System

Abstract

A system is provided.SOLUTION: A system includes a means for receiving image data transmitted from a user terminal having a camera, a means for analyzing the received image data and recognizing a road, an obstacle, a sign, a pedestrian signal, or the like by using a generation AI model, a means for generating guide information for a user on the basis of contents recognized by the generation AI model, and a means for converting the generated guide information into voice and notifying the user of the voice.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] When visually impaired people walk, they face challenges, such as difficulty detecting obstacles and unexpected situations, even in familiar places. Existing assistance methods include guide dogs and accompanying support workers, but their numbers are in short supply, so new solutions are needed to enable visually impaired people to move around independently. Furthermore, as the number of visually impaired people is expected to increase in the future, it is necessary to develop effective and widely available walking assistance systems. [Means for solving the problem]

[0005] The present invention provides a system that includes a means for receiving image data sent from a user terminal equipped with a camera, a means for analyzing the received image data and using a generative AI model that recognizes roads, obstacles, signs, pedestrian signals, etc., a means for generating guide information for the user based on the content recognized by the generative AI model, and a means for converting the generated guide information into audio and notifying the user. This system provides an environment in which visually impaired people can walk independently with peace of mind.

[0006] A "user terminal with a camera" refers to a device that can be carried by a user and has a built-in camera for capturing images of the surroundings.

[0007] "Means for receiving image data" refers to software or hardware functions for receiving image files or streaming data sent from a user terminal via a network.

[0008] "Means for analyzing image data" refers to the algorithms and software functions used to process received image data and extract specific information.

[0009] A "generative AI model" refers to an artificial intelligence model that uses machine learning and deep learning to automatically generate and analyze useful information from image data.

[0010] "Means for recognizing roads, obstacles, signs, pedestrian signals, etc." refers to the function of using a generative AI model to identify specific objects from image data and grasp their location and status.

[0011] "Means for generating guide information for the user" refers to a software function that generates appropriate instructions and warnings for the user based on information about the recognized object.

[0012] The "means for converting guide information into voice" refers to a software or hardware function for converting the generated guide information into voice format using voice synthesis technology.

[0013] "Means for converting into audio and notifying the user" refers to a function for conveying guide information converted into audio to the user through an audio output device such as a speaker or earphones. [Brief explanation of the drawings]

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

[0015] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

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

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

[0018] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0019] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

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

[0021] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0022] [First embodiment]

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

[0024] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

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

[0026] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

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

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

[0029] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0031] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0032] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

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

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

[0035] System Overview

[0036] This system provides audio guidance to help visually impaired people walk safely and independently. The server receives image data sent from the user's camera-equipped device, analyzes the images, and generates guide information based on the results, which is then sent to the user via audio.

[0037] User terminal operation

[0038] A user carries a device with a camera and activates the device's camera while walking. The device periodically captures images of the surroundings and creates image data. This image data is then sent to a server via a network.

[0039] Server Operation

[0040] The server receives image data sent from the user's device. The received image data is analyzed using a generative AI model. This AI model is trained using machine learning and deep learning techniques to identify roads, obstacles, signs, pedestrian signals, and other objects, and understand their location and status.

[0041] Based on the analysis results, the server generates guide information for the user, including instructions such as "There is an obstacle ahead," "Turn left," and "The traffic light is red."

[0042] The generated guide information is converted into voice using a voice synthesis technique, and the converted guide information is transmitted again to the user terminal via the network.

[0043] User Notification

[0044] The user device plays back the audio guide information received from the server, allowing the user to understand their surroundings in real time and walk safely.

[0045] Specific examples

[0046] For example, when a user approaches an intersection, the device's camera captures the image. This image data is sent to the server and analyzed. The server recognizes that "there is an obstacle ahead of the intersection," and based on this information generates guide information instructing the user to "go five meters to the right." This guide information is converted into audio and sent to the user's device. The user's device then notifies the user by audio, saying, "There is an obstacle ahead. Go five meters to the right." This allows the user to pass through the intersection safely.

[0047] Usage scenarios

[0048] By utilizing this system, visually impaired product delivery personnel can safely walk to their destination. Visually impaired people visiting tourist spots can also enjoy sightseeing independently while receiving guidance. Users can not only receive audio guidance, but also request specific instructions using their own voice input. This makes it a flexible assistance system that can adapt to a variety of situations.

[0049] The processing flow will be explained below.

[0050] Step 1:

[0051] The user activates the camera

[0052] The user activates the device's camera and prepares to capture video of the surroundings.

[0053] Step 2:

[0054] The device generates image data.

[0055] The device converts the video captured in real time into image data for each frame and temporarily stores it in a buffer.

[0056] Step 3:

[0057] The device sends the image data to the server.

[0058] The terminal periodically transmits the stored image data to the server as a POST request.

[0059] Step 4:

[0060] The server receives the image data.

[0061] The server receives the image data sent from the terminal and expands the data in memory.

[0062] Step 5:

[0063] The server decodes the image data

[0064] The server decodes the received image data using libraries such as OpenCV and converts it into a format that can be analyzed as a color image.

[0065] Step 6:

[0066] The server inputs the image data into the generative AI model

[0067] The server preprocesses the decoded image data into the format required by the generative AI model and inputs it into the generative AI model as tensor data.

[0068] Step 7:

[0069] Server-generated captions

[0070] Using a generative AI model, the system generates captions from input image data, including information about roads, obstacles, signs, traffic lights, and other features in the image.

[0071] Step 8:

[0072] The server converts the captions into audio

[0073] The server converts the generated captions into audio data using a text-to-speech engine (TTS).

[0074] Step 9:

[0075] The server sends the voice data to the user terminal.

[0076] The server returns the generated voice data to the user terminal via the network.

[0077] Step 10:

[0078] The device receives the audio data.

[0079] The user terminal receives the voice data transmitted from the server.

[0080] Step 11:

[0081] The device plays the audio data.

[0082] The user terminal reproduces the received audio data to the user through a speaker or earphone.

[0083] Step 12:

[0084] The user walks according to the audio guide

[0085] The user listens to the audio guidance from the device and continues walking safely.

[0086] Example 1

[0087] 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."

[0088] For visually impaired people to walk safely and independently, they need a way to grasp their surroundings in real time and take appropriate action. However, conventional technology lacks the accuracy and real-time capabilities of analyzing image data, and the ability to reflect user feedback, making it difficult to provide sufficiently safe guidance. Furthermore, the inability to respond to voice input instructions means that the technology is unable to respond flexibly to the situation.

[0089] 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.

[0090] In this invention, the server includes: means for receiving image data transmitted from a user terminal equipped with a camera; means for analyzing the received image data and using a generative AI model to recognize roads, obstacles, signs, pedestrian signals, etc.; means for generating guide information for the user based on the content recognized by the generative AI model; means for converting the generated guide information into voice and notifying the user; and means for receiving voice input transmitted from the user terminal and correcting the guide information based on the voice input. This allows visually impaired people to receive safe guide information in real time and walk independently. Furthermore, specific instructions can be requested by voice input, enabling flexible responses according to the situation.

[0091] A "user terminal with a camera" is a portable electronic device that has the function of capturing video, processing it as image data, and transmitting it to a server via a network.

[0092] The "means for receiving image data" is a function that allows the server to receive image data transmitted via the network.

[0093] "Means for analyzing image data" refers to a function that uses a generative AI model to analyze image data received by the server.

[0094] A "generative AI model" is an artificial intelligence model trained using machine learning and deep learning technologies to recognize roads, obstacles, signs, pedestrian signals, etc. from image data and understand their location and status.

[0095] The "means for generating guide information" is a function for creating appropriate instructions for the user in text format based on the analyzed data.

[0096] The "means for converting guide information into voice" is a function for converting the generated text-format guide information into voice data using voice synthesis technology.

[0097] The "means for notifying the user" is a function that transmits guide information converted into voice to the user terminal so that the user can hear it.

[0098] "Means for receiving voice input" refers to a function that allows a user to send specific instructions or questions to the server by voice.

[0099] The "means for correcting guide information based on voice input" is a function for analyzing voice input from the user and updating or correcting the guide information to be provided in accordance with the contents of the voice input.

[0100] System Overview

[0101] This invention is a system that provides audio guidance to help visually impaired people walk safely and independently. The server receives image data sent from the user's camera-equipped device and analyzes it using a generative AI model. Based on the analysis results, it generates guide information, converts it into audio, and notifies the user.

[0102] User terminal operation

[0103] A user carries a device with a camera and activates the device's camera while walking. The device periodically captures images of the surroundings and creates image data. This image data is sent to a server via a network. The device also has a function to accept voice input from the user.

[0104] Server Operation

[0105] The server receives image data sent from the user device using an HTTP request. The received image data is converted into an appropriate format and input into the generative AI model. This generative AI model is trained using machine learning and deep learning techniques, and is built using, for example, TENSORFLOW (registered trademark) or PyTorch.

[0106] The server uses the generative AI model to analyze image data, identify roads, obstacles, signs, pedestrian signals, and other objects, and determine their location and status. Based on the results of this analysis, it generates guide information for the user. This guide information includes instructions such as "There is an obstacle ahead," "Turn left," and "The traffic light is red."

[0107] The server converts the generated guide information into speech using speech synthesis technology such as Google® Text-to-Speech (TTS) or Amazon Polly. The converted guide information is then sent back to the user terminal via the network. The server can also receive speech input sent from the user terminal and modify the guide information based on the received input.

[0108] User Notification

[0109] The user device plays back the audio guide information received from the server. This allows the user to understand their surroundings in real time and walk safely. For example, when the user approaches an intersection, the device's camera captures the image and the server recognizes that "there is an obstacle before the intersection." Based on this information, guide information instructing "please go 5 meters to the right" is generated, converted into voice and notified to the user. This allows the user to pass through the intersection safely.

[0110] Examples of prompt statements

[0111] For example, the following prompts might be used when building a generative AI model:

[0112] "Design a system that uses machine learning and deep learning techniques to recognize and analyze obstacles in captured images in real time and provide audio guidance to the user."

[0113] This invention provides comprehensive support for the visually impaired to walk safely and independently, greatly improving safety and convenience, and aiming to make mobility in daily life for the visually impaired more flexible and comfortable.

[0114] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0115] Step 1:

[0116] A user activates a device equipped with a camera and uses the device's camera to capture images of the surroundings while walking. The device captures the images as still images at regular intervals and generates image data.

[0117] Input: Video captured by the camera

[0118] Output: Generated image data

[0119] Step 2:

[0120] The terminal transmits the generated image data to the server via the network using an HTTP request.

[0121] Input: Generated image data

[0122] Output: Image data sent to the server

[0123] Step 3:

[0124] The server receives the image data sent from the terminal, converts the received data into an appropriate format, and prepares it for the next analysis step.

[0125] Input: Image data sent from the device

[0126] Output: Format-converted image data

[0127] Step 4:

[0128] The server inputs the received image data into a generative AI model. The generative AI model is built using libraries such as TensorFlow and PyTorch, and analyzes the image data to identify roads, obstacles, signs, pedestrian signals, etc. Specifically, it detects different objects in the image and understands their respective positions and states.

[0129] Input: Format-converted image data

[0130] Output: Analysis results (location and status of roads, obstacles, signs, pedestrian signals, etc.)

[0131] Step 5:

[0132] The server generates guide information for the user based on the analysis results of the generative AI model, including instructions such as "There is an obstacle ahead," "Turn left," and "The traffic light is red."

[0133] Input: Analysis results

[0134] Output: Generated guide information

[0135] Step 6:

[0136] The server converts the generated guide information into speech using speech synthesis technology, using services such as Google Text-to-Speech (TTS) and Amazon Polly.

[0137] Input: Generated guide information

[0138] Output: Audio data

[0139] Step 7:

[0140] The server then transmits the guide information converted into audio to the user terminal via the network using the HTTPS protocol.

[0141] Input: Audio data

[0142] Output: Audio data sent to the user's device

[0143] Step 8:

[0144] The user device plays back the received voice guidance information. As a result, the user can grasp the surrounding situation in real time and take appropriate action. For example, if the voice guidance instructs, "There is an obstacle ahead. Please move 5 meters to the right," the user will act accordingly.

[0145] Input: Audio data sent to the user's device

[0146] Output: Played audio description information

[0147] Step 9:

[0148] The user can use their voice input to request specific instructions or questions from the server, such as "What's the next intersection like?", and the device transmits this voice input to the server.

[0149] Input: User voice input

[0150] Output: Audio input sent to the server

[0151] Step 10:

[0152] The server receives voice input sent from the user terminal, analyzes the content, and updates or modifies the guide information. For example, it generates new guide information such as "There is a traffic light at the next intersection. Turn left."

[0153] Input: Voice input from the user

[0154] Output: Updated or corrected guide information

[0155] Step 11:

[0156] The updated guide information is converted back into audio and sent to the user terminal via the network. The user terminal then plays back the new guide information and notifies the user. This allows the user to respond flexibly according to the situation.

[0157] Input: Updated guide information

[0158] Output: New audio guide information sent to the user's device.

[0159] (Application example 1)

[0160] 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."

[0161] It is extremely difficult for visually impaired people to navigate safely on public transportation or in crowded places, and conventional technologies may not be able to provide sufficient support. Recognizing roads, obstacles, signs, pedestrian signals, and other obstacles is particularly difficult, requiring accurate information. Additionally, there is a lack of systems that provide users with real-time guidance information. The purpose of this invention is to solve these problems and provide an audio guidance system that enables visually impaired people to navigate safely.

[0162] 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.

[0163] In this invention, the server includes means for receiving image data transmitted from a user device having a camera, means for analyzing the received image data and using a generative AI model that recognizes roads, obstacles, signs, pedestrian signals, etc., means for generating guide information for the user based on the content recognized by the generative AI model, means for converting the generated guide information into audio and notifying the user, means for periodically transmitting image data from the user device to the server to support safe behavior on public transportation and in crowded places, and means for the generative AI model to provide analysis results and notify the user by audio. This enables visually impaired people to grasp their surroundings in real time and act safely.

[0164] A "user device" is a terminal that has a camera and can be carried around by a visually impaired person.

[0165] "Image data" is digital data of visual information captured by a camera on a user device and transmitted to a server.

[0166] A "generative AI model" is a model trained using machine learning and deep learning techniques that analyzes image data to recognize roads, obstacles, signs, pedestrian signals, etc.

[0167] "Guide information" is information that provides instructions and advice to the user, which is generated based on the analysis results, and is notified by voice.

[0168] The "means for converting into voice" is a function for converting the generated guide information into voice data using voice synthesis technology.

[0169] "Public transport and crowded places" refers to places and situations where many people use or gather, and where safe movement is particularly important for people with visual impairments.

[0170] The "periodic transmission means" is a function that allows the user device to transmit image data to the server at regular time intervals.

[0171] "Means of providing analysis results and notifying by voice" refers to a function that uses voice synthesis technology to notify the user in real time based on the analysis results of the generated AI model on the received image data.

[0172] The present invention relates to an audio guide system that supports visually impaired people to navigate safely on public transportation and in crowded places. Hereinafter, an embodiment will be described in detail.

[0173] System Overview

[0174] The system is composed of a user device, a server, and a generative AI model.

[0175] User Device Operation

[0176] The user device has a built-in camera and can be carried around by visually impaired people. The user activates the camera while walking and periodically captures images of the surroundings. The captured image data is sent to a server at regular intervals.

[0177] Server Operation

[0178] The server receives image data sent from the user device. The received image data is analyzed using a generative AI model. The generative AI model is trained using machine learning and deep learning techniques to accurately identify roads, obstacles, signs, pedestrian signals, and other objects. Based on the analysis results, the server generates guide information for the user. The generated guide information is converted into voice data using speech synthesis technology and transmitted to the user device.

[0179] User Notification

[0180] The user device plays back the audio data received from the server, providing real-time audio notification of the surrounding situation to the visually impaired, allowing the user to act safely.

[0181] Hardware and Software

[0182] The hardware required to implement this system is a user device with a camera (e.g., smart glasses) and a server. The software uses Python, OpenCV, the Requests library, etc. The generative AI model performs analysis based on tensor-formatted data, and the speech synthesis technology uses an API.

[0183] Specific examples

[0184] For example, if a user is walking down a busy street, they can send the following prompt to the generative AI model:

[0185] Example prompt sentence:

[0186] "Please generate guide information for walking safely in busy areas with many pedestrians. Please also include information about obstacles, dangerous areas, and traffic light status in the notification content."

[0187] Example result:

[0188] The server analyzes the prompt text and generates guide information such as, "There are many pedestrians ahead, so please move to the right. Be careful, there is a traffic light ahead, but it is red." and notifies the user device by voice.

[0189] As described above, the present invention enables visually impaired people to grasp the situation around them in real time and act safely.

[0190] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0191] Step 1:

[0192] A user activates the camera of a user device and periodically captures images of the surroundings. As input, the user device acquires real-time image data through the camera. This image data is saved as image data in JPEG format, etc. As output, the image data is stored in the user device.

[0193] Step 2:

[0194] The user device sends the accumulated image data to the server at regular intervals. As input, the latest image data stored on the user device is used to send an HTTP POST request to the server's API endpoint. As output, the image data is uploaded to the server.

[0195] Step 3:

[0196] The server receives image data sent from the user device. As input, it obtains the image data in the received HTTP POST request and stores it internally. As output, it obtains the image data newly stored in the server.

[0197] Step 4:

[0198] The server inputs the received image data into the generative AI model for analysis. As input, it converts the stored image data into tensor format and passes it to the generative AI model. The generative AI model analyzes the image data and identifies objects such as roads, obstacles, signs, and pedestrian signals. The results obtained from this analysis are the output.

[0199] Step 5:

[0200] The server generates guide information for the user based on the analysis results. It uses the analysis result data output by the generative AI model as input and generates guide information according to predefined rules and patterns. For example, text information such as "There is an obstacle ahead" or "Please move to the right and proceed" is generated. The generated guide information text data is obtained as output.

