system

Abstract

We provide the system. [Solution] A receiving means for receiving user actions, An analysis means that analyzes the received motion data and identifies the classification and content of the motion, An estimation means for estimating the user's intent based on the analyzed behavior, A generation means for generating a natural language response based on an estimated intention, A means of providing the generated response to the user or other party, A means to communicate specific intentions to caregivers using user actions, and to provide additional functions to support communication. A system that includes this.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, and includes steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, 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

Summary of the Invention

Problems to be Solved by the Invention

[0004] People with visual, auditory, or speech impairments, or those who are anxious about conversations, find it difficult to effectively convey their intentions verbally in daily life, and as a result, often feel a sense of loneliness and alienation. In such situations, smooth communication with society is restricted, and the quality of life may decline. The present invention aims to remove such communication barriers and support these people in appropriately expressing their intentions and leading a rich social life.

Means for Solving the Problems

[0005] The present invention provides a communication support system comprising: receiving means for receiving gestures and inputs made by a user; analysis means for analyzing the data to identify gestures and input content; estimation means for estimating the user's intentions based on the analyzed data; and generation means for generating natural language responses based on the estimated intentions. Furthermore, it provides an output means for outputting the generated responses to the user or the person being communicated with. This enables users to effectively communicate their intentions in real time using simple gestures and inputs, thereby reducing barriers to communication.

[0006] "Receiving means" refers to a device or function that senses gestures or inputs made by the user and incorporates that data into the system.

[0007] "Analysis means" refers to a device or function that processes data acquired by the receiving means and performs a process to identify gestures or input content.

[0008] An "estimation means" is a device or function that performs a process of inferring and deriving the user's intent based on analyzed gestures and input content.

[0009] "Generating means" refers to a device or function that performs the process of creating an appropriate natural language response based on the estimated user intent.

[0010] "Output means" refers to a device or function for presenting the generated natural language response to the user or recipient in audio or visual form.

[0011] A "communication support system" refers to a comprehensive set of devices and functions that enable users to communicate effectively with others using gestures and input. [Brief explanation of the drawing]

[0012] [Figure 1]This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2, which incorporates an emotion engine. [Figure 14] This is a sequence diagram showing the processing flow of the data processing system in Application Example 2, which combines an emotion engine. [Modes for carrying out the invention]

[0013] Hereinafter, an example of an embodiment of the system relating to the technology of this disclosure will be described with reference to the attached drawings.

[0014] First, the terms used in the following description will be explained.

[0015] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0016] In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0017] In the following embodiments, the numbered storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, etc.

[0018] In the following embodiments, the numbered communication I / F (Interface) is an interface including a communication processor and 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), or Bluetooth (registered trademark), etc.

[0019] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0020] [First Embodiment]

[0021] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0022] As shown in Figure 1, the 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.

[0023] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0024] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0025] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and 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.

[0026] 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 perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0027] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0028] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0029] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

[0030] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0031] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0032] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0033] This invention provides a communication support system that allows users to communicate their intentions with confidence. This system is implemented by a program that accurately receives user gestures and inputs and generates responses in real time.

[0034] When a user performs a specific gesture, such as a greeting or a question, the device receives this information through its camera or touch interface. The acquired data is sent to a server for analysis. The server uses machine learning algorithms to analyze the data and identify the user's intent from their gestures.

[0035] A crucial part of this system is that the server generates the user's intent in natural language. Based on the estimated intent, the generative AI creates a response that fits the context of a natural conversation. In this process, the generated text is adjusted to be faithful to the user's intended meaning.

[0036] The generated response is sent to the device as audio or text and presented to the user and the other party. For example, if the user makes a waving gesture, the system will output "Hello" as audio or text, helping the recipient communicate smoothly.

[0037] Thus, the system based on the present invention is an embodiment that can be used easily and intuitively by people with disabilities or those who have anxiety about conversation, and promotes social connection.

[0038] The following describes the processing flow.

[0039] Step 1:

[0040] The user performs a gesture or input. This action is initiated using the camera or touch interface.

[0041] Step 2:

[0042] The device receives user gestures and input in real time. The received data is temporarily recorded and converted into a format that can be sent to the server.

[0043] Step 3:

[0044] The device converts the data to a specific format and sends it to the server via the internet.

[0045] Step 4:

[0046] The server analyzes the received data to identify the type of gesture and the input content. This analysis utilizes image recognition algorithms and machine learning models.

[0047] Step 5:

[0048] The server estimates the user's intent based on the analysis results. It evaluates the meaning of gestures and inputs in context and determines the appropriate response.

[0049] Step 6:

[0050] The server generates natural language responses based on estimated intent. Generative AI is used to create contextually appropriate and natural communication.

[0051] Step 7:

[0052] The server sends the generated response data to the terminal. The data is converted to the appropriate format and provided as audio output or text display.

[0053] Step 8:

[0054] The terminal presents the received response data to the user or the other party. The generated content is then transmitted through the speaker or display to constitute actual communication.

[0055] (Example 1)

[0056] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0057] Conventional communication support technologies have struggled to accurately recognize users' intentions and generate appropriate responses based on them. Therefore, a key challenge has been mitigating communication barriers, particularly faced by people with disabilities and those who experience anxiety in conversation.

