mobile terminals and wearable devices that are connected to the Internet and have the ability to run applications.

[0159] A "generative AI model" is an algorithm or machine learning model that uses artificial intelligence to generate optimal advertisements from input data.

[0160] A "prompt sentence" is an input sentence used to request the generative AI model to generate an ad.

[0161] "Attributes" are information that indicates the age, sex, interests, behavior, etc. of the person being analyzed.

[0162] "Situation" is information that indicates the environment in which the analysis target is placed and the situation at that time.

[0163] The present invention relates to a system that generates and distributes advertisements in real time that are optimized for the current situation based on video and audio data acquired from a camera and a microphone. Specific embodiments for implementing the present invention will be described below.

[0164] Data collection

[0165] The server collects video and audio data in real time from cameras and microphones installed in the store. The video data consists of successive image frames sent from the camera, and the audio data is digitized sound waves sent from the microphone. The collected data is temporarily stored in a database for analysis. The hardware used is a standard camera (e.g., a 1080p webcam) and a high-sensitivity microphone (e.g., a USB-connected condenser microphone).

[0166] Data analysis

[0167] The server analyzes the collected video data and identifies people using a facial recognition algorithm. It then infers their attributes (e.g., age and gender) from the identified people. It also converts the audio data into text using a speech recognition engine and analyzes the content and context of the conversation. The software used includes OpenCV (for video analysis) and the SpeechRecognition library (for audio analysis).

[0168] Ad Generation

[0169] Based on the information obtained through the analysis, the server compares it with pre-prepared ad generation information to generate the optimal ad. Specifically, it selects advertising materials based on the analysis results and creates a customized ad video using a generative AI model. The generative AI model uses OpenAI's API. This ad generation process makes it possible to quickly create personalized ads for each user. An example of a prompt sentence when generating a generated ad is as follows:

[0170] Generate a personalized advertisement for a 25-year-old female interested in new fashion trends.

[0171] Ad serving

[0172] The terminal (smart device) receives the advertising video generated from the server and immediately displays it. The terminal then plays the advertising video on a screen for passersby and users to view. Specific smart devices include smartphones and head-mounted displays.

[0173] Pay-as-you-go model

[0174] The server collects ad playback data, recording the number of plays and playback time. Based on this data, a pay-per-view model is applied to calculate fees for advertisers. This allows advertisers to deliver ads in a cost-effective manner. In addition, user interactions with ads displayed on smart devices can be analyzed, which can be used to further analyze the effectiveness of advertising.

[0175] Specific examples

[0176] Example 1: Supermarket

[0177] Cameras and microphones installed at the entrance of supermarkets capture video and audio recordings of families entering the store in real time. The server identifies parents and children from the video and analyzes audio conversations about "children's toys." Based on this, the server generates promotional advertisements for children's toys and instantly distributes them to in-store digital signage. The same advertisements are also displayed in real time on smartphone apps.

[0178] Example 2: Inside a station

[0179] Cameras and microphones installed inside the station capture images of passersby and audio of them talking about "travel." The server analyzes the data, identifies age groups and genders, and generates advertisements for travel agencies. These advertisements are displayed on digital signage, promoting travel products to passersby, while also being displayed in real time on head-mounted displays.

[0180] As explained above, the present invention makes it possible to generate and distribute advertisements that are optimal for each situation in real time, thereby maximizing the effectiveness of advertisements.

[0181] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0182] Step 1:

[0183] The server acquires video data from cameras installed inside the store. The server acquires successive image frames and temporarily stores them in a database for analysis. At this time, the data input from the camera is acquired as a series of still image data, which is output as video data.

[0184] Step 2:

[0185] The server simultaneously captures audio data from a microphone. The sound waves coming from the microphone are digitized and stored as an audio file in a database on the server. This audio data is used in subsequent analysis steps.

[0186] Step 3:

[0187] The server applies a facial recognition algorithm to the captured video data. Specifically, it uses the OpenCV library to perform face detection for each frame, identify specific people, and infer their attributes. The input is the video data, and the output is the attribute information of the identified people (e.g., age, gender).

[0188] Step 4:

[0189] The server converts the voice data into text using the SpeechRecognition library. The voice data is input into a speech recognition engine, which then outputs analyzed text data. Based on this text data, the content of the conversation and the situation are analyzed to understand the subject's interests and behavior.

[0190] Step 5:

[0191] The server compares the results of video and audio analysis to identify the overall situation and the target's interests and behavior. This determines what type of advertising will be most effective. The input is the results of video and audio analysis, and the output is a situational judgment based on the analysis results.

[0192] Step 6:

[0193] Based on the analysis results, the server generates the optimal advertisement from pre-prepared information for advertisement generation. Here, a generative AI model is used to generate a customized advertisement by providing a prompt. For example, an advertisement is generated based on the prompt "Generate a personalized advertisement for a 25-year-old female interested in new fashion trends." The input is the prompt and information for advertisement generation, and the output is the generated advertisement content.

[0194] Step 7:

[0195] The server sends the generated advertisement to the terminal (e.g., smart device). The terminal immediately displays the advertisement content received from the server. The input is the generated advertisement content, and the output is the advertisement displayed on the terminal.

[0196] Step 8:

[0197] When a user interacts with an advertisement on their device, the device sends this interaction data to a server. The server analyzes this data and evaluates the effectiveness of the advertisement and the user's response. The input is the interaction data, and the output is the analyzed evaluation results.

[0198] Step 9:

[0199] The server finally applies a pay-per-view model based on the ad playback data and interaction data to calculate the fee for the advertiser. The input is the playback data and interaction data, and the output is the fee calculation result.

[0200] 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.

[0201] The present invention relates to a system that generates and distributes advertisements in real time that are optimized for the situation and user emotions based on video and audio data acquired from a camera and a microphone. Specific embodiments for carrying out the present invention will be described below.

[0202] Data collection

[0203] Subject: Server

[0204] The server captures video and audio data in real time from the installed cameras and microphones. These data are temporarily stored in a database for analysis. Video data consists of a series of image frames, and audio data is digitized sound waves.

[0205] Data analysis

[0206] Subject: Server

[0207] The server analyzes the collected video data and identifies people using a facial recognition algorithm. It then infers their attributes (e.g., age and gender) from the detected people. It also sends the audio data to a speech recognition engine, which converts it into text. This allows the content and context of the conversation to be analyzed.

[0208] emotion recognition

[0209] Subject: Server

[0210] The server recognizes the user's emotions by analyzing facial expressions from video data and tone of voice from audio data. The emotion recognition engine identifies emotions such as anger, joy, and surprise, and uses the results for further analysis.

[0211] Ad Generation

[0212] Subject: Server

[0213] The server generates the optimal advertisement based on the results of attribute and situation analysis and emotion recognition, comparing them with pre-prepared advertisement generation information. Specifically, it selects advertising materials based on the analysis results and emotion recognition results, and creates a customized advertising video using a generative AI model. In this process, the content that best suits the user's emotional state is reflected.

[0214] Ad serving

[0215] Subject: Device

[0216] The terminal (digital signage) receives the advertising video generated from the server and immediately displays it. The terminal then plays the advertising video on the display screen for the user to view. This allows the optimal advertising to be delivered in real time according to the situation.

[0217] Pay-as-you-go model

[0218] Subject: Server

[0219] The server collects ad playback data, recording the number of plays and the playback time. Based on this data, a pay-per-view model is applied to calculate the fees for advertisers, enabling them to deliver ads in a cost-effective manner.

[0220] Specific examples

[0221] Example 1: Supermarket

[0222] Cameras and microphones installed at the entrance of the supermarket capture video and audio of families entering the store in real time. The server identifies parents and children from the video and analyzes the audio content of conversations about "toys." At the same time, an emotion recognition engine detects whether the child is excited. Based on this, the server generates a promotional advertisement for children's toys and instantly distributes it to digital signage within the store.

[0223] Example 2: Inside a station

[0224] Cameras and microphones installed inside the station capture images of passersby and audio of them talking about "travel." The server analyzes the data, identifies their age group and gender, and uses an emotion recognition engine to assess their excitement level. Based on this, the server generates advertisements for travel agencies and displays them on digital signage.

[0225] As explained above, the present invention makes it possible to generate and distribute optimal advertisements in real time according to the situation and user emotions, thereby maximizing the effectiveness of advertisements.

[0226] The processing flow will be explained below.

[0227] Step 1:

[0228] Subject: Server

[0229] The camera and microphone begin streaming video and audio data. Video data is collected frame by frame, and audio data is digitized and captured in real time. This data is temporarily stored in a database for analysis.

[0230] Step 2:

[0231] Subject: Server

[0232] The server analyzes the collected video data and uses facial recognition algorithms to identify people. Using a face detection library such as OpenCV, it detects faces in the video frames and estimates attributes such as age and gender from the faces.

[0233] Step 3:

[0234] Subject: Server

[0235] Voice data is sent to a speech recognition engine and converted to text in real time. Speech recognition services such as Google Cloud Speech-to-Text API or Watson Speech to Text are used to transcribe the conversation and perform text analysis to identify interests and behaviors.

[0236] Step 4:

[0237] Subject: Server

[0238] The results of the analysis of video and audio data are integrated to identify the target's attributes and situation, thereby clarifying the user's basic profile and current behavior and interests.

[0239] Step 5:

[0240] Subject: Server

[0241] Based on video and audio data, the system analyzes the user's facial expressions and tone of voice to recognize their emotions. Using an emotion recognition engine, it identifies emotions such as anger, joy, and surprise, and uses the results in the next ad generation process.

[0242] Step 6:

[0243] Subject: Server

[0244] Based on the analysis results and emotion recognition results, the optimal advertising materials are selected by comparing them with pre-prepared information for advertising generation.Then, the selected materials and analysis results are input into the generation AI model to generate a customized advertising video that matches the emotions.

[0245] Step 7:

[0246] Subject: Device

[0247] The digital signage terminal receives the generated advertising video from the server, and immediately plays the downloaded advertising video using a file transfer protocol and displays it on the terminal screen.

[0248] Step 8:

[0249] Subject: Server

[0250] The device collects data on the ads played, recording the number of plays and the play time, and periodically sends this playback data to the server.

[0251] Step 9:

[0252] Subject: Server

[0253] The server applies a pay-per-view model based on the collected playback data to calculate the fees for advertisers, calculates the total amount based on a fee schedule according to the number of plays and the playback time, and provides a fee report to the advertiser.

[0254] The above steps enable the creation and distribution of optimal advertisements in real time according to the situation and user emotions, maximizing the effectiveness of the advertisements.

[0255] Example 2

[0256] 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."

[0257] In modern advertising delivery systems, it is difficult to generate and deliver optimal advertisements in real time that match user attributes and emotions. Maximizing the cost-effectiveness of advertising by introducing an efficient pay-per-use model for advertisers is also a challenge. Conventional systems are unable to perform such real-time data analysis and advertisement generation, making it impossible to deliver effective advertisements that match user interests and emotions.

[0258] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for acquiring video data, means for acquiring audio data, means for analyzing the video data and the audio data to identify person attributes and situations, means for recognizing a user's emotions from the video data and the audio data, means for generating an optimal advertisement from pre-prepared advertisement generation information based on the analysis results and the user's emotion recognition results, and means for delivering the generated advertisement via a display device. This enables real-time advertisement generation and delivery based on user attributes and emotions, maximizing the effectiveness of advertisements. In addition, by applying a pay-per-view model based on the number of impressions and playback time, advertisers can be effectively charged advertising fees.

[0259] "Video data" is a digital representation of visual information captured using a device such as a camera.

[0260] "Audio data" refers to digitized sound waves recorded using a device such as a microphone.

[0261] "Person attributes" are characteristics about an individual, such as age, gender, or identity, that are inferred using facial recognition algorithms or other methods.

[0262] "User emotion" refers to an emotional state, such as anger, joy, or surprise, that is recognized through facial expression analysis from video data and tone analysis from audio data.

[0263] "Information for advertisement generation" refers to materials such as text, images, and videos prepared in advance for generating advertisements, as well as related data.

[0264] A "generative AI model" is an artificial intelligence model that generates customized output data based on input data and specified prompts.

[0265] A "prompt sentence" is a textual input used to give specific instructions to a generative AI model.

[0266] A "pay-as-you-go model" is a pricing model in which the fee is calculated based on the number of times an advertisement is displayed and the duration of its playback.

[0267] A "display device" refers to a display or digital signage used to allow users to view the generated advertisement.

[0268] "Playback data" refers to data relating to playback status, such as the number of times a generated advertisement is played and the playback time.

[0269] This invention relates to a system that generates and distributes advertisements in real time that are optimized for the situation and user emotions based on video and audio data acquired from a camera and microphone.

[0270] The server collects video and audio data in real time from cameras and microphones installed via the network. The video data consists of successive image frames, and the audio data is digitized sound waves. These data are temporarily stored in a database for analysis. Specifically, the camera collects video at 30 frames per second, and the microphone collects audio data at 44.1 kHz per second.

[0271] The server analyzes the collected video data and identifies people using a facial recognition algorithm. For example, it uses image processing software (e.g., OpenCV). Attributes such as age and gender are inferred by analyzing the feature points of the detected face. At the same time, the audio data is sent to a speech recognition engine (e.g., Google Speech-to-Text API) and converted into text. This allows the content and situation of the conversation to be extracted. For example, the keyword "travel" can be extracted from the audio data.

[0272] The server then analyzes the user's facial expressions from the video data and the tone of voice from the audio data to recognize the user's emotions. It uses an emotion recognition engine (e.g., Amazon Rekognition or Google Cloud Natural Language API) to determine emotional states such as anger, joy, and surprise. For example, it can detect whether the user is smiling and recognize excitement from the tone of their voice.

[0273] Next, the server generates the optimal advertisement based on the analysis results and emotion recognition results, comparing them with pre-prepared advertisement generation information. It selects advertising materials (text, images, videos) and uses a generative AI model (e.g., GPT-4) to create a customized advertising video. During this process, it issues instructions by sending prompt text to the generative AI model. For example, a prompt text such as "An excited man in his 60s who is interested in traveling" could be sent to the generative AI model to generate a customized advertising video for a travel agency.

[0274] The generated advertising video is sent to a terminal (digital signage) and played on the display, allowing users to view it. For example, a travel advertising video can be played on a display in a store to attract the attention of passersby.

[0275] Finally, the server collects ad play data, recording the number of plays and play time. Based on this data, it applies a pay-per-view model and calculates the fee to the advertiser. For example, it records that a particular ad was played 100 times and the total play time was 50 minutes, and calculates the cost to be charged to the advertiser.

[0276] As a result, the system of the present invention enables real-time generation and distribution of advertisements based on user attributes and emotions, maximizing the effectiveness of advertisements. In addition, by applying a pay-per-view model based on the number of impressions and playback time, advertisers can be effectively charged advertising fees.

[0277] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0278] Step 1:

[0279] Data collection:

[0280] The server receives real-time video and audio data from cameras and microphones via the network. Specifically, the cameras capture video data at 30 frames per second, and the microphones capture audio data at 44.1 kHz per second. This data is temporarily stored in a database for analysis.

[0281] Input: Video and audio data obtained from the camera and microphone.

[0282] Output: Video and audio data stored in a database for analysis.

[0283] Step 2:

[0284] Data Analysis:

[0285] The server analyzes the collected video data. Specifically, it uses image processing software (e.g., OpenCV) to run a facial recognition algorithm to identify people. Attributes such as age and gender are inferred by analyzing facial feature points. At the same time, the server sends the audio data to a speech recognition engine (e.g., Google Speech-to-Text API) to convert it into text.

[0286] Input: Video and audio data obtained from the analysis database.

[0287] Output: Identified person's attributes (age, gender) and transcribed audio data.

[0288] Step 3:

[0289] Emotion recognition:

[0290] The server analyzes the user's facial expressions from the video data and the tone of voice from the audio data to recognize the user's emotions. An emotion recognition engine (e.g., Amazon Rekognition or Google Cloud Natural Language API) is used to determine the user's emotional state, such as anger, joy, or surprise.

[0291] Input: Identified person attributes and transcribed audio data.

[0292] Output: The user's emotional state (e.g., anger, joy, surprise).

[0293] Step 4:

[0294] Ad Generation:

[0295] The server generates the optimal ad based on the analysis results and emotion recognition results, comparing them with pre-prepared ad generation information. It selects ad materials (text, images, videos) and creates a customized ad video using a generative AI model (e.g., GPT-4). During this process, it sends a prompt to the generative AI model.

[0296] Input: User's emotional state, identified person attributes, and information for ad generation.

[0297] Output: A customized advertising video.

[0298] Step 5:

[0299] Ad serving:

[0300] The terminal (digital signage) receives the advertising video generated from the server and immediately displays it. The terminal plays the advertising video on the display so that the user can view it.

[0301] Input: The customized ad video sent from the server.

[0302] Output: Ad video played on the display.

[0303] Step 6:

[0304] Pay-as-you-go:

[0305] The server collects ad playback data, records the number of plays and the playback time, and calculates the fee to the advertiser based on this data using a pay-per-view model.

[0306] Input: Playback data of the ad video played on the display.

[0307] Output: Fees charged to advertisers.