[0201] Step 6:

[0202] The server converts the generated guide information into voice data using voice synthesis technology. As input, the text data of the guide information is passed to the voice synthesis API, which generates voice data. As output, the generated voice data is obtained.

[0203] Step 7:

[0204] The server transmits the generated voice data to the user device. As input, the generated voice data is transmitted to the user device as an HTTP response. As output, new voice data to be received by the user device is obtained.

[0205] Step 8:

[0206] The user device plays the audio data received from the server and notifies the user. As input, the received audio data is passed to an internal audio player and played. As output, the user receives audio information and can understand the surrounding situation.

[0207] 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.

[0208] System Overview

[0209] This system provides audio guidance to help visually impaired people walk safely and independently. The server receives image data sent from the user's camera-equipped device, analyzes the images, and generates guide information based on the results, which is then notified to the user via audio. It also incorporates an emotion engine that recognizes the user's emotions, providing guidance tailored to each individual situation, thereby achieving more personalized assistance.

[0210] User terminal operation

[0211] A user carries a device with a camera and activates the device's camera while walking. The device periodically captures images of the surroundings and creates image data. This image data is sent to a server via a network. The user's voice input is also captured and sent to the server at the same time.

[0212] Server Operation

[0213] The server receives image data and voice input data sent from the user's device. The received image data is analyzed using a generative AI model. This AI model is trained using machine learning and deep learning techniques to identify roads, obstacles, signs, pedestrian signals, and other objects, and understand their location and status.

[0214] Based on the analysis results, the server generates guide information for the user, including instructions such as "There is an obstacle ahead," "Turn left," and "The traffic light is red."

[0215] Emotion Engine Operation

[0216] The server inputs the user's voice input into an emotion engine to recognize the user's emotions. The emotion engine analyzes the tone, rate, and context of the voice input to identify the user's emotional state (e.g., nervousness, anxiety, impatience).

[0217] Guide information adjustments

[0218] Based on the user's recognized emotions, the server adjusts the content of the guide information accordingly. For example, if the user is nervous, the server will provide more detailed and careful guide information. Conversely, if the user is relaxed, the server will provide simpler guide information.

[0219] The generated guide information is converted into voice using a voice synthesis technique, and the converted guide information is transmitted again to the user terminal via the network.

[0220] User Notification

[0221] The user device plays back the audio guide information received from the server, allowing the user to understand their surroundings in real time and walk safely.

[0222] Specific examples

[0223] For example, when a user approaches an intersection, the device's camera captures the image. This image data is sent to the server and analyzed. The server recognizes that there is an obstacle before the intersection and generates guide information based on this information, such as "Go 5 meters to the right."

[0224] At the same time, the emotion engine analyzes the user's voice input and recognizes that the user is anxious. Taking this into account, the guidance information is adjusted to be more detailed, such as "There is an obstacle ahead. Move 5 meters to the right. Please proceed slowly and without rushing."

[0225] This guide information is converted into audio and sent to the user's device. The user device then notifies the user by audio, saying, "There is an obstacle ahead. Please move 5 meters to the right. Please proceed slowly and without rushing." This allows the user to pass through the intersection safely and with peace of mind.

[0226] The processing flow will be explained below.

[0227] Step 1:

[0228] The user activates the camera

[0229] The user activates the device's camera and prepares to capture video of the surroundings.

[0230] Step 2:

[0231] The device generates image data and audio data.

[0232] The device converts the captured video into image data for each frame in real time and stores it temporarily in a buffer. It also captures the user's voice input and stores it as audio data.

[0233] Step 3:

[0234] The device sends image data and audio data to the server.

[0235] The terminal periodically transmits the stored image data and audio data to the server as a POST request.

[0236] Step 4:

[0237] The server receives the image data and audio data.

[0238] The server receives the image data and audio data sent from the terminal and expands each data in memory.

[0239] Step 5:

[0240] The server decodes the image data

[0241] The server decodes the received image data using libraries such as OpenCV and converts it into a format that can be analyzed as a color image.

[0242] Step 6:

[0243] The server inputs the image data into the generative AI model

[0244] The server preprocesses the decoded image data into the format required by the generative AI model and inputs it into the generative AI model as tensor data.

[0245] Step 7:

[0246] Server-generated captions

[0247] Using a generative AI model, the system generates captions from input image data, including information about roads, obstacles, signs, traffic lights, and other features in the image.

[0248] Step 8:

[0249] The server inputs the voice data into the emotion engine.

[0250] The server inputs the received voice data into an emotion engine that analyzes the user's voice tone, speed, and context.

[0251] Step 9:

[0252] The server analyzes the user's emotions

[0253] The emotion engine analyzes the audio data to identify the user's emotional state (e.g., tension, anxiety, impatience).

[0254] Step 10:

[0255] The server generates guide information

[0256] The server generates guidance information for the user based on the generated captions and the analyzed user's emotional information. For example, if the user is nervous, it provides more detailed and careful guidance information.

[0257] Step 11:

[0258] The server converts the guide information into voice

[0259] The server converts the guide information into voice data using a text-to-speech engine (TTS).

[0260] Step 12:

[0261] The server sends the voice data to the user terminal.

[0262] The server transmits the generated voice data to the user terminal via the network.

[0263] Step 13:

[0264] The device receives the audio data.

[0265] The user terminal receives the audio data sent from the server and prepares to play the data.

[0266] Step 14:

[0267] The device plays the audio data.

[0268] The user terminal reproduces the received audio data to the user through a speaker or earphone.

[0269] Step 15:

[0270] The user walks according to the audio guide

[0271] The user continues walking safely by following the audio guidance played from the device, which is adjusted appropriately according to the user's emotions.

[0272] Example 2

[0273] 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."

[0274] There is a need for technology to assist visually impaired people in walking safely and independently. Conventional voice guidance systems have difficulty grasping the surrounding situation in real time and providing appropriate instructions. Furthermore, because they do not take the user's emotional state into consideration, they are insufficient in providing support to improve the user's sense of security. Furthermore, there is a need for technology that can receive voice input, analyze emotions, and adjust guide information based on that.

[0275] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0276] In this invention, the server includes means for receiving image data transmitted from a user terminal having a camera, means for analyzing the received image data and using a generative AI model that recognizes roads, obstacles, signs, pedestrian signals, etc., means for generating guide information for the user based on the content recognized by the generative AI model, means for converting the generated guide information into voice and notifying the user, means for receiving voice input from the user and analyzing their emotional state, and means for adjusting the guide information based on the analyzed emotional state. This not only enables visually impaired people to grasp their surroundings in real time and walk safely, but also enables personalized support according to the user's emotional state.

[0277] A "user device with a camera" refers to a device that is equipped with a camera for acquiring visual information and that can be carried by the user. This includes smartphones and dedicated mobile devices.

[0278] The "means for receiving image data" refers to the communication technology and protocol for receiving image data sent from a user terminal having a camera and importing it into the server.

[0279] A "generative AI model" is a model trained using machine learning and deep learning techniques and used to analyze image data to recognize roads, obstacles, signs, pedestrian signals, etc.

[0280] "Means for generating guide information" refers to the process of generating information that instructs the user on the surrounding situation and direction of travel based on the analysis results of the generative AI model.

[0281] "Means for converting into voice and notifying the user" refers to technology for converting the generated guide information into voice data using voice synthesis technology and notifying the user of that voice in real time.

[0282] "Means for receiving voice input and analyzing emotional state" refers to processes and techniques for receiving a user's voice input and analyzing the voice data to identify the user's emotional state (e.g., tension, anxiety, impatience, etc.).

[0283] "Means for adjusting guide information" refers to the process of appropriately changing or adjusting the generated guide information based on the analyzed user's emotional state, for example, by adding detailed instructions or simplifying them.

[0284] This invention is a system that supports visually impaired people to walk safely and independently. The system uses a server to analyze data sent from a camera-equipped device held by the user, and provides audio guidance based on the results.

[0285] A user carries a device with a camera and activates the device's camera while walking. The device periodically captures images of the surroundings and creates image data. This image data is sent to a server via a network. The user's voice input is also captured and sent to the server at the same time. Specifically, a smartphone or a dedicated mobile device is used.

[0286] The server receives image data and audio data sent from the user's device. The received image data is analyzed using a generative AI model. This generative AI model utilizes a deep learning model trained using, for example, TensorFlow or PyTorch. This model identifies objects such as roads, obstacles, signs, and pedestrian signals, and determines their location and status.

[0287] Based on the analysis results, the server generates guide information for the user. This guide information includes instructions such as "There is an obstacle ahead," "Turn left," and "The traffic light is red." Furthermore, the server inputs the user's voice into an emotion engine to analyze the user's emotional state. This emotion engine uses, for example, Google Dialogflow or IBM Watson (registered trademark).

[0288] Based on the recognized user's emotions, the server adjusts the content of the guide information accordingly. For example, if the user is nervous, the server provides more detailed and careful guide information. Conversely, if the user is relaxed, the server provides concise guide information. The generated guide information is converted into audio using speech synthesis technology (e.g., Google Text-to-Speech or Amazon Polly). The converted audio guide information is then sent back to the user's device via the network.

[0289] The user device plays back the audio guide information received from the server, allowing the user to understand their surroundings in real time and walk safely.

[0290] For example, when a user approaches an intersection, the device's camera captures the image. This image data is sent to the server and analyzed. The server recognizes that there is an obstacle before the intersection and generates guide information based on this information, such as "Go 5 meters to the right."

[0291] At the same time, the emotion engine analyzes the user's voice input and recognizes that the user is in an anxious state. Taking this into account, the guide information is adjusted to be more detailed, such as "There is an obstacle ahead. Please move 5 meters to the right. Please proceed slowly and without rushing." This guide information is converted into voice and sent to the user's device. The user's device then notifies the user by voice, "There is an obstacle ahead. Please move 5 meters to the right. Please proceed slowly and without rushing." This allows the user to pass through the intersection safely and with peace of mind.

[0292] An example of a prompt sentence is "There is an obstacle ahead. Please guide me in which direction I should go." This is used as input to the generative AI model, forming the basis for generating optimal guidance information.

[0293] This system enables visually impaired people to accurately grasp their surroundings based on visual information and walk safely through audio guidance. It also provides individualized support based on the user's emotional state, helping them live independently and with peace of mind.

[0294] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0295] Step 1:

[0296] A user carries a device with a camera and activates the device's camera while walking. An application on the device periodically uses the camera to capture images of the surroundings and create image data. Specifically, the application captures images twice per second and saves the captured data as JPEG image files. The inputs are user operations and camera images, and the output is JPEG image data.

[0297] Step 2:

[0298] The user's device sends the captured image and audio data to the server. The device compresses the generated JPEG image file and the simultaneously collected audio input data, and transfers them to the server via the Internet. In particular, it performs a process to convert the data into an appropriate format for transmission. The input is JPEG image data and audio data, and the compressed data is sent to the server as output.

[0299] Step 3:

[0300] The server receives image data and audio data sent from the user's device. The server decompresses the received data and converts the image data into tensor format for input into the generative AI model. Specifically, the image data is converted into numerical data for each pixel and processed into tensor format. The input is compressed image data and audio data, and the output is tensor-formatted image data.

[0301] Step 4:

[0302] The server analyzes tensor-formatted image data using a generative AI model. This model identifies objects such as roads, obstacles, signs, and pedestrian signals, and outputs their location and status. For example, a deep learning framework is used to recognize and classify objects. The input is tensor-formatted image data, and the output is the object identification results.

[0303] Step 5:

[0304] The server generates guide information for the user based on the results of image analysis. For example, it creates guide information that includes specific instructions such as "There is an obstacle ahead," "Turn left," and "The traffic light is red." The input is the object identification result, and the output is guide information.

[0305] Step 6:

[0306] The server analyzes the received voice data and inputs it into an emotion engine that recognizes the user's emotional state. The emotion engine analyzes the tone, speed, and context of the voice to identify emotional states such as tension, anxiety, and impatience. Specifically, it extracts features from the voice data and inputs them into an emotion classification model. The input is the voice data, and the output is the user's emotional state.

[0307] Step 7:

[0308] Based on the analyzed user's emotional state, the server adjusts the guide information accordingly. For example, if the user is nervous, the server makes the guide information more detailed and careful. The inputs are the emotional state and the initial guide information, and the output is the adjusted guide information.

[0309] Step 8:

[0310] The server converts the adjusted guide information into speech using speech synthesis technology. For example, it uses a text-to-speech system to create a natural-sounding audio guide. The adjusted guide information is input, and audio data is generated as output.

[0311] Step 9:

[0312] The server transmits the generated voice data to the user terminal. The voice data is transferred again in real time via the network. The voice data is input, and transmission to the user terminal is output.

[0313] Step 10:

[0314] The user device plays the audio guide information received from the server. The user continues walking safely by following the played audio guide. The input is audio data, and the output is the playback of audio guide. This allows the user to understand the surrounding situation in real time and walk safely.

[0315] (Application example 2)

[0316] 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."

[0317] Visually impaired people face difficulties when searching for products in physical stores because they cannot see their surroundings. They also need the help of others to understand the location of shelves and products, making it difficult for them to enjoy shopping independently.

[0318] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for receiving image data transmitted from a user terminal having a camera; means for analyzing the received image data and using a generative AI model that recognizes roads, obstacles, signs, pedestrian signals, etc.; means for generating guide information for the user based on the content recognized by the generative AI model; means for converting the generated guide information into voice and notifying the user; means for capturing the situation in a physical store using a camera in smart glasses and identifying product shelves and products; and means for locating the user in the store and generating guide information that guides them to the products. This enables visually impaired people to find products in a physical store and shop independently.

[0319] A "camera" is a device for capturing images, and is installed in smart glasses or a user terminal.

[0320] A "user terminal" is a device that has a camera and can send image data to a server.

[0321] "Image data" refers to video information of the surroundings captured by a camera.

[0322] A "generative AI model" is an artificial intelligence model that analyzes received image data and identifies specific objects.

[0323] "Guide information" refers to guidance and instruction information for users that is generated based on the content recognized by the generative AI model.

[0324] "Convert to voice" is a process for conveying text-format guide information to the user as voice.

[0325] "Smart glasses" are head-mounted display devices that incorporate cameras and other sensors and are worn by users.

[0326] A "brick and mortar store" is a physical sales location where products are displayed and customers can visit and purchase in person.

[0327] A "product shelf" is a shelf used to display products in a physical store.

[0328] "Determining the location" means determining the user's current location and obtaining accurate location information.

[0329] "Guidance" refers to the provision of information to assist a user in reaching a destination.

[0330] As an embodiment of this invention, we will explain a support system for visually impaired people to search for and purchase products in a physical store. The system is realized by combining smart glasses worn by the user, a server, a generative AI model, and an audio guide.

[0331] System configuration

[0332] Smart glasses and user terminal

[0333] Smart glasses have a built-in camera and are worn by the user. The camera captures the surrounding situation in real time and generates image data. The user device (such as a smartphone or tablet) periodically transmits this image data to a server.

[0334] Server and Generative AI Model

[0335] The server receives image data sent from the user's device. The received image data is input into the generative AI model for analysis. The generative AI model is trained using machine learning and deep learning techniques to identify product shelves, directions, obstacles, and other items in a physical store.

[0336] Guide information generation

[0337] The server generates guide information for the user based on the content recognized by the generative AI model. This information includes the user's location, the location of the desired product, and the route to take within the store. The guide information is generated in text format.

[0338] Text-to-Speech and Notifications

[0339] The generated guide information is converted into audio on the server side using speech synthesis technology. Specifically, Google Text-to-Speech (gTTS) is used. The converted audio guide information is then sent back to the user's device via the network. The user's device (smart glasses) then plays back this audio guide and notifies the user.

[0340] Operation example

[0341] When a user searches for a product in a store, the smart glasses capture the surroundings with a camera and send the image data to the server. For example, if a user says, "I'm looking for the condiments section," the following prompt sentences are generated as guide information:

[0342] "There is a shelf about three meters ahead. The item you are looking for is on the second shelf on the left. Please walk slowly."

[0343] System Operation

[0344] The server receives image data sent from the smart glasses with a camera and first performs image analysis. It processes the image using libraries such as OpenCV, extracts specific objects, and converts them into tensor format. This data is input into a generative AI model to identify shelf and product locations, the user's current location, and other information. Based on the recognized content, guide information for the user is generated in text format and converted into audio using gTTS. The converted audio guide information is sent to the user's device, and an audio notification is given to the user via the smart glasses. By following these audio notifications, visually impaired people can shop independently.

[0345] This system enables visually impaired people to safely and efficiently search for and purchase products in physical stores.

[0346] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0347] Step 1:

[0348] A user wears smart glasses and walks around a physical store. The camera in the smart glasses captures the surroundings and generates image data. This image data is sent to the user device. The input is the captured image, and the output is the image data.

[0349] Step 2:

[0350] The user terminal periodically transmits the captured image data to the server, where the input is the image data received from the smart glasses and the output is the image data transmitted to the server.

[0351] Step 3:

[0352] The server analyzes the image data received from the user device, converts this image data into tensor format, and inputs it into the generative AI model. The input is the received image data, and the output is tensor format data.

[0353] Step 4:

[0354] The generative AI model identifies product shelves, product locations, obstacles, etc. in a physical store based on tensor-formatted data. The input is tensor-formatted data, and the output is information about the identified content.

[0355] Step 5:

[0356] The server generates guide information for the user based on the content identified by the generative AI model. For example, this includes information about the product's location and travel route. The input is the identified content, and the output is text data of the guide information.

[0357] Step 6:

[0358] The server converts the generated guide information into audio using speech synthesis technology. Google Text-to-Speech (gTTS) is used here. The input is the text data of the guide information, and the output is an audio file.