[0058] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0059] In this invention, the server includes an imaging device means for receiving user actions, a computing device means for analyzing the received data and identifying the actions, and an evaluation device means for estimating the user's intent based on the analyzed actions. This makes it possible to accurately recognize the user's intent and generate and present a rapid and appropriate natural language response.

[0060] An "imaging device" is a device equipped with optical sensors or contact surfaces for acquiring user movements and gestures.

[0061] A "processing unit" is a device that has information processing capabilities to analyze received data and identify its contents.

[0062] An "evaluation device" is a device that performs processing to estimate the user's intent based on the analyzed data.

[0063] A "response generation device" is a device that has the function of generating a response in natural language based on an estimated intention.

[0064] A "presentation device" is a device that provides a generated response to the user or other party visually or audibly.

[0065] A "learning model" refers to an algorithm that is trained using a large amount of data and generates an output that is appropriate to the given input information.

[0066] This invention provides a communication support system that recognizes user gestures and generates and provides natural language responses based on those gestures. The system mainly consists of a terminal operated by the user and a server that supports it.

[0067] Terminal role

[0068] The device is equipped with a camera as an imaging device and a touch interface as a contact surface to capture user movements and gestures. For example, if a user performs a "wave" motion, that motion is received by the device.

[0069] Server Role

[0070] Data received by the terminal is sent to the server. Here, the server functions as a computing device, analyzing the received data to determine its operation. Computer vision technology and machine learning algorithms are used for this purpose. During the analysis process, the server also functions as an evaluation device, accurately estimating the user's intent.

[0071] Next, based on the estimated intent, the server uses a generative AI model to generate a natural language response as a response generator. For example, if the wavering motion is recognized as the greeting "hello," the server will create a response saying "hello."

[0072] In addition, the server sends the generated response to the presentation device, which then provides it to the user or recipient as voice or text. Voice responses are produced using speech synthesis technology, while text responses are displayed on the screen.

[0073] Specific example

[0074] The user makes a "thumbs up" gesture. The device receives this action and sends data to the server. The server recognizes the action as an "OK" intent and uses a generative AI model to generate a "thumbs up" response. Finally, the device either plays this response aloud or displays it as text.

[0075] Example of a prompt

[0076] "The user has detected a waving gesture. Please generate an appropriate greeting based on this gesture."

[0077] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0078] Step 1:

[0079] The user performs a specific gesture, and the device receives this action. The device records this action using imaging devices such as a camera or touch interface. User action information is acquired as input, and image data of the action or contact location data is generated as output.

[0080] Step 2:

[0081] The terminal transmits acquired operational data to a server via the internet. The input is operational data stored within the terminal, and the output is the transmission of that data to the server. Specifically, data transmission is performed using a wireless communication module.

[0082] Step 3:

[0083] The server analyzes the received motion data. The input is motion data from the terminal, and the output is the result of gesture identification. The server uses computer vision technology and machine learning algorithms to analyze the data and identify gestures. Specifically, feature extraction is performed using an image analysis program.

[0084] Step 4:

[0085] The server estimates the user's intent based on the analyzed actions. The input is the result of identifying the gesture, and the output is the estimated intent. The server utilizes a learning model to evaluate intent based on past data. Specifically, an intent estimation algorithm is applied.

[0086] Step 5:

[0087] The server generates natural language responses using a generative AI model based on estimated intent. The input is the estimated intent, and the output is the generated natural language response. Specifically, the generative AI model creates appropriate prompts.

[0088] Step 6:

[0089] The server sends the generated response to the terminal. The input is the generated natural language response, and the output is the transmission of the response to the terminal. Specifically, data transfer takes place through a communication protocol.

[0090] Step 7:

[0091] The terminal presents the received response to the user or the other party in audio or text format. Input is a natural language response from the server, and output is audio playback or display. Specifically, this involves audio output via the speaker or text display on the screen.

[0092] (Application Example 1)

[0093] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0094] Currently, many elderly people and individuals requiring care face difficulties in communicating smoothly with others in their daily lives. Communication is a particularly significant challenge for those with physical limitations or anxieties about communication. Therefore, there is a need to improve their quality of life and provide them with means to express themselves with confidence.

[0095] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0096] In this invention, the server includes receiving means for receiving user actions, analysis means for analyzing the received action data and identifying the classification and content of the actions, and estimation means for estimating the user's intentions based on the analyzed actions. This enables elderly people and caregivers to communicate quickly and accurately through actions.

[0097] "User actions" refer to physical movements and gestures performed by the user, and these actions are what the system recognizes.

[0098] A "receiving means" is a device or interface for capturing user actions, and has the function of acquiring data using a video acquisition device or a contact interface.

[0099] "Analysis means" refers to means for analyzing motion data captured by the receiving means and classifying and identifying the intent behind the motion.

[0100] "Estimation means" are methods used to determine and identify the intentions indicated by the analyzed user actions.

[0101] A "generative means" is a means for forming and creating a response in natural language based on the intention identified by the estimation means.

[0102] A "presentation means" is a means that transmits the response formed by the generation means to the user or recipient in a physical or digital form.