[0308] (Application example 2)

[0309] 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."

[0310] Conventional ad delivery systems generate and deliver ads based on user attributes and circumstances, but they do not fully consider the user's real-time emotions or specific circumstances, which means that the effectiveness of the ads is not maximized.In addition, there is a lack of technology for generating and delivering effective ads, and the cost-effectiveness for advertisers is also insufficient.

[0311] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0312] In this invention, the server includes means for acquiring video data, means for acquiring audio data, means for analyzing the video data and the audio data to identify target attributes and situations, means for analyzing user emotions, means for generating an optimal advertisement from pre-prepared information for advertisement generation based on the results of the analysis and the emotion analysis, and means for delivering the generated advertisement via a display device. This makes it possible to generate and deliver an optimal advertisement according to the user's real-time emotions and specific situations, thereby maximizing advertising effectiveness.

[0313] "Video data" refers to visual information acquired by an image sensor such as a camera.

[0314] "Audio data" refers to auditory information acquired by an acoustic sensor such as a microphone.

[0315] "Analysis" refers to computational processing to extract or infer specific information from acquired data.

[0316] "Attributes" refer to the characteristics or properties of a subject that are determined through analysis, and include, for example, age and gender.

[0317] "Situation" refers to the subject's current environmental and behavioral state as determined by analysis.

[0318] "Emotion" refers to the subject's psychological state as determined by analysis, and includes anger, joy, surprise, etc.

[0319] "Information for advertisement generation" refers to materials and data prepared in advance for generating an advertisement.

[0320] The term "display device" refers to a device for visually presenting the generated advertisement to a user.

[0321] "Advertising materials" refers to the individual elements and content that make up an advertisement.

[0322] "Customized" refers to advertising that is generated and tailored to specific requirements and conditions.

[0323] A "pay-as-you-go model" refers to a pricing system in which charges are based on the amount of usage or the number of times it is used.

[0324] A system for implementing the present invention has the following configuration: A server acquires and analyzes video data and audio data, and generates and distributes optimal advertisements. A detailed explanation is provided below with specific examples.

[0325] First, the server acquires video and audio data in real time from cameras and microphones. Video data is visual information acquired by image sensors such as cameras, and audio data is auditory information acquired by acoustic sensors such as microphones. These data are temporarily stored in a database for analysis.

[0326] The server then analyzes the collected video and audio data. The video data uses facial recognition algorithms to identify people and infer their attributes (e.g., age, gender, etc.), while the audio data is sent to a speech recognition engine, which converts it into text and analyzes it. This analysis process uses software libraries such as OpenCV and SpeechRecognition.

[0327] Furthermore, the server uses an emotion recognition engine to analyze the user's emotions. By analyzing the user's facial expressions from the video data and the tone of voice from the audio data, emotions such as anger, joy, and surprise can be recognized. This emotion analysis combines multiple algorithms.

[0328] The server generates optimal advertisements based on the results of these analyses and sentiment analysis. Specifically, it selects the optimal advertising materials for the user's attributes and emotional state from pre-prepared advertisement generation information, and creates customized advertisements using a generative AI model.

[0329] Finally, the generated advertisement is delivered to a display device, such as a smartphone or digital signage. The user can watch the advertisement in real time, optimized for the situation and emotions of the moment. The terminal device plays the advertisement video and provides it visually to the user.

[0330] For example, when a user is using a smartphone, the smartphone's camera and microphone collect data and detect that the user is having a pleasant conversation. The server generates an advertisement for an entertainment app that matches that pleasant emotion and displays it on the smartphone. This application maximizes the effectiveness of advertising.

[0331] Example prompt sentence:

[0332] "The user is showing happy emotions. Please generate an ad for an entertainment app. User demographics: Age 25, Gender Male"

[0333] As described above, the system of the present invention is capable of generating and delivering optimal advertisements that correspond to the user's real-time emotions and specific situations, thereby maximizing the effectiveness of advertisements and improving cost-effectiveness for advertisers.

[0334] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0335] Step 1:

[0336] The server acquires video and audio data from the camera and microphone in real time. As input, it receives the real-time output of the camera and microphone and temporarily stores it as video and audio data in a database within the server. In this process, the camera captures successive image frames and the microphone generates a digital audio signal.

[0337] Step 2:

[0338] The server analyzes the acquired video data. As input, it receives the video data acquired in step 1, identifies the person using a facial recognition algorithm, and then estimates attributes such as age and gender. For example, it uses OpenCV to detect faces from image frames and estimates attributes based on the results. As output, it obtains attribute information for the identified person.

[0339] Step 3:

[0340] The server analyzes the acquired voice data. As input, it receives the voice data acquired in step 1 and converts it into text using a voice recognition engine. For example, it uses the SpeechRecognition library to convert the voice data into text, and analyzes the topic and situation of the conversation from the text content. As output, it obtains text data and the analysis results.

[0341] Step 4:

[0342] The server analyzes the user's emotions from the video and audio data. It receives the outputs of steps 2 and 3 as input and uses an emotion recognition engine to analyze facial expressions and tone of voice. Specifically, it reads facial expressions from image frames and identifies emotions from audio tones. The output is emotional information such as anger, joy, and surprise.

[0343] Step 5:

[0344] The server generates the optimal advertisement based on the analysis results and the results of sentiment analysis. It receives the outputs of steps 2, 3, and 4 as input and selects the optimal advertising material from pre-prepared information for advertisement generation. It uses a generative AI model to generate a customized advertisement that best suits the user's attributes and emotions. The output is a customized advertising video.

[0345] Step 6:

[0346] The server delivers the generated advertisement to the terminal. It receives the customized advertisement video output from step 5 as input and sends it to the terminal. The terminal plays the received advertisement video on a display device (e.g., a smartphone or digital signage). This process allows the user to watch the optimized advertisement in real time.

[0347] Step 7:

[0348] The server collects playback data of the displayed advertisements and calculates fees based on a pay-per-view model. As input, it receives data on the number of plays and the playback time sent from the terminal, and calculates the fee for the advertiser based on that data. As output, it obtains the final fee calculation result.

[0349] 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.

[0350] 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.

[0351] 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.

[0352] [Second embodiment]

[0353] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0354] 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.

[0355] 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).

[0356] 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.

[0357] 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.

[0358] 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).

[0359] 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.

[0360] 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.

[0361] 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.

[0362] 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.

[0363] 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.

[0364] 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."

[0365] The present invention relates to a system that generates and distributes advertisements in real time that are optimized for the current situation based on video and audio data acquired from a camera and a microphone. Specific embodiments for implementing the present invention will be described below.

[0366] Data collection

[0367] Subject: Server

[0368] The server collects real-time video and audio data from the installed cameras and microphones. The video data consists of successive image frames sent from the cameras, and the audio data is digitized sound waves sent from the microphones. The collected data is temporarily stored in a database for analysis.

[0369] Data analysis

[0370] Subject: Server

[0371] The server analyzes the collected video data and identifies people using a facial recognition algorithm. From this identified person, the server infers their attributes (e.g., age and gender). It also uses a voice recognition engine to convert the voice data into text and analyzes the conversation content and situation. This makes it possible to understand the interests and behavior of the target.

[0372] Ad Generation

[0373] Subject: Server

[0374] The server generates optimal advertisements based on the information obtained through the analysis and compares it with pre-prepared advertisement generation information. Specifically, it selects advertising materials based on the analysis results and creates customized advertising videos using a generative AI model. This advertisement generation process makes it possible to create personalized advertisements for each user in a short amount of time.

[0375] Ad serving

[0376] Subject: Device

[0377] The terminal (digital signage) receives the advertising video generated from the server and displays it instantly. The terminal then plays this advertising video on the screen for passersby and users to view. This allows for the delivery of advertising that is appropriate for the situation in real time, which can be expected to be more effective.

[0378] Pay-as-you-go model

[0379] Subject: Server

[0380] The server collects ad playback data, recording the number of plays and the playback time. Based on this data, a pay-per-view model is applied to calculate the fees for advertisers, enabling them to deliver ads in a cost-effective manner.

[0381] Specific examples

[0382] Example 1: Supermarket

[0383] Cameras and microphones installed at the entrance of supermarkets capture video and audio of families entering the store in real time. The server identifies parents and children from the video and analyzes the audio for conversations about "children's toys." Based on this, the server generates promotional advertisements for children's toys and instantly distributes them to digital signage in the store.

[0384] Example 2: Inside a station

[0385] Cameras and microphones installed inside the station capture images of passersby and audio of them talking about "travel." The server analyzes the data, identifies age groups and genders, and generates advertisements for travel agencies. These advertisements are displayed on digital signage, promoting travel products to passersby.

[0386] As explained above, the present invention makes it possible to generate and distribute advertisements that are optimal for each situation in real time, thereby maximizing the effectiveness of advertisements.

[0387] The processing flow will be explained below.

[0388] Step 1:

[0389] Subject: Server

[0390] The server starts streaming video and audio data from the installed cameras and microphones. The video data is captured frame by frame, and the audio data is collected in real time. These data are temporarily stored in a database for analysis.

[0391] Step 2:

[0392] Subject: Server

[0393] The server analyzes the video data and identifies people using a facial recognition algorithm. It uses a face detection library such as OpenCV to detect faces in the frame and estimate attributes such as age and gender for the detected faces.

[0394] Step 3:

[0395] Subject: Server

[0396] The server sends the voice data to a speech recognition engine, which converts it into text in real time. It then transcribes the conversation using speech recognition services such as Google Cloud Speech-to-Text API or Watson Speech to Text, and uses text analysis to identify interests and behaviors.

[0397] Step 4:

[0398] Subject: Server

[0399] The server combines the results of facial and voice recognition analysis to identify target attributes and context, which serves as the basis for generating customized advertisements that reflect the user's interests and current behavior.

[0400] Step 5:

[0401] Subject: Server

[0402] The server compares the analysis results with pre-prepared information for generating advertisements and selects the most suitable advertising materials. For example, if the customer is a family, it will select advertising materials for products aimed at children. This selected material is then input into the generative AI model.

[0403] Step 6:

[0404] Subject: Server

[0405] The generative AI model generates customized ad videos based on the selected ad materials and analysis results, and uses facial recognition results to quickly create personalized ads relevant to users.

[0406] Step 7:

[0407] Subject: Device

[0408] The terminal (digital signage) receives the generated advertising video from the server and immediately displays it. The video file is downloaded using a file transfer protocol and played on the display via a video player application.

[0409] Step 8:

[0410] Subject: Server

[0411] The server collects ad playback data, records the number of plays and the playback time, and periodically transmits ad playback event data from the digital signage terminal.

[0412] Step 9:

[0413] Subject: Server

[0414] The server calculates the fee for the advertiser based on the recorded playback data using a pay-per-view model, refers to a fee schedule based on the number of playbacks and duration, calculates the total amount, and provides a fee report to the advertiser.

[0415] The above steps enable the generation and distribution of optimal advertisements in real time according to the situation, maximizing the effectiveness of advertising.

[0416] Example 1

[0417] 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."

[0418] Conventional ad delivery systems have had difficulty generating and delivering optimal ads in real time based on target attributes and circumstances. In particular, they have been unable to effectively utilize video and audio data to instantly provide personalized ads based on analysis results. Furthermore, they lacked an efficient method for accurately calculating fees when applying a pay-per-use model to advertisers. As a result, maximizing advertising effectiveness and improving cost-effectiveness have not been fully achieved.

[0419] 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.

[0420] In this invention, the server includes a means for acquiring video data, a means for acquiring audio data, and an analysis means, which allows the server to generate and distribute optimal advertisements in real time according to the attributes and circumstances of the target audience, thereby maximizing the effectiveness of the advertisements.

[0421] "Video data" refers to digital data consisting of successive image frames captured in real time from an image capture device such as a camera.

[0422] "Audio data" refers to data obtained by digitizing sound waves acquired from a sound acquisition device such as a microphone.

[0423] The "analysis results" are the results of calculations performed using the acquired video and audio data to identify the attributes and circumstances of the target.

[0424] "Information for advertisement generation" refers to materials and data prepared in advance that will be the basis for the advertisement to be generated.

[0425] A "generative AI model" is an artificial intelligence model that uses machine learning algorithms to generate content based on input prompts.

[0426] A "prompt" is an instruction or question input to a generative AI model that provides the requirements for generating a specific output.

[0427] A "display device" is a screen or display for visually displaying the generated advertisement.

[0428] The "pay-as-you-go model" is a billing method in which fees are calculated based on the number of times an advertisement is displayed and the duration of its playback.

[0429] The present invention is a system that generates and distributes advertisements in real time that are optimized for the current situation based on video and audio data acquired from a camera and microphone. Specific embodiments for implementing the present invention will be described below.

[0430] Data collection

[0431] server

[0432] The server acquires video and audio data in real time from the installed cameras and microphones. This video data consists of successive image frames sent from the cameras, and the audio data is digitized sound waves sent from the microphones. Specifically, the server acquires the video stream from the cameras using RTSP (Real-Time Streaming Protocol), and the audio data is received via the Internet. This data is temporarily stored in a database for analysis.

[0433] Data analysis

[0434] server

[0435] The server analyzes the collected video and audio data. Using a facial recognition algorithm, it detects the faces of people in the video frames and infers their attributes (age, gender). Specifically, it uses OpenCV (an open-source computer vision library). It also uses the Google Cloud Speech-to-Text API as a speech recognition engine to convert the audio data into text. The text data can then be analyzed to understand the content and context of the conversation.

[0436] Ad Generation

[0437] server

[0438] Based on the results of the data analysis, the server uses a generative AI model (e.g., GPT-4) to generate the optimal advertisement from pre-prepared information for advertisement generation. The server inputs the prompt text into the generative AI model, which generates the advertisement script as a result. The generated script is then used in video editing software such as Adobe Premiere Pro to edit and complete the advertisement video. An example of a specific prompt text is, "Based on the camera footage and microphone audio data, generate an advertisement that is optimal for the current situation. As a specific scenario, please create an advertisement for a family entering a supermarket and talking about children's toys."

[0439] Ad serving

[0440] Terminal

[0441] Devices such as digital signage receive advertising videos generated from the server and instantly display them on the screen. The device uses a basic HTTP request to download the latest advertising video from the server and then plays the video file for passersby or users to view, enabling real-time advertising.

[0442] Pay-as-you-go model

[0443] server

[0444] The server collects ad playback data, recording the number of plays and the playback time. It receives access logs and playback logs sent from the device and uses this data to apply a pay-per-view model to calculate fees for advertisers. This allows advertisers to deliver ads in a cost-effective manner.

[0445] Specific examples

[0446] Example 1: Supermarket

[0447] Cameras and microphones installed at the entrance of the supermarket capture video and audio of families entering the store in real time. The server identifies parents and children from the video and analyzes the conversation about "children's toys" from the audio. Based on the analysis results, prompts are input into a generative AI model to generate promotional advertisements for children's toys. The completed advertising video is distributed to and displayed on digital signage terminals in the store.

[0448] Example 2: Inside a station

[0449] Cameras and microphones installed inside the station capture images of passersby and audio of them talking about "travel." The server analyzes this data, identifies age groups and genders, and then generates advertisements for travel agencies. These advertisements are displayed on digital signage, promoting travel products to passersby.

[0450] This makes it possible to generate and deliver optimal advertisements in real time according to the situation, maximizing the effectiveness of advertising.

[0451] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0452] Step 1:

[0453] Data collection

[0454] The server acquires video and audio data from the camera and microphone in real time. Specifically, the server receives a video stream of continuous frames from the camera using RTSP (Real-Time Streaming Protocol). It also receives audio data, which is digitized sound waves, from the microphone via the Internet. This data is temporarily stored in a database for analysis.

[0455] Input: Video stream from camera, audio data from microphone

[0456] Output: Video and audio data stored in a database for analysis

[0457] Step 2:

[0458] Data analysis

[0459] The server runs a facial recognition algorithm (e.g., OpenCV) to detect faces from the collected video data. This algorithm detects faces in the video frames and infers their attributes (age, gender). Additionally, the audio data is converted to text using the Google Cloud Speech-to-Text API. This allows the audio to be analyzed for conversation content and context, providing information to understand the subject's interests and behavior.

[0460] Input: Video data, audio data

[0461] Output: Attributes of identified person, audio data converted to text

[0462] Step 3:

[0463] Ad Generation

[0464] The server uses a generative AI model (e.g., GPT-4) to generate optimal advertisements based on the results of data analysis. The server inputs prompt sentences into the generative AI model and obtains the resulting advertisement script. It then uses video editing software such as Adobe Premiere Pro to edit and generate advertisement videos based on the generated script.

[0465] Input: Identified person's attributes, transcribed voice data, prompt

[0466] Output: Ad script, finished ad video

[0467] Step 4:

[0468] Ad serving

[0469] The terminal (digital signage) receives the generated advertising video from the server and plays it on the screen for immediate display. The terminal uses a basic HTTP request to download the latest advertising video from the server and plays the received video file.

[0470] Input: Ad video

[0471] Output: Ad video displayed on the screen

[0472] Step 5:

[0473] Pay-as-you-go model

[0474] The server collects data each time an advertisement is played, recording the number of plays and the play time. Specifically, it receives access logs and play logs sent from the terminal and applies a pay-per-view model based on that data. This is used to calculate the fee for the advertiser.