[0359] Step 7:

[0360] The server converts the guide information into audio and sends it to the user terminal. The input is an audio file, and the output is the audio file sent to the user terminal.

[0361] Step 8:

[0362] The user terminal (smart glasses) plays the audio guide information received from the server and notifies the user. The input is an audio file, and the output is the played audio guide. This allows the user to search for and move around products in the store by following the audio guide.

[0363] 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.

[0364] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (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.

[0365] 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.

[0366] [Second embodiment]

[0367] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0368] 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.

[0369] 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).

[0370] 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.

[0371] 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.

[0372] 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).

[0373] 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.

[0374] 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.

[0375] 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.

[0376] 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.

[0377] 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.

[0378] 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."

[0379] System Overview

[0380] This system provides audio guidance to help visually impaired people walk safely and independently. The server receives image data sent from the user's camera-equipped device, analyzes the images, and generates guide information based on the results, which is then sent to the user via audio.

[0381] User terminal operation

[0382] A user carries a device with a camera and activates the device's camera while walking. The device periodically captures images of the surroundings and creates image data. This image data is then sent to a server via a network.

[0383] Server Operation

[0384] The server receives image data sent from the user's device. The received image data is analyzed using a generative AI model. This AI model is trained using machine learning and deep learning techniques to identify roads, obstacles, signs, pedestrian signals, and other objects, and understand their location and status.

[0385] Based on the analysis results, the server generates guide information for the user, including instructions such as "There is an obstacle ahead," "Turn left," and "The traffic light is red."

[0386] The generated guide information is converted into voice using a voice synthesis technique, and the converted guide information is transmitted again to the user terminal via the network.

[0387] User Notification

[0388] The user device plays back the audio guide information received from the server, allowing the user to understand their surroundings in real time and walk safely.

[0389] Specific examples

[0390] For example, when a user approaches an intersection, the device's camera captures the image. This image data is sent to the server and analyzed. The server recognizes that "there is an obstacle ahead of the intersection," and based on this information generates guide information instructing the user to "go five meters to the right." This guide information is converted into audio and sent to the user's device. The user's device then notifies the user by audio, saying, "There is an obstacle ahead. Go five meters to the right." This allows the user to pass through the intersection safely.

[0391] Usage scenarios

[0392] By utilizing this system, visually impaired product delivery personnel can safely walk to their destination. Visually impaired people visiting tourist spots can also enjoy sightseeing independently while receiving guidance. Users can not only receive audio guidance, but also request specific instructions using their own voice input. This makes it a flexible assistance system that can adapt to a variety of situations.

[0393] The processing flow will be explained below.

[0394] Step 1:

[0395] The user activates the camera

[0396] The user activates the device's camera and prepares to capture video of the surroundings.

[0397] Step 2:

[0398] The device generates image data.

[0399] The device converts the video captured in real time into image data for each frame and temporarily stores it in a buffer.

[0400] Step 3:

[0401] The device sends the image data to the server.

[0402] The terminal periodically transmits the stored image data to the server as a POST request.

[0403] Step 4:

[0404] The server receives the image data.

[0405] The server receives the image data sent from the terminal and expands the data in memory.

[0406] Step 5:

[0407] The server decodes the image data

[0408] The server decodes the received image data using libraries such as OpenCV and converts it into a format that can be analyzed as a color image.

[0409] Step 6:

[0410] The server inputs the image data into the generative AI model

[0411] The server preprocesses the decoded image data into the format required by the generative AI model and inputs it into the generative AI model as tensor data.

[0412] Step 7:

[0413] Server-generated captions

[0414] Using a generative AI model, the system generates captions from input image data, including information about roads, obstacles, signs, traffic lights, and other features in the image.

[0415] Step 8:

[0416] The server converts the captions into audio

[0417] The server converts the generated captions into audio data using a text-to-speech engine (TTS).

[0418] Step 9:

[0419] The server sends the voice data to the user terminal.

[0420] The server returns the generated voice data to the user terminal via the network.

[0421] Step 10:

[0422] The device receives the audio data.

[0423] The user terminal receives the voice data transmitted from the server.

[0424] Step 11:

[0425] The device plays the audio data.

[0426] The user terminal reproduces the received audio data to the user through a speaker or earphone.

[0427] Step 12:

[0428] The user walks according to the audio guide

[0429] The user listens to the audio guidance from the device and continues walking safely.

[0430] Example 1

[0431] 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."

[0432] For visually impaired people to walk safely and independently, they need a way to grasp their surroundings in real time and take appropriate action. However, conventional technology lacks the accuracy and real-time capabilities of analyzing image data, and the ability to reflect user feedback, making it difficult to provide sufficiently safe guidance. Furthermore, the inability to respond to voice input instructions means that the technology is unable to respond flexibly to the situation.

[0433] 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.

[0434] In this invention, the server includes: means for receiving image data transmitted from a user terminal equipped with a camera; means for analyzing the received image data and using a generative AI model to recognize roads, obstacles, signs, pedestrian signals, etc.; means for generating guide information for the user based on the content recognized by the generative AI model; means for converting the generated guide information into voice and notifying the user; and means for receiving voice input transmitted from the user terminal and correcting the guide information based on the voice input. This allows visually impaired people to receive safe guide information in real time and walk independently. Furthermore, specific instructions can be requested by voice input, enabling flexible responses according to the situation.

[0435] A "user terminal with a camera" is a portable electronic device that has the function of capturing video, processing it as image data, and transmitting it to a server via a network.

[0436] The "means for receiving image data" is a function that allows the server to receive image data transmitted via the network.

[0437] "Means for analyzing image data" refers to a function that uses a generative AI model to analyze image data received by the server.

[0438] A "generative AI model" is an artificial intelligence model trained using machine learning and deep learning technologies to recognize roads, obstacles, signs, pedestrian signals, etc. from image data and understand their location and status.

[0439] The "means for generating guide information" is a function for creating appropriate instructions for the user in text format based on the analyzed data.

[0440] The "means for converting guide information into voice" is a function for converting the generated text-format guide information into voice data using voice synthesis technology.

[0441] The "means for notifying the user" is a function that transmits guide information converted into voice to the user terminal so that the user can hear it.

[0442] "Means for receiving voice input" refers to a function that allows a user to send specific instructions or questions to the server by voice.

[0443] The "means for correcting guide information based on voice input" is a function for analyzing voice input from the user and updating or correcting the guide information to be provided in accordance with the contents of the voice input.

[0444] System Overview

[0445] This invention is a system that provides audio guidance to help visually impaired people walk safely and independently. The server receives image data sent from the user's camera-equipped device and analyzes it using a generative AI model. Based on the analysis results, it generates guide information, converts it into audio, and notifies the user.

[0446] User terminal operation

[0447] A user carries a device with a camera and activates the device's camera while walking. The device periodically captures images of the surroundings and creates image data. This image data is sent to a server via a network. The device also has a function to accept voice input from the user.

[0448] Server Operation

[0449] The server receives image data sent from the user device using an HTTP request. The received image data is converted into an appropriate format and input into the generative AI model. This generative AI model is trained using machine learning and deep learning techniques, such as TensorFlow and PyTorch.

[0450] The server uses the generative AI model to analyze image data, identify roads, obstacles, signs, pedestrian signals, and other objects, and determine their location and status. Based on the results of this analysis, it generates guide information for the user. This guide information includes instructions such as "There is an obstacle ahead," "Turn left," and "The traffic light is red."

[0451] The server converts the generated guide information into speech using speech synthesis technology such as Google Text-to-Speech (TTS) or Amazon Polly. The converted speech guide information is then sent back to the user's device via the network. The server can also receive speech input from the user's device and modify the guide information based on that input.

[0452] User Notification

[0453] The user device plays back the audio guide information received from the server. This allows the user to understand their surroundings in real time and walk safely. For example, when the user approaches an intersection, the device's camera captures the image and the server recognizes that "there is an obstacle before the intersection." Based on this information, guide information instructing "please go 5 meters to the right" is generated, converted into voice and notified to the user. This allows the user to pass through the intersection safely.

[0454] Examples of prompt statements

[0455] For example, the following prompts might be used when building a generative AI model:

[0456] "Design a system that uses machine learning and deep learning techniques to recognize and analyze obstacles in captured images in real time and provide audio guidance to the user."

[0457] This invention provides comprehensive support for the visually impaired to walk safely and independently, greatly improving safety and convenience, and aiming to make mobility in daily life for the visually impaired more flexible and comfortable.

[0458] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0459] Step 1:

[0460] A user activates a device equipped with a camera and uses the device's camera to capture images of the surroundings while walking. The device captures the images as still images at regular intervals and generates image data.

[0461] Input: Video captured by the camera

[0462] Output: Generated image data

[0463] Step 2:

[0464] The terminal transmits the generated image data to the server via the network using an HTTP request.

[0465] Input: Generated image data

[0466] Output: Image data sent to the server

[0467] Step 3:

[0468] The server receives the image data sent from the terminal, converts the received data into an appropriate format, and prepares it for the next analysis step.

[0469] Input: Image data sent from the device

[0470] Output: Format-converted image data

[0471] Step 4:

[0472] The server inputs the received image data into a generative AI model. The generative AI model is built using libraries such as TensorFlow and PyTorch, and analyzes the image data to identify roads, obstacles, signs, pedestrian signals, etc. Specifically, it detects different objects in the image and understands their respective positions and states.

[0473] Input: Format-converted image data

[0474] Output: Analysis results (location and status of roads, obstacles, signs, pedestrian signals, etc.)

[0475] Step 5:

[0476] The server generates guide information for the user based on the analysis results of the generative AI model, including instructions such as "There is an obstacle ahead," "Turn left," and "The traffic light is red."

[0477] Input: Analysis results

[0478] Output: Generated guide information

[0479] Step 6:

[0480] The server converts the generated guide information into speech using speech synthesis technology, using services such as Google Text-to-Speech (TTS) and Amazon Polly.

[0481] Input: Generated guide information

[0482] Output: Audio data

[0483] Step 7:

[0484] The server then transmits the guide information converted into audio to the user terminal via the network using the HTTPS protocol.

[0485] Input: Audio data

[0486] Output: Audio data sent to the user's device

[0487] Step 8:

[0488] The user device plays back the received voice guidance information. As a result, the user can grasp the surrounding situation in real time and take appropriate action. For example, if the voice guidance instructs, "There is an obstacle ahead. Please move 5 meters to the right," the user will act accordingly.

[0489] Input: Audio data sent to the user's device

[0490] Output: Played audio description information

[0491] Step 9:

[0492] The user can use their voice input to request specific instructions or questions from the server, such as "What's the next intersection like?", and the device transmits this voice input to the server.

[0493] Input: User voice input

[0494] Output: Audio input sent to the server

[0495] Step 10:

[0496] The server receives voice input sent from the user terminal, analyzes the content, and updates or modifies the guide information. For example, it generates new guide information such as "There is a traffic light at the next intersection. Turn left."

[0497] Input: Voice input from the user

[0498] Output: Updated or corrected guide information

[0499] Step 11:

[0500] The updated guide information is converted back into audio and sent to the user terminal via the network. The user terminal then plays back the new guide information and notifies the user. This allows the user to respond flexibly according to the situation.

[0501] Input: Updated guide information

[0502] Output: New audio guide information sent to the user's device.

[0503] (Application example 1)

[0504] 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."

[0505] It is extremely difficult for visually impaired people to navigate safely on public transportation or in crowded places, and conventional technologies may not be able to provide sufficient support. Recognizing roads, obstacles, signs, pedestrian signals, and other obstacles is particularly difficult, requiring accurate information. Additionally, there is a lack of systems that provide users with real-time guidance information. The purpose of this invention is to solve these problems and provide an audio guidance system that enables visually impaired people to navigate safely.

[0506] 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.

[0507] In this invention, the server includes means for receiving image data transmitted from a user device having a camera, means for analyzing the received image data and using a generative AI model that recognizes roads, obstacles, signs, pedestrian signals, etc., means for generating guide information for the user based on the content recognized by the generative AI model, means for converting the generated guide information into audio and notifying the user, means for periodically transmitting image data from the user device to the server to support safe behavior on public transportation and in crowded places, and means for the generative AI model to provide analysis results and notify the user by audio. This enables visually impaired people to grasp their surroundings in real time and act safely.

[0508] A "user device" is a terminal that has a camera and can be carried around by a visually impaired person.

[0509] "Image data" is digital data of visual information captured by a camera on a user device and transmitted to a server.

[0510] A "generative AI model" is a model trained using machine learning and deep learning techniques that analyzes image data to recognize roads, obstacles, signs, pedestrian signals, etc.

[0511] "Guide information" is information that provides instructions and advice to the user, which is generated based on the analysis results, and is notified by voice.

[0512] The "means for converting into voice" is a function for converting the generated guide information into voice data using voice synthesis technology.

[0513] "Public transport and crowded places" refers to places and situations where many people use or gather, and where safe movement is particularly important for people with visual impairments.

[0514] The "periodic transmission means" is a function that allows the user device to transmit image data to the server at regular time intervals.

[0515] "Means of providing analysis results and notifying by voice" refers to a function that uses voice synthesis technology to notify the user in real time based on the analysis results of the generated AI model on the received image data.

[0516] The present invention relates to an audio guide system that supports visually impaired people to navigate safely on public transportation and in crowded places. Hereinafter, an embodiment will be described in detail.

[0517] System Overview

[0518] The system is composed of a user device, a server, and a generative AI model.

[0519] User Device Operation

[0520] The user device has a built-in camera and can be carried around by visually impaired people. The user activates the camera while walking and periodically captures images of the surroundings. The captured image data is sent to a server at regular intervals.

[0521] Server Operation

[0522] The server receives image data sent from the user device. The received image data is analyzed using a generative AI model. The generative AI model is trained using machine learning and deep learning techniques to accurately identify roads, obstacles, signs, pedestrian signals, and other objects. Based on the analysis results, the server generates guide information for the user. The generated guide information is converted into voice data using speech synthesis technology and transmitted to the user device.

[0523] User Notification

[0524] The user device plays back the audio data received from the server, providing real-time audio notification of the surrounding situation to the visually impaired, allowing the user to act safely.

[0525] Hardware and Software

[0526] The hardware required to implement this system is a user device with a camera (e.g., smart glasses) and a server. The software uses Python, OpenCV, the Requests library, etc. The generative AI model performs analysis based on tensor-formatted data, and the speech synthesis technology uses an API.

[0527] Specific examples

[0528] For example, if a user is walking down a busy street, they can send the following prompt to the generative AI model:

[0529] Example prompt sentence:

[0530] "Please generate guide information for walking safely in busy areas with many pedestrians. Please also include information about obstacles, dangerous areas, and traffic light status in the notification content."

[0531] Example result:

[0532] The server analyzes the prompt text and generates guide information such as, "There are many pedestrians ahead, so please move to the right. Be careful, there is a traffic light ahead, but it is red." and notifies the user device by voice.

[0533] As described above, the present invention enables visually impaired people to grasp the situation around them in real time and act safely.

[0534] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0535] Step 1:

[0536] A user activates the camera of a user device and periodically captures images of the surroundings. As input, the user device acquires real-time image data through the camera. This image data is saved as image data in JPEG format, etc. As output, the image data is stored in the user device.

[0537] Step 2:

[0538] The user device sends the accumulated image data to the server at regular intervals. As input, the latest image data stored on the user device is used to send an HTTP POST request to the server's API endpoint. As output, the image data is uploaded to the server.

[0539] Step 3:

[0540] The server receives image data sent from the user device. As input, it obtains the image data in the received HTTP POST request and stores it internally. As output, it obtains the image data newly stored in the server.

[0541] Step 4:

[0542] The server inputs the received image data into the generative AI model for analysis. As input, it converts the stored image data into tensor format and passes it to the generative AI model. The generative AI model analyzes the image data and identifies objects such as roads, obstacles, signs, and pedestrian signals. The results obtained from this analysis are the output.

[0543] Step 5:

[0544] The server generates guide information for the user based on the analysis results. It uses the analysis result data output by the generative AI model as input and generates guide information according to predefined rules and patterns. For example, text information such as "There is an obstacle ahead" or "Please move to the right and proceed" is generated. The generated guide information text data is obtained as output.

[0545] Step 6:

[0546] The server converts the generated guide information into voice data using voice synthesis technology. As input, the text data of the guide information is passed to the voice synthesis API, which generates voice data. As output, the generated voice data is obtained.

[0547] Step 7:

[0548] The server transmits the generated voice data to the user device. As input, the generated voice data is transmitted to the user device as an HTTP response. As output, new voice data to be received by the user device is obtained.

[0549] Step 8:

[0550] The user device plays the audio data received from the server and notifies the user. As input, the received audio data is passed to an internal audio player and played. As output, the user receives audio information and can understand the surrounding situation.

[0551] 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.

[0552] System Overview

[0553] This system provides audio guidance to help visually impaired people walk safely and independently. The server receives image data sent from the user's camera-equipped device, analyzes the images, and generates guide information based on the results, which is then notified to the user via audio. It also incorporates an emotion engine that recognizes the user's emotions, providing guidance tailored to each individual situation, thereby achieving more personalized assistance.

[0554] User terminal operation

[0555] A user carries a device with a camera and activates the device's camera while walking. The device periodically captures images of the surroundings and creates image data. This image data is sent to a server via a network. The user's voice input is also captured and sent to the server at the same time.

[0556] Server Operation

[0557] The server receives image data and voice input data sent from the user's device. The received image data is analyzed using a generative AI model. This AI model is trained using machine learning and deep learning techniques to identify roads, obstacles, signs, pedestrian signals, and other objects, and understand their location and status.

[0558] Based on the analysis results, the server generates guide information for the user, including instructions such as "There is an obstacle ahead," "Turn left," and "The traffic light is red."