[0103] A "video acquisition device" is a device that records a user's movements as digital data, and cameras are an example of such devices.

[0104] A "contact interface" is an interface designed to capture user touch, and touchscreens are an example of this.

[0105] A "learning model" refers to an algorithm or program that has been trained to predict and generate user intent and appropriate responses based on data.

[0106] To realize this invention, the following system configuration and program are required. The system mainly consists of a server, a terminal, and a user interface, and each component works in cooperation with each other.

[0107] The server utilizes computing resources in the cloud to process data for analyzing user actions. Specifically, the server first receives video and contact data received via video acquisition devices and contact interfaces. Next, it analyzes the received data using a learning model (e.g., TENSORFLOW®) to classify and identify user actions. Based on the analysis results, it estimates the user's intent and generates a natural language response using a generation AI model (e.g., GPT). The generated response is then adjusted to provide the user with the necessary information clearly and quickly.

[0108] The terminal functions as an interface between the user and the system, presenting natural language responses sent from the server to the user in voice or text. This allows the user to engage in meaningful communication through their own actions.

[0109] Users primarily use smart glasses or smartphones to perform actions. For example, if a user waves their hand, this is sent to the cloud, interpreted as an intention to "want water," and a generated voice message saying "I want water" is output from the device.

[0110] As a concrete example, the following prompt statement is possible:

[0111] Prompt: "Please describe the process of interpreting intent and generating a response when a user makes a hand-waving gesture."

[0112] Such systems can remove communication barriers and provide a richer communication experience for the elderly and those who have difficulty conversing.

[0113] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0114] Step 1:

[0115] The user performs actions using smart glasses or a smartphone. This allows video acquisition devices and contact interfaces to capture the user's motion data. The input is the user's physical movements, and the output is motion data based on these movements. A concrete example of such an action is the user waving their hand.

[0116] Step 2:

[0117] The server receives behavioral data sent from the terminal. Based on the received data, it performs analysis using a machine learning model (e.g., TensorFlow). In this step, the input is behavioral data, and the output is the classification or identification of the behavior's content. Data processing includes executing a machine learning algorithm using the features of the behavior.

[0118] Step 3:

[0119] The server estimates the user's intent from the analyzed results. This estimation process uses the data obtained through analysis as input to perform data calculations that predict what intent it represents. The output is data representing the user's intent. Specifically, this involves determining that "this action indicates the intention 'I want water'."

[0120] Step 4:

[0121] The server generates natural language responses using a generative AI model (e.g., GPT) based on the estimated intent. Here, user intent data is input, and a response in natural language format is output. The generated response is then adapted to human language appropriate to the specific context. A concrete example of this process is when the response to the intent "I want water" is "I want water."

[0122] Step 5:

[0123] The device presents the user with the response received from the server. The response is displayed as audio or text. The input is natural language response data, and the output is the actual notification or voice message provided to the user. A specific example of this operation is using the speech synthesis function of smart glasses to say "I want water."

[0124] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0125] This invention provides a system that effectively supports communication by receiving and analyzing user gestures and inputs. Furthermore, by combining it with an emotion engine, it enables the generation of responses that take the user's emotions into consideration.

[0126] When a user performs a specific gesture or inputs via the touch interface during use, the device receives this information. The device then sends this data to a server where it is analyzed. During the analysis process, the gesture or input content is first identified, and then the user's emotional state is recognized using an emotion engine.

[0127] The server estimates the user's intent while considering the emotion recognition results. It then uses generative AI to create a natural language response, employing a tone and expression that matches each user's emotions. For example, if the server recognizes that the user is confused, it adjusts the response to provide a gentler and more specific explanation.

[0128] The device receives the final generated response and presents it to the user or the other party in voice or text format. By utilizing speech synthesis technology and display functions, communication support optimized for the user's current emotional state is achieved.

[0129] As a concrete example, imagine a user making a questioning gesture while showing a "troubled" expression in front of the camera. In this case, the system recognizes the user's emotion as "confusion" from the expression and generates a response that includes kindness and encouragement, such as "Are you okay?". This response is then fed back to the user via voice, resulting in a more human-like conversation.

[0130] Thus, by integrating an emotion engine, the present invention provides high-quality communication support that takes into account not only normal input analysis but also the user's emotions. This enables a more reassuring conversational experience for the user.

[0131] The following describes the processing flow.

[0132] Step 1:

[0133] The user makes gestures or touch inputs. For example, they might express their intentions by waving their hand or making a confused facial expression.

[0134] Step 2:

[0135] The device receives user gestures and facial expressions in real time via the camera and touch interface. The received data is temporarily stored as image and signal data.

[0136] Step 3:

[0137] The terminal sends the received data to the server for analysis. The data is transferred quickly using a secure protocol.

[0138] Step 4:

[0139] The server begins analyzing the received data. First, it identifies the type of gesture or input, and then uses the emotion engine to recognize the user's emotional state from their facial expressions.

[0140] Step 5:

[0141] The server estimates the user's intent based on the analysis results. The recognized emotional state is used to understand the intent and adjust the response accordingly.

[0142] Step 6:

[0143] The server uses generative AI to construct natural language responses based on estimated intentions and emotional states. The responses are appropriately adjusted in tone and content according to the recognized emotions.