[0475] Input: Access log and playback log from the device

[0476] Output: Pricing results based on number of plays and duration

[0477] (Application example 1)

[0478] 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."

[0479] Modern commercial facilities and brick-and-mortar stores require real-time advertising that meets customer needs and interests. However, existing systems have difficulty accurately analyzing customer attributes and conversation content, and quickly generating and delivering individually customized advertisements. Another issue is that they are unable to meet the need for direct advertising display using smart devices.

[0480] 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.

[0481] In this invention, the server includes means for acquiring video data, means for acquiring audio data, means for analyzing the video data and the audio data and identifying target attributes and situations, means for generating an optimal advertisement from pre-prepared advertisement generation information based on the analysis results, means for delivering the generated advertisement via a display device, and means for displaying the advertisement on a smart device in real time. This makes it possible to quickly deliver an optimized advertisement to a smart device based on the real-time attributes and conversation content of a customer.

[0482] "Video data" refers to data made up of successive image frames captured using a photographic device such as a camera.

[0483] "Audio data" refers to data obtained by digitizing sound waves acquired using an acoustic device such as a microphone.

[0484] The "analyzing means" is a means for executing a process to determine a person's attributes and situation based on the acquired video data and audio data.

[0485] "Information for advertisement generation" is a collection of materials and information prepared in advance and used to generate optimal advertisements based on the analysis results.

[0486] A "display device" is a hardware device for providing the generated advertisement to the viewer, and includes digital signage, smartphones, etc.

[0487] "Smart devices" is a general term for mobile terminals and wearable devices that are connected to the Internet and have the ability to run applications.

[0488] A "generative AI model" is an algorithm or machine learning model that uses artificial intelligence to generate optimal advertisements from input data.

[0489] A "prompt sentence" is an input sentence used to request the generative AI model to generate an ad.

[0490] "Attributes" are information that indicates the age, sex, interests, behavior, etc. of the person being analyzed.

[0491] "Situation" is information that indicates the environment in which the analysis target is placed and the situation at that time.

[0492] The present invention relates to a system that generates and distributes advertisements in real time that are optimized for the current situation based on video and audio data acquired from a camera and a microphone. Specific embodiments for implementing the present invention will be described below.

[0493] Data collection

[0494] The server collects video and audio data in real time from cameras and microphones installed in the store. The video data consists of successive image frames sent from the camera, and the audio data is digitized sound waves sent from the microphone. The collected data is temporarily stored in a database for analysis. The hardware used is a standard camera (e.g., a 1080p webcam) and a high-sensitivity microphone (e.g., a USB-connected condenser microphone).

[0495] Data analysis

[0496] The server analyzes the collected video data and identifies people using a facial recognition algorithm. It then infers their attributes (e.g., age and gender) from the identified people. It also converts the audio data into text using a speech recognition engine and analyzes the content and context of the conversation. The software used includes OpenCV (for video analysis) and the SpeechRecognition library (for audio analysis).

[0497] Ad Generation

[0498] Based on the information obtained through the analysis, the server compares it with pre-prepared ad generation information to generate the optimal ad. Specifically, it selects advertising materials based on the analysis results and creates a customized ad video using a generative AI model. The generative AI model uses OpenAI's API. This ad generation process makes it possible to quickly create personalized ads for each user. An example of a prompt sentence when generating a generated ad is as follows:

[0499] Generate a personalized advertisement for a 25-year-old female interested in new fashion trends.

[0500] Ad serving

[0501] The terminal (smart device) receives the advertising video generated from the server and immediately displays it. The terminal then plays the advertising video on a screen for passersby and users to view. Specific smart devices include smartphones and head-mounted displays.

[0502] Pay-as-you-go model

[0503] The server collects ad playback data, recording the number of plays and playback time. Based on this data, a pay-per-view model is applied to calculate fees for advertisers. This allows advertisers to deliver ads in a cost-effective manner. In addition, user interactions with ads displayed on smart devices can be analyzed, which can be used to further analyze the effectiveness of advertising.

[0504] Specific examples

[0505] Example 1: Supermarket

[0506] Cameras and microphones installed at the entrance of supermarkets capture video and audio recordings of families entering the store in real time. The server identifies parents and children from the video and analyzes audio conversations about "children's toys." Based on this, the server generates promotional advertisements for children's toys and instantly distributes them to in-store digital signage. The same advertisements are also displayed in real time on smartphone apps.

[0507] Example 2: Inside a station

[0508] Cameras and microphones installed inside the station capture images of passersby and audio of them talking about "travel." The server analyzes the data, identifies age groups and genders, and generates advertisements for travel agencies. These advertisements are displayed on digital signage, promoting travel products to passersby, while also being displayed in real time on head-mounted displays.

[0509] As explained above, the present invention makes it possible to generate and distribute advertisements that are optimal for each situation in real time, thereby maximizing the effectiveness of advertisements.

[0510] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0511] Step 1:

[0512] The server acquires video data from cameras installed inside the store. The server acquires successive image frames and temporarily stores them in a database for analysis. At this time, the data input from the camera is acquired as a series of still image data, which is output as video data.

[0513] Step 2:

[0514] The server simultaneously captures audio data from a microphone. The sound waves coming from the microphone are digitized and stored as an audio file in a database on the server. This audio data is used in subsequent analysis steps.

[0515] Step 3:

[0516] The server applies a facial recognition algorithm to the captured video data. Specifically, it uses the OpenCV library to perform face detection for each frame, identify specific people, and infer their attributes. The input is the video data, and the output is the attribute information of the identified people (e.g., age, gender).

[0517] Step 4:

[0518] The server converts the voice data into text using the SpeechRecognition library. The voice data is input into a speech recognition engine, which then outputs analyzed text data. Based on this text data, the content of the conversation and the situation are analyzed to understand the subject's interests and behavior.

[0519] Step 5:

[0520] The server compares the results of video and audio analysis to identify the overall situation and the target's interests and behavior. This determines what type of advertising will be most effective. The input is the results of video and audio analysis, and the output is a situational judgment based on the analysis results.

[0521] Step 6:

[0522] Based on the analysis results, the server generates the optimal advertisement from pre-prepared information for advertisement generation. Here, a generative AI model is used to generate a customized advertisement by providing a prompt. For example, an advertisement is generated based on the prompt "Generate a personalized advertisement for a 25-year-old female interested in new fashion trends." The input is the prompt and information for advertisement generation, and the output is the generated advertisement content.

[0523] Step 7:

[0524] The server sends the generated advertisement to the terminal (e.g., smart device). The terminal immediately displays the advertisement content received from the server. The input is the generated advertisement content, and the output is the advertisement displayed on the terminal.

[0525] Step 8:

[0526] When a user interacts with an advertisement on their device, the device sends this interaction data to a server. The server analyzes this data and evaluates the effectiveness of the advertisement and the user's response. The input is the interaction data, and the output is the analyzed evaluation results.

[0527] Step 9:

[0528] The server finally applies a pay-per-view model based on the ad playback data and interaction data to calculate the fee for the advertiser. The input is the playback data and interaction data, and the output is the fee calculation result.

[0529] 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.

[0530] The present invention relates to a system that generates and distributes advertisements in real time that are optimized for the situation and user emotions based on video and audio data acquired from a camera and a microphone. Specific embodiments for carrying out the present invention will be described below.

[0531] Data collection

[0532] Subject: Server

[0533] The server captures video and audio data in real time from the installed cameras and microphones. These data are temporarily stored in a database for analysis. Video data consists of a series of image frames, and audio data is digitized sound waves.

[0534] Data analysis

[0535] Subject: Server

[0536] The server analyzes the collected video data and identifies people using a facial recognition algorithm. It then infers their attributes (e.g., age and gender) from the detected people. It also sends the audio data to a speech recognition engine, which converts it into text. This allows the content and context of the conversation to be analyzed.

[0537] emotion recognition

[0538] Subject: Server

[0539] The server recognizes the user's emotions by analyzing facial expressions from video data and tone of voice from audio data. The emotion recognition engine identifies emotions such as anger, joy, and surprise, and uses the results for further analysis.

[0540] Ad Generation

[0541] Subject: Server

[0542] The server generates the optimal advertisement based on the results of attribute and situation analysis and emotion recognition, comparing them with pre-prepared advertisement generation information. Specifically, it selects advertising materials based on the analysis results and emotion recognition results, and creates a customized advertising video using a generative AI model. In this process, the content that best suits the user's emotional state is reflected.

[0543] Ad serving

[0544] Subject: Device

[0545] The terminal (digital signage) receives the advertising video generated from the server and immediately displays it. The terminal then plays the advertising video on the display screen for the user to view. This allows the optimal advertising to be delivered in real time according to the situation.

[0546] Pay-as-you-go model

[0547] Subject: Server

[0548] The server collects ad playback data, recording the number of plays and the playback time. Based on this data, a pay-per-view model is applied to calculate the fees for advertisers, enabling them to deliver ads in a cost-effective manner.

[0549] Specific examples

[0550] Example 1: Supermarket

[0551] Cameras and microphones installed at the entrance of the supermarket capture video and audio of families entering the store in real time. The server identifies parents and children from the video and analyzes the audio content of conversations about "toys." At the same time, an emotion recognition engine detects whether the child is excited. Based on this, the server generates a promotional advertisement for children's toys and instantly distributes it to digital signage within the store.

[0552] Example 2: Inside a station

[0553] Cameras and microphones installed inside the station capture images of passersby and audio of them talking about "travel." The server analyzes the data, identifies their age group and gender, and uses an emotion recognition engine to assess their excitement level. Based on this, the server generates advertisements for travel agencies and displays them on digital signage.

[0554] As explained above, the present invention makes it possible to generate and distribute optimal advertisements in real time according to the situation and user emotions, thereby maximizing the effectiveness of advertisements.

[0555] The processing flow will be explained below.

[0556] Step 1:

[0557] Subject: Server

[0558] The camera and microphone begin streaming video and audio data. Video data is collected frame by frame, and audio data is digitized and captured in real time. This data is temporarily stored in a database for analysis.

[0559] Step 2:

[0560] Subject: Server

[0561] The server analyzes the collected video data and uses facial recognition algorithms to identify people. Using a face detection library such as OpenCV, it detects faces in the video frames and estimates attributes such as age and gender from the faces.

[0562] Step 3:

[0563] Subject: Server

[0564] Voice data is sent to a speech recognition engine and converted to text in real time. Speech recognition services such as Google Cloud Speech-to-Text API or Watson Speech to Text are used to transcribe the conversation and perform text analysis to identify interests and behaviors.

[0565] Step 4:

[0566] Subject: Server

[0567] The results of the analysis of video and audio data are integrated to identify the target's attributes and situation, thereby clarifying the user's basic profile and current behavior and interests.

[0568] Step 5:

[0569] Subject: Server

[0570] Based on video and audio data, the system analyzes the user's facial expressions and tone of voice to recognize their emotions. Using an emotion recognition engine, it identifies emotions such as anger, joy, and surprise, and uses the results in the next ad generation process.

[0571] Step 6:

[0572] Subject: Server

[0573] Based on the analysis results and emotion recognition results, the optimal advertising materials are selected by comparing them with pre-prepared information for advertising generation.Then, the selected materials and analysis results are input into the generation AI model to generate a customized advertising video that matches the emotions.

[0574] Step 7:

[0575] Subject: Device

[0576] The digital signage terminal receives the generated advertising video from the server, and immediately plays the downloaded advertising video using a file transfer protocol and displays it on the terminal screen.

[0577] Step 8:

[0578] Subject: Server

[0579] The device collects data on the ads played, recording the number of plays and the play time, and periodically sends this playback data to the server.

[0580] Step 9:

[0581] Subject: Server

[0582] The server applies a pay-per-view model based on the collected playback data to calculate the fees for advertisers, calculates the total amount based on a fee schedule according to the number of plays and the playback time, and provides a fee report to the advertiser.

[0583] The above steps enable the creation and distribution of optimal advertisements in real time according to the situation and user emotions, maximizing the effectiveness of the advertisements.

[0584] Example 2

[0585] 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."

[0586] In modern advertising delivery systems, it is difficult to generate and deliver optimal advertisements in real time that match user attributes and emotions. Maximizing the cost-effectiveness of advertising by introducing an efficient pay-per-use model for advertisers is also a challenge. Conventional systems are unable to perform such real-time data analysis and advertisement generation, making it impossible to deliver effective advertisements that match user interests and emotions.

[0587] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for acquiring video data, means for acquiring audio data, means for analyzing the video data and the audio data to identify person attributes and situations, means for recognizing a user's emotions from the video data and the audio data, means for generating an optimal advertisement from pre-prepared advertisement generation information based on the analysis results and the user's emotion recognition results, and means for delivering the generated advertisement via a display device. This enables real-time advertisement generation and delivery based on user attributes and emotions, maximizing the effectiveness of advertisements. In addition, by applying a pay-per-view model based on the number of impressions and playback time, advertisers can be effectively charged advertising fees.

[0588] "Video data" is a digital representation of visual information captured using a device such as a camera.

[0589] "Audio data" refers to digitized sound waves recorded using a device such as a microphone.

[0590] "Person attributes" are characteristics about an individual, such as age, gender, or identity, that are inferred using facial recognition algorithms or other methods.

[0591] "User emotion" refers to an emotional state, such as anger, joy, or surprise, that is recognized through facial expression analysis from video data and tone analysis from audio data.

[0592] "Information for advertisement generation" refers to materials such as text, images, and videos prepared in advance for generating advertisements, as well as related data.

[0593] A "generative AI model" is an artificial intelligence model that generates customized output data based on input data and specified prompts.

[0594] A "prompt sentence" is a textual input used to give specific instructions to a generative AI model.

[0595] A "pay-as-you-go model" is a pricing model in which the fee is calculated based on the number of times an advertisement is displayed and the duration of its playback.

[0596] A "display device" refers to a display or digital signage used to allow users to view the generated advertisement.

[0597] "Playback data" refers to data relating to playback status, such as the number of times a generated advertisement is played and the playback time.

[0598] This invention relates to a system that generates and distributes advertisements in real time that are optimized for the situation and user emotions based on video and audio data acquired from a camera and microphone.

[0599] The server collects video and audio data in real time from cameras and microphones installed via the network. The video data consists of successive image frames, and the audio data is digitized sound waves. These data are temporarily stored in a database for analysis. Specifically, the camera collects video at 30 frames per second, and the microphone collects audio data at 44.1 kHz per second.

[0600] The server analyzes the collected video data and identifies people using a facial recognition algorithm. For example, it uses image processing software (e.g., OpenCV). Attributes such as age and gender are inferred by analyzing the feature points of the detected face. At the same time, the audio data is sent to a speech recognition engine (e.g., Google Speech-to-Text API) and converted into text. This allows the content and situation of the conversation to be extracted. For example, the keyword "travel" can be extracted from the audio data.

[0601] The server then analyzes the user's facial expressions from the video data and the tone of voice from the audio data to recognize the user's emotions. It uses an emotion recognition engine (e.g., Amazon Rekognition or Google Cloud Natural Language API) to determine emotional states such as anger, joy, and surprise. For example, it can detect whether the user is smiling and recognize excitement from the tone of their voice.

[0602] Next, the server generates the optimal advertisement based on the analysis results and emotion recognition results, comparing them with pre-prepared advertisement generation information. It selects advertising materials (text, images, videos) and uses a generative AI model (e.g., GPT-4) to create a customized advertising video. During this process, it issues instructions by sending prompt text to the generative AI model. For example, a prompt text such as "An excited man in his 60s who is interested in traveling" could be sent to the generative AI model to generate a customized advertising video for a travel agency.

[0603] The generated advertising video is sent to a terminal (digital signage) and played on the display, allowing users to view it. For example, a travel advertising video can be played on a display in a store to attract the attention of passersby.

[0604] Finally, the server collects ad play data, recording the number of plays and play time. Based on this data, it applies a pay-per-view model and calculates the fee to the advertiser. For example, it records that a particular ad was played 100 times and the total play time was 50 minutes, and calculates the cost to be charged to the advertiser.

[0605] As a result, the system of the present invention enables real-time generation and distribution of advertisements based on user attributes and emotions, maximizing the effectiveness of advertisements. In addition, by applying a pay-per-view model based on the number of impressions and playback time, advertisers can be effectively charged advertising fees.

[0606] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0607] Step 1:

[0608] Data collection:

[0609] The server receives real-time video and audio data from cameras and microphones via the network. Specifically, the cameras capture video data at 30 frames per second, and the microphones capture audio data at 44.1 kHz per second. This data is temporarily stored in a database for analysis.

[0610] Input: Video and audio data obtained from the camera and microphone.

[0611] Output: Video and audio data stored in a database for analysis.

[0612] Step 2:

[0613] Data Analysis:

[0614] The server analyzes the collected video data. Specifically, it uses image processing software (e.g., OpenCV) to run a facial recognition algorithm to identify people. Attributes such as age and gender are inferred by analyzing facial feature points. At the same time, the server sends the audio data to a speech recognition engine (e.g., Google Speech-to-Text API) to convert it into text.