[0559] Emotion Engine Operation

[0560] The server inputs the user's voice input into an emotion engine to recognize the user's emotions. The emotion engine analyzes the tone, rate, and context of the voice input to identify the user's emotional state (e.g., nervousness, anxiety, impatience).

[0561] Guide information adjustments

[0562] Based on the user's recognized emotions, the server adjusts the content of the guide information accordingly. For example, if the user is nervous, the server will provide more detailed and careful guide information. Conversely, if the user is relaxed, the server will provide simpler guide information.

[0563] The generated guide information is converted into voice using a voice synthesis technique, and the converted guide information is transmitted again to the user terminal via the network.

[0564] User Notification

[0565] The user device plays back the audio guide information received from the server, allowing the user to understand their surroundings in real time and walk safely.

[0566] Specific examples

[0567] For example, when a user approaches an intersection, the device's camera captures the image. This image data is sent to the server and analyzed. The server recognizes that there is an obstacle before the intersection and generates guide information based on this information, such as "Go 5 meters to the right."

[0568] At the same time, the emotion engine analyzes the user's voice input and recognizes that the user is anxious. Taking this into account, the guidance information is adjusted to be more detailed, such as "There is an obstacle ahead. Move 5 meters to the right. Please proceed slowly and without rushing."

[0569] This guide information is converted into audio and sent to the user's device. The user device then notifies the user by audio, saying, "There is an obstacle ahead. Please move 5 meters to the right. Please proceed slowly and without rushing." This allows the user to pass through the intersection safely and with peace of mind.

[0570] The processing flow will be explained below.

[0571] Step 1:

[0572] The user activates the camera

[0573] The user activates the device's camera and prepares to capture video of the surroundings.

[0574] Step 2:

[0575] The device generates image data and audio data.

[0576] The device converts the captured video into image data for each frame in real time and stores it temporarily in a buffer. It also captures the user's voice input and stores it as audio data.

[0577] Step 3:

[0578] The device sends image data and audio data to the server.

[0579] The terminal periodically transmits the stored image data and audio data to the server as a POST request.

[0580] Step 4:

[0581] The server receives the image data and audio data.

[0582] The server receives the image data and audio data sent from the terminal and expands each data in memory.

[0583] Step 5:

[0584] The server decodes the image data

[0585] The server decodes the received image data using libraries such as OpenCV and converts it into a format that can be analyzed as a color image.

[0586] Step 6:

[0587] The server inputs the image data into the generative AI model

[0588] The server preprocesses the decoded image data into the format required by the generative AI model and inputs it into the generative AI model as tensor data.

[0589] Step 7:

[0590] Server-generated captions

[0591] Using a generative AI model, the system generates captions from input image data, including information about roads, obstacles, signs, traffic lights, and other features in the image.

[0592] Step 8:

[0593] The server inputs the voice data into the emotion engine.

[0594] The server inputs the received voice data into an emotion engine that analyzes the user's voice tone, speed, and context.

[0595] Step 9:

[0596] The server analyzes the user's emotions

[0597] The emotion engine analyzes the audio data to identify the user's emotional state (e.g., tension, anxiety, impatience).

[0598] Step 10:

[0599] The server generates guide information

[0600] The server generates guidance information for the user based on the generated captions and the analyzed user's emotional information. For example, if the user is nervous, it provides more detailed and careful guidance information.

[0601] Step 11:

[0602] The server converts the guide information into voice

[0603] The server converts the guide information into voice data using a text-to-speech engine (TTS).

[0604] Step 12:

[0605] The server sends the voice data to the user terminal.

[0606] The server transmits the generated voice data to the user terminal via the network.

[0607] Step 13:

[0608] The device receives the audio data.

[0609] The user terminal receives the audio data sent from the server and prepares to play the data.

[0610] Step 14:

[0611] The device plays the audio data.

[0612] The user terminal reproduces the received audio data to the user through a speaker or earphone.

[0613] Step 15:

[0614] The user walks according to the audio guide

[0615] The user continues walking safely by following the audio guidance played from the device, which is adjusted appropriately according to the user's emotions.

[0616] Example 2

[0617] 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."

[0618] There is a need for technology to assist visually impaired people in walking safely and independently. Conventional voice guidance systems have difficulty grasping the surrounding situation in real time and providing appropriate instructions. Furthermore, because they do not take the user's emotional state into consideration, they are insufficient in providing support to improve the user's sense of security. Furthermore, there is a need for technology that can receive voice input, analyze emotions, and adjust guide information based on that.

[0619] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0620] In this invention, the server includes means for receiving image data transmitted from a user terminal having a camera, means for analyzing the received image data and using a generative AI model that recognizes roads, obstacles, signs, pedestrian signals, etc., means for generating guide information for the user based on the content recognized by the generative AI model, means for converting the generated guide information into voice and notifying the user, means for receiving voice input from the user and analyzing their emotional state, and means for adjusting the guide information based on the analyzed emotional state. This not only enables visually impaired people to grasp their surroundings in real time and walk safely, but also enables personalized support according to the user's emotional state.

[0621] A "user device with a camera" refers to a device that is equipped with a camera for acquiring visual information and that can be carried by the user. This includes smartphones and dedicated mobile devices.

[0622] The "means for receiving image data" refers to the communication technology and protocol for receiving image data sent from a user terminal having a camera and importing it into the server.

[0623] A "generative AI model" is a model trained using machine learning and deep learning techniques and used to analyze image data to recognize roads, obstacles, signs, pedestrian signals, etc.

[0624] "Means for generating guide information" refers to the process of generating information that instructs the user on the surrounding situation and direction of travel based on the analysis results of the generative AI model.

[0625] "Means for converting into voice and notifying the user" refers to technology for converting the generated guide information into voice data using voice synthesis technology and notifying the user of that voice in real time.

[0626] "Means for receiving voice input and analyzing emotional state" refers to processes and techniques for receiving a user's voice input and analyzing the voice data to identify the user's emotional state (e.g., tension, anxiety, impatience, etc.).

[0627] "Means for adjusting guide information" refers to the process of appropriately changing or adjusting the generated guide information based on the analyzed user's emotional state, for example, by adding detailed instructions or simplifying them.

[0628] This invention is a system that supports visually impaired people to walk safely and independently. The system uses a server to analyze data sent from a camera-equipped device held by the user, and provides audio guidance based on the results.

[0629] A user carries a device with a camera and activates the device's camera while walking. The device periodically captures images of the surroundings and creates image data. This image data is sent to a server via a network. The user's voice input is also captured and sent to the server at the same time. Specifically, a smartphone or a dedicated mobile device is used.

[0630] The server receives image data and audio data sent from the user's device. The received image data is analyzed using a generative AI model. This generative AI model utilizes a deep learning model trained using, for example, TensorFlow or PyTorch. This model identifies objects such as roads, obstacles, signs, and pedestrian signals, and determines their location and status.

[0631] Based on the analysis results, the server generates guide information for the user. This guide information includes instructions such as "There is an obstacle ahead," "Turn left," and "The traffic light is red." The server also inputs the user's voice input into an emotion engine to analyze the user's emotional state. This emotion engine uses, for example, Google Dialogflow or IBM Watson.

[0632] Based on the recognized user's emotions, the server adjusts the content of the guide information accordingly. For example, if the user is nervous, the server provides more detailed and careful guide information. Conversely, if the user is relaxed, the server provides concise guide information. The generated guide information is converted into audio using speech synthesis technology (e.g., Google Text-to-Speech or Amazon Polly). The converted audio guide information is then sent back to the user's device via the network.

[0633] The user device plays back the audio guide information received from the server, allowing the user to understand their surroundings in real time and walk safely.

[0634] For example, when a user approaches an intersection, the device's camera captures the image. This image data is sent to the server and analyzed. The server recognizes that there is an obstacle before the intersection and generates guide information based on this information, such as "Go 5 meters to the right."

[0635] At the same time, the emotion engine analyzes the user's voice input and recognizes that the user is in an anxious state. Taking this into account, the guide information is adjusted to be more detailed, such as "There is an obstacle ahead. Please move 5 meters to the right. Please proceed slowly and without rushing." This guide information is converted into voice and sent to the user's device. The user's device then notifies the user by voice, "There is an obstacle ahead. Please move 5 meters to the right. Please proceed slowly and without rushing." This allows the user to pass through the intersection safely and with peace of mind.

[0636] An example of a prompt sentence is "There is an obstacle ahead. Please guide me in which direction I should go." This is used as input to the generative AI model, forming the basis for generating optimal guidance information.

[0637] This system enables visually impaired people to accurately grasp their surroundings based on visual information and walk safely through audio guidance. It also provides individualized support based on the user's emotional state, helping them live independently and with peace of mind.

[0638] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0639] Step 1:

[0640] A user carries a device with a camera and activates the device's camera while walking. An application on the device periodically uses the camera to capture images of the surroundings and create image data. Specifically, the application captures images twice per second and saves the captured data as JPEG image files. The inputs are user operations and camera images, and the output is JPEG image data.

[0641] Step 2:

[0642] The user's device sends the captured image and audio data to the server. The device compresses the generated JPEG image file and the simultaneously collected audio input data, and transfers them to the server via the Internet. In particular, it performs a process to convert the data into an appropriate format for transmission. The input is JPEG image data and audio data, and the compressed data is sent to the server as output.

[0643] Step 3:

[0644] The server receives image data and audio data sent from the user's device. The server decompresses the received data and converts the image data into tensor format for input into the generative AI model. Specifically, the image data is converted into numerical data for each pixel and processed into tensor format. The input is compressed image data and audio data, and the output is tensor-formatted image data.

[0645] Step 4:

[0646] The server analyzes tensor-formatted image data using a generative AI model. This model identifies objects such as roads, obstacles, signs, and pedestrian signals, and outputs their location and status. For example, a deep learning framework is used to recognize and classify objects. The input is tensor-formatted image data, and the output is the object identification results.

[0647] Step 5:

[0648] The server generates guide information for the user based on the results of image analysis. For example, it creates guide information that includes specific instructions such as "There is an obstacle ahead," "Turn left," and "The traffic light is red." The input is the object identification result, and the output is guide information.

[0649] Step 6:

[0650] The server analyzes the received voice data and inputs it into an emotion engine that recognizes the user's emotional state. The emotion engine analyzes the tone, speed, and context of the voice to identify emotional states such as tension, anxiety, and impatience. Specifically, it extracts features from the voice data and inputs them into an emotion classification model. The input is the voice data, and the output is the user's emotional state.

[0651] Step 7:

[0652] Based on the analyzed user's emotional state, the server adjusts the guide information accordingly. For example, if the user is nervous, the server makes the guide information more detailed and careful. The inputs are the emotional state and the initial guide information, and the output is the adjusted guide information.

[0653] Step 8:

[0654] The server converts the adjusted guide information into speech using speech synthesis technology. For example, it uses a text-to-speech system to create a natural-sounding audio guide. The adjusted guide information is input, and audio data is generated as output.

[0655] Step 9:

[0656] The server transmits the generated voice data to the user terminal. The voice data is transferred again in real time via the network. The voice data is input, and transmission to the user terminal is output.

[0657] Step 10:

[0658] The user device plays the audio guide information received from the server. The user continues walking safely by following the played audio guide. The input is audio data, and the output is the playback of audio guide. This allows the user to understand the surrounding situation in real time and walk safely.

[0659] (Application example 2)

[0660] 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."

[0661] Visually impaired people face difficulties when searching for products in physical stores because they cannot see their surroundings. They also need the help of others to understand the location of shelves and products, making it difficult for them to enjoy shopping independently.

[0662] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for receiving image data transmitted from a user terminal having a camera; means for analyzing the received image data and using a generative AI model that recognizes roads, obstacles, signs, pedestrian signals, etc.; means for generating guide information for the user based on the content recognized by the generative AI model; means for converting the generated guide information into voice and notifying the user; means for capturing the situation in a physical store using a camera in smart glasses and identifying product shelves and products; and means for locating the user in the store and generating guide information that guides them to the products. This enables visually impaired people to find products in a physical store and shop independently.

[0663] A "camera" is a device for capturing images, and is installed in smart glasses or a user terminal.

[0664] A "user terminal" is a device that has a camera and can send image data to a server.

[0665] "Image data" refers to video information of the surroundings captured by a camera.

[0666] A "generative AI model" is an artificial intelligence model that analyzes received image data and identifies specific objects.

[0667] "Guide information" refers to guidance and instruction information for users that is generated based on the content recognized by the generative AI model.

[0668] "Convert to voice" is a process for conveying text-format guide information to the user as voice.

[0669] "Smart glasses" are head-mounted display devices that incorporate cameras and other sensors and are worn by users.

[0670] A "brick and mortar store" is a physical sales location where products are displayed and customers can visit and purchase in person.

[0671] A "product shelf" is a shelf used to display products in a physical store.

[0672] "Determining the location" means determining the user's current location and obtaining accurate location information.

[0673] "Guidance" refers to the provision of information to assist a user in reaching a destination.

[0674] As an embodiment of this invention, we will explain a support system for visually impaired people to search for and purchase products in a physical store. The system is realized by combining smart glasses worn by the user, a server, a generative AI model, and an audio guide.

[0675] System configuration

[0676] Smart glasses and user terminal

[0677] Smart glasses have a built-in camera and are worn by the user. The camera captures the surrounding situation in real time and generates image data. The user device (such as a smartphone or tablet) periodically transmits this image data to a server.

[0678] Server and Generative AI Model

[0679] The server receives image data sent from the user's device. The received image data is input into the generative AI model for analysis. The generative AI model is trained using machine learning and deep learning techniques to identify product shelves, directions, obstacles, and other items in a physical store.

[0680] Guide information generation

[0681] The server generates guide information for the user based on the content recognized by the generative AI model. This information includes the user's location, the location of the desired product, and the route to take within the store. The guide information is generated in text format.

[0682] Text-to-Speech and Notifications

[0683] The generated guide information is converted into audio on the server side using speech synthesis technology. Specifically, Google Text-to-Speech (gTTS) is used. The converted audio guide information is then sent back to the user's device via the network. The user's device (smart glasses) then plays back this audio guide and notifies the user.

[0684] Operation example

[0685] When a user searches for a product in a store, the smart glasses capture the surroundings with a camera and send the image data to the server. For example, if a user says, "I'm looking for the condiments section," the following prompt sentences are generated as guide information:

[0686] "There is a shelf about three meters ahead. The item you are looking for is on the second shelf on the left. Please walk slowly."

[0687] System Operation

[0688] The server receives image data sent from the smart glasses with a camera and first performs image analysis. It processes the image using libraries such as OpenCV, extracts specific objects, and converts them into tensor format. This data is input into a generative AI model to identify shelf and product locations, the user's current location, and other information. Based on the recognized content, guide information for the user is generated in text format and converted into audio using gTTS. The converted audio guide information is sent to the user's device, and an audio notification is given to the user via the smart glasses. By following these audio notifications, visually impaired people can shop independently.

[0689] This system enables visually impaired people to safely and efficiently search for and purchase products in physical stores.

[0690] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0691] Step 1:

[0692] A user wears smart glasses and walks around a physical store. The camera in the smart glasses captures the surroundings and generates image data. This image data is sent to the user device. The input is the captured image, and the output is the image data.

[0693] Step 2:

[0694] The user terminal periodically transmits the captured image data to the server, where the input is the image data received from the smart glasses and the output is the image data transmitted to the server.

[0695] Step 3:

[0696] The server analyzes the image data received from the user device, converts this image data into tensor format, and inputs it into the generative AI model. The input is the received image data, and the output is tensor format data.

[0697] Step 4:

[0698] The generative AI model identifies product shelves, product locations, obstacles, etc. in a physical store based on tensor-formatted data. The input is tensor-formatted data, and the output is information about the identified content.

[0699] Step 5:

[0700] The server generates guide information for the user based on the content identified by the generative AI model. For example, this includes information about the product's location and travel route. The input is the identified content, and the output is text data of the guide information.

[0701] Step 6:

[0702] The server converts the generated guide information into audio using speech synthesis technology. Google Text-to-Speech (gTTS) is used here. The input is the text data of the guide information, and the output is an audio file.

[0703] Step 7:

[0704] The server converts the guide information into audio and sends it to the user terminal. The input is an audio file, and the output is the audio file sent to the user terminal.

[0705] Step 8:

[0706] The user terminal (smart glasses) plays the audio guide information received from the server and notifies the user. The input is an audio file, and the output is the played audio guide. This allows the user to search for and move around products in the store by following the audio guide.

[0707] 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.

[0708] 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.

[0709] 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.

[0710] [Third embodiment]

[0711] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0712] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0713] 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).

[0714] 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.

[0715] 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.

[0716] 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).

[0717] 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.

[0718] 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.

[0719] 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.

[0720] 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.

[0721] 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.

[0722] 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."

[0723] System Overview

[0724] This system provides audio guidance to help visually impaired people walk safely and independently. The server receives image data sent from the user's camera-equipped device, analyzes the images, and generates guide information based on the results, which is then sent to the user via audio.

[0725] User terminal operation

[0726] A user carries a device with a camera and activates the device's camera while walking. The device periodically captures images of the surroundings and creates image data. This image data is then sent to a server via a network.

[0727] Server Operation

[0728] The server receives image data sent from the user's device. The received image data is analyzed using a generative AI model. This AI model is trained using machine learning and deep learning techniques to identify roads, obstacles, signs, pedestrian signals, and other objects, and understand their location and status.

[0729] Based on the analysis results, the server generates guide information for the user, including instructions such as "There is an obstacle ahead," "Turn left," and "The traffic light is red."

[0730] The generated guide information is converted into voice using a voice synthesis technique, and the converted guide information is transmitted again to the user terminal via the network.

[0731] User Notification

[0732] The user device plays back the audio guide information received from the server, allowing the user to understand their surroundings in real time and walk safely.