[0144] Step 7:

[0145] The server sends the generated response to the terminal. The response data is prepared as audio or text.

[0146] Step 8:

[0147] The device presents the received response to the user or the other party through a speaker or display. In the case of voice output, natural and emotionally sensitive expressions are reproduced using synthesized speech technology.

[0148] (Example 2)

[0149] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0150] Conventional communication support systems simply receive and analyze user gestures and input, making it impossible to generate responses that take user emotions into account. This makes natural and user-friendly communication difficult, and often results in inadequate responses, especially in interactions involving emotional nuances.

[0151] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0152] In this invention, the server includes a receiving means, an analysis means, and an emotion recognition means. This makes it possible to generate a response that takes into account the user's emotional state in addition to their gestures and input content.

[0153] "Receiving means" refers to the function of receiving gestures and touch inputs from the user through sensors and interfaces.

[0154] "Analysis means" refers to a function that analyzes received data to identify gestures and input content.

[0155] "Emotion recognition means" refers to a function that recognizes the user's emotional state based on analyzed data.

[0156] "Estimation method" refers to a function that estimates the user's intentions from the recognized emotional state and the analyzed input content.

[0157] "Generation method" refers to a function that generates responses in natural language based on the estimated user's intentions and emotional state.

[0158] "Output means" refers to a function that provides the generated response to the user or the other party in text or voice.

[0159] The communication support system of the present invention aims to receive and analyze user gestures and inputs to generate appropriate responses. This system mainly consists of two components: a terminal and a server.

[0160] When a user performs a specific gesture or inputs via a touch interface, that input is received by the terminal. Hardware such as image sensors and touch displays are utilized here. The terminal processes this input data in real time and transfers it to a server. The data is transmitted via a secure protocol over a communication network.

[0161] To analyze the received data, the server first applies pattern recognition algorithms to identify gestures and input content. Open-source libraries such as OpenCV can be used for this purpose. Based on the analyzed data, the server utilizes an emotion recognition engine to recognize the user's emotional state from text and facial expression data. Deep learning models are suitable for emotion recognition. For example, emotion classification can be performed using a model based on TensorFlow.

[0162] Next, the server utilizes a generative AI model to generate natural language responses that are optimal for the user's intent and emotional state. By using prompts, flexible and user-friendly responses are achieved. For example, by using a prompt such as "This user is confused. How do you support them?", the generative AI will generate a human-like response such as "Don't worry, I'll explain in detail."

[0163] Finally, the device receives this generated response and presents it to the user in either audio or text format. Text-to-speech technology is used for speech synthesis, and it is desirable to utilize techniques such as WaveNet to enhance the naturalness of the speech. When displaying text via a screen, adopting a clear and intuitive UI design can further improve the user experience.

[0164] By implementing this system, users can receive flexible and considerate responses that are tailored to their emotions, resulting in a more natural communication experience.

[0165] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0166] Step 1:

[0167] The user performs gestures and touch inputs. For example, they might tap the smartphone screen or wave their hand in front of the camera. This is the initial input for the system. The device receives this input using its built-in image sensor and touch interface. The data received at this time includes the video of the gesture and the touch coordinates.

[0168] Step 2:

[0169] The terminal sends the user's input data to the server. During this process, pre-processing is performed on the terminal, and the data is compressed or encrypted before transmission. The server decompresses or decrypts the received data and prepares it for analysis. The input consists of gesture images and coordinate data, while the output is data formatted for analysis.

[0170] Step 3:

[0171] The server performs pattern recognition on the received data to identify gestures and input content. Specifically, it uses the OpenCV library to perform image processing and classify gestures. In this process, the input is video data, and the output is the label of the identified gesture and the type of touch.

[0172] Step 4:

[0173] The server performs emotion recognition on identified gestures. Here, a deep learning model is used to estimate emotions from the user's facial expressions and voice. A pre-trained emotion classification model using TensorFlow is used. The input is analyzed data, and the output is an emotion label (e.g., joy, sadness, confusion).

[0174] Step 5:

[0175] The server uses a generative AI model to generate prompts based on the user's emotional state and gestures. For example, it might generate a prompt such as, "This user is confused. How do you support them?" Furthermore, this prompt is used as input to generate a natural language response. The output is a customized text response tailored to the user's state.

[0176] Step 6:

[0177] The terminal receives a response from the server. This response includes contextual tone and expression. The received data is fed back to the user through a speech synthesizer or display. Specifically, text-to-speech technology is used to provide voice feedback such as, "Don't worry." The output is voice or text information in a format that is easy for the user to understand.

[0178] (Application Example 2)

[0179] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0180] Many modern communication devices and systems lack the ability to accurately understand a user's intentions and emotional state and generate appropriate responses accordingly. As a result, users often don't receive the responses they expect and become dissatisfied. Furthermore, current technology struggles to engage in emotionally sensitive dialogue like a human being. There is a need to solve this problem and realize more human-like communication that takes emotions into account.

[0181] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0182] In this invention, the server includes a receiving device that receives user gestures or inputs, an analysis device that analyzes the received data to identify gestures and input content, and an emotion recognition engine that recognizes the user's emotional state based on the analyzed gestures and input content. This enables intent estimation that takes the user's emotional state into account and the generation of appropriate responses in natural language.