[0615] Input: Video and audio data obtained from the analysis database.

[0616] Output: Identified person's attributes (age, gender) and transcribed audio data.

[0617] Step 3:

[0618] Emotion recognition:

[0619] The server analyzes the user's facial expressions from the video data and the tone of voice from the audio data to recognize the user's emotions. An emotion recognition engine (e.g., Amazon Rekognition or Google Cloud Natural Language API) is used to determine the user's emotional state, such as anger, joy, or surprise.

[0620] Input: Identified person attributes and transcribed audio data.

[0621] Output: The user's emotional state (e.g., anger, joy, surprise).

[0622] Step 4:

[0623] Ad Generation:

[0624] The server generates the optimal ad based on the analysis results and emotion recognition results, comparing them with pre-prepared ad generation information. It selects ad materials (text, images, videos) and creates a customized ad video using a generative AI model (e.g., GPT-4). During this process, it sends a prompt to the generative AI model.

[0625] Input: User's emotional state, identified person attributes, and information for ad generation.

[0626] Output: A customized advertising video.

[0627] Step 5:

[0628] Ad serving:

[0629] The terminal (digital signage) receives the advertising video generated from the server and immediately displays it. The terminal plays the advertising video on the display so that the user can view it.

[0630] Input: The customized ad video sent from the server.

[0631] Output: Ad video played on the display.

[0632] Step 6:

[0633] Pay-as-you-go:

[0634] The server collects ad playback data, records the number of plays and the playback time, and calculates the fee to the advertiser based on this data using a pay-per-view model.

[0635] Input: Playback data of the ad video played on the display.

[0636] Output: Fees charged to advertisers.

[0637] (Application example 2)

[0638] 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."

[0639] Conventional ad delivery systems generate and deliver ads based on user attributes and circumstances, but they do not fully consider the user's real-time emotions or specific circumstances, which means that the effectiveness of the ads is not maximized.In addition, there is a lack of technology for generating and delivering effective ads, and the cost-effectiveness for advertisers is also insufficient.

[0640] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0641] In this invention, the server includes means for acquiring video data, means for acquiring audio data, means for analyzing the video data and the audio data to identify target attributes and situations, means for analyzing user emotions, means for generating an optimal advertisement from pre-prepared information for advertisement generation based on the results of the analysis and the emotion analysis, and means for delivering the generated advertisement via a display device. This makes it possible to generate and deliver an optimal advertisement according to the user's real-time emotions and specific situations, thereby maximizing advertising effectiveness.

[0642] "Video data" refers to visual information acquired by an image sensor such as a camera.

[0643] "Audio data" refers to auditory information acquired by an acoustic sensor such as a microphone.

[0644] "Analysis" refers to computational processing to extract or infer specific information from acquired data.

[0645] "Attributes" refer to the characteristics or properties of a subject that are determined through analysis, and include, for example, age and gender.

[0646] "Situation" refers to the subject's current environmental and behavioral state as determined by analysis.

[0647] "Emotion" refers to the subject's psychological state as determined by analysis, and includes anger, joy, surprise, etc.

[0648] "Information for advertisement generation" refers to materials and data prepared in advance for generating an advertisement.

[0649] The term "display device" refers to a device for visually presenting the generated advertisement to a user.

[0650] "Advertising materials" refers to the individual elements and content that make up an advertisement.

[0651] "Customized" refers to advertising that is generated and tailored to specific requirements and conditions.

[0652] A "pay-as-you-go model" refers to a pricing system in which charges are based on the amount of usage or the number of times it is used.

[0653] A system for implementing the present invention has the following configuration: A server acquires and analyzes video data and audio data, and generates and distributes optimal advertisements. A detailed explanation is provided below with specific examples.

[0654] First, the server acquires video and audio data in real time from cameras and microphones. Video data is visual information acquired by image sensors such as cameras, and audio data is auditory information acquired by acoustic sensors such as microphones. These data are temporarily stored in a database for analysis.

[0655] The server then analyzes the collected video and audio data. The video data uses facial recognition algorithms to identify people and infer their attributes (e.g., age, gender, etc.), while the audio data is sent to a speech recognition engine, which converts it into text and analyzes it. This analysis process uses software libraries such as OpenCV and SpeechRecognition.

[0656] Furthermore, the server uses an emotion recognition engine to analyze the user's emotions. By analyzing the user's facial expressions from the video data and the tone of voice from the audio data, emotions such as anger, joy, and surprise can be recognized. This emotion analysis combines multiple algorithms.

[0657] The server generates optimal advertisements based on the results of these analyses and sentiment analysis. Specifically, it selects the optimal advertising materials for the user's attributes and emotional state from pre-prepared advertisement generation information, and creates customized advertisements using a generative AI model.

[0658] Finally, the generated advertisement is delivered to a display device, such as a smartphone or digital signage. The user can watch the advertisement in real time, optimized for the situation and emotions of the moment. The terminal device plays the advertisement video and provides it visually to the user.

[0659] For example, when a user is using a smartphone, the smartphone's camera and microphone collect data and detect that the user is having a pleasant conversation. The server generates an advertisement for an entertainment app that matches that pleasant emotion and displays it on the smartphone. This application maximizes the effectiveness of advertising.

[0660] Example prompt sentence:

[0661] "The user is showing happy emotions. Please generate an ad for an entertainment app. User demographics: Age 25, Gender Male"

[0662] As described above, the system of the present invention is capable of generating and delivering optimal advertisements that correspond to the user's real-time emotions and specific situations, thereby maximizing the effectiveness of advertisements and improving cost-effectiveness for advertisers.

[0663] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0664] Step 1:

[0665] The server acquires video and audio data from the camera and microphone in real time. As input, it receives the real-time output of the camera and microphone and temporarily stores it as video and audio data in a database within the server. In this process, the camera captures successive image frames and the microphone generates a digital audio signal.

[0666] Step 2:

[0667] The server analyzes the acquired video data. As input, it receives the video data acquired in step 1, identifies the person using a facial recognition algorithm, and then estimates attributes such as age and gender. For example, it uses OpenCV to detect faces from image frames and estimates attributes based on the results. As output, it obtains attribute information for the identified person.

[0668] Step 3:

[0669] The server analyzes the acquired voice data. As input, it receives the voice data acquired in step 1 and converts it into text using a voice recognition engine. For example, it uses the SpeechRecognition library to convert the voice data into text, and analyzes the topic and situation of the conversation from the text content. As output, it obtains text data and the analysis results.

[0670] Step 4:

[0671] The server analyzes the user's emotions from the video and audio data. It receives the outputs of steps 2 and 3 as input and uses an emotion recognition engine to analyze facial expressions and tone of voice. Specifically, it reads facial expressions from image frames and identifies emotions from audio tones. The output is emotional information such as anger, joy, and surprise.

[0672] Step 5:

[0673] The server generates the optimal advertisement based on the analysis results and the results of sentiment analysis. It receives the outputs of steps 2, 3, and 4 as input and selects the optimal advertising material from pre-prepared information for advertisement generation. It uses a generative AI model to generate a customized advertisement that best suits the user's attributes and emotions. The output is a customized advertising video.

[0674] Step 6:

[0675] The server delivers the generated advertisement to the terminal. It receives the customized advertisement video output from step 5 as input and sends it to the terminal. The terminal plays the received advertisement video on a display device (e.g., a smartphone or digital signage). This process allows the user to watch the optimized advertisement in real time.

[0676] Step 7:

[0677] The server collects playback data of the displayed advertisements and calculates fees based on a pay-per-view model. As input, it receives data on the number of plays and the playback time sent from the terminal, and calculates the fee for the advertiser based on that data. As output, it obtains the final fee calculation result.

[0678] 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.

[0679] 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.

[0680] 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.

[0681] [Third embodiment]

[0682] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0683] 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.

[0684] 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).

[0685] 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.

[0686] 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.

[0687] 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).

[0688] 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.

[0689] 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.

[0690] 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.

[0691] 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.

[0692] 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.

[0693] 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."

[0694] The present invention relates to a system that generates and distributes advertisements in real time that are optimized for the current situation based on video and audio data acquired from a camera and a microphone. Specific embodiments for implementing the present invention will be described below.

[0695] Data collection

[0696] Subject: Server

[0697] The server collects real-time video and audio data from the installed cameras and microphones. The video data consists of successive image frames sent from the cameras, and the audio data is digitized sound waves sent from the microphones. The collected data is temporarily stored in a database for analysis.

[0698] Data analysis

[0699] Subject: Server

[0700] The server analyzes the collected video data and identifies people using a facial recognition algorithm. From this identified person, the server infers their attributes (e.g., age and gender). It also uses a voice recognition engine to convert the voice data into text and analyzes the conversation content and situation. This makes it possible to understand the interests and behavior of the target.

[0701] Ad Generation

[0702] Subject: Server

[0703] The server generates optimal advertisements based on the information obtained through the analysis and compares it with pre-prepared advertisement generation information. Specifically, it selects advertising materials based on the analysis results and creates customized advertising videos using a generative AI model. This advertisement generation process makes it possible to create personalized advertisements for each user in a short amount of time.

[0704] Ad serving

[0705] Subject: Device

[0706] The terminal (digital signage) receives the advertising video generated from the server and displays it instantly. The terminal then plays this advertising video on the screen for passersby and users to view. This allows for the delivery of advertising that is appropriate for the situation in real time, which can be expected to be more effective.

[0707] Pay-as-you-go model

[0708] Subject: Server

[0709] The server collects ad playback data, recording the number of plays and the playback time. Based on this data, a pay-per-view model is applied to calculate the fees for advertisers, enabling them to deliver ads in a cost-effective manner.

[0710] Specific examples

[0711] Example 1: Supermarket

[0712] Cameras and microphones installed at the entrance of supermarkets capture video and audio of families entering the store in real time. The server identifies parents and children from the video and analyzes the audio for conversations about "children's toys." Based on this, the server generates promotional advertisements for children's toys and instantly distributes them to digital signage in the store.

[0713] Example 2: Inside a station

[0714] Cameras and microphones installed inside the station capture images of passersby and audio of them talking about "travel." The server analyzes the data, identifies age groups and genders, and generates advertisements for travel agencies. These advertisements are displayed on digital signage, promoting travel products to passersby.

[0715] As explained above, the present invention makes it possible to generate and distribute advertisements that are optimal for each situation in real time, thereby maximizing the effectiveness of advertisements.

[0716] The processing flow will be explained below.

[0717] Step 1:

[0718] Subject: Server

[0719] The server starts streaming video and audio data from the installed cameras and microphones. The video data is captured frame by frame, and the audio data is collected in real time. These data are temporarily stored in a database for analysis.

[0720] Step 2:

[0721] Subject: Server

[0722] The server analyzes the video data and identifies people using a facial recognition algorithm. It uses a face detection library such as OpenCV to detect faces in the frame and estimate attributes such as age and gender for the detected faces.

[0723] Step 3:

[0724] Subject: Server

[0725] The server sends the voice data to a speech recognition engine, which converts it into text in real time. It then transcribes the conversation using speech recognition services such as Google Cloud Speech-to-Text API or Watson Speech to Text, and uses text analysis to identify interests and behaviors.

[0726] Step 4:

[0727] Subject: Server

[0728] The server combines the results of facial and voice recognition analysis to identify target attributes and context, which serves as the basis for generating customized advertisements that reflect the user's interests and current behavior.

[0729] Step 5:

[0730] Subject: Server

[0731] The server compares the analysis results with pre-prepared information for generating advertisements and selects the most suitable advertising materials. For example, if the customer is a family, it will select advertising materials for products aimed at children. This selected material is then input into the generative AI model.

[0732] Step 6:

[0733] Subject: Server

[0734] The generative AI model generates customized ad videos based on the selected ad materials and analysis results, and uses facial recognition results to quickly create personalized ads relevant to users.

[0735] Step 7:

[0736] Subject: Device

[0737] The terminal (digital signage) receives the generated advertising video from the server and immediately displays it. The video file is downloaded using a file transfer protocol and played on the display via a video player application.

[0738] Step 8:

[0739] Subject: Server

[0740] The server collects ad playback data, records the number of plays and the playback time, and periodically transmits ad playback event data from the digital signage terminal.

[0741] Step 9:

[0742] Subject: Server

[0743] The server calculates the fee for the advertiser based on the recorded playback data using a pay-per-view model, refers to a fee schedule based on the number of playbacks and duration, calculates the total amount, and provides a fee report to the advertiser.

[0744] The above steps enable the generation and distribution of optimal advertisements in real time according to the situation, maximizing the effectiveness of advertising.

[0745] Example 1

[0746] 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."

[0747] Conventional ad delivery systems have had difficulty generating and delivering optimal ads in real time based on target attributes and circumstances. In particular, they have been unable to effectively utilize video and audio data to instantly provide personalized ads based on analysis results. Furthermore, they lacked an efficient method for accurately calculating fees when applying a pay-per-use model to advertisers. As a result, maximizing advertising effectiveness and improving cost-effectiveness have not been fully achieved.

[0748] 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.

[0749] In this invention, the server includes a means for acquiring video data, a means for acquiring audio data, and an analysis means, which allows the server to generate and distribute optimal advertisements in real time according to the attributes and circumstances of the target audience, thereby maximizing the effectiveness of the advertisements.

[0750] "Video data" refers to digital data consisting of successive image frames captured in real time from an image capture device such as a camera.

[0751] "Audio data" refers to data obtained by digitizing sound waves acquired from a sound acquisition device such as a microphone.

[0752] The "analysis results" are the results of calculations performed using the acquired video and audio data to identify the attributes and circumstances of the target.

[0753] "Information for advertisement generation" refers to materials and data prepared in advance that will be the basis for the advertisement to be generated.

[0754] A "generative AI model" is an artificial intelligence model that uses machine learning algorithms to generate content based on input prompts.

[0755] A "prompt" is an instruction or question input to a generative AI model that provides the requirements for generating a specific output.

[0756] A "display device" is a screen or display for visually displaying the generated advertisement.

[0757] The "pay-as-you-go model" is a billing method in which fees are calculated based on the number of times an advertisement is displayed and the duration of its playback.

[0758] The present invention is a system that generates and distributes advertisements in real time that are optimized for the current situation based on video and audio data acquired from a camera and microphone. Specific embodiments for implementing the present invention will be described below.

[0759] Data collection

[0760] server

[0761] The server acquires video and audio data in real time from the installed cameras and microphones. This video data consists of successive image frames sent from the cameras, and the audio data is digitized sound waves sent from the microphones. Specifically, the server acquires the video stream from the cameras using RTSP (Real-Time Streaming Protocol), and the audio data is received via the Internet. This data is temporarily stored in a database for analysis.

[0762] Data analysis

[0763] server

[0764] The server analyzes the collected video and audio data. Using a facial recognition algorithm, it detects the faces of people in the video frames and infers their attributes (age, gender). Specifically, it uses OpenCV (an open-source computer vision library). It also uses the Google Cloud Speech-to-Text API as a speech recognition engine to convert the audio data into text. The text data can then be analyzed to understand the content and context of the conversation.

[0765] Ad Generation

[0766] server

[0767] Based on the results of the data analysis, the server uses a generative AI model (e.g., GPT-4) to generate the optimal advertisement from pre-prepared information for advertisement generation. The server inputs the prompt text into the generative AI model, which generates the advertisement script as a result. The generated script is then used in video editing software such as Adobe Premiere Pro to edit and complete the advertisement video. An example of a specific prompt text is, "Based on the camera footage and microphone audio data, generate an advertisement that is optimal for the current situation. As a specific scenario, please create an advertisement for a family entering a supermarket and talking about children's toys."

[0768] Ad serving

[0769] Terminal

[0770] Devices such as digital signage receive advertising videos generated from the server and instantly display them on the screen. The device uses a basic HTTP request to download the latest advertising video from the server and then plays the video file for passersby or users to view, enabling real-time advertising.

[0771] Pay-as-you-go model

[0772] server

[0773] The server collects ad playback data, recording the number of plays and the playback time. It receives access logs and playback logs sent from the device and uses this data to apply a pay-per-view model to calculate fees for advertisers. This allows advertisers to deliver ads in a cost-effective manner.

[0774] Specific examples

[0775] Example 1: Supermarket

[0776] Cameras and microphones installed at the entrance of the supermarket capture video and audio of families entering the store in real time. The server identifies parents and children from the video and analyzes the conversation about "children's toys" from the audio. Based on the analysis results, prompts are input into a generative AI model to generate promotional advertisements for children's toys. The completed advertising video is distributed to and displayed on digital signage terminals in the store.

[0777] Example 2: Inside a station

[0778] Cameras and microphones installed inside the station capture images of passersby and audio of them talking about "travel." The server analyzes this data, identifies age groups and genders, and then generates advertisements for travel agencies. These advertisements are displayed on digital signage, promoting travel products to passersby.

[0779] This makes it possible to generate and deliver optimal advertisements in real time according to the situation, maximizing the effectiveness of advertising.

[0780] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0781] Step 1:

[0782] Data collection

[0783] The server acquires video and audio data from the camera and microphone in real time. Specifically, the server receives a video stream of continuous frames from the camera using RTSP (Real-Time Streaming Protocol). It also receives audio data, which is digitized sound waves, from the microphone via the Internet. This data is temporarily stored in a database for analysis.