[0733] Specific examples

[0734] For example, when a user approaches an intersection, the device's camera captures the image. This image data is sent to the server and analyzed. The server recognizes that "there is an obstacle ahead of the intersection," and based on this information generates guide information instructing the user to "go five meters to the right." This guide information is converted into audio and sent to the user's device. The user's device then notifies the user by audio, saying, "There is an obstacle ahead. Go five meters to the right." This allows the user to pass through the intersection safely.

[0735] Usage scenarios

[0736] By utilizing this system, visually impaired product delivery personnel can safely walk to their destination. Visually impaired people visiting tourist spots can also enjoy sightseeing independently while receiving guidance. Users can not only receive audio guidance, but also request specific instructions using their own voice input. This makes it a flexible assistance system that can adapt to a variety of situations.

[0737] The processing flow will be explained below.

[0738] Step 1:

[0739] The user activates the camera

[0740] The user activates the device's camera and prepares to capture video of the surroundings.

[0741] Step 2:

[0742] The device generates image data.

[0743] The device converts the video captured in real time into image data for each frame and temporarily stores it in a buffer.

[0744] Step 3:

[0745] The device sends the image data to the server.

[0746] The terminal periodically transmits the stored image data to the server as a POST request.

[0747] Step 4:

[0748] The server receives the image data.

[0749] The server receives the image data sent from the terminal and expands the data in memory.

[0750] Step 5:

[0751] The server decodes the image data

[0752] The server decodes the received image data using libraries such as OpenCV and converts it into a format that can be analyzed as a color image.

[0753] Step 6:

[0754] The server inputs the image data into the generative AI model

[0755] The server preprocesses the decoded image data into the format required by the generative AI model and inputs it into the generative AI model as tensor data.

[0756] Step 7:

[0757] Server-generated captions

[0758] Using a generative AI model, the system generates captions from input image data, including information about roads, obstacles, signs, traffic lights, and other features in the image.

[0759] Step 8:

[0760] The server converts the captions into audio

[0761] The server converts the generated captions into audio data using a text-to-speech engine (TTS).

[0762] Step 9:

[0763] The server sends the voice data to the user terminal.

[0764] The server returns the generated voice data to the user terminal via the network.

[0765] Step 10:

[0766] The device receives the audio data.

[0767] The user terminal receives the voice data transmitted from the server.

[0768] Step 11:

[0769] The device plays the audio data.

[0770] The user terminal reproduces the received audio data to the user through a speaker or earphone.

[0771] Step 12:

[0772] The user walks according to the audio guide

[0773] The user listens to the audio guidance from the device and continues walking safely.

[0774] Example 1

[0775] 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."

[0776] For visually impaired people to walk safely and independently, they need a way to grasp their surroundings in real time and take appropriate action. However, conventional technology lacks the accuracy and real-time capabilities of analyzing image data, and the ability to reflect user feedback, making it difficult to provide sufficiently safe guidance. Furthermore, the inability to respond to voice input instructions means that the technology is unable to respond flexibly to the situation.

[0777] 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.

[0778] In this invention, the server includes: means for receiving image data transmitted from a user terminal equipped with a camera; means for analyzing the received image data and using a generative AI model to recognize roads, obstacles, signs, pedestrian signals, etc.; means for generating guide information for the user based on the content recognized by the generative AI model; means for converting the generated guide information into voice and notifying the user; and means for receiving voice input transmitted from the user terminal and correcting the guide information based on the voice input. This allows visually impaired people to receive safe guide information in real time and walk independently. Furthermore, specific instructions can be requested by voice input, enabling flexible responses according to the situation.

[0779] A "user terminal with a camera" is a portable electronic device that has the function of capturing video, processing it as image data, and transmitting it to a server via a network.

[0780] The "means for receiving image data" is a function that allows the server to receive image data transmitted via the network.

[0781] "Means for analyzing image data" refers to a function that uses a generative AI model to analyze image data received by the server.

[0782] A "generative AI model" is an artificial intelligence model trained using machine learning and deep learning technologies to recognize roads, obstacles, signs, pedestrian signals, etc. from image data and understand their location and status.

[0783] The "means for generating guide information" is a function for creating appropriate instructions for the user in text format based on the analyzed data.

[0784] The "means for converting guide information into voice" is a function for converting the generated text-format guide information into voice data using voice synthesis technology.

[0785] The "means for notifying the user" is a function that transmits guide information converted into voice to the user terminal so that the user can hear it.

[0786] "Means for receiving voice input" refers to a function that allows a user to send specific instructions or questions to the server by voice.

[0787] The "means for correcting guide information based on voice input" is a function for analyzing voice input from the user and updating or correcting the guide information to be provided in accordance with the contents of the voice input.

[0788] System Overview

[0789] This invention is a system that provides audio guidance to help visually impaired people walk safely and independently. The server receives image data sent from the user's camera-equipped device and analyzes it using a generative AI model. Based on the analysis results, it generates guide information, converts it into audio, and notifies the user.

[0790] User terminal operation

[0791] A user carries a device with a camera and activates the device's camera while walking. The device periodically captures images of the surroundings and creates image data. This image data is sent to a server via a network. The device also has a function to accept voice input from the user.

[0792] Server Operation

[0793] The server receives image data sent from the user device using an HTTP request. The received image data is converted into an appropriate format and input into the generative AI model. This generative AI model is trained using machine learning and deep learning techniques, such as TensorFlow and PyTorch.

[0794] The server uses the generative AI model to analyze image data, identify roads, obstacles, signs, pedestrian signals, and other objects, and determine their location and status. Based on the results of this analysis, it generates guide information for the user. This guide information includes instructions such as "There is an obstacle ahead," "Turn left," and "The traffic light is red."

[0795] The server converts the generated guide information into speech using speech synthesis technology such as Google Text-to-Speech (TTS) or Amazon Polly. The converted speech guide information is then sent back to the user's device via the network. The server can also receive speech input from the user's device and modify the guide information based on that input.

[0796] User Notification

[0797] The user device plays back the audio guide information received from the server. This allows the user to understand their surroundings in real time and walk safely. For example, when the user approaches an intersection, the device's camera captures the image and the server recognizes that "there is an obstacle before the intersection." Based on this information, guide information instructing "please go 5 meters to the right" is generated, converted into voice and notified to the user. This allows the user to pass through the intersection safely.

[0798] Examples of prompt statements

[0799] For example, the following prompts might be used when building a generative AI model:

[0800] "Design a system that uses machine learning and deep learning techniques to recognize and analyze obstacles in captured images in real time and provide audio guidance to the user."

[0801] This invention provides comprehensive support for the visually impaired to walk safely and independently, greatly improving safety and convenience, and aiming to make mobility in daily life for the visually impaired more flexible and comfortable.

[0802] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0803] Step 1:

[0804] A user activates a device equipped with a camera and uses the device's camera to capture images of the surroundings while walking. The device captures the images as still images at regular intervals and generates image data.

[0805] Input: Video captured by the camera

[0806] Output: Generated image data

[0807] Step 2:

[0808] The terminal transmits the generated image data to the server via the network using an HTTP request.

[0809] Input: Generated image data

[0810] Output: Image data sent to the server

[0811] Step 3:

[0812] The server receives the image data sent from the terminal, converts the received data into an appropriate format, and prepares it for the next analysis step.

[0813] Input: Image data sent from the device

[0814] Output: Format-converted image data

[0815] Step 4:

[0816] The server inputs the received image data into a generative AI model. The generative AI model is built using libraries such as TensorFlow and PyTorch, and analyzes the image data to identify roads, obstacles, signs, pedestrian signals, etc. Specifically, it detects different objects in the image and understands their respective positions and states.

[0817] Input: Format-converted image data

[0818] Output: Analysis results (location and status of roads, obstacles, signs, pedestrian signals, etc.)

[0819] Step 5:

[0820] The server generates guide information for the user based on the analysis results of the generative AI model, including instructions such as "There is an obstacle ahead," "Turn left," and "The traffic light is red."

[0821] Input: Analysis results

[0822] Output: Generated guide information

[0823] Step 6:

[0824] The server converts the generated guide information into speech using speech synthesis technology, using services such as Google Text-to-Speech (TTS) and Amazon Polly.

[0825] Input: Generated guide information

[0826] Output: Audio data

[0827] Step 7:

[0828] The server then transmits the guide information converted into audio to the user terminal via the network using the HTTPS protocol.

[0829] Input: Audio data

[0830] Output: Audio data sent to the user's device

[0831] Step 8:

[0832] The user device plays back the received voice guidance information. As a result, the user can grasp the surrounding situation in real time and take appropriate action. For example, if the voice guidance instructs, "There is an obstacle ahead. Please move 5 meters to the right," the user will act accordingly.

[0833] Input: Audio data sent to the user's device

[0834] Output: Played audio description information

[0835] Step 9:

[0836] The user can use their voice input to request specific instructions or questions from the server, such as "What's the next intersection like?", and the device transmits this voice input to the server.

[0837] Input: User voice input

[0838] Output: Audio input sent to the server

[0839] Step 10:

[0840] The server receives voice input sent from the user terminal, analyzes the content, and updates or modifies the guide information. For example, it generates new guide information such as "There is a traffic light at the next intersection. Turn left."

[0841] Input: Voice input from the user

[0842] Output: Updated or corrected guide information

[0843] Step 11:

[0844] The updated guide information is converted back into audio and sent to the user terminal via the network. The user terminal then plays back the new guide information and notifies the user. This allows the user to respond flexibly according to the situation.

[0845] Input: Updated guide information

[0846] Output: New audio guide information sent to the user's device.

[0847] (Application example 1)

[0848] 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."

[0849] It is extremely difficult for visually impaired people to navigate safely on public transportation or in crowded places, and conventional technologies may not be able to provide sufficient support. Recognizing roads, obstacles, signs, pedestrian signals, and other obstacles is particularly difficult, requiring accurate information. Additionally, there is a lack of systems that provide users with real-time guidance information. The purpose of this invention is to solve these problems and provide an audio guidance system that enables visually impaired people to navigate safely.

[0850] 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.

[0851] In this invention, the server includes means for receiving image data transmitted from a user device having a camera, means for analyzing the received image data and using a generative AI model that recognizes roads, obstacles, signs, pedestrian signals, etc., means for generating guide information for the user based on the content recognized by the generative AI model, means for converting the generated guide information into audio and notifying the user, means for periodically transmitting image data from the user device to the server to support safe behavior on public transportation and in crowded places, and means for the generative AI model to provide analysis results and notify the user by audio. This enables visually impaired people to grasp their surroundings in real time and act safely.

[0852] A "user device" is a terminal that has a camera and can be carried around by a visually impaired person.

[0853] "Image data" is digital data of visual information captured by a camera on a user device and transmitted to a server.

[0854] A "generative AI model" is a model trained using machine learning and deep learning techniques that analyzes image data to recognize roads, obstacles, signs, pedestrian signals, etc.

[0855] "Guide information" is information that provides instructions and advice to the user, which is generated based on the analysis results, and is notified by voice.

[0856] The "means for converting into voice" is a function for converting the generated guide information into voice data using voice synthesis technology.

[0857] "Public transport and crowded places" refers to places and situations where many people use or gather, and where safe movement is particularly important for people with visual impairments.

[0858] The "periodic transmission means" is a function that allows the user device to transmit image data to the server at regular time intervals.

[0859] "Means of providing analysis results and notifying by voice" refers to a function that uses voice synthesis technology to notify the user in real time based on the analysis results of the generated AI model on the received image data.

[0860] The present invention relates to an audio guide system that supports visually impaired people to navigate safely on public transportation and in crowded places. Hereinafter, an embodiment will be described in detail.

[0861] System Overview

[0862] The system is composed of a user device, a server, and a generative AI model.

[0863] User Device Operation

[0864] The user device has a built-in camera and can be carried around by visually impaired people. The user activates the camera while walking and periodically captures images of the surroundings. The captured image data is sent to a server at regular intervals.

[0865] Server Operation

[0866] The server receives image data sent from the user device. The received image data is analyzed using a generative AI model. The generative AI model is trained using machine learning and deep learning techniques to accurately identify roads, obstacles, signs, pedestrian signals, and other objects. Based on the analysis results, the server generates guide information for the user. The generated guide information is converted into voice data using speech synthesis technology and transmitted to the user device.

[0867] User Notification

[0868] The user device plays back the audio data received from the server, providing real-time audio notification of the surrounding situation to the visually impaired, allowing the user to act safely.

[0869] Hardware and Software

[0870] The hardware required to implement this system is a user device with a camera (e.g., smart glasses) and a server. The software uses Python, OpenCV, the Requests library, etc. The generative AI model performs analysis based on tensor-formatted data, and the speech synthesis technology uses an API.

[0871] Specific examples

[0872] For example, if a user is walking down a busy street, they can send the following prompt to the generative AI model:

[0873] Example prompt sentence:

[0874] "Please generate guide information for walking safely in busy areas with many pedestrians. Please also include information about obstacles, dangerous areas, and traffic light status in the notification content."

[0875] Example result:

[0876] The server analyzes the prompt text and generates guide information such as, "There are many pedestrians ahead, so please move to the right. Be careful, there is a traffic light ahead, but it is red." and notifies the user device by voice.

[0877] As described above, the present invention enables visually impaired people to grasp the situation around them in real time and act safely.

[0878] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0879] Step 1:

[0880] A user activates the camera of a user device and periodically captures images of the surroundings. As input, the user device acquires real-time image data through the camera. This image data is saved as image data in JPEG format, etc. As output, the image data is stored in the user device.

[0881] Step 2:

[0882] The user device sends the accumulated image data to the server at regular intervals. As input, the latest image data stored on the user device is used to send an HTTP POST request to the server's API endpoint. As output, the image data is uploaded to the server.

[0883] Step 3:

[0884] The server receives image data sent from the user device. As input, it obtains the image data in the received HTTP POST request and stores it internally. As output, it obtains the image data newly stored in the server.

[0885] Step 4:

[0886] The server inputs the received image data into the generative AI model for analysis. As input, it converts the stored image data into tensor format and passes it to the generative AI model. The generative AI model analyzes the image data and identifies objects such as roads, obstacles, signs, and pedestrian signals. The results obtained from this analysis are the output.

[0887] Step 5:

[0888] The server generates guide information for the user based on the analysis results. It uses the analysis result data output by the generative AI model as input and generates guide information according to predefined rules and patterns. For example, text information such as "There is an obstacle ahead" or "Please move to the right and proceed" is generated. The generated guide information text data is obtained as output.

[0889] Step 6:

[0890] The server converts the generated guide information into voice data using voice synthesis technology. As input, the text data of the guide information is passed to the voice synthesis API, which generates voice data. As output, the generated voice data is obtained.

[0891] Step 7:

[0892] The server transmits the generated voice data to the user device. As input, the generated voice data is transmitted to the user device as an HTTP response. As output, new voice data to be received by the user device is obtained.

[0893] Step 8:

[0894] The user device plays the audio data received from the server and notifies the user. As input, the received audio data is passed to an internal audio player and played. As output, the user receives audio information and can understand the surrounding situation.

[0895] 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.

[0896] System Overview

[0897] This system provides audio guidance to help visually impaired people walk safely and independently. The server receives image data sent from the user's camera-equipped device, analyzes the images, and generates guide information based on the results, which is then notified to the user via audio. It also incorporates an emotion engine that recognizes the user's emotions, providing guidance tailored to each individual situation, thereby achieving more personalized assistance.

[0898] User terminal operation

[0899] A user carries a device with a camera and activates the device's camera while walking. The device periodically captures images of the surroundings and creates image data. This image data is sent to a server via a network. The user's voice input is also captured and sent to the server at the same time.

[0900] Server Operation

[0901] The server receives image data and voice input data sent from the user's device. The received image data is analyzed using a generative AI model. This AI model is trained using machine learning and deep learning techniques to identify roads, obstacles, signs, pedestrian signals, and other objects, and understand their location and status.

[0902] Based on the analysis results, the server generates guide information for the user, including instructions such as "There is an obstacle ahead," "Turn left," and "The traffic light is red."

[0903] Emotion Engine Operation

[0904] The server inputs the user's voice input into an emotion engine to recognize the user's emotions. The emotion engine analyzes the tone, rate, and context of the voice input to identify the user's emotional state (e.g., nervousness, anxiety, impatience).

[0905] Guide information adjustments

[0906] Based on the user's recognized emotions, the server adjusts the content of the guide information accordingly. For example, if the user is nervous, the server will provide more detailed and careful guide information. Conversely, if the user is relaxed, the server will provide simpler guide information.

[0907] The generated guide information is converted into voice using a voice synthesis technique, and the converted guide information is transmitted again to the user terminal via the network.

[0908] User Notification

[0909] The user device plays back the audio guide information received from the server, allowing the user to understand their surroundings in real time and walk safely.

[0910] Specific examples

[0911] For example, when a user approaches an intersection, the device's camera captures the image. This image data is sent to the server and analyzed. The server recognizes that there is an obstacle before the intersection and generates guide information based on this information, such as "Go 5 meters to the right."

[0912] At the same time, the emotion engine analyzes the user's voice input and recognizes that the user is anxious. Taking this into account, the guidance information is adjusted to be more detailed, such as "There is an obstacle ahead. Move 5 meters to the right. Please proceed slowly and without rushing."

[0913] This guide information is converted into audio and sent to the user's device. The user device then notifies the user by audio, saying, "There is an obstacle ahead. Please move 5 meters to the right. Please proceed slowly and without rushing." This allows the user to pass through the intersection safely and with peace of mind.

[0914] The processing flow will be explained below.

[0915] Step 1:

[0916] The user activates the camera

[0917] The user activates the device's camera and prepares to capture video of the surroundings.

[0918] Step 2:

[0919] The device generates image data and audio data.

[0920] The device converts the captured video into image data for each frame in real time and stores it temporarily in a buffer. It also captures the user's voice input and stores it as audio data.

[0921] Step 3:

[0922] The device sends image data and audio data to the server.