[0183] A "receiving device" is a device that detects user gestures and inputs and acquires that data.

[0184] An "analysis device" is a device that analyzes data obtained from a receiving device to identify gestures and distinguish input content.

[0185] An "emotion recognition engine" is an engine that determines the user's emotional state based on analyzed gestures and input content.

[0186] An "estimation device" is a device that infers what a user intends by considering the user's emotional state and analysis results.

[0187] A "generator" is a device that creates appropriate responses in natural language based on the estimated user's intentions and emotional state.

[0188] An "output device" is a device that provides the user with a generated natural language response in either voice or text format.

[0189] The system for realizing this invention comprises multiple devices for processing user gestures and input. The server uses a receiving device to receive gestures and voice input from the user. The received data is analyzed using an analysis device to identify gestures and input content. Furthermore, based on the analyzed data, an emotion recognition engine determines the user's emotional state. Information on the emotional state is used by an estimation device to estimate the user's intentions. Subsequently, a generation device generates a natural language response based on the estimated intentions and emotional state. This response is provided to the user in voice or text through an output device.

[0190] The hardware includes image processing sensors and microphones to capture user gestures and facial expressions. A server with high processing power is also required to analyze this data. The software uses image recognition algorithms for analysis and emotion recognition engines (e.g., Microsoft® Azure® Face API) for emotion recognition. Generative AI (e.g., OpenAI® GPT) is used for generating natural language responses.

[0191] For example, if a user with a tired expression asks the robot, "Tell me today's news," the server will select news in a calm tone that reflects the user's fatigue, and use a generative AI to form a response such as, "We have some calm news today. For example..." and transmit it to the robot. This allows the user to experience a dialogue that reflects their own emotional state.

[0192] An example of a prompt message would be, "When the user looks tired, adjust the content and tone of the news message to respond."

[0193] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0194] Step 1:

[0195] The server acquires user gestures and voice input via a receiving device. The input data collected includes video data from a camera and audio data from a microphone. This data is stored on the server for subsequent analysis.

[0196] Step 2:

[0197] The server analyzes video and audio data using an analysis device. Image recognition algorithms are used to identify gestures and transcribe audio input. The output obtained here contains information about the user's specific actions and statements.

[0198] Step 3:

[0199] The server uses an emotion recognition engine to determine the user's emotional state from the analyzed gesture and voice input data. This process outputs an emotional state (e.g., "fatigue" or "confusion") based on changes in facial expression and tone of voice.

[0200] Step 4:

[0201] The server uses an estimation device to estimate the user's intentions based on the obtained emotional state and the content of the user's statements. At this stage, the server identifies the user's specific purpose or request from the combination of emotion and content, and outputs that information.

[0202] Step 5:

[0203] The server generates natural language responses based on the intent and emotional state estimated by the generator. It utilizes a generative AI model to create responses with appropriate context and tone. The final output is optimized messaging for the user.

[0204] Step 6:

[0205] The terminal provides the user with a response generated using an output device. The output is either in audio format via speech synthesis or in text format via a display. This allows the user to receive emotionally resonant responses.

[0206] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0207] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0208] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0209] [Second Embodiment]

[0210] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0211] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0212] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0213] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0214] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0215] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0216] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0217] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0218] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0219] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0220] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0221] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. 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".

[0222] This invention provides a communication support system that allows users to communicate their intentions with confidence. This system is implemented by a program that accurately receives user gestures and inputs and generates responses in real time.

[0223] When a user performs a specific gesture, such as a greeting or a question, the device receives this information through its camera or touch interface. The acquired data is sent to a server for analysis. The server uses machine learning algorithms to analyze the data and identify the user's intent from their gestures.

[0224] A crucial part of this system is that the server generates the user's intent in natural language. Based on the estimated intent, the generative AI creates a response that fits the context of a natural conversation. In this process, the generated text is adjusted to be faithful to the user's intended meaning.

[0225] The generated response is sent to the device as audio or text and presented to the user and the other party. For example, if the user makes a waving gesture, the system will output "Hello" as audio or text, helping the recipient communicate smoothly.

[0226] Thus, the system based on the present invention is an embodiment that can be used easily and intuitively by people with disabilities or those who have anxiety about conversation, and promotes social connection.

[0227] The following describes the processing flow.

[0228] Step 1:

[0229] The user performs a gesture or input. This action is initiated using the camera or touch interface.

[0230] Step 2:

[0231] The device receives user gestures and input in real time. The received data is temporarily recorded and converted into a format that can be sent to the server.

[0232] Step 3:

[0233] The device converts the data to a specific format and sends it to the server via the internet.

[0234] Step 4:

[0235] The server analyzes the received data to identify the type of gesture and the input content. This analysis utilizes image recognition algorithms and machine learning models.

[0236] Step 5:

[0237] The server estimates the user's intent based on the analysis results. It evaluates the meaning of gestures and inputs in context and determines the appropriate response.

[0238] Step 6:

[0239] The server generates natural language responses based on estimated intent. Generative AI is used to create contextually appropriate and natural communication.

[0240] Step 7:

[0241] The server sends the generated response data to the terminal. The data is converted to the appropriate format and provided as audio output or text display.