[0784] Input: Video stream from camera, audio data from microphone

[0785] Output: Video and audio data stored in a database for analysis

[0786] Step 2:

[0787] Data analysis

[0788] The server runs a facial recognition algorithm (e.g., OpenCV) to detect faces from the collected video data. This algorithm detects faces in the video frames and infers their attributes (age, gender). Additionally, the audio data is converted to text using the Google Cloud Speech-to-Text API. This allows the audio to be analyzed for conversation content and context, providing information to understand the subject's interests and behavior.

[0789] Input: Video data, audio data

[0790] Output: Attributes of identified person, audio data converted to text

[0791] Step 3:

[0792] Ad Generation

[0793] The server uses a generative AI model (e.g., GPT-4) to generate optimal advertisements based on the results of data analysis. The server inputs prompt sentences into the generative AI model and obtains the resulting advertisement script. It then uses video editing software such as Adobe Premiere Pro to edit and generate advertisement videos based on the generated script.

[0794] Input: Identified person's attributes, transcribed voice data, prompt

[0795] Output: Ad script, finished ad video

[0796] Step 4:

[0797] Ad serving

[0798] The terminal (digital signage) receives the generated advertising video from the server and plays it on the screen for immediate display. The terminal uses a basic HTTP request to download the latest advertising video from the server and plays the received video file.

[0799] Input: Ad video

[0800] Output: Ad video displayed on the screen

[0801] Step 5:

[0802] Pay-as-you-go model

[0803] The server collects data each time an advertisement is played, recording the number of plays and the play time. Specifically, it receives access logs and play logs sent from the terminal and applies a pay-per-view model based on that data. This is used to calculate the fee for the advertiser.

[0804] Input: Access log and playback log from the device

[0805] Output: Pricing results based on number of plays and duration

[0806] (Application example 1)

[0807] 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."

[0808] Modern commercial facilities and brick-and-mortar stores require real-time advertising that meets customer needs and interests. However, existing systems have difficulty accurately analyzing customer attributes and conversation content, and quickly generating and delivering individually customized advertisements. Another issue is that they are unable to meet the need for direct advertising display using smart devices.

[0809] 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.

[0810] In this invention, the server includes means for acquiring video data, means for acquiring audio data, means for analyzing the video data and the audio data and identifying target attributes and situations, means for generating an optimal advertisement from pre-prepared advertisement generation information based on the analysis results, means for delivering the generated advertisement via a display device, and means for displaying the advertisement on a smart device in real time. This makes it possible to quickly deliver an optimized advertisement to a smart device based on the real-time attributes and conversation content of a customer.

[0811] "Video data" refers to data made up of successive image frames captured using a photographic device such as a camera.

[0812] "Audio data" refers to data obtained by digitizing sound waves acquired using an acoustic device such as a microphone.

[0813] The "analyzing means" is a means for executing a process to determine a person's attributes and situation based on the acquired video data and audio data.

[0814] "Information for advertisement generation" is a collection of materials and information prepared in advance and used to generate optimal advertisements based on the analysis results.

[0815] A "display device" is a hardware device for providing the generated advertisement to the viewer, and includes digital signage, smartphones, etc.

[0816] "Smart devices" is a general term for mobile terminals and wearable devices that are connected to the Internet and have the ability to run applications.

[0817] A "generative AI model" is an algorithm or machine learning model that uses artificial intelligence to generate optimal advertisements from input data.

[0818] A "prompt sentence" is an input sentence used to request the generative AI model to generate an ad.

[0819] "Attributes" are information that indicates the age, sex, interests, behavior, etc. of the person being analyzed.

[0820] "Situation" is information that indicates the environment in which the analysis target is placed and the situation at that time.

[0821] The present invention relates to a system that generates and distributes advertisements in real time that are optimized for the current situation based on video and audio data acquired from a camera and a microphone. Specific embodiments for implementing the present invention will be described below.

[0822] Data collection

[0823] The server collects video and audio data in real time from cameras and microphones installed in the store. The video data consists of successive image frames sent from the camera, and the audio data is digitized sound waves sent from the microphone. The collected data is temporarily stored in a database for analysis. The hardware used is a standard camera (e.g., a 1080p webcam) and a high-sensitivity microphone (e.g., a USB-connected condenser microphone).

[0824] Data analysis

[0825] The server analyzes the collected video data and identifies people using a facial recognition algorithm. It then infers their attributes (e.g., age and gender) from the identified people. It also converts the audio data into text using a speech recognition engine and analyzes the content and context of the conversation. The software used includes OpenCV (for video analysis) and the SpeechRecognition library (for audio analysis).

[0826] Ad Generation

[0827] Based on the information obtained through the analysis, the server compares it with pre-prepared ad generation information to generate the optimal ad. Specifically, it selects advertising materials based on the analysis results and creates a customized ad video using a generative AI model. The generative AI model uses OpenAI's API. This ad generation process makes it possible to quickly create personalized ads for each user. An example of a prompt sentence when generating a generated ad is as follows:

[0828] Generate a personalized advertisement for a 25-year-old female interested in new fashion trends.

[0829] Ad serving

[0830] The terminal (smart device) receives the advertising video generated from the server and immediately displays it. The terminal then plays the advertising video on a screen for passersby and users to view. Specific smart devices include smartphones and head-mounted displays.

[0831] Pay-as-you-go model

[0832] The server collects ad playback data, recording the number of plays and playback time. Based on this data, a pay-per-view model is applied to calculate fees for advertisers. This allows advertisers to deliver ads in a cost-effective manner. In addition, user interactions with ads displayed on smart devices can be analyzed, which can be used to further analyze the effectiveness of advertising.

[0833] Specific examples

[0834] Example 1: Supermarket

[0835] Cameras and microphones installed at the entrance of supermarkets capture video and audio recordings of families entering the store in real time. The server identifies parents and children from the video and analyzes audio conversations about "children's toys." Based on this, the server generates promotional advertisements for children's toys and instantly distributes them to in-store digital signage. The same advertisements are also displayed in real time on smartphone apps.

[0836] Example 2: Inside a station

[0837] Cameras and microphones installed inside the station capture images of passersby and audio of them talking about "travel." The server analyzes the data, identifies age groups and genders, and generates advertisements for travel agencies. These advertisements are displayed on digital signage, promoting travel products to passersby, while also being displayed in real time on head-mounted displays.

[0838] As explained above, the present invention makes it possible to generate and distribute advertisements that are optimal for each situation in real time, thereby maximizing the effectiveness of advertisements.

[0839] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0840] Step 1:

[0841] The server acquires video data from cameras installed inside the store. The server acquires successive image frames and temporarily stores them in a database for analysis. At this time, the data input from the camera is acquired as a series of still image data, which is output as video data.

[0842] Step 2:

[0843] The server simultaneously captures audio data from a microphone. The sound waves coming from the microphone are digitized and stored as an audio file in a database on the server. This audio data is used in subsequent analysis steps.

[0844] Step 3:

[0845] The server applies a facial recognition algorithm to the captured video data. Specifically, it uses the OpenCV library to perform face detection for each frame, identify specific people, and infer their attributes. The input is the video data, and the output is the attribute information of the identified people (e.g., age, gender).

[0846] Step 4:

[0847] The server converts the voice data into text using the SpeechRecognition library. The voice data is input into a speech recognition engine, which then outputs analyzed text data. Based on this text data, the content of the conversation and the situation are analyzed to understand the subject's interests and behavior.

[0848] Step 5:

[0849] The server compares the results of video and audio analysis to identify the overall situation and the target's interests and behavior. This determines what type of advertising will be most effective. The input is the results of video and audio analysis, and the output is a situational judgment based on the analysis results.

[0850] Step 6:

[0851] Based on the analysis results, the server generates the optimal advertisement from pre-prepared information for advertisement generation. Here, a generative AI model is used to generate a customized advertisement by providing a prompt. For example, an advertisement is generated based on the prompt "Generate a personalized advertisement for a 25-year-old female interested in new fashion trends." The input is the prompt and information for advertisement generation, and the output is the generated advertisement content.

[0852] Step 7:

[0853] The server sends the generated advertisement to the terminal (e.g., smart device). The terminal immediately displays the advertisement content received from the server. The input is the generated advertisement content, and the output is the advertisement displayed on the terminal.

[0854] Step 8:

[0855] When a user interacts with an advertisement on their device, the device sends this interaction data to a server. The server analyzes this data and evaluates the effectiveness of the advertisement and the user's response. The input is the interaction data, and the output is the analyzed evaluation results.

[0856] Step 9:

[0857] The server finally applies a pay-per-view model based on the ad playback data and interaction data to calculate the fee for the advertiser. The input is the playback data and interaction data, and the output is the fee calculation result.

[0858] 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.

[0859] The present invention relates to a system that generates and distributes advertisements in real time that are optimized for the situation and user emotions based on video and audio data acquired from a camera and a microphone. Specific embodiments for carrying out the present invention will be described below.

[0860] Data collection

[0861] Subject: Server

[0862] The server captures video and audio data in real time from the installed cameras and microphones. These data are temporarily stored in a database for analysis. Video data consists of a series of image frames, and audio data is digitized sound waves.

[0863] Data analysis

[0864] Subject: Server

[0865] The server analyzes the collected video data and identifies people using a facial recognition algorithm. It then infers their attributes (e.g., age and gender) from the detected people. It also sends the audio data to a speech recognition engine, which converts it into text. This allows the content and context of the conversation to be analyzed.

[0866] emotion recognition

[0867] Subject: Server

[0868] The server recognizes the user's emotions by analyzing facial expressions from video data and tone of voice from audio data. The emotion recognition engine identifies emotions such as anger, joy, and surprise, and uses the results for further analysis.

[0869] Ad Generation

[0870] Subject: Server

[0871] The server generates the optimal advertisement based on the results of attribute and situation analysis and emotion recognition, comparing them with pre-prepared advertisement generation information. Specifically, it selects advertising materials based on the analysis results and emotion recognition results, and creates a customized advertising video using a generative AI model. In this process, the content that best suits the user's emotional state is reflected.

[0872] Ad serving

[0873] Subject: Device

[0874] The terminal (digital signage) receives the advertising video generated from the server and immediately displays it. The terminal then plays the advertising video on the display screen for the user to view. This allows the optimal advertising to be delivered in real time according to the situation.

[0875] Pay-as-you-go model

[0876] Subject: Server

[0877] The server collects ad playback data, recording the number of plays and the playback time. Based on this data, a pay-per-view model is applied to calculate the fees for advertisers, enabling them to deliver ads in a cost-effective manner.

[0878] Specific examples

[0879] Example 1: Supermarket

[0880] Cameras and microphones installed at the entrance of the supermarket capture video and audio of families entering the store in real time. The server identifies parents and children from the video and analyzes the audio content of conversations about "toys." At the same time, an emotion recognition engine detects whether the child is excited. Based on this, the server generates a promotional advertisement for children's toys and instantly distributes it to digital signage within the store.

[0881] Example 2: Inside a station

[0882] Cameras and microphones installed inside the station capture images of passersby and audio of them talking about "travel." The server analyzes the data, identifies their age group and gender, and uses an emotion recognition engine to assess their excitement level. Based on this, the server generates advertisements for travel agencies and displays them on digital signage.

[0883] As explained above, the present invention makes it possible to generate and distribute optimal advertisements in real time according to the situation and user emotions, thereby maximizing the effectiveness of advertisements.

[0884] The processing flow will be explained below.

[0885] Step 1:

[0886] Subject: Server

[0887] The camera and microphone begin streaming video and audio data. Video data is collected frame by frame, and audio data is digitized and captured in real time. This data is temporarily stored in a database for analysis.

[0888] Step 2:

[0889] Subject: Server

[0890] The server analyzes the collected video data and uses facial recognition algorithms to identify people. Using a face detection library such as OpenCV, it detects faces in the video frames and estimates attributes such as age and gender from the faces.

[0891] Step 3:

[0892] Subject: Server

[0893] Voice data is sent to a speech recognition engine and converted to text in real time. Speech recognition services such as Google Cloud Speech-to-Text API or Watson Speech to Text are used to transcribe the conversation and perform text analysis to identify interests and behaviors.

[0894] Step 4:

[0895] Subject: Server

[0896] The results of the analysis of video and audio data are integrated to identify the target's attributes and situation, thereby clarifying the user's basic profile and current behavior and interests.

[0897] Step 5:

[0898] Subject: Server

[0899] Based on video and audio data, the system analyzes the user's facial expressions and tone of voice to recognize their emotions. Using an emotion recognition engine, it identifies emotions such as anger, joy, and surprise, and uses the results in the next ad generation process.

[0900] Step 6:

[0901] Subject: Server

[0902] Based on the analysis results and emotion recognition results, the optimal advertising materials are selected by comparing them with pre-prepared information for advertising generation.Then, the selected materials and analysis results are input into the generation AI model to generate a customized advertising video that matches the emotions.

[0903] Step 7:

[0904] Subject: Device

[0905] The digital signage terminal receives the generated advertising video from the server, and immediately plays the downloaded advertising video using a file transfer protocol and displays it on the terminal screen.

[0906] Step 8:

[0907] Subject: Server

[0908] The device collects data on the ads played, recording the number of plays and the play time, and periodically sends this playback data to the server.

[0909] Step 9:

[0910] Subject: Server

[0911] The server applies a pay-per-view model based on the collected playback data to calculate the fees for advertisers, calculates the total amount based on a fee schedule according to the number of plays and the playback time, and provides a fee report to the advertiser.

[0912] The above steps enable the creation and distribution of optimal advertisements in real time according to the situation and user emotions, maximizing the effectiveness of the advertisements.

[0913] Example 2

[0914] 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."

[0915] In modern advertising delivery systems, it is difficult to generate and deliver optimal advertisements in real time that match user attributes and emotions. Maximizing the cost-effectiveness of advertising by introducing an efficient pay-per-use model for advertisers is also a challenge. Conventional systems are unable to perform such real-time data analysis and advertisement generation, making it impossible to deliver effective advertisements that match user interests and emotions.

[0916] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for acquiring video data, means for acquiring audio data, means for analyzing the video data and the audio data to identify person attributes and situations, means for recognizing a user's emotions from the video data and the audio data, means for generating an optimal advertisement from pre-prepared advertisement generation information based on the analysis results and the user's emotion recognition results, and means for delivering the generated advertisement via a display device. This enables real-time advertisement generation and delivery based on user attributes and emotions, maximizing the effectiveness of advertisements. In addition, by applying a pay-per-view model based on the number of impressions and playback time, advertisers can be effectively charged advertising fees.

[0917] "Video data" is a digital representation of visual information captured using a device such as a camera.

[0918] "Audio data" refers to digitized sound waves recorded using a device such as a microphone.

[0919] "Person attributes" are characteristics about an individual, such as age, gender, or identity, that are inferred using facial recognition algorithms or other methods.

[0920] "User emotion" refers to an emotional state, such as anger, joy, or surprise, that is recognized through facial expression analysis from video data and tone analysis from audio data.

[0921] "Information for advertisement generation" refers to materials such as text, images, and videos prepared in advance for generating advertisements, as well as related data.

[0922] A "generative AI model" is an artificial intelligence model that generates customized output data based on input data and specified prompts.

[0923] A "prompt sentence" is a textual input used to give specific instructions to a generative AI model.

[0924] A "pay-as-you-go model" is a pricing model in which the fee is calculated based on the number of times an advertisement is displayed and the duration of its playback.

[0925] A "display device" refers to a display or digital signage used to allow users to view the generated advertisement.

[0926] "Playback data" refers to data relating to playback status, such as the number of times a generated advertisement is played and the playback time.

[0927] This invention relates to a system that generates and distributes advertisements in real time that are optimized for the situation and user emotions based on video and audio data acquired from a camera and microphone.

[0928] The server collects video and audio data in real time from cameras and microphones installed via the network. The video data consists of successive image frames, and the audio data is digitized sound waves. These data are temporarily stored in a database for analysis. Specifically, the camera collects video at 30 frames per second, and the microphone collects audio data at 44.1 kHz per second.

[0929] The server analyzes the collected video data and identifies people using a facial recognition algorithm. For example, it uses image processing software (e.g., OpenCV). Attributes such as age and gender are inferred by analyzing the feature points of the detected face. At the same time, the audio data is sent to a speech recognition engine (e.g., Google Speech-to-Text API) and converted into text. This allows the content and situation of the conversation to be extracted. For example, the keyword "travel" can be extracted from the audio data.

[0930] The server then analyzes the user's facial expressions from the video data and the tone of voice from the audio data to recognize the user's emotions. It uses an emotion recognition engine (e.g., Amazon Rekognition or Google Cloud Natural Language API) to determine emotional states such as anger, joy, and surprise. For example, it can detect whether the user is smiling and recognize excitement from the tone of their voice.

[0931] Next, the server generates the optimal advertisement based on the analysis results and emotion recognition results, comparing them with pre-prepared advertisement generation information. It selects advertising materials (text, images, videos) and uses a generative AI model (e.g., GPT-4) to create a customized advertising video. During this process, it issues instructions by sending prompt text to the generative AI model. For example, a prompt text such as "An excited man in his 60s who is interested in traveling" could be sent to the generative AI model to generate a customized advertising video for a travel agency.