[0923] The terminal periodically transmits the stored image data and audio data to the server as a POST request.

[0924] Step 4:

[0925] The server receives the image data and audio data.

[0926] The server receives the image data and audio data sent from the terminal and expands each data in memory.

[0927] Step 5:

[0928] The server decodes the image data

[0929] The server decodes the received image data using libraries such as OpenCV and converts it into a format that can be analyzed as a color image.

[0930] Step 6:

[0931] The server inputs the image data into the generative AI model

[0932] The server preprocesses the decoded image data into the format required by the generative AI model and inputs it into the generative AI model as tensor data.

[0933] Step 7:

[0934] Server-generated captions

[0935] Using a generative AI model, the system generates captions from input image data, including information about roads, obstacles, signs, traffic lights, and other features in the image.

[0936] Step 8:

[0937] The server inputs the voice data into the emotion engine.

[0938] The server inputs the received voice data into an emotion engine that analyzes the user's voice tone, speed, and context.

[0939] Step 9:

[0940] The server analyzes the user's emotions

[0941] The emotion engine analyzes the audio data to identify the user's emotional state (e.g., tension, anxiety, impatience).

[0942] Step 10:

[0943] The server generates guide information

[0944] The server generates guidance information for the user based on the generated captions and the analyzed user's emotional information. For example, if the user is nervous, it provides more detailed and careful guidance information.

[0945] Step 11:

[0946] The server converts the guide information into voice

[0947] The server converts the guide information into voice data using a text-to-speech engine (TTS).

[0948] Step 12:

[0949] The server sends the voice data to the user terminal.

[0950] The server transmits the generated voice data to the user terminal via the network.

[0951] Step 13:

[0952] The device receives the audio data.

[0953] The user terminal receives the audio data sent from the server and prepares to play the data.

[0954] Step 14:

[0955] The device plays the audio data.

[0956] The user terminal reproduces the received audio data to the user through a speaker or earphone.

[0957] Step 15:

[0958] The user walks according to the audio guide

[0959] The user continues walking safely by following the audio guidance played from the device, which is adjusted appropriately according to the user's emotions.

[0960] Example 2

[0961] 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."

[0962] There is a need for technology to assist visually impaired people in walking safely and independently. Conventional voice guidance systems have difficulty grasping the surrounding situation in real time and providing appropriate instructions. Furthermore, because they do not take the user's emotional state into consideration, they are insufficient in providing support to improve the user's sense of security. Furthermore, there is a need for technology that can receive voice input, analyze emotions, and adjust guide information based on that.

[0963] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0964] In this invention, the server includes means for receiving image data transmitted from a user terminal having a camera, means for analyzing the received image data and using a generative AI model that recognizes roads, obstacles, signs, pedestrian signals, etc., means for generating guide information for the user based on the content recognized by the generative AI model, means for converting the generated guide information into voice and notifying the user, means for receiving voice input from the user and analyzing their emotional state, and means for adjusting the guide information based on the analyzed emotional state. This not only enables visually impaired people to grasp their surroundings in real time and walk safely, but also enables personalized support according to the user's emotional state.

[0965] A "user device with a camera" refers to a device that is equipped with a camera for acquiring visual information and that can be carried by the user. This includes smartphones and dedicated mobile devices.

[0966] The "means for receiving image data" refers to the communication technology and protocol for receiving image data sent from a user terminal having a camera and importing it into the server.

[0967] A "generative AI model" is a model trained using machine learning and deep learning techniques and used to analyze image data to recognize roads, obstacles, signs, pedestrian signals, etc.

[0968] "Means for generating guide information" refers to the process of generating information that instructs the user on the surrounding situation and direction of travel based on the analysis results of the generative AI model.

[0969] "Means for converting into voice and notifying the user" refers to technology for converting the generated guide information into voice data using voice synthesis technology and notifying the user of that voice in real time.

[0970] "Means for receiving voice input and analyzing emotional state" refers to processes and techniques for receiving a user's voice input and analyzing the voice data to identify the user's emotional state (e.g., tension, anxiety, impatience, etc.).

[0971] "Means for adjusting guide information" refers to the process of appropriately changing or adjusting the generated guide information based on the analyzed user's emotional state, for example, by adding detailed instructions or simplifying them.

[0972] This invention is a system that supports visually impaired people to walk safely and independently. The system uses a server to analyze data sent from a camera-equipped device held by the user, and provides audio guidance based on the results.

[0973] A user carries a device with a camera and activates the device's camera while walking. The device periodically captures images of the surroundings and creates image data. This image data is sent to a server via a network. The user's voice input is also captured and sent to the server at the same time. Specifically, a smartphone or a dedicated mobile device is used.

[0974] The server receives image data and audio data sent from the user's device. The received image data is analyzed using a generative AI model. This generative AI model utilizes a deep learning model trained using, for example, TensorFlow or PyTorch. This model identifies objects such as roads, obstacles, signs, and pedestrian signals, and determines their location and status.

[0975] Based on the analysis results, the server generates guide information for the user. This guide information includes instructions such as "There is an obstacle ahead," "Turn left," and "The traffic light is red." The server also inputs the user's voice input into an emotion engine to analyze the user's emotional state. This emotion engine uses, for example, Google Dialogflow or IBM Watson.

[0976] Based on the recognized user's emotions, the server adjusts the content of the guide information accordingly. For example, if the user is nervous, the server provides more detailed and careful guide information. Conversely, if the user is relaxed, the server provides concise guide information. The generated guide information is converted into audio using speech synthesis technology (e.g., Google Text-to-Speech or Amazon Polly). The converted audio guide information is then sent back to the user's device via the network.

[0977] The user device plays back the audio guide information received from the server, allowing the user to understand their surroundings in real time and walk safely.

[0978] For example, when a user approaches an intersection, the device's camera captures the image. This image data is sent to the server and analyzed. The server recognizes that there is an obstacle before the intersection and generates guide information based on this information, such as "Go 5 meters to the right."

[0979] At the same time, the emotion engine analyzes the user's voice input and recognizes that the user is in an anxious state. Taking this into account, the guide information is adjusted to be more detailed, such as "There is an obstacle ahead. Please move 5 meters to the right. Please proceed slowly and without rushing." This guide information is converted into voice and sent to the user's device. The user's device then notifies the user by voice, "There is an obstacle ahead. Please move 5 meters to the right. Please proceed slowly and without rushing." This allows the user to pass through the intersection safely and with peace of mind.

[0980] An example of a prompt sentence is "There is an obstacle ahead. Please guide me in which direction I should go." This is used as input to the generative AI model, forming the basis for generating optimal guidance information.

[0981] This system enables visually impaired people to accurately grasp their surroundings based on visual information and walk safely through audio guidance. It also provides individualized support based on the user's emotional state, helping them live independently and with peace of mind.

[0982] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0983] Step 1:

[0984] A user carries a device with a camera and activates the device's camera while walking. An application on the device periodically uses the camera to capture images of the surroundings and create image data. Specifically, the application captures images twice per second and saves the captured data as JPEG image files. The inputs are user operations and camera images, and the output is JPEG image data.

[0985] Step 2:

[0986] The user's device sends the captured image and audio data to the server. The device compresses the generated JPEG image file and the simultaneously collected audio input data, and transfers them to the server via the Internet. In particular, it performs a process to convert the data into an appropriate format for transmission. The input is JPEG image data and audio data, and the compressed data is sent to the server as output.

[0987] Step 3:

[0988] The server receives image data and audio data sent from the user's device. The server decompresses the received data and converts the image data into tensor format for input into the generative AI model. Specifically, the image data is converted into numerical data for each pixel and processed into tensor format. The input is compressed image data and audio data, and the output is tensor-formatted image data.

[0989] Step 4:

[0990] The server analyzes tensor-formatted image data using a generative AI model. This model identifies objects such as roads, obstacles, signs, and pedestrian signals, and outputs their location and status. For example, a deep learning framework is used to recognize and classify objects. The input is tensor-formatted image data, and the output is the object identification results.

[0991] Step 5:

[0992] The server generates guide information for the user based on the results of image analysis. For example, it creates guide information that includes specific instructions such as "There is an obstacle ahead," "Turn left," and "The traffic light is red." The input is the object identification result, and the output is guide information.

[0993] Step 6:

[0994] The server analyzes the received voice data and inputs it into an emotion engine that recognizes the user's emotional state. The emotion engine analyzes the tone, speed, and context of the voice to identify emotional states such as tension, anxiety, and impatience. Specifically, it extracts features from the voice data and inputs them into an emotion classification model. The input is the voice data, and the output is the user's emotional state.

[0995] Step 7:

[0996] Based on the analyzed user's emotional state, the server adjusts the guide information accordingly. For example, if the user is nervous, the server makes the guide information more detailed and careful. The inputs are the emotional state and the initial guide information, and the output is the adjusted guide information.

[0997] Step 8:

[0998] The server converts the adjusted guide information into speech using speech synthesis technology. For example, it uses a text-to-speech system to create a natural-sounding audio guide. The adjusted guide information is input, and audio data is generated as output.

[0999] Step 9:

[1000] The server transmits the generated voice data to the user terminal. The voice data is transferred again in real time via the network. The voice data is input, and transmission to the user terminal is output.

[1001] Step 10:

[1002] The user device plays the audio guide information received from the server. The user continues walking safely by following the played audio guide. The input is audio data, and the output is the playback of audio guide. This allows the user to understand the surrounding situation in real time and walk safely.

[1003] (Application example 2)

[1004] 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."

[1005] Visually impaired people face difficulties when searching for products in physical stores because they cannot see their surroundings. They also need the help of others to understand the location of shelves and products, making it difficult for them to enjoy shopping independently.

[1006] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for receiving image data transmitted from a user terminal having a camera; means for analyzing the received image data and using a generative AI model that recognizes roads, obstacles, signs, pedestrian signals, etc.; means for generating guide information for the user based on the content recognized by the generative AI model; means for converting the generated guide information into voice and notifying the user; means for capturing the situation in a physical store using a camera in smart glasses and identifying product shelves and products; and means for locating the user in the store and generating guide information that guides them to the products. This enables visually impaired people to find products in a physical store and shop independently.

[1007] A "camera" is a device for capturing images, and is installed in smart glasses or a user terminal.

[1008] A "user terminal" is a device that has a camera and can send image data to a server.

[1009] "Image data" refers to video information of the surroundings captured by a camera.

[1010] A "generative AI model" is an artificial intelligence model that analyzes received image data and identifies specific objects.

[1011] "Guide information" refers to guidance and instruction information for users that is generated based on the content recognized by the generative AI model.

[1012] "Convert to voice" is a process for conveying text-format guide information to the user as voice.

[1013] "Smart glasses" are head-mounted display devices that incorporate cameras and other sensors and are worn by users.

[1014] A "brick and mortar store" is a physical sales location where products are displayed and customers can visit and purchase in person.

[1015] A "product shelf" is a shelf used to display products in a physical store.

[1016] "Determining the location" means determining the user's current location and obtaining accurate location information.

[1017] "Guidance" refers to the provision of information to assist a user in reaching a destination.

[1018] As an embodiment of this invention, we will explain a support system for visually impaired people to search for and purchase products in a physical store. The system is realized by combining smart glasses worn by the user, a server, a generative AI model, and an audio guide.

[1019] System configuration

[1020] Smart glasses and user terminal

[1021] Smart glasses have a built-in camera and are worn by the user. The camera captures the surrounding situation in real time and generates image data. The user device (such as a smartphone or tablet) periodically transmits this image data to a server.

[1022] Server and Generative AI Model

[1023] The server receives image data sent from the user's device. The received image data is input into the generative AI model for analysis. The generative AI model is trained using machine learning and deep learning techniques to identify product shelves, directions, obstacles, and other items in a physical store.

[1024] Guide information generation

[1025] The server generates guide information for the user based on the content recognized by the generative AI model. This information includes the user's location, the location of the desired product, and the route to take within the store. The guide information is generated in text format.

[1026] Text-to-Speech and Notifications

[1027] The generated guide information is converted into audio on the server side using speech synthesis technology. Specifically, Google Text-to-Speech (gTTS) is used. The converted audio guide information is then sent back to the user's device via the network. The user's device (smart glasses) then plays back this audio guide and notifies the user.

[1028] Operation example

[1029] When a user searches for a product in a store, the smart glasses capture the surroundings with a camera and send the image data to the server. For example, if a user says, "I'm looking for the condiments section," the following prompt sentences are generated as guide information:

[1030] "There is a shelf about three meters ahead. The item you are looking for is on the second shelf on the left. Please walk slowly."

[1031] System Operation

[1032] The server receives image data sent from the smart glasses with a camera and first performs image analysis. It processes the image using libraries such as OpenCV, extracts specific objects, and converts them into tensor format. This data is input into a generative AI model to identify shelf and product locations, the user's current location, and other information. Based on the recognized content, guide information for the user is generated in text format and converted into audio using gTTS. The converted audio guide information is sent to the user's device, and an audio notification is given to the user via the smart glasses. By following these audio notifications, visually impaired people can shop independently.

[1033] This system enables visually impaired people to safely and efficiently search for and purchase products in physical stores.

[1034] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1035] Step 1:

[1036] A user wears smart glasses and walks around a physical store. The camera in the smart glasses captures the surroundings and generates image data. This image data is sent to the user device. The input is the captured image, and the output is the image data.

[1037] Step 2:

[1038] The user terminal periodically transmits the captured image data to the server, where the input is the image data received from the smart glasses and the output is the image data transmitted to the server.

[1039] Step 3:

[1040] The server analyzes the image data received from the user device, converts this image data into tensor format, and inputs it into the generative AI model. The input is the received image data, and the output is tensor format data.

[1041] Step 4:

[1042] The generative AI model identifies product shelves, product locations, obstacles, etc. in a physical store based on tensor-formatted data. The input is tensor-formatted data, and the output is information about the identified content.

[1043] Step 5:

[1044] The server generates guide information for the user based on the content identified by the generative AI model. For example, this includes information about the product's location and travel route. The input is the identified content, and the output is text data of the guide information.

[1045] Step 6:

[1046] The server converts the generated guide information into audio using speech synthesis technology. Google Text-to-Speech (gTTS) is used here. The input is the text data of the guide information, and the output is an audio file.

[1047] Step 7:

[1048] The server converts the guide information into audio and sends it to the user terminal. The input is an audio file, and the output is the audio file sent to the user terminal.

[1049] Step 8:

[1050] The user terminal (smart glasses) plays the audio guide information received from the server and notifies the user. The input is an audio file, and the output is the played audio guide. This allows the user to search for and move around products in the store by following the audio guide.

[1051] 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.

[1052] 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.

[1053] 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.

[1054] [Fourth embodiment]

[1055] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1056] 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.

[1057] 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).

[1058] 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.

[1059] 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.

[1060] 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).

[1061] 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.

[1062] 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.

[1063] 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.

[1064] 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.

[1065] 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.

[1066] 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.

[1067] 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."

[1068] System Overview

[1069] This system provides audio guidance to help visually impaired people walk safely and independently. The server receives image data sent from the user's camera-equipped device, analyzes the images, and generates guide information based on the results, which is then sent to the user via audio.

[1070] User terminal operation

[1071] A user carries a device with a camera and activates the device's camera while walking. The device periodically captures images of the surroundings and creates image data. This image data is then sent to a server via a network.

[1072] Server Operation

[1073] The server receives image data sent from the user's device. The received image data is analyzed using a generative AI model. This AI model is trained using machine learning and deep learning techniques to identify roads, obstacles, signs, pedestrian signals, and other objects, and understand their location and status.

[1074] Based on the analysis results, the server generates guide information for the user, including instructions such as "There is an obstacle ahead," "Turn left," and "The traffic light is red."

[1075] The generated guide information is converted into voice using a voice synthesis technique, and the converted guide information is transmitted again to the user terminal via the network.

[1076] User Notification

[1077] The user device plays back the audio guide information received from the server, allowing the user to understand their surroundings in real time and walk safely.

[1078] Specific examples

[1079] For example, when a user approaches an intersection, the device's camera captures the image. This image data is sent to the server and analyzed. The server recognizes that "there is an obstacle ahead of the intersection," and based on this information generates guide information instructing the user to "go five meters to the right." This guide information is converted into audio and sent to the user's device. The user's device then notifies the user by audio, saying, "There is an obstacle ahead. Go five meters to the right." This allows the user to pass through the intersection safely.

[1080] Usage scenarios

[1081] By utilizing this system, visually impaired product delivery personnel can safely walk to their destination. Visually impaired people visiting tourist spots can also enjoy sightseeing independently while receiving guidance. Users can not only receive audio guidance, but also request specific instructions using their own voice input. This makes it a flexible assistance system that can adapt to a variety of situations.

[1082] The processing flow will be explained below.

[1083] Step 1:

[1084] The user activates the camera

[1085] The user activates the device's camera and prepares to capture video of the surroundings.

[1086] Step 2:

[1087] The device generates image data.

[1088] The device converts the video captured in real time into image data for each frame and temporarily stores it in a buffer.

[1089] Step 3:

[1090] The device sends the image data to the server.

[1091] The terminal periodically transmits the stored image data to the server as a POST request.

[1092] Step 4:

[1093] The server receives the image data.

[1094] The server receives the image data sent from the terminal and expands the data in memory.

[1095] Step 5:

[1096] The server decodes the image data

[1097] The server decodes the received image data using libraries such as OpenCV and converts it into a format that can be analyzed as a color image.

[1098] Step 6:

[1099] The server inputs the image data into the generative AI model

[1100] The server preprocesses the decoded image data into the format required by the generative AI model and inputs it into the generative AI model as tensor data.

[1101] Step 7:

[1102] Server-generated captions

[1103] Using a generative AI model, the system generates captions from input image data, including information about roads, obstacles, signs, traffic lights, and other features in the image.

[1104] Step 8:

[1105] The server converts the captions into audio

[1106] The server converts the generated captions into audio data using a text-to-speech engine (TTS).