[0242] Step 8:

[0243] The terminal presents the received response data to the user or the other party. The generated content is then transmitted through the speaker or display to constitute actual communication.

[0244] (Example 1)

[0245] Next, we will describe Example 1. 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."

[0246] Conventional communication support technologies have struggled to accurately recognize users' intentions and generate appropriate responses based on them. Therefore, a key challenge has been mitigating communication barriers, particularly faced by people with disabilities and those who experience anxiety in conversation.

[0247] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0248] In this invention, the server includes an imaging device means for receiving user actions, a computing device means for analyzing the received data and identifying the actions, and an evaluation device means for estimating the user's intent based on the analyzed actions. This makes it possible to accurately recognize the user's intent and generate and present a rapid and appropriate natural language response.

[0249] An "imaging device" is a device equipped with optical sensors or contact surfaces for acquiring user movements and gestures.

[0250] A "processing unit" is a device that has information processing capabilities to analyze received data and identify its contents.

[0251] An "evaluation device" is a device that performs processing to estimate the user's intent based on the analyzed data.

[0252] A "response generation device" is a device that has the function of generating a response in natural language based on an estimated intention.

[0253] A "presentation device" is a device that provides a generated response to the user or other party visually or audibly.

[0254] A "learning model" refers to an algorithm that is trained using a large amount of data and generates an output that is appropriate to the given input information.

[0255] This invention provides a communication support system that recognizes user gestures and generates and provides natural language responses based on those gestures. The system mainly consists of a terminal operated by the user and a server that supports it.

[0256] Terminal role

[0257] The device is equipped with a camera as an imaging device and a touch interface as a contact surface to capture user movements and gestures. For example, if a user performs a "wave" motion, that motion is received by the device.

[0258] Server Role

[0259] Data received by the terminal is sent to the server. Here, the server functions as a computing device, analyzing the received data to determine its operation. Computer vision technology and machine learning algorithms are used for this purpose. During the analysis process, the server also functions as an evaluation device, accurately estimating the user's intent.

[0260] Next, based on the estimated intent, the server uses a generative AI model to generate a natural language response as a response generator. For example, if the wavering motion is recognized as the greeting "hello," the server will create a response saying "hello."

[0261] In addition, the server sends the generated response to the presentation device, which then provides it to the user or recipient as voice or text. Voice responses are produced using speech synthesis technology, while text responses are displayed on the screen.

[0262] Specific example

[0263] The user makes a "thumbs up" gesture. The device receives this action and sends data to the server. The server recognizes the action as an "OK" intent and uses a generative AI model to generate a "thumbs up" response. Finally, the device either plays this response aloud or displays it as text.

[0264] Example of a prompt

[0265] "The user has detected a waving gesture. Please generate an appropriate greeting based on this gesture."

[0266] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0267] Step 1:

[0268] The user performs a specific gesture, and the device receives this action. The device records this action using imaging devices such as a camera or touch interface. User action information is acquired as input, and image data of the action or contact location data is generated as output.

[0269] Step 2:

[0270] The terminal transmits acquired operational data to a server via the internet. The input is operational data stored within the terminal, and the output is the transmission of that data to the server. Specifically, data transmission is performed using a wireless communication module.

[0271] Step 3:

[0272] The server analyzes the received motion data. The input is motion data from the terminal, and the output is the result of gesture identification. The server uses computer vision technology and machine learning algorithms to analyze the data and identify gestures. Specifically, feature extraction is performed using an image analysis program.

[0273] Step 4:

[0274] The server estimates the user's intent based on the analyzed actions. The input is the result of identifying the gesture, and the output is the estimated intent. The server utilizes a learning model to evaluate intent based on past data. Specifically, an intent estimation algorithm is applied.

[0275] Step 5:

[0276] The server generates a natural language response using a generative AI model based on the estimated intention. The input is the result of the intention estimation, and the output is the generated natural language response. As a specific operation, an appropriate prompt is created by the generative AI model.

[0277] Step 6:

[0278] The server transmits the generated response to the terminal. The input is the generated natural language response, and the output is the transmission of the response to the terminal. As a specific operation, data transfer is performed through a communication protocol.

[0279] Step 7:

[0280] The terminal presents the received response to the user or the other party in voice or text. The input is the natural language response from the server, and the output is voice playback or display on the display. As a specific operation, voice output through a speaker or text display on the screen is performed.

[0281] (Application Example 1)

[0282] Next, Application Example 1 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart glasses 214 are referred to as the "terminal".

[0283] Currently, many elderly people and individuals who need care are in a situation where it is difficult to communicate smoothly with others in their daily lives. In particular, for people with physical limitations or anxiety about communication, communication is a major issue. Therefore, it is required to improve their quality of life and provide a means for them to express themselves with confidence.

[0284] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0285] In this invention, the server includes a receiving means for receiving the user's actions, an analyzing means for analyzing the received action data to identify the classification and content of the actions, and an estimating means for estimating the user's intention based on the analyzed actions. As a result, it becomes possible for elderly people and caregivers to communicate quickly and accurately through actions.

[0286] "The user's actions" refers to the physical movements and gestures performed by the user, which are the targets recognized by the system.