[0932] The generated advertising video is sent to a terminal (digital signage) and played on the display, allowing users to view it. For example, a travel advertising video can be played on a display in a store to attract the attention of passersby.

[0933] Finally, the server collects ad play data, recording the number of plays and play time. Based on this data, it applies a pay-per-view model and calculates the fee to the advertiser. For example, it records that a particular ad was played 100 times and the total play time was 50 minutes, and calculates the cost to be charged to the advertiser.

[0934] As a result, the system of the present invention enables real-time generation and distribution of advertisements based on user attributes and emotions, maximizing the effectiveness of advertisements. In addition, by applying a pay-per-view model based on the number of impressions and playback time, advertisers can be effectively charged advertising fees.

[0935] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0936] Step 1:

[0937] Data collection:

[0938] The server receives real-time video and audio data from cameras and microphones via the network. Specifically, the cameras capture video data at 30 frames per second, and the microphones capture audio data at 44.1 kHz per second. This data is temporarily stored in a database for analysis.

[0939] Input: Video and audio data obtained from the camera and microphone.

[0940] Output: Video and audio data stored in a database for analysis.

[0941] Step 2:

[0942] Data Analysis:

[0943] The server analyzes the collected video data. Specifically, it uses image processing software (e.g., OpenCV) to run a facial recognition algorithm to identify people. Attributes such as age and gender are inferred by analyzing facial feature points. At the same time, the server sends the audio data to a speech recognition engine (e.g., Google Speech-to-Text API) to convert it into text.

[0944] Input: Video and audio data obtained from the analysis database.

[0945] Output: Identified person's attributes (age, gender) and transcribed audio data.

[0946] Step 3:

[0947] Emotion recognition:

[0948] The server analyzes the user's facial expressions from the video data and the tone of voice from the audio data to recognize the user's emotions. An emotion recognition engine (e.g., Amazon Rekognition or Google Cloud Natural Language API) is used to determine the user's emotional state, such as anger, joy, or surprise.

[0949] Input: Identified person attributes and transcribed audio data.

[0950] Output: The user's emotional state (e.g., anger, joy, surprise).

[0951] Step 4:

[0952] Ad Generation:

[0953] The server generates the optimal ad based on the analysis results and emotion recognition results, comparing them with pre-prepared ad generation information. It selects ad materials (text, images, videos) and creates a customized ad video using a generative AI model (e.g., GPT-4). During this process, it sends a prompt to the generative AI model.

[0954] Input: User's emotional state, identified person attributes, and information for ad generation.

[0955] Output: A customized advertising video.

[0956] Step 5:

[0957] Ad serving:

[0958] The terminal (digital signage) receives the advertising video generated from the server and immediately displays it. The terminal plays the advertising video on the display so that the user can view it.

[0959] Input: The customized ad video sent from the server.

[0960] Output: Ad video played on the display.

[0961] Step 6:

[0962] Pay-as-you-go:

[0963] The server collects ad playback data, records the number of plays and the playback time, and calculates the fee to the advertiser based on this data using a pay-per-view model.

[0964] Input: Playback data of the ad video played on the display.

[0965] Output: Fees charged to advertisers.

[0966] (Application example 2)

[0967] 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."

[0968] Conventional ad delivery systems generate and deliver ads based on user attributes and circumstances, but they do not fully consider the user's real-time emotions or specific circumstances, which means that the effectiveness of the ads is not maximized.In addition, there is a lack of technology for generating and delivering effective ads, and the cost-effectiveness for advertisers is also insufficient.

[0969] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0970] In this invention, the server includes means for acquiring video data, means for acquiring audio data, means for analyzing the video data and the audio data to identify target attributes and situations, means for analyzing user emotions, means for generating an optimal advertisement from pre-prepared information for advertisement generation based on the results of the analysis and the emotion analysis, and means for delivering the generated advertisement via a display device. This makes it possible to generate and deliver an optimal advertisement according to the user's real-time emotions and specific situations, thereby maximizing advertising effectiveness.

[0971] "Video data" refers to visual information acquired by an image sensor such as a camera.

[0972] "Audio data" refers to auditory information acquired by an acoustic sensor such as a microphone.

[0973] "Analysis" refers to computational processing to extract or infer specific information from acquired data.

[0974] "Attributes" refer to the characteristics or properties of a subject that are determined through analysis, and include, for example, age and gender.

[0975] "Situation" refers to the subject's current environmental and behavioral state as determined by analysis.

[0976] "Emotion" refers to the subject's psychological state as determined by analysis, and includes anger, joy, surprise, etc.

[0977] "Information for advertisement generation" refers to materials and data prepared in advance for generating an advertisement.

[0978] The term "display device" refers to a device for visually presenting the generated advertisement to a user.

[0979] "Advertising materials" refers to the individual elements and content that make up an advertisement.

[0980] "Customized" refers to advertising that is generated and tailored to specific requirements and conditions.

[0981] A "pay-as-you-go model" refers to a pricing system in which charges are based on the amount of usage or the number of times it is used.

[0982] A system for implementing the present invention has the following configuration: A server acquires and analyzes video data and audio data, and generates and distributes optimal advertisements. A detailed explanation is provided below with specific examples.

[0983] First, the server acquires video and audio data in real time from cameras and microphones. Video data is visual information acquired by image sensors such as cameras, and audio data is auditory information acquired by acoustic sensors such as microphones. These data are temporarily stored in a database for analysis.

[0984] The server then analyzes the collected video and audio data. The video data uses facial recognition algorithms to identify people and infer their attributes (e.g., age, gender, etc.), while the audio data is sent to a speech recognition engine, which converts it into text and analyzes it. This analysis process uses software libraries such as OpenCV and SpeechRecognition.

[0985] Furthermore, the server uses an emotion recognition engine to analyze the user's emotions. By analyzing the user's facial expressions from the video data and the tone of voice from the audio data, emotions such as anger, joy, and surprise can be recognized. This emotion analysis combines multiple algorithms.

[0986] The server generates optimal advertisements based on the results of these analyses and sentiment analysis. Specifically, it selects the optimal advertising materials for the user's attributes and emotional state from pre-prepared advertisement generation information, and creates customized advertisements using a generative AI model.

[0987] Finally, the generated advertisement is delivered to a display device, such as a smartphone or digital signage. The user can watch the advertisement in real time, optimized for the situation and emotions of the moment. The terminal device plays the advertisement video and provides it visually to the user.

[0988] For example, when a user is using a smartphone, the smartphone's camera and microphone collect data and detect that the user is having a pleasant conversation. The server generates an advertisement for an entertainment app that matches that pleasant emotion and displays it on the smartphone. This application maximizes the effectiveness of advertising.

[0989] Example prompt sentence:

[0990] "The user is showing happy emotions. Please generate an ad for an entertainment app. User demographics: Age 25, Gender Male"

[0991] As described above, the system of the present invention is capable of generating and delivering optimal advertisements that correspond to the user's real-time emotions and specific situations, thereby maximizing the effectiveness of advertisements and improving cost-effectiveness for advertisers.

[0992] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0993] Step 1:

[0994] The server acquires video and audio data from the camera and microphone in real time. As input, it receives the real-time output of the camera and microphone and temporarily stores it as video and audio data in a database within the server. In this process, the camera captures successive image frames and the microphone generates a digital audio signal.

[0995] Step 2:

[0996] The server analyzes the acquired video data. As input, it receives the video data acquired in step 1, identifies the person using a facial recognition algorithm, and then estimates attributes such as age and gender. For example, it uses OpenCV to detect faces from image frames and estimates attributes based on the results. As output, it obtains attribute information for the identified person.

[0997] Step 3:

[0998] The server analyzes the acquired voice data. As input, it receives the voice data acquired in step 1 and converts it into text using a voice recognition engine. For example, it uses the SpeechRecognition library to convert the voice data into text, and analyzes the topic and situation of the conversation from the text content. As output, it obtains text data and the analysis results.

[0999] Step 4:

[1000] The server analyzes the user's emotions from the video and audio data. It receives the outputs of steps 2 and 3 as input and uses an emotion recognition engine to analyze facial expressions and tone of voice. Specifically, it reads facial expressions from image frames and identifies emotions from audio tones. The output is emotional information such as anger, joy, and surprise.

[1001] Step 5:

[1002] The server generates the optimal advertisement based on the analysis results and the results of sentiment analysis. It receives the outputs of steps 2, 3, and 4 as input and selects the optimal advertising material from pre-prepared information for advertisement generation. It uses a generative AI model to generate a customized advertisement that best suits the user's attributes and emotions. The output is a customized advertising video.

[1003] Step 6:

[1004] The server delivers the generated advertisement to the terminal. It receives the customized advertisement video output from step 5 as input and sends it to the terminal. The terminal plays the received advertisement video on a display device (e.g., a smartphone or digital signage). This process allows the user to watch the optimized advertisement in real time.

[1005] Step 7:

[1006] The server collects playback data of the displayed advertisements and calculates fees based on a pay-per-view model. As input, it receives data on the number of plays and the playback time sent from the terminal, and calculates the fee for the advertiser based on that data. As output, it obtains the final fee calculation result.

[1007] 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.

[1008] 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.

[1009] 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.

[1010] [Fourth embodiment]

[1011] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1012] 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.

[1013] 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).

[1014] 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.

[1015] 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.

[1016] 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).

[1017] 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.

[1018] 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.

[1019] 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.

[1020] 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.

[1021] 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.

[1022] 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.

[1023] 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."

[1024] The present invention relates to a system that generates and distributes advertisements in real time that are optimized for the current situation based on video and audio data acquired from a camera and a microphone. Specific embodiments for implementing the present invention will be described below.

[1025] Data collection

[1026] Subject: Server

[1027] The server collects real-time video and audio data from the installed cameras and microphones. The video data consists of successive image frames sent from the cameras, and the audio data is digitized sound waves sent from the microphones. The collected data is temporarily stored in a database for analysis.

[1028] Data analysis

[1029] Subject: Server

[1030] The server analyzes the collected video data and identifies people using a facial recognition algorithm. From this identified person, the server infers their attributes (e.g., age and gender). It also uses a voice recognition engine to convert the voice data into text and analyzes the conversation content and situation. This makes it possible to understand the interests and behavior of the target.

[1031] Ad Generation

[1032] Subject: Server

[1033] The server generates optimal advertisements based on the information obtained through the analysis and compares it with pre-prepared advertisement generation information. Specifically, it selects advertising materials based on the analysis results and creates customized advertising videos using a generative AI model. This advertisement generation process makes it possible to create personalized advertisements for each user in a short amount of time.

[1034] Ad serving

[1035] Subject: Device

[1036] The terminal (digital signage) receives the advertising video generated from the server and displays it instantly. The terminal then plays this advertising video on the screen for passersby and users to view. This allows for the delivery of advertising that is appropriate for the situation in real time, which can be expected to be more effective.

[1037] Pay-as-you-go model

[1038] Subject: Server

[1039] The server collects ad playback data, recording the number of plays and the playback time. Based on this data, a pay-per-view model is applied to calculate the fees for advertisers, enabling them to deliver ads in a cost-effective manner.

[1040] Specific examples

[1041] Example 1: Supermarket

[1042] Cameras and microphones installed at the entrance of supermarkets capture video and audio of families entering the store in real time. The server identifies parents and children from the video and analyzes the audio for conversations about "children's toys." Based on this, the server generates promotional advertisements for children's toys and instantly distributes them to digital signage in the store.

[1043] Example 2: Inside a station

[1044] Cameras and microphones installed inside the station capture images of passersby and audio of them talking about "travel." The server analyzes the data, identifies age groups and genders, and generates advertisements for travel agencies. These advertisements are displayed on digital signage, promoting travel products to passersby.

[1045] As explained above, the present invention makes it possible to generate and distribute advertisements that are optimal for each situation in real time, thereby maximizing the effectiveness of advertisements.

[1046] The processing flow will be explained below.

[1047] Step 1:

[1048] Subject: Server

[1049] The server starts streaming video and audio data from the installed cameras and microphones. The video data is captured frame by frame, and the audio data is collected in real time. These data are temporarily stored in a database for analysis.

[1050] Step 2:

[1051] Subject: Server

[1052] The server analyzes the video data and identifies people using a facial recognition algorithm. It uses a face detection library such as OpenCV to detect faces in the frame and estimate attributes such as age and gender for the detected faces.

[1053] Step 3:

[1054] Subject: Server

[1055] The server sends the voice data to a speech recognition engine, which converts it into text in real time. It then transcribes the conversation using speech recognition services such as Google Cloud Speech-to-Text API or Watson Speech to Text, and uses text analysis to identify interests and behaviors.

[1056] Step 4:

[1057] Subject: Server

[1058] The server combines the results of facial and voice recognition analysis to identify target attributes and context, which serves as the basis for generating customized advertisements that reflect the user's interests and current behavior.

[1059] Step 5:

[1060] Subject: Server

[1061] The server compares the analysis results with pre-prepared information for generating advertisements and selects the most suitable advertising materials. For example, if the customer is a family, it will select advertising materials for products aimed at children. This selected material is then input into the generative AI model.

[1062] Step 6:

[1063] Subject: Server

[1064] The generative AI model generates customized ad videos based on the selected ad materials and analysis results, and uses facial recognition results to quickly create personalized ads relevant to users.

[1065] Step 7:

[1066] Subject: Device

[1067] The terminal (digital signage) receives the generated advertising video from the server and immediately displays it. The video file is downloaded using a file transfer protocol and played on the display via a video player application.

[1068] Step 8:

[1069] Subject: Server

[1070] The server collects ad playback data, records the number of plays and the playback time, and periodically transmits ad playback event data from the digital signage terminal.

[1071] Step 9:

[1072] Subject: Server

[1073] The server calculates the fee for the advertiser based on the recorded playback data using a pay-per-view model, refers to a fee schedule based on the number of playbacks and duration, calculates the total amount, and provides a fee report to the advertiser.

[1074] The above steps enable the generation and distribution of optimal advertisements in real time according to the situation, maximizing the effectiveness of advertising.

[1075] Example 1

[1076] 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."

[1077] Conventional ad delivery systems have had difficulty generating and delivering optimal ads in real time based on target attributes and circumstances. In particular, they have been unable to effectively utilize video and audio data to instantly provide personalized ads based on analysis results. Furthermore, they lacked an efficient method for accurately calculating fees when applying a pay-per-use model to advertisers. As a result, maximizing advertising effectiveness and improving cost-effectiveness have not been fully achieved.

[1078] 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.

[1079] In this invention, the server includes a means for acquiring video data, a means for acquiring audio data, and an analysis means, which allows the server to generate and distribute optimal advertisements in real time according to the attributes and circumstances of the target audience, thereby maximizing the effectiveness of the advertisements.

[1080] "Video data" refers to digital data consisting of successive image frames captured in real time from an image capture device such as a camera.

[1081] "Audio data" refers to data obtained by digitizing sound waves acquired from a sound acquisition device such as a microphone.

[1082] The "analysis results" are the results of calculations performed using the acquired video and audio data to identify the attributes and circumstances of the target.

[1083] "Information for advertisement generation" refers to materials and data prepared in advance that will be the basis for the advertisement to be generated.

[1084] A "generative AI model" is an artificial intelligence model that uses machine learning algorithms to generate content based on input prompts.

[1085] A "prompt" is an instruction or question input to a generative AI model that provides the requirements for generating a specific output.

[1086] A "display device" is a screen or display for visually displaying the generated advertisement.

[1087] The "pay-as-you-go model" is a billing method in which fees are calculated based on the number of times an advertisement is displayed and the duration of its playback.

[1088] The present invention is a system that generates and distributes advertisements in real time that are optimized for the current situation based on video and audio data acquired from a camera and microphone. Specific embodiments for implementing the present invention will be described below.

[1089] Data collection

[1090] server

[1091] The server acquires video and audio data in real time from the installed cameras and microphones. This video data consists of successive image frames sent from the cameras, and the audio data is digitized sound waves sent from the microphones. Specifically, the server acquires the video stream from the cameras using RTSP (Real-Time Streaming Protocol), and the audio data is received via the Internet. This data is temporarily stored in a database for analysis.

[1092] Data analysis

[1093] server

[1094] The server analyzes the collected video and audio data. Using a facial recognition algorithm, it detects the faces of people in the video frames and infers their attributes (age, gender). Specifically, it uses OpenCV (an open-source computer vision library). It also uses the Google Cloud Speech-to-Text API as a speech recognition engine to convert the audio data into text. The text data can then be analyzed to understand the content and context of the conversation.

[1095] Ad Generation

[1096] server

[1097] Based on the results of the data analysis, the server uses a generative AI model (e.g., GPT-4) to generate the optimal advertisement from pre-prepared information for advertisement generation. The server inputs the prompt text into the generative AI model, which generates the advertisement script as a result. The generated script is then used in video editing software such as Adobe Premiere Pro to edit and complete the advertisement video. An example of a specific prompt text is, "Based on the camera footage and microphone audio data, generate an advertisement that is optimal for the current situation. As a specific scenario, please create an advertisement for a family entering a supermarket and talking about children's toys."