[1107] Step 9:

[1108] The server sends the voice data to the user terminal.

[1109] The server returns the generated voice data to the user terminal via the network.

[1110] Step 10:

[1111] The device receives the audio data.

[1112] The user terminal receives the voice data transmitted from the server.

[1113] Step 11:

[1114] The device plays the audio data.

[1115] The user terminal reproduces the received audio data to the user through a speaker or earphone.

[1116] Step 12:

[1117] The user walks according to the audio guide

[1118] The user listens to the audio guidance from the device and continues walking safely.

[1119] Example 1

[1120] 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."

[1121] For visually impaired people to walk safely and independently, they need a way to grasp their surroundings in real time and take appropriate action. However, conventional technology lacks the accuracy and real-time capabilities of analyzing image data, and the ability to reflect user feedback, making it difficult to provide sufficiently safe guidance. Furthermore, the inability to respond to voice input instructions means that the technology is unable to respond flexibly to the situation.

[1122] 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.

[1123] In this invention, the server includes: means for receiving image data transmitted from a user terminal equipped with a camera; means for analyzing the received image data and using a generative AI model to recognize roads, obstacles, signs, pedestrian signals, etc.; means for generating guide information for the user based on the content recognized by the generative AI model; means for converting the generated guide information into voice and notifying the user; and means for receiving voice input transmitted from the user terminal and correcting the guide information based on the voice input. This allows visually impaired people to receive safe guide information in real time and walk independently. Furthermore, specific instructions can be requested by voice input, enabling flexible responses according to the situation.

[1124] A "user terminal with a camera" is a portable electronic device that has the function of capturing video, processing it as image data, and transmitting it to a server via a network.

[1125] The "means for receiving image data" is a function that allows the server to receive image data transmitted via the network.

[1126] "Means for analyzing image data" refers to a function that uses a generative AI model to analyze image data received by the server.

[1127] A "generative AI model" is an artificial intelligence model trained using machine learning and deep learning technologies to recognize roads, obstacles, signs, pedestrian signals, etc. from image data and understand their location and status.

[1128] The "means for generating guide information" is a function for creating appropriate instructions for the user in text format based on the analyzed data.

[1129] The "means for converting guide information into voice" is a function for converting the generated text-format guide information into voice data using voice synthesis technology.

[1130] The "means for notifying the user" is a function that transmits guide information converted into voice to the user terminal so that the user can hear it.

[1131] "Means for receiving voice input" refers to a function that allows a user to send specific instructions or questions to the server by voice.

[1132] The "means for correcting guide information based on voice input" is a function for analyzing voice input from the user and updating or correcting the guide information to be provided in accordance with the contents of the voice input.

[1133] System Overview

[1134] This invention is a system that provides audio guidance to help visually impaired people walk safely and independently. The server receives image data sent from the user's camera-equipped device and analyzes it using a generative AI model. Based on the analysis results, it generates guide information, converts it into audio, and notifies the user.

[1135] User terminal operation

[1136] A user carries a device with a camera and activates the device's camera while walking. The device periodically captures images of the surroundings and creates image data. This image data is sent to a server via a network. The device also has a function to accept voice input from the user.

[1137] Server Operation

[1138] The server receives image data sent from the user device using an HTTP request. The received image data is converted into an appropriate format and input into the generative AI model. This generative AI model is trained using machine learning and deep learning techniques, such as TensorFlow and PyTorch.

[1139] The server uses the generative AI model to analyze image data, identify roads, obstacles, signs, pedestrian signals, and other objects, and determine their location and status. Based on the results of this analysis, it generates guide information for the user. This guide information includes instructions such as "There is an obstacle ahead," "Turn left," and "The traffic light is red."

[1140] The server converts the generated guide information into speech using speech synthesis technology such as Google Text-to-Speech (TTS) or Amazon Polly. The converted speech guide information is then sent back to the user's device via the network. The server can also receive speech input from the user's device and modify the guide information based on that input.

[1141] User Notification

[1142] The user device plays back the audio guide information received from the server. This allows the user to understand their surroundings in real time and walk safely. For example, when the user approaches an intersection, the device's camera captures the image and the server recognizes that "there is an obstacle before the intersection." Based on this information, guide information instructing "please go 5 meters to the right" is generated, converted into voice and notified to the user. This allows the user to pass through the intersection safely.

[1143] Examples of prompt statements

[1144] For example, the following prompts might be used when building a generative AI model:

[1145] "Design a system that uses machine learning and deep learning techniques to recognize and analyze obstacles in captured images in real time and provide audio guidance to the user."

[1146] This invention provides comprehensive support for the visually impaired to walk safely and independently, greatly improving safety and convenience, and aiming to make mobility in daily life for the visually impaired more flexible and comfortable.

[1147] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1148] Step 1:

[1149] A user activates a device equipped with a camera and uses the device's camera to capture images of the surroundings while walking. The device captures the images as still images at regular intervals and generates image data.

[1150] Input: Video captured by the camera

[1151] Output: Generated image data

[1152] Step 2:

[1153] The terminal transmits the generated image data to the server via the network using an HTTP request.

[1154] Input: Generated image data

[1155] Output: Image data sent to the server

[1156] Step 3:

[1157] The server receives the image data sent from the terminal, converts the received data into an appropriate format, and prepares it for the next analysis step.

[1158] Input: Image data sent from the device

[1159] Output: Format-converted image data

[1160] Step 4:

[1161] The server inputs the received image data into a generative AI model. The generative AI model is built using libraries such as TensorFlow and PyTorch, and analyzes the image data to identify roads, obstacles, signs, pedestrian signals, etc. Specifically, it detects different objects in the image and understands their respective positions and states.

[1162] Input: Format-converted image data

[1163] Output: Analysis results (location and status of roads, obstacles, signs, pedestrian signals, etc.)

[1164] Step 5:

[1165] The server generates guide information for the user based on the analysis results of the generative AI model, including instructions such as "There is an obstacle ahead," "Turn left," and "The traffic light is red."

[1166] Input: Analysis results

[1167] Output: Generated guide information

[1168] Step 6:

[1169] The server converts the generated guide information into speech using speech synthesis technology, using services such as Google Text-to-Speech (TTS) and Amazon Polly.

[1170] Input: Generated guide information

[1171] Output: Audio data

[1172] Step 7:

[1173] The server then transmits the guide information converted into audio to the user terminal via the network using the HTTPS protocol.

[1174] Input: Audio data

[1175] Output: Audio data sent to the user's device

[1176] Step 8:

[1177] The user device plays back the received voice guidance information. As a result, the user can grasp the surrounding situation in real time and take appropriate action. For example, if the voice guidance instructs, "There is an obstacle ahead. Please move 5 meters to the right," the user will act accordingly.

[1178] Input: Audio data sent to the user's device

[1179] Output: Played audio description information

[1180] Step 9:

[1181] The user can use their voice input to request specific instructions or questions from the server, such as "What's the next intersection like?", and the device transmits this voice input to the server.

[1182] Input: User voice input

[1183] Output: Audio input sent to the server

[1184] Step 10:

[1185] The server receives voice input sent from the user terminal, analyzes the content, and updates or modifies the guide information. For example, it generates new guide information such as "There is a traffic light at the next intersection. Turn left."

[1186] Input: Voice input from the user

[1187] Output: Updated or corrected guide information

[1188] Step 11:

[1189] The updated guide information is converted back into audio and sent to the user terminal via the network. The user terminal then plays back the new guide information and notifies the user. This allows the user to respond flexibly according to the situation.

[1190] Input: Updated guide information

[1191] Output: New audio guide information sent to the user's device.

[1192] (Application example 1)

[1193] 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."

[1194] It is extremely difficult for visually impaired people to navigate safely on public transportation or in crowded places, and conventional technologies may not be able to provide sufficient support. Recognizing roads, obstacles, signs, pedestrian signals, and other obstacles is particularly difficult, requiring accurate information. Additionally, there is a lack of systems that provide users with real-time guidance information. The purpose of this invention is to solve these problems and provide an audio guidance system that enables visually impaired people to navigate safely.

[1195] 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.

[1196] In this invention, the server includes means for receiving image data transmitted from a user device having a camera, means for analyzing the received image data and using a generative AI model that recognizes roads, obstacles, signs, pedestrian signals, etc., means for generating guide information for the user based on the content recognized by the generative AI model, means for converting the generated guide information into audio and notifying the user, means for periodically transmitting image data from the user device to the server to support safe behavior on public transportation and in crowded places, and means for the generative AI model to provide analysis results and notify the user by audio. This enables visually impaired people to grasp their surroundings in real time and act safely.

[1197] A "user device" is a terminal that has a camera and can be carried around by a visually impaired person.

[1198] "Image data" is digital data of visual information captured by a camera on a user device and transmitted to a server.

[1199] A "generative AI model" is a model trained using machine learning and deep learning techniques that analyzes image data to recognize roads, obstacles, signs, pedestrian signals, etc.

[1200] "Guide information" is information that provides instructions and advice to the user, which is generated based on the analysis results, and is notified by voice.

[1201] The "means for converting into voice" is a function for converting the generated guide information into voice data using voice synthesis technology.

[1202] "Public transport and crowded places" refers to places and situations where many people use or gather, and where safe movement is particularly important for people with visual impairments.

[1203] The "periodic transmission means" is a function that allows the user device to transmit image data to the server at regular time intervals.

[1204] "Means of providing analysis results and notifying by voice" refers to a function that uses voice synthesis technology to notify the user in real time based on the analysis results of the generated AI model on the received image data.

[1205] The present invention relates to an audio guide system that supports visually impaired people to navigate safely on public transportation and in crowded places. Hereinafter, an embodiment will be described in detail.

[1206] System Overview

[1207] The system is composed of a user device, a server, and a generative AI model.

[1208] User Device Operation

[1209] The user device has a built-in camera and can be carried around by visually impaired people. The user activates the camera while walking and periodically captures images of the surroundings. The captured image data is sent to a server at regular intervals.

[1210] Server Operation

[1211] The server receives image data sent from the user device. The received image data is analyzed using a generative AI model. The generative AI model is trained using machine learning and deep learning techniques to accurately identify roads, obstacles, signs, pedestrian signals, and other objects. Based on the analysis results, the server generates guide information for the user. The generated guide information is converted into voice data using speech synthesis technology and transmitted to the user device.

[1212] User Notification

[1213] The user device plays back the audio data received from the server, providing real-time audio notification of the surrounding situation to the visually impaired, allowing the user to act safely.

[1214] Hardware and Software

[1215] The hardware required to implement this system is a user device with a camera (e.g., smart glasses) and a server. The software uses Python, OpenCV, the Requests library, etc. The generative AI model performs analysis based on tensor-formatted data, and the speech synthesis technology uses an API.

[1216] Specific examples

[1217] For example, if a user is walking down a busy street, they can send the following prompt to the generative AI model:

[1218] Example prompt sentence:

[1219] "Please generate guide information for walking safely in busy areas with many pedestrians. Please also include information about obstacles, dangerous areas, and traffic light status in the notification content."

[1220] Example result:

[1221] The server analyzes the prompt text and generates guide information such as, "There are many pedestrians ahead, so please move to the right. Be careful, there is a traffic light ahead, but it is red." and notifies the user device by voice.

[1222] As described above, the present invention enables visually impaired people to grasp the situation around them in real time and act safely.

[1223] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1224] Step 1:

[1225] A user activates the camera of a user device and periodically captures images of the surroundings. As input, the user device acquires real-time image data through the camera. This image data is saved as image data in JPEG format, etc. As output, the image data is stored in the user device.

[1226] Step 2:

[1227] The user device sends the accumulated image data to the server at regular intervals. As input, the latest image data stored on the user device is used to send an HTTP POST request to the server's API endpoint. As output, the image data is uploaded to the server.

[1228] Step 3:

[1229] The server receives image data sent from the user device. As input, it obtains the image data in the received HTTP POST request and stores it internally. As output, it obtains the image data newly stored in the server.

[1230] Step 4:

[1231] The server inputs the received image data into the generative AI model for analysis. As input, it converts the stored image data into tensor format and passes it to the generative AI model. The generative AI model analyzes the image data and identifies objects such as roads, obstacles, signs, and pedestrian signals. The results obtained from this analysis are the output.

[1232] Step 5:

[1233] The server generates guide information for the user based on the analysis results. It uses the analysis result data output by the generative AI model as input and generates guide information according to predefined rules and patterns. For example, text information such as "There is an obstacle ahead" or "Please move to the right and proceed" is generated. The generated guide information text data is obtained as output.

[1234] Step 6:

[1235] The server converts the generated guide information into voice data using voice synthesis technology. As input, the text data of the guide information is passed to the voice synthesis API, which generates voice data. As output, the generated voice data is obtained.

[1236] Step 7:

[1237] The server transmits the generated voice data to the user device. As input, the generated voice data is transmitted to the user device as an HTTP response. As output, new voice data to be received by the user device is obtained.

[1238] Step 8:

[1239] The user device plays the audio data received from the server and notifies the user. As input, the received audio data is passed to an internal audio player and played. As output, the user receives audio information and can understand the surrounding situation.

[1240] 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.

[1241] System Overview

[1242] This system provides audio guidance to help visually impaired people walk safely and independently. The server receives image data sent from the user's camera-equipped device, analyzes the images, and generates guide information based on the results, which is then notified to the user via audio. It also incorporates an emotion engine that recognizes the user's emotions, providing guidance tailored to each individual situation, thereby achieving more personalized assistance.

[1243] User terminal operation

[1244] A user carries a device with a camera and activates the device's camera while walking. The device periodically captures images of the surroundings and creates image data. This image data is sent to a server via a network. The user's voice input is also captured and sent to the server at the same time.

[1245] Server Operation

[1246] The server receives image data and voice input data sent from the user's device. The received image data is analyzed using a generative AI model. This AI model is trained using machine learning and deep learning techniques to identify roads, obstacles, signs, pedestrian signals, and other objects, and understand their location and status.

[1247] Based on the analysis results, the server generates guide information for the user, including instructions such as "There is an obstacle ahead," "Turn left," and "The traffic light is red."

[1248] Emotion Engine Operation

[1249] The server inputs the user's voice input into an emotion engine to recognize the user's emotions. The emotion engine analyzes the tone, rate, and context of the voice input to identify the user's emotional state (e.g., nervousness, anxiety, impatience).

[1250] Guide information adjustments

[1251] Based on the user's recognized emotions, the server adjusts the content of the guide information accordingly. For example, if the user is nervous, the server will provide more detailed and careful guide information. Conversely, if the user is relaxed, the server will provide simpler guide information.

[1252] The generated guide information is converted into voice using a voice synthesis technique, and the converted guide information is transmitted again to the user terminal via the network.

[1253] User Notification

[1254] The user device plays back the audio guide information received from the server, allowing the user to understand their surroundings in real time and walk safely.

[1255] Specific examples

[1256] For example, when a user approaches an intersection, the device's camera captures the image. This image data is sent to the server and analyzed. The server recognizes that there is an obstacle before the intersection and generates guide information based on this information, such as "Go 5 meters to the right."

[1257] At the same time, the emotion engine analyzes the user's voice input and recognizes that the user is anxious. Taking this into account, the guidance information is adjusted to be more detailed, such as "There is an obstacle ahead. Move 5 meters to the right. Please proceed slowly and without rushing."

[1258] This guide information is converted into audio and sent to the user's device. The user device then notifies the user by audio, saying, "There is an obstacle ahead. Please move 5 meters to the right. Please proceed slowly and without rushing." This allows the user to pass through the intersection safely and with peace of mind.

[1259] The processing flow will be explained below.

[1260] Step 1:

[1261] The user activates the camera

[1262] The user activates the device's camera and prepares to capture video of the surroundings.

[1263] Step 2:

[1264] The device generates image data and audio data.

[1265] The device converts the captured video into image data for each frame in real time and stores it temporarily in a buffer. It also captures the user's voice input and stores it as audio data.

[1266] Step 3:

[1267] The device sends image data and audio data to the server.

[1268] The terminal periodically transmits the stored image data and audio data to the server as a POST request.

[1269] Step 4:

[1270] The server receives the image data and audio data.

[1271] The server receives the image data and audio data sent from the terminal and expands each data in memory.

[1272] Step 5:

[1273] The server decodes the image data

[1274] The server decodes the received image data using libraries such as OpenCV and converts it into a format that can be analyzed as a color image.

[1275] Step 6:

[1276] The server inputs the image data into the generative AI model

[1277] The server preprocesses the decoded image data into the format required by the generative AI model and inputs it into the generative AI model as tensor data.

[1278] Step 7:

[1279] Server-generated captions

[1280] Using a generative AI model, the system generates captions from input image data, including information about roads, obstacles, signs, traffic lights, and other features in the image.

[1281] Step 8:

[1282] The server inputs the voice data into the emotion engine.

[1283] The server inputs the received voice data into an emotion engine that analyzes the user's voice tone, speed, and context.

[1284] Step 9:

[1285] The server analyzes the user's emotions

[1286] The emotion engine analyzes the audio data to identify the user's emotional state (e.g., tension, anxiety, impatience).

[1287] Step 10:

[1288] The server generates guide information

[1289] The server generates guidance information for the user based on the generated captions and the analyzed user's emotional information. For example, if the user is nervous, it provides more detailed and careful guidance information.

[1290] Step 11:

[1291] The server converts the guide information into voice

[1292] The server converts the guide information into voice data using a text-to-speech engine (TTS).

[1293] Step 12:

[1294] The server sends the voice data to the user terminal.

[1295] The server transmits the generated voice data to the user terminal via the network.

[1296] Step 13:

[1297] The device receives the audio data.

[1298] The user terminal receives the audio data sent from the server and prepares to play the data.

[1299] Step 14:

[1300] The device plays the audio data.