[0287] "The receiving means" is a device or interface for capturing the user's actions, and has the function of capturing data using a video acquisition device or a contact interface.

[0288] "The analyzing means" is a means for analyzing the action data captured by the receiving means and classifying / identifying what kind of intention the action has.

[0289] "The estimating means" is a means used to judge and identify the intention indicated by the analyzed user actions.

[0290] "The generating means" is a means for forming and creating a response in natural language based on the intention specified by the estimating means.

[0291] "The presenting means" is a means having the role of transmitting the response formed by the generating means to the user or the other party in a physical or digital form.

[0292] "The video acquisition device" is a device for recording the user's movements as digital data, such as a camera.

[0293] "The contact interface" is an interface for capturing the user's touching actions, such as a touch screen.

[0294] A "learning model" refers to an algorithm or program that has been trained to predict and generate user intent and appropriate responses based on data.

[0295] To realize this invention, the following system configuration and program are required. The system mainly consists of a server, a terminal, and a user interface, and each component works in cooperation with each other.

[0296] The server utilizes computing resources in the cloud to process data for analyzing user actions. Specifically, the server first receives video and contact data received via video acquisition devices and contact interfaces. Next, it analyzes the received data using a learning model (e.g., TensorFlow) to classify and identify user actions. Based on the analysis results, it estimates the user's intent and generates a natural language response using a generation AI model (e.g., GPT). The generated response is then adjusted to provide the user with the necessary information clearly and quickly.

[0297] The terminal functions as an interface between the user and the system, presenting natural language responses sent from the server to the user in voice or text. This allows the user to engage in meaningful communication through their own actions.

[0298] Users primarily use smart glasses or smartphones to perform actions. For example, if a user waves their hand, this is sent to the cloud, interpreted as an intention to "want water," and a generated voice message saying "I want water" is output from the device.

[0299] As a concrete example, the following prompt statement is possible:

[0300] Prompt: "Please describe the process of interpreting intent and generating a response when a user makes a hand-waving gesture."

[0301] Such a system will serve to remove the communication barriers for the elderly and those who are anxious about conversations, and provide a richer communication experience.

[0302] The flow of the specific process in Application Example 1 will be described using FIG. 12.

[0303] Step 1:

[0304] The user performs operations using smart glasses or a smartphone. As a result, the user's operation data is captured by the video acquisition device or the contact interface. The input is the user's physical operation, and the output is the operation data based on this. As a specific operation, an example is the user performing a waving motion with the hand.

[0305] Step 2:

[0306] The server receives the operation data transmitted from the terminal. Based on the received data, analysis is performed using a machine learning model (e.g., TensorFlow). In this step, the operation data is the input, and the classification of the operation and the identification result of the content are the outputs. Data processing includes operations of executing a machine learning algorithm using the features of the operation.

[0307] Step 3:

[0308] The server estimates the user's intention from the analyzed results. In the estimation, data calculation is performed to estimate what intention the data obtained by the analysis indicates, with the input being the data obtained by the analysis. The output is the data representing the user's intention. As a specific operation, judgments such as "this operation indicates the intention of 'wanting water'" are made.

[0309] Step 4:

[0310] The server generates natural language responses using a generative AI model (e.g., GPT) based on the estimated intent. Here, user intent data is input, and a response in natural language format is output. The generated response is then adapted to human language appropriate to the specific context. A concrete example of this process is when the response to the intent "I want water" is "I want water."

[0311] Step 5:

[0312] The device presents the user with the response received from the server. The response is displayed as audio or text. The input is natural language response data, and the output is the actual notification or voice message provided to the user. A specific example of this operation is using the speech synthesis function of smart glasses to say "I want water."

[0313] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0314] This invention provides a system that effectively supports communication by receiving and analyzing user gestures and inputs. Furthermore, by combining it with an emotion engine, it enables the generation of responses that take the user's emotions into consideration.

[0315] When a user performs a specific gesture or inputs via the touch interface during use, the device receives this information. The device then sends this data to a server where it is analyzed. During the analysis process, the gesture or input content is first identified, and then the user's emotional state is recognized using an emotion engine.

[0316] The server estimates the user's intent while considering the emotion recognition results. It then uses generative AI to create a natural language response, employing a tone and expression that matches each user's emotions. For example, if the server recognizes that the user is confused, it adjusts the response to provide a gentler and more specific explanation.

[0317] The device receives the final generated response and presents it to the user or the other party in voice or text format. By utilizing speech synthesis technology and display functions, communication support optimized for the user's current emotional state is achieved.

[0318] As a concrete example, imagine a user making a questioning gesture while showing a "troubled" expression in front of the camera. In this case, the system recognizes the user's emotion as "confusion" from the expression and generates a response that includes kindness and encouragement, such as "Are you okay?". This response is then fed back to the user via voice, resulting in a more human-like conversation.

[0319] Thus, by integrating an emotion engine, the present invention provides high-quality communication support that takes into account not only normal input analysis but also the user's emotions. This enables a more reassuring conversational experience for the user.

[0320] The following describes the processing flow.

[0321] Step 1:

[0322] The user makes gestures or touch inputs. For example, they might express their intentions by waving their hand or making a confused facial expression.