[1098] Ad serving

[1099] Terminal

[1100] Devices such as digital signage receive advertising videos generated from the server and instantly display them on the screen. The device uses a basic HTTP request to download the latest advertising video from the server and then plays the video file for passersby or users to view, enabling real-time advertising.

[1101] Pay-as-you-go model

[1102] server

[1103] The server collects ad playback data, recording the number of plays and the playback time. It receives access logs and playback logs sent from the device and uses this data to apply a pay-per-view model to calculate fees for advertisers. This allows advertisers to deliver ads in a cost-effective manner.

[1104] Specific examples

[1105] Example 1: Supermarket

[1106] Cameras and microphones installed at the entrance of the supermarket capture video and audio of families entering the store in real time. The server identifies parents and children from the video and analyzes the conversation about "children's toys" from the audio. Based on the analysis results, prompts are input into a generative AI model to generate promotional advertisements for children's toys. The completed advertising video is distributed to and displayed on digital signage terminals in the store.

[1107] Example 2: Inside a station

[1108] Cameras and microphones installed inside the station capture images of passersby and audio of them talking about "travel." The server analyzes this data, identifies age groups and genders, and then generates advertisements for travel agencies. These advertisements are displayed on digital signage, promoting travel products to passersby.

[1109] This makes it possible to generate and deliver optimal advertisements in real time according to the situation, maximizing the effectiveness of advertising.

[1110] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1111] Step 1:

[1112] Data collection

[1113] The server acquires video and audio data from the camera and microphone in real time. Specifically, the server receives a video stream of continuous frames from the camera using RTSP (Real-Time Streaming Protocol). It also receives audio data, which is digitized sound waves, from the microphone via the Internet. This data is temporarily stored in a database for analysis.

[1114] Input: Video stream from camera, audio data from microphone

[1115] Output: Video and audio data stored in a database for analysis

[1116] Step 2:

[1117] Data analysis

[1118] The server runs a facial recognition algorithm (e.g., OpenCV) to detect faces from the collected video data. This algorithm detects faces in the video frames and infers their attributes (age, gender). Additionally, the audio data is converted to text using the Google Cloud Speech-to-Text API. This allows the audio to be analyzed for conversation content and context, providing information to understand the subject's interests and behavior.

[1119] Input: Video data, audio data

[1120] Output: Attributes of identified person, audio data converted to text

[1121] Step 3:

[1122] Ad Generation

[1123] The server uses a generative AI model (e.g., GPT-4) to generate optimal advertisements based on the results of data analysis. The server inputs prompt sentences into the generative AI model and obtains the resulting advertisement script. It then uses video editing software such as Adobe Premiere Pro to edit and generate advertisement videos based on the generated script.

[1124] Input: Identified person's attributes, transcribed voice data, prompt

[1125] Output: Ad script, finished ad video

[1126] Step 4:

[1127] Ad serving

[1128] The terminal (digital signage) receives the generated advertising video from the server and plays it on the screen for immediate display. The terminal uses a basic HTTP request to download the latest advertising video from the server and plays the received video file.

[1129] Input: Ad video

[1130] Output: Ad video displayed on the screen

[1131] Step 5:

[1132] Pay-as-you-go model

[1133] The server collects data each time an advertisement is played, recording the number of plays and the play time. Specifically, it receives access logs and play logs sent from the terminal and applies a pay-per-view model based on that data. This is used to calculate the fee for the advertiser.

[1134] Input: Access log and playback log from the device

[1135] Output: Pricing results based on number of plays and duration

[1136] (Application example 1)

[1137] 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."

[1138] Modern commercial facilities and brick-and-mortar stores require real-time advertising that meets customer needs and interests. However, existing systems have difficulty accurately analyzing customer attributes and conversation content, and quickly generating and delivering individually customized advertisements. Another issue is that they are unable to meet the need for direct advertising display using smart devices.

[1139] 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.

[1140] In this invention, the server includes means for acquiring video data, means for acquiring audio data, means for analyzing the video data and the audio data and identifying target attributes and situations, means for generating an optimal advertisement from pre-prepared advertisement generation information based on the analysis results, means for delivering the generated advertisement via a display device, and means for displaying the advertisement on a smart device in real time. This makes it possible to quickly deliver an optimized advertisement to a smart device based on the real-time attributes and conversation content of a customer.

[1141] "Video data" refers to data made up of successive image frames captured using a photographic device such as a camera.

[1142] "Audio data" refers to data obtained by digitizing sound waves acquired using an acoustic device such as a microphone.

[1143] The "analyzing means" is a means for executing a process to determine a person's attributes and situation based on the acquired video data and audio data.

[1144] "Information for advertisement generation" is a collection of materials and information prepared in advance and used to generate optimal advertisements based on the analysis results.

[1145] A "display device" is a hardware device for providing the generated advertisement to the viewer, and includes digital signage, smartphones, etc.

[1146] "Smart devices" is a general term for mobile terminals and wearable devices that are connected to the Internet and have the ability to run applications.

[1147] A "generative AI model" is an algorithm or machine learning model that uses artificial intelligence to generate optimal advertisements from input data.

[1148] A "prompt sentence" is an input sentence used to request the generative AI model to generate an ad.

[1149] "Attributes" are information that indicates the age, sex, interests, behavior, etc. of the person being analyzed.

[1150] "Situation" is information that indicates the environment in which the analysis target is placed and the situation at that time.

[1151] The present invention relates to a system that generates and distributes advertisements in real time that are optimized for the current situation based on video and audio data acquired from a camera and a microphone. Specific embodiments for implementing the present invention will be described below.

[1152] Data collection

[1153] The server collects video and audio data in real time from cameras and microphones installed in the store. The video data consists of successive image frames sent from the camera, and the audio data is digitized sound waves sent from the microphone. The collected data is temporarily stored in a database for analysis. The hardware used is a standard camera (e.g., a 1080p webcam) and a high-sensitivity microphone (e.g., a USB-connected condenser microphone).

[1154] Data analysis

[1155] The server analyzes the collected video data and identifies people using a facial recognition algorithm. It then infers their attributes (e.g., age and gender) from the identified people. It also converts the audio data into text using a speech recognition engine and analyzes the content and context of the conversation. The software used includes OpenCV (for video analysis) and the SpeechRecognition library (for audio analysis).

[1156] Ad Generation

[1157] Based on the information obtained through the analysis, the server compares it with pre-prepared ad generation information to generate the optimal ad. Specifically, it selects advertising materials based on the analysis results and creates a customized ad video using a generative AI model. The generative AI model uses OpenAI's API. This ad generation process makes it possible to quickly create personalized ads for each user. An example of a prompt sentence when generating a generated ad is as follows:

[1158] Generate a personalized advertisement for a 25-year-old female interested in new fashion trends.

[1159] Ad serving

[1160] The terminal (smart device) receives the advertising video generated from the server and immediately displays it. The terminal then plays the advertising video on a screen for passersby and users to view. Specific smart devices include smartphones and head-mounted displays.

[1161] Pay-as-you-go model

[1162] The server collects ad playback data, recording the number of plays and playback time. Based on this data, a pay-per-view model is applied to calculate fees for advertisers. This allows advertisers to deliver ads in a cost-effective manner. In addition, user interactions with ads displayed on smart devices can be analyzed, which can be used to further analyze the effectiveness of advertising.

[1163] Specific examples

[1164] Example 1: Supermarket

[1165] Cameras and microphones installed at the entrance of supermarkets capture video and audio recordings of families entering the store in real time. The server identifies parents and children from the video and analyzes audio conversations about "children's toys." Based on this, the server generates promotional advertisements for children's toys and instantly distributes them to in-store digital signage. The same advertisements are also displayed in real time on smartphone apps.

[1166] Example 2: Inside a station

[1167] Cameras and microphones installed inside the station capture images of passersby and audio of them talking about "travel." The server analyzes the data, identifies age groups and genders, and generates advertisements for travel agencies. These advertisements are displayed on digital signage, promoting travel products to passersby, while also being displayed in real time on head-mounted displays.

[1168] As explained above, the present invention makes it possible to generate and distribute advertisements that are optimal for each situation in real time, thereby maximizing the effectiveness of advertisements.

[1169] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1170] Step 1:

[1171] The server acquires video data from cameras installed inside the store. The server acquires successive image frames and temporarily stores them in a database for analysis. At this time, the data input from the camera is acquired as a series of still image data, which is output as video data.

[1172] Step 2:

[1173] The server simultaneously captures audio data from a microphone. The sound waves coming from the microphone are digitized and stored as an audio file in a database on the server. This audio data is used in subsequent analysis steps.

[1174] Step 3:

[1175] The server applies a facial recognition algorithm to the captured video data. Specifically, it uses the OpenCV library to perform face detection for each frame, identify specific people, and infer their attributes. The input is the video data, and the output is the attribute information of the identified people (e.g., age, gender).

[1176] Step 4:

[1177] The server converts the voice data into text using the SpeechRecognition library. The voice data is input into a speech recognition engine, which then outputs analyzed text data. Based on this text data, the content of the conversation and the situation are analyzed to understand the subject's interests and behavior.

[1178] Step 5:

[1179] The server compares the results of video and audio analysis to identify the overall situation and the target's interests and behavior. This determines what type of advertising will be most effective. The input is the results of video and audio analysis, and the output is a situational judgment based on the analysis results.

[1180] Step 6:

[1181] Based on the analysis results, the server generates the optimal advertisement from pre-prepared information for advertisement generation. Here, a generative AI model is used to generate a customized advertisement by providing a prompt. For example, an advertisement is generated based on the prompt "Generate a personalized advertisement for a 25-year-old female interested in new fashion trends." The input is the prompt and information for advertisement generation, and the output is the generated advertisement content.

[1182] Step 7:

[1183] The server sends the generated advertisement to the terminal (e.g., smart device). The terminal immediately displays the advertisement content received from the server. The input is the generated advertisement content, and the output is the advertisement displayed on the terminal.

[1184] Step 8:

[1185] When a user interacts with an advertisement on their device, the device sends this interaction data to a server. The server analyzes this data and evaluates the effectiveness of the advertisement and the user's response. The input is the interaction data, and the output is the analyzed evaluation results.

[1186] Step 9:

[1187] The server finally applies a pay-per-view model based on the ad playback data and interaction data to calculate the fee for the advertiser. The input is the playback data and interaction data, and the output is the fee calculation result.

[1188] 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.

[1189] The present invention relates to a system that generates and distributes advertisements in real time that are optimized for the situation and user emotions based on video and audio data acquired from a camera and a microphone. Specific embodiments for carrying out the present invention will be described below.

[1190] Data collection

[1191] Subject: Server

[1192] The server captures video and audio data in real time from the installed cameras and microphones. These data are temporarily stored in a database for analysis. Video data consists of a series of image frames, and audio data is digitized sound waves.

[1193] Data analysis

[1194] Subject: Server

[1195] The server analyzes the collected video data and identifies people using a facial recognition algorithm. It then infers their attributes (e.g., age and gender) from the detected people. It also sends the audio data to a speech recognition engine, which converts it into text. This allows the content and context of the conversation to be analyzed.

[1196] emotion recognition

[1197] Subject: Server

[1198] The server recognizes the user's emotions by analyzing facial expressions from video data and tone of voice from audio data. The emotion recognition engine identifies emotions such as anger, joy, and surprise, and uses the results for further analysis.

[1199] Ad Generation

[1200] Subject: Server

[1201] The server generates the optimal advertisement based on the results of attribute and situation analysis and emotion recognition, comparing them with pre-prepared advertisement generation information. Specifically, it selects advertising materials based on the analysis results and emotion recognition results, and creates a customized advertising video using a generative AI model. In this process, the content that best suits the user's emotional state is reflected.

[1202] Ad serving

[1203] Subject: Device

[1204] The terminal (digital signage) receives the advertising video generated from the server and immediately displays it. The terminal then plays the advertising video on the display screen for the user to view. This allows the optimal advertising to be delivered in real time according to the situation.

[1205] Pay-as-you-go model

[1206] Subject: Server

[1207] The server collects ad playback data, recording the number of plays and the playback time. Based on this data, a pay-per-view model is applied to calculate the fees for advertisers, enabling them to deliver ads in a cost-effective manner.

[1208] Specific examples

[1209] Example 1: Supermarket

[1210] Cameras and microphones installed at the entrance of the supermarket capture video and audio of families entering the store in real time. The server identifies parents and children from the video and analyzes the audio content of conversations about "toys." At the same time, an emotion recognition engine detects whether the child is excited. Based on this, the server generates a promotional advertisement for children's toys and instantly distributes it to digital signage within the store.

[1211] Example 2: Inside a station

[1212] Cameras and microphones installed inside the station capture images of passersby and audio of them talking about "travel." The server analyzes the data, identifies their age group and gender, and uses an emotion recognition engine to assess their excitement level. Based on this, the server generates advertisements for travel agencies and displays them on digital signage.

[1213] As explained above, the present invention makes it possible to generate and distribute optimal advertisements in real time according to the situation and user emotions, thereby maximizing the effectiveness of advertisements.

[1214] The processing flow will be explained below.

[1215] Step 1:

[1216] Subject: Server

[1217] The camera and microphone begin streaming video and audio data. Video data is collected frame by frame, and audio data is digitized and captured in real time. This data is temporarily stored in a database for analysis.

[1218] Step 2:

[1219] Subject: Server

[1220] The server analyzes the collected video data and uses facial recognition algorithms to identify people. Using a face detection library such as OpenCV, it detects faces in the video frames and estimates attributes such as age and gender from the faces.

[1221] Step 3:

[1222] Subject: Server

[1223] Voice data is sent to a speech recognition engine and converted to text in real time. Speech recognition services such as Google Cloud Speech-to-Text API or Watson Speech to Text are used to transcribe the conversation and perform text analysis to identify interests and behaviors.

[1224] Step 4:

[1225] Subject: Server

[1226] The results of the analysis of video and audio data are integrated to identify the target's attributes and situation, thereby clarifying the user's basic profile and current behavior and interests.

[1227] Step 5:

[1228] Subject: Server

[1229] Based on video and audio data, the system analyzes the user's facial expressions and tone of voice to recognize their emotions. Using an emotion recognition engine, it identifies emotions such as anger, joy, and surprise, and uses the results in the next ad generation process.

[1230] Step 6:

[1231] Subject: Server

[1232] Based on the analysis results and emotion recognition results, the optimal advertising materials are selected by comparing them with pre-prepared information for advertising generation.Then, the selected materials and analysis results are input into the generation AI model to generate a customized advertising video that matches the emotions.

[1233] Step 7:

[1234] Subject: Device

[1235] The digital signage terminal receives the generated advertising video from the server, and immediately plays the downloaded advertising video using a file transfer protocol and displays it on the terminal screen.

[1236] Step 8:

[1237] Subject: Server

[1238] The device collects data on the ads played, recording the number of plays and the play time, and periodically sends this playback data to the server.

[1239] Step 9:

[1240] Subject: Server

[1241] The server applies a pay-per-view model based on the collected playback data to calculate the fees for advertisers, calculates the total amount based on a fee schedule according to the number of plays and the playback time, and provides a fee report to the advertiser.

[1242] The above steps enable the creation and distribution of optimal advertisements in real time according to the situation and user emotions, maximizing the effectiveness of the advertisements.

[1243] Example 2

[1244] 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."

[1245] In modern advertising delivery systems, it is difficult to generate and deliver optimal advertisements in real time that match user attributes and emotions. Maximizing the cost-effectiveness of advertising by introducing an efficient pay-per-use model for advertisers is also a challenge. Conventional systems are unable to perform such real-time data analysis and advertisement generation, making it impossible to deliver effective advertisements that match user interests and emotions.

[1246] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for acquiring video data, means for acquiring audio data, means for analyzing the video data and the audio data to identify person attributes and situations, means for recognizing a user's emotions from the video data and the audio data, means for generating an optimal advertisement from pre-prepared advertisement generation information based on the analysis results and the user's emotion recognition results, and means for delivering the generated advertisement via a display device. This enables real-time advertisement generation and delivery based on user attributes and emotions, maximizing the effectiveness of advertisements. In addition, by applying a pay-per-view model based on the number of impressions and playback time, advertisers can be effectively charged advertising fees.

[1247] "Video data" is a digital representation of visual information captured using a device such as a camera.

[1248] "Audio data" refers to digitized sound waves recorded using a device such as a microphone.

[1249] "Person attributes" are characteristics about an individual, such as age, gender, or identity, that are inferred using facial recognition algorithms or other methods.

[1250] "User emotion" refers to an emotional state, such as anger, joy, or surprise, that is recognized through facial expression analysis from video data and tone analysis from audio data.

[1251] "Information for advertisement generation" refers to materials such as text, images, and videos prepared in advance for generating advertisements, as well as related data.

[1252] A "generative AI model" is an artificial intelligence model that generates customized output data based on input data and specified prompts.

[1253] A "prompt sentence" is a textual input used to give specific instructions to a generative AI model.

[1254] A "pay-as-you-go model" is a pricing model in which the fee is calculated based on the number of times an advertisement is displayed and the duration of its playback.