[1301] The user terminal reproduces the received audio data to the user through a speaker or earphone.

[1302] Step 15:

[1303] The user walks according to the audio guide

[1304] The user continues walking safely by following the audio guidance played from the device, which is adjusted appropriately according to the user's emotions.

[1305] Example 2

[1306] 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."

[1307] There is a need for technology to assist visually impaired people in walking safely and independently. Conventional voice guidance systems have difficulty grasping the surrounding situation in real time and providing appropriate instructions. Furthermore, because they do not take the user's emotional state into consideration, they are insufficient in providing support to improve the user's sense of security. Furthermore, there is a need for technology that can receive voice input, analyze emotions, and adjust guide information based on that.

[1308] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1309] In this invention, the server includes means for receiving image data transmitted from a user terminal having a camera, means for analyzing the received image data and using a generative AI model that recognizes roads, obstacles, signs, pedestrian signals, etc., means for generating guide information for the user based on the content recognized by the generative AI model, means for converting the generated guide information into voice and notifying the user, means for receiving voice input from the user and analyzing their emotional state, and means for adjusting the guide information based on the analyzed emotional state. This not only enables visually impaired people to grasp their surroundings in real time and walk safely, but also enables personalized support according to the user's emotional state.

[1310] A "user device with a camera" refers to a device that is equipped with a camera for acquiring visual information and that can be carried by the user. This includes smartphones and dedicated mobile devices.

[1311] The "means for receiving image data" refers to the communication technology and protocol for receiving image data sent from a user terminal having a camera and importing it into the server.

[1312] A "generative AI model" is a model trained using machine learning and deep learning techniques and used to analyze image data to recognize roads, obstacles, signs, pedestrian signals, etc.

[1313] "Means for generating guide information" refers to the process of generating information that instructs the user on the surrounding situation and direction of travel based on the analysis results of the generative AI model.

[1314] "Means for converting into voice and notifying the user" refers to technology for converting the generated guide information into voice data using voice synthesis technology and notifying the user of that voice in real time.

[1315] "Means for receiving voice input and analyzing emotional state" refers to processes and techniques for receiving a user's voice input and analyzing the voice data to identify the user's emotional state (e.g., tension, anxiety, impatience, etc.).

[1316] "Means for adjusting guide information" refers to the process of appropriately changing or adjusting the generated guide information based on the analyzed user's emotional state, for example, by adding detailed instructions or simplifying them.

[1317] This invention is a system that supports visually impaired people to walk safely and independently. The system uses a server to analyze data sent from a camera-equipped device held by the user, and provides audio guidance based on the results.

[1318] A user carries a device with a camera and activates the device's camera while walking. The device periodically captures images of the surroundings and creates image data. This image data is sent to a server via a network. The user's voice input is also captured and sent to the server at the same time. Specifically, a smartphone or a dedicated mobile device is used.

[1319] The server receives image data and audio data sent from the user's device. The received image data is analyzed using a generative AI model. This generative AI model utilizes a deep learning model trained using, for example, TensorFlow or PyTorch. This model identifies objects such as roads, obstacles, signs, and pedestrian signals, and determines their location and status.

[1320] Based on the analysis results, the server generates guide information for the user. This guide information includes instructions such as "There is an obstacle ahead," "Turn left," and "The traffic light is red." The server also inputs the user's voice input into an emotion engine to analyze the user's emotional state. This emotion engine uses, for example, Google Dialogflow or IBM Watson.

[1321] Based on the recognized user's emotions, the server adjusts the content of the guide information accordingly. For example, if the user is nervous, the server provides more detailed and careful guide information. Conversely, if the user is relaxed, the server provides concise guide information. The generated guide information is converted into audio using speech synthesis technology (e.g., Google Text-to-Speech or Amazon Polly). The converted audio guide information is then sent back to the user's device via the network.

[1322] The user device plays back the audio guide information received from the server, allowing the user to understand their surroundings in real time and walk safely.

[1323] For example, when a user approaches an intersection, the device's camera captures the image. This image data is sent to the server and analyzed. The server recognizes that there is an obstacle before the intersection and generates guide information based on this information, such as "Go 5 meters to the right."

[1324] At the same time, the emotion engine analyzes the user's voice input and recognizes that the user is in an anxious state. Taking this into account, the guide information is adjusted to be more detailed, such as "There is an obstacle ahead. Please move 5 meters to the right. Please proceed slowly and without rushing." This guide information is converted into voice and sent to the user's device. The user's device then notifies the user by voice, "There is an obstacle ahead. Please move 5 meters to the right. Please proceed slowly and without rushing." This allows the user to pass through the intersection safely and with peace of mind.

[1325] An example of a prompt sentence is "There is an obstacle ahead. Please guide me in which direction I should go." This is used as input to the generative AI model, forming the basis for generating optimal guidance information.

[1326] This system enables visually impaired people to accurately grasp their surroundings based on visual information and walk safely through audio guidance. It also provides individualized support based on the user's emotional state, helping them live independently and with peace of mind.

[1327] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1328] Step 1:

[1329] A user carries a device with a camera and activates the device's camera while walking. An application on the device periodically uses the camera to capture images of the surroundings and create image data. Specifically, the application captures images twice per second and saves the captured data as JPEG image files. The inputs are user operations and camera images, and the output is JPEG image data.

[1330] Step 2:

[1331] The user's device sends the captured image and audio data to the server. The device compresses the generated JPEG image file and the simultaneously collected audio input data, and transfers them to the server via the Internet. In particular, it performs a process to convert the data into an appropriate format for transmission. The input is JPEG image data and audio data, and the compressed data is sent to the server as output.

[1332] Step 3:

[1333] The server receives image data and audio data sent from the user's device. The server decompresses the received data and converts the image data into tensor format for input into the generative AI model. Specifically, the image data is converted into numerical data for each pixel and processed into tensor format. The input is compressed image data and audio data, and the output is tensor-formatted image data.

[1334] Step 4:

[1335] The server analyzes tensor-formatted image data using a generative AI model. This model identifies objects such as roads, obstacles, signs, and pedestrian signals, and outputs their location and status. For example, a deep learning framework is used to recognize and classify objects. The input is tensor-formatted image data, and the output is the object identification results.

[1336] Step 5:

[1337] The server generates guide information for the user based on the results of image analysis. For example, it creates guide information that includes specific instructions such as "There is an obstacle ahead," "Turn left," and "The traffic light is red." The input is the object identification result, and the output is guide information.

[1338] Step 6:

[1339] The server analyzes the received voice data and inputs it into an emotion engine that recognizes the user's emotional state. The emotion engine analyzes the tone, speed, and context of the voice to identify emotional states such as tension, anxiety, and impatience. Specifically, it extracts features from the voice data and inputs them into an emotion classification model. The input is the voice data, and the output is the user's emotional state.

[1340] Step 7:

[1341] Based on the analyzed user's emotional state, the server adjusts the guide information accordingly. For example, if the user is nervous, the server makes the guide information more detailed and careful. The inputs are the emotional state and the initial guide information, and the output is the adjusted guide information.

[1342] Step 8:

[1343] The server converts the adjusted guide information into speech using speech synthesis technology. For example, it uses a text-to-speech system to create a natural-sounding audio guide. The adjusted guide information is input, and audio data is generated as output.

[1344] Step 9:

[1345] The server transmits the generated voice data to the user terminal. The voice data is transferred again in real time via the network. The voice data is input, and transmission to the user terminal is output.

[1346] Step 10:

[1347] The user device plays the audio guide information received from the server. The user continues walking safely by following the played audio guide. The input is audio data, and the output is the playback of audio guide. This allows the user to understand the surrounding situation in real time and walk safely.

[1348] (Application example 2)

[1349] 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."

[1350] Visually impaired people face difficulties when searching for products in physical stores because they cannot see their surroundings. They also need the help of others to understand the location of shelves and products, making it difficult for them to enjoy shopping independently.

[1351] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for receiving image data transmitted from a user terminal having a camera; means for analyzing the received image data and using a generative AI model that recognizes roads, obstacles, signs, pedestrian signals, etc.; means for generating guide information for the user based on the content recognized by the generative AI model; means for converting the generated guide information into voice and notifying the user; means for capturing the situation in a physical store using a camera in smart glasses and identifying product shelves and products; and means for locating the user in the store and generating guide information that guides them to the products. This enables visually impaired people to find products in a physical store and shop independently.

[1352] A "camera" is a device for capturing images, and is installed in smart glasses or a user terminal.

[1353] A "user terminal" is a device that has a camera and can send image data to a server.

[1354] "Image data" refers to video information of the surroundings captured by a camera.

[1355] A "generative AI model" is an artificial intelligence model that analyzes received image data and identifies specific objects.

[1356] "Guide information" refers to guidance and instruction information for users that is generated based on the content recognized by the generative AI model.

[1357] "Convert to voice" is a process for conveying text-format guide information to the user as voice.

[1358] "Smart glasses" are head-mounted display devices that incorporate cameras and other sensors and are worn by users.

[1359] A "brick and mortar store" is a physical sales location where products are displayed and customers can visit and purchase in person.

[1360] A "product shelf" is a shelf used to display products in a physical store.

[1361] "Determining the location" means determining the user's current location and obtaining accurate location information.

[1362] "Guidance" refers to the provision of information to assist a user in reaching a destination.

[1363] As an embodiment of this invention, we will explain a support system for visually impaired people to search for and purchase products in a physical store. The system is realized by combining smart glasses worn by the user, a server, a generative AI model, and an audio guide.

[1364] System configuration

[1365] Smart glasses and user terminal

[1366] Smart glasses have a built-in camera and are worn by the user. The camera captures the surrounding situation in real time and generates image data. The user device (such as a smartphone or tablet) periodically transmits this image data to a server.

[1367] Server and Generative AI Model

[1368] The server receives image data sent from the user's device. The received image data is input into the generative AI model for analysis. The generative AI model is trained using machine learning and deep learning techniques to identify product shelves, directions, obstacles, and other items in a physical store.

[1369] Guide information generation

[1370] The server generates guide information for the user based on the content recognized by the generative AI model. This information includes the user's location, the location of the desired product, and the route to take within the store. The guide information is generated in text format.

[1371] Text-to-Speech and Notifications

[1372] The generated guide information is converted into audio on the server side using speech synthesis technology. Specifically, Google Text-to-Speech (gTTS) is used. The converted audio guide information is then sent back to the user's device via the network. The user's device (smart glasses) then plays back this audio guide and notifies the user.

[1373] Operation example

[1374] When a user searches for a product in a store, the smart glasses capture the surroundings with a camera and send the image data to the server. For example, if a user says, "I'm looking for the condiments section," the following prompt sentences are generated as guide information:

[1375] "There is a shelf about three meters ahead. The item you are looking for is on the second shelf on the left. Please walk slowly."

[1376] System Operation

[1377] The server receives image data sent from the smart glasses with a camera and first performs image analysis. It processes the image using libraries such as OpenCV, extracts specific objects, and converts them into tensor format. This data is input into a generative AI model to identify shelf and product locations, the user's current location, and other information. Based on the recognized content, guide information for the user is generated in text format and converted into audio using gTTS. The converted audio guide information is sent to the user's device, and an audio notification is given to the user via the smart glasses. By following these audio notifications, visually impaired people can shop independently.

[1378] This system enables visually impaired people to safely and efficiently search for and purchase products in physical stores.

[1379] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1380] Step 1:

[1381] A user wears smart glasses and walks around a physical store. The camera in the smart glasses captures the surroundings and generates image data. This image data is sent to the user device. The input is the captured image, and the output is the image data.

[1382] Step 2:

[1383] The user terminal periodically transmits the captured image data to the server, where the input is the image data received from the smart glasses and the output is the image data transmitted to the server.

[1384] Step 3:

[1385] The server analyzes the image data received from the user device, converts this image data into tensor format, and inputs it into the generative AI model. The input is the received image data, and the output is tensor format data.

[1386] Step 4:

[1387] The generative AI model identifies product shelves, product locations, obstacles, etc. in a physical store based on tensor-formatted data. The input is tensor-formatted data, and the output is information about the identified content.

[1388] Step 5:

[1389] The server generates guide information for the user based on the content identified by the generative AI model. For example, this includes information about the product's location and travel route. The input is the identified content, and the output is text data of the guide information.

[1390] Step 6:

[1391] The server converts the generated guide information into audio using speech synthesis technology. Google Text-to-Speech (gTTS) is used here. The input is the text data of the guide information, and the output is an audio file.

[1392] Step 7:

[1393] The server converts the guide information into audio and sends it to the user terminal. The input is an audio file, and the output is the audio file sent to the user terminal.

[1394] Step 8:

[1395] The user terminal (smart glasses) plays the audio guide information received from the server and notifies the user. The input is an audio file, and the output is the played audio guide. This allows the user to search for and move around products in the store by following the audio guide.

[1396] 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.

[1397] 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.

[1398] 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.

[1399] 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.

[1400] 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.

[1401] 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.

[1402] 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).

[1403] 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.

[1404] 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."

[1405] 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.

[1406] 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).

[1407] 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.

[1408] 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.

[1409] 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.

[1410] 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.

[1411] 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.

[1412] 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.

[1413] 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.

[1414] 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.

[1415] 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.

[1416] 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.

[1417] The following is further disclosed regarding the above embodiment.

[1418] (Claim 1)

[1419] means for receiving image data transmitted from a user terminal having a camera;

[1420] A means of analyzing received image data and using a generative AI model to recognize roads, obstacles, signs, pedestrian signals, etc.;

[1421] A means for generating guide information for a user based on the content recognized by the generation AI model;

[1422] means for converting the generated guide information into voice and notifying the user of the voice;

[1423] A system including:

[1424] (Claim 2)

[1425] 2. The system according to claim 1, wherein the user terminal having the camera further comprises means for periodically transmitting image data to a server.

[1426] (Claim 3)

[1427] 10. The system of claim 1, further comprising means for converting received image data into tensors and inputting them into the generative AI model.

[1428] "Example 1"

[1429] (Claim 1)

[1430] means for receiving image data transmitted from a user terminal having a camera;

[1431] A means of analyzing received image data and using a generative AI model to recognize roads, obstacles, signs, pedestrian signals, etc.;

[1432] A means for generating guide information for a user based on the content recognized by the generation AI model;

[1433] means for converting the generated guide information into voice and notifying the user of the voice;

[1434] means for receiving a voice input transmitted from a user terminal and modifying guide information based on the voice input;

[1435] A system including:

[1436] (Claim 2)

[1437] 2. The system according to claim 1, wherein the user terminal having the camera further comprises means for periodically transmitting image data to a server.

[1438] (Claim 3)

[1439] 10. The system of claim 1, further comprising means for converting the received image data into an array and inputting it into the generative AI model.

[1440] "Application Example 1"

[1441] (Claim 1)

[1442] means for receiving image data transmitted from a user device having a camera;

[1443] A means of analyzing received image data and using a generative AI model to recognize roads, obstacles, signs, pedestrian signals, etc.;

[1444] A means for generating guide information for a user based on the content recognized by the generation AI model;

[1445] means for converting the generated guide information into voice and notifying the user of the voice;

[1446] means for periodically transmitting image data from the user device to a server to assist safe behavior on public transportation and in crowded places;

[1447] A means for the generative AI model to provide an analysis result and notify the user by voice;

[1448] A system including:

[1449] (Claim 2)

[1450] 10. The system of claim 1, further comprising means for converting the image data into tensors and inputting them into the generative AI model.

[1451] (Claim 3)

[1452] 10. The system of claim 1, wherein the camera-equipped user device further comprises means for periodically transmitting image data to the server.

[1453] "Example 2: Combining Emotion Engines"

[1454] (Claim 1)

[1455] means for receiving image data transmitted from a user terminal having a camera;

[1456] A means of analyzing received image data and using a generative AI model to recognize roads, obstacles, signs, pedestrian signals, etc.;

[1457] A means for generating guide information for a user based on the content recognized by the generation AI model;

[1458] means for converting the generated guide information into voice and notifying the user of the voice;

[1459] means for receiving a user's voice input and analyzing the user's emotional state;

[1460] means for adjusting the guide information based on the analyzed emotional state;

[1461] A system including:

[1462] (Claim 2)

[1463] 2. The system according to claim 1, wherein the user terminal having the camera further comprises means for periodically transmitting image data and audio data to a server.

[1464] (Claim 3)

[1465] 10. The system of claim 1, further comprising means for converting received image data into tensors and inputting them into the generative AI model.

[1466] "Application example 2 when combining emotion engines"

[1467] (Claim 1)

[1468] means for receiving image data transmitted from a user terminal having a camera;

[1469] A means of analyzing received image data and using a generative AI model to recognize roads, obstacles, signs, pedestrian signals, etc.;

[1470] A means for generating guide information for a user based on the content recognized by the generation AI model;

[1471] means for converting the generated guide information into voice and notifying the user of the voice;

[1472] A means for capturing the situation in a physical store using a camera in smart glasses and identifying shelves and products;

[1473] a means for generating guide information that identifies the user's location in the store and guides the user to the product;

[1474] A system including:

[1475] (Claim 2)

[1476] 2. The system according to claim 1, wherein the user terminal having the camera further comprises means for periodically transmitting image data to the server.

[1477] (Claim 3)

[1478] 10. The system of claim 1, further comprising means for converting received image data into tensors and inputting them into the generative AI model. [Explanation of symbols]

[1479] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for receiving image data transmitted from a user terminal having a camera; A means of analyzing received image data and using a generative AI model to recognize roads, obstacles, signs, pedestrian signals, etc.; A means for generating guide information for a user based on the content recognized by the generation AI model; means for converting the generated guide information into voice and notifying the user of the voice; A system including:

2. 2. The system according to claim 1, wherein the user terminal having the camera further comprises means for periodically transmitting image data to a server.

3. 10. The system of claim 1, further comprising means for converting received image data into tensors and inputting them into the generative AI model.

Images