[1255] A "display device" refers to a display or digital signage used to allow users to view the generated advertisement.

[1256] "Playback data" refers to data relating to playback status, such as the number of times a generated advertisement is played and the playback time.

[1257] This invention relates to a system that generates and distributes advertisements in real time that are optimized for the situation and user emotions based on video and audio data acquired from a camera and microphone.

[1258] The server collects video and audio data in real time from cameras and microphones installed via the network. The video data consists of successive image frames, and the audio data is digitized sound waves. These data are temporarily stored in a database for analysis. Specifically, the camera collects video at 30 frames per second, and the microphone collects audio data at 44.1 kHz per second.

[1259] The server analyzes the collected video data and identifies people using a facial recognition algorithm. For example, it uses image processing software (e.g., OpenCV). Attributes such as age and gender are inferred by analyzing the feature points of the detected face. At the same time, the audio data is sent to a speech recognition engine (e.g., Google Speech-to-Text API) and converted into text. This allows the content and situation of the conversation to be extracted. For example, the keyword "travel" can be extracted from the audio data.

[1260] The server then analyzes the user's facial expressions from the video data and the tone of voice from the audio data to recognize the user's emotions. It uses an emotion recognition engine (e.g., Amazon Rekognition or Google Cloud Natural Language API) to determine emotional states such as anger, joy, and surprise. For example, it can detect whether the user is smiling and recognize excitement from the tone of their voice.

[1261] Next, the server generates the optimal advertisement based on the analysis results and emotion recognition results, comparing them with pre-prepared advertisement generation information. It selects advertising materials (text, images, videos) and uses a generative AI model (e.g., GPT-4) to create a customized advertising video. During this process, it issues instructions by sending prompt text to the generative AI model. For example, a prompt text such as "An excited man in his 60s who is interested in traveling" could be sent to the generative AI model to generate a customized advertising video for a travel agency.

[1262] The generated advertising video is sent to a terminal (digital signage) and played on the display, allowing users to view it. For example, a travel advertising video can be played on a display in a store to attract the attention of passersby.

[1263] Finally, the server collects ad play data, recording the number of plays and play time. Based on this data, it applies a pay-per-view model and calculates the fee to the advertiser. For example, it records that a particular ad was played 100 times and the total play time was 50 minutes, and calculates the cost to be charged to the advertiser.

[1264] As a result, the system of the present invention enables real-time generation and distribution of advertisements based on user attributes and emotions, maximizing the effectiveness of advertisements. In addition, by applying a pay-per-view model based on the number of impressions and playback time, advertisers can be effectively charged advertising fees.

[1265] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1266] Step 1:

[1267] Data collection:

[1268] The server receives real-time video and audio data from cameras and microphones via the network. Specifically, the cameras capture video data at 30 frames per second, and the microphones capture audio data at 44.1 kHz per second. This data is temporarily stored in a database for analysis.

[1269] Input: Video and audio data obtained from the camera and microphone.

[1270] Output: Video and audio data stored in a database for analysis.

[1271] Step 2:

[1272] Data Analysis:

[1273] The server analyzes the collected video data. Specifically, it uses image processing software (e.g., OpenCV) to run a facial recognition algorithm to identify people. Attributes such as age and gender are inferred by analyzing facial feature points. At the same time, the server sends the audio data to a speech recognition engine (e.g., Google Speech-to-Text API) to convert it into text.

[1274] Input: Video and audio data obtained from the analysis database.

[1275] Output: Identified person's attributes (age, gender) and transcribed audio data.

[1276] Step 3:

[1277] Emotion recognition:

[1278] The server analyzes the user's facial expressions from the video data and the tone of voice from the audio data to recognize the user's emotions. An emotion recognition engine (e.g., Amazon Rekognition or Google Cloud Natural Language API) is used to determine the user's emotional state, such as anger, joy, or surprise.

[1279] Input: Identified person attributes and transcribed audio data.

[1280] Output: The user's emotional state (e.g., anger, joy, surprise).

[1281] Step 4:

[1282] Ad Generation:

[1283] The server generates the optimal ad based on the analysis results and emotion recognition results, comparing them with pre-prepared ad generation information. It selects ad materials (text, images, videos) and creates a customized ad video using a generative AI model (e.g., GPT-4). During this process, it sends a prompt to the generative AI model.

[1284] Input: User's emotional state, identified person attributes, and information for ad generation.

[1285] Output: A customized advertising video.

[1286] Step 5:

[1287] Ad serving:

[1288] The terminal (digital signage) receives the advertising video generated from the server and immediately displays it. The terminal plays the advertising video on the display so that the user can view it.

[1289] Input: The customized ad video sent from the server.

[1290] Output: Ad video played on the display.

[1291] Step 6:

[1292] Pay-as-you-go:

[1293] The server collects ad playback data, records the number of plays and the playback time, and calculates the fee to the advertiser based on this data using a pay-per-view model.

[1294] Input: Playback data of the ad video played on the display.

[1295] Output: Fees charged to advertisers.

[1296] (Application example 2)

[1297] 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."

[1298] Conventional ad delivery systems generate and deliver ads based on user attributes and circumstances, but they do not fully consider the user's real-time emotions or specific circumstances, which means that the effectiveness of the ads is not maximized.In addition, there is a lack of technology for generating and delivering effective ads, and the cost-effectiveness for advertisers is also insufficient.

[1299] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1300] In this invention, the server includes means for acquiring video data, means for acquiring audio data, means for analyzing the video data and the audio data to identify target attributes and situations, means for analyzing user emotions, means for generating an optimal advertisement from pre-prepared information for advertisement generation based on the results of the analysis and the emotion analysis, and means for delivering the generated advertisement via a display device. This makes it possible to generate and deliver an optimal advertisement according to the user's real-time emotions and specific situations, thereby maximizing advertising effectiveness.

[1301] "Video data" refers to visual information acquired by an image sensor such as a camera.

[1302] "Audio data" refers to auditory information acquired by an acoustic sensor such as a microphone.

[1303] "Analysis" refers to computational processing to extract or infer specific information from acquired data.

[1304] "Attributes" refer to the characteristics or properties of a subject that are determined through analysis, and include, for example, age and gender.

[1305] "Situation" refers to the subject's current environmental and behavioral state as determined by analysis.

[1306] "Emotion" refers to the subject's psychological state as determined by analysis, and includes anger, joy, surprise, etc.

[1307] "Information for advertisement generation" refers to materials and data prepared in advance for generating an advertisement.

[1308] The term "display device" refers to a device for visually presenting the generated advertisement to a user.

[1309] "Advertising materials" refers to the individual elements and content that make up an advertisement.

[1310] "Customized" refers to advertising that is generated and tailored to specific requirements and conditions.

[1311] A "pay-as-you-go model" refers to a pricing system in which charges are based on the amount of usage or the number of times it is used.

[1312] A system for implementing the present invention has the following configuration: A server acquires and analyzes video data and audio data, and generates and distributes optimal advertisements. A detailed explanation is provided below with specific examples.

[1313] First, the server acquires video and audio data in real time from cameras and microphones. Video data is visual information acquired by image sensors such as cameras, and audio data is auditory information acquired by acoustic sensors such as microphones. These data are temporarily stored in a database for analysis.

[1314] The server then analyzes the collected video and audio data. The video data uses facial recognition algorithms to identify people and infer their attributes (e.g., age, gender, etc.), while the audio data is sent to a speech recognition engine, which converts it into text and analyzes it. This analysis process uses software libraries such as OpenCV and SpeechRecognition.

[1315] Furthermore, the server uses an emotion recognition engine to analyze the user's emotions. By analyzing the user's facial expressions from the video data and the tone of voice from the audio data, emotions such as anger, joy, and surprise can be recognized. This emotion analysis combines multiple algorithms.

[1316] The server generates optimal advertisements based on the results of these analyses and sentiment analysis. Specifically, it selects the optimal advertising materials for the user's attributes and emotional state from pre-prepared advertisement generation information, and creates customized advertisements using a generative AI model.

[1317] Finally, the generated advertisement is delivered to a display device, such as a smartphone or digital signage. The user can watch the advertisement in real time, optimized for the situation and emotions of the moment. The terminal device plays the advertisement video and provides it visually to the user.

[1318] For example, when a user is using a smartphone, the smartphone's camera and microphone collect data and detect that the user is having a pleasant conversation. The server generates an advertisement for an entertainment app that matches that pleasant emotion and displays it on the smartphone. This application maximizes the effectiveness of advertising.

[1319] Example prompt sentence:

[1320] "The user is showing happy emotions. Please generate an ad for an entertainment app. User demographics: Age 25, Gender Male"

[1321] As described above, the system of the present invention is capable of generating and delivering optimal advertisements that correspond to the user's real-time emotions and specific situations, thereby maximizing the effectiveness of advertisements and improving cost-effectiveness for advertisers.

[1322] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1323] Step 1:

[1324] The server acquires video and audio data from the camera and microphone in real time. As input, it receives the real-time output of the camera and microphone and temporarily stores it as video and audio data in a database within the server. In this process, the camera captures successive image frames and the microphone generates a digital audio signal.

[1325] Step 2:

[1326] The server analyzes the acquired video data. As input, it receives the video data acquired in step 1, identifies the person using a facial recognition algorithm, and then estimates attributes such as age and gender. For example, it uses OpenCV to detect faces from image frames and estimates attributes based on the results. As output, it obtains attribute information for the identified person.

[1327] Step 3:

[1328] The server analyzes the acquired voice data. As input, it receives the voice data acquired in step 1 and converts it into text using a voice recognition engine. For example, it uses the SpeechRecognition library to convert the voice data into text, and analyzes the topic and situation of the conversation from the text content. As output, it obtains text data and the analysis results.

[1329] Step 4:

[1330] The server analyzes the user's emotions from the video and audio data. It receives the outputs of steps 2 and 3 as input and uses an emotion recognition engine to analyze facial expressions and tone of voice. Specifically, it reads facial expressions from image frames and identifies emotions from audio tones. The output is emotional information such as anger, joy, and surprise.

[1331] Step 5:

[1332] The server generates the optimal advertisement based on the analysis results and the results of sentiment analysis. It receives the outputs of steps 2, 3, and 4 as input and selects the optimal advertising material from pre-prepared information for advertisement generation. It uses a generative AI model to generate a customized advertisement that best suits the user's attributes and emotions. The output is a customized advertising video.

[1333] Step 6:

[1334] The server delivers the generated advertisement to the terminal. It receives the customized advertisement video output from step 5 as input and sends it to the terminal. The terminal plays the received advertisement video on a display device (e.g., a smartphone or digital signage). This process allows the user to watch the optimized advertisement in real time.

[1335] Step 7:

[1336] The server collects playback data of the displayed advertisements and calculates fees based on a pay-per-view model. As input, it receives data on the number of plays and the playback time sent from the terminal, and calculates the fee for the advertiser based on that data. As output, it obtains the final fee calculation result.

[1337] 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.

[1338] 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.

[1339] 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.

[1340] 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.

[1341] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1342] 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.

[1343] 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).

[1344] 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.

[1345] 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."

[1346] 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.

[1347] 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).

[1348] 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.

[1349] 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.

[1350] 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.

[1351] 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.

[1352] 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.

[1353] 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.

[1354] 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.

[1355] 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.

[1356] 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.

[1357] 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.

[1358] The following is further disclosed regarding the above embodiment.

[1359] (Claim 1)

[1360] a means for acquiring video data;

[1361] means for acquiring audio data;

[1362] means for analyzing the video data and the audio data and identifying the attributes and circumstances of a target;

[1363] a means for generating an optimal advertisement from pre-prepared advertisement generation information based on the analysis result;

[1364] means for delivering the generated advertisement via a display device;

[1365] A system including:

[1366] (Claim 2)

[1367] A means for selecting an optimal advertisement from a plurality of advertisement materials based on the analysis result, and a means for linking the selected advertisement materials to generate a customized advertisement;

[1368] The system of claim 1 further comprising:

[1369] (Claim 3)

[1370] means for applying a pay-per-view model according to the number of times the video data and the audio data are displayed and the playback time based on the analysis results of the video data and the audio data;

[1371] means for calculating fees based on the pay-per-use model;

[1372] The system of claim 1 further comprising:

[1373] "Example 1"

[1374] (Claim 1)

[1375] a means for acquiring video data;

[1376] means for acquiring audio data;

[1377] means for analyzing the video data and the audio data and identifying the attributes and circumstances of a target;

[1378] a means for generating an optimal advertisement from pre-prepared advertisement generation information based on the analysis result;

[1379] a means for customizing the generated advertisements using a generative AI model; and

[1380] means for delivering the generated advertisement via a display device;

[1381] A system including:

[1382] (Claim 2)

[1383] A means for selecting an optimal advertisement from a plurality of advertisement materials based on the analysis result, and a means for linking the selected advertisement materials to generate a customized advertisement;

[1384] A means for inputting a prompt sentence into a generative AI model to generate advertising content;

[1385] The system of claim 1 further comprising:

[1386] (Claim 3)

[1387] means for applying a pay-per-view model according to the number of times the video data and the audio data are displayed and the playback time based on the analysis results of the video data and the audio data;

[1388] means for calculating fees based on the pay-per-use model;

[1389] The system of claim 1 further comprising:

[1390] "Application Example 1"

[1391] (Claim 1)

[1392] a means for acquiring video data;

[1393] means for acquiring audio data;

[1394] means for analyzing the video data and the audio data and identifying the attributes and circumstances of a target;

[1395] a means for generating an optimal advertisement from pre-prepared advertisement generation information based on the analysis result;

[1396] means for delivering the generated advertisement via a display device;

[1397] A means for displaying advertisements on smart devices in real time;

[1398] A system including:

[1399] (Claim 2)

[1400] A means for selecting an optimal advertisement from a plurality of advertisement materials based on the analysis results, and a means for combining the selected advertisement materials and generating a customized advertisement using a generative AI model;

[1401] The system of claim 1 further comprising:

[1402] (Claim 3)

[1403] means for applying a pay-per-view model according to the number of times the video data and the audio data are displayed and the playback time based on the analysis results of the video data and the audio data;

[1404] means for calculating fees based on the pay-per-use model;

[1405] A means for analyzing user interactions with advertisements displayed on a smart device;

[1406] The system of claim 1 further comprising:

[1407] "Example 2: Combining Emotion Engines"

[1408] (Claim 1)

[1409] a means for acquiring video data;

[1410] means for acquiring audio data;

[1411] means for analyzing the video data and the audio data to identify the attributes and circumstances of a person;

[1412] means for recognizing a user's emotion from the video data and the audio data;

[1413] means for generating an optimal advertisement from pre-prepared advertisement generation information based on the analysis result and the user emotion recognition result;

[1414] means for delivering the generated advertisement via a display device;

[1415] A system including:

[1416] (Claim 2)

[1417] A means for selecting an optimal advertisement from a plurality of advertisement materials based on the analysis result and the user emotion recognition result, and a means for linking the selected advertisement materials and generating a customized advertisement using a generative AI model;

[1418] means for sending a prompt sentence to the generative AI model;

[1419] The system of claim 1 further comprising:

[1420] (Claim 3)

[1421] means for applying a pay-per-view model according to the number of times the video data and the audio data are displayed and the playback time based on the analysis results of the video data and the audio data;

[1422] means for calculating fees based on the pay-per-use model;

[1423] means for collecting playback data of said advertisements and recording the number of playbacks and playback times;

[1424] The system of claim 1 further comprising:

[1425] "Application example 2 when combining emotion engines"

[1426] (Claim 1)

[1427] a means for acquiring video data;

[1428] means for acquiring audio data;

[1429] means for analyzing the video data and the audio data and identifying the attributes and circumstances of a target;

[1430] means for analyzing user emotions;

[1431] a means for generating an optimal advertisement from pre-prepared advertisement generation information based on the analysis result and the emotion analysis result;

[1432] means for delivering the generated advertisement via a display device;

[1433] A system including:

[1434] (Claim 2)

[1435] a means for selecting an optimal advertisement from a plurality of advertisement materials based on the analysis result and the emotion analysis result; and a means for linking the selected advertisement materials to generate a customized advertisement.

[1436] The system of claim 1 further comprising:

[1437] (Claim 3)

[1438] means for applying a pay-per-view model according to the number of times the video data and the audio data are displayed and the playback time based on the analysis results of the video data and the audio data;

[1439] means for calculating fees based on the pay-per-use model;

[1440] The system of claim 1 further comprising: [Explanation of symbols]

[1441] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means for acquiring video data; means for acquiring audio data; means for analyzing the video data and the audio data and identifying the attributes and circumstances of a target; a means for generating an optimal advertisement from pre-prepared advertisement generation information based on the analysis result; means for delivering the generated advertisement via a display device; A system including:

2. A means for selecting an optimal advertisement from a plurality of advertisement materials based on the analysis result, and a means for linking the selected advertisement materials to generate a customized advertisement; The system of claim 1 further comprising:

3. means for applying a pay-per-view model according to the number of times the video data and the audio data are displayed and the playback time based on the analysis results of the video data and the audio data; means for calculating fees based on the pay-per-use model; The system of claim 1 further comprising:

Images