[0323] Step 2:

[0324] The device receives user gestures and facial expressions in real time via the camera and touch interface. The received data is temporarily stored as image and signal data.

[0325] Step 3:

[0326] The terminal sends the received data to the server for analysis. The data is transferred quickly using a secure protocol.

[0327] Step 4:

[0328] The server begins analyzing the received data. First, it identifies the type of gesture or input, and then uses the emotion engine to recognize the user's emotional state from their facial expressions.

[0329] Step 5:

[0330] The server estimates the user's intent based on the analysis results. The recognized emotional state is used to understand the intent and adjust the response accordingly.

[0331] Step 6:

[0332] The server uses generative AI to construct natural language responses based on estimated intentions and emotional states. The responses are appropriately adjusted in tone and content according to the recognized emotions.

[0333] Step 7:

[0334] The server sends the generated response to the terminal. The response data is prepared as audio or text.

[0335] Step 8:

[0336] The device presents the received response to the user or the other party through a speaker or display. In the case of voice output, natural and emotionally sensitive expressions are reproduced using synthesized speech technology.

[0337] (Example 2)

[0338] Next, we will describe Example 2. 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".

[0339] Conventional communication support systems simply receive and analyze user gestures and input, making it impossible to generate responses that take user emotions into account. This makes natural and user-friendly communication difficult, and often results in inadequate responses, especially in interactions involving emotional nuances.

[0340] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0341] In this invention, the server includes a receiving means, an analysis means, and an emotion recognition means. This makes it possible to generate a response that takes into account the user's emotional state in addition to their gestures and input content.

[0342] "Receiving means" refers to the function of receiving gestures and touch inputs from the user through sensors and interfaces.

[0343] "Analysis means" refers to a function that analyzes received data to identify gestures and input content.

[0344] "Emotion recognition means" refers to a function that recognizes the user's emotional state based on analyzed data.

[0345] "Estimation method" refers to a function that estimates the user's intentions from the recognized emotional state and the analyzed input content.

[0346] "Generation method" refers to a function that generates responses in natural language based on the estimated user's intentions and emotional state.

[0347] "Output means" refers to a function that provides the generated response to the user or the other party in text or voice.

[0348] The communication support system of the present invention aims to receive and analyze user gestures and inputs to generate appropriate responses. This system mainly consists of two components: a terminal and a server.

[0349] When a user performs a specific gesture or inputs via a touch interface, that input is received by the terminal. Hardware such as image sensors and touch displays are utilized here. The terminal processes this input data in real time and transfers it to a server. The data is transmitted via a secure protocol over a communication network.

[0350] To analyze the received data, the server first applies pattern recognition algorithms to identify gestures and input content. Open-source libraries such as OpenCV can be used for this purpose. Based on the analyzed data, the server utilizes an emotion recognition engine to recognize the user's emotional state from text and facial expression data. Deep learning models are suitable for emotion recognition. For example, emotion classification can be performed using a model based on TensorFlow.

[0351] Next, the server utilizes a generative AI model to generate natural language responses that are optimal for the user's intent and emotional state. By using prompts, flexible and user-friendly responses are achieved. For example, by using a prompt such as "This user is confused. How do you support them?", the generative AI will generate a human-like response such as "Don't worry, I'll explain in detail."

[0352] Finally, the device receives this generated response and presents it to the user in either audio or text format. Text-to-speech technology is used for speech synthesis, and it is desirable to utilize techniques such as WaveNet to enhance the naturalness of the speech. When displaying text via a screen, adopting a clear and intuitive UI design can further improve the user experience.

[0353] By implementing this system, users can receive flexible and considerate responses that are tailored to their emotions, resulting in a more natural communication experience.

[0354] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0355] Step 1:

[0356] The user performs gestures and touch inputs. For example, they might tap the smartphone screen or wave their hand in front of the camera. This is the initial input for the system. The device receives this input using its built-in image sensor and touch interface. The data received at this time includes the video of the gesture and the touch coordinates.

[0357] Step 2:

[0358] The terminal sends the user's input data to the server. During this process, pre-processing is performed on the terminal, and the data is compressed or encrypted before transmission. The server decompresses or decrypts the received data and prepares it for analysis. The input consists of gesture images and coordinate data, while the output is data formatted for analysis.

[0359] Step 3:

[0360] The server performs pattern recognition on the received data to identify gestures and input content. Specifically, it uses the OpenCV library to perform image processing and classify gestures. In this process, the input is video data, and the output is the label of the identified gesture and the type of touch.

[0361] Step 4:

[0362] The server performs emotion recognition on identified gestures. Here, a deep learning model is used to estimate emotions from the user's facial expressions and voice. A pre-trained emotion classification model using TensorFlow is used. The input is analyzed data, and the output is an emotion label (e.g., joy, sadness, confusion).

[0363] Step 5:

[0364] The server uses a generative AI model to generate prompts based on the user's emotional state and gestures. For example, it might generate a prompt such as, "This user is confused. How do you support them?" Furthermore, this prompt is used as input to generate a natural language response. The output is a customized text response tailored to the user's state.

[0365] Step 6:

[0366] The terminal receives a response from the server. This response includes contextual tone and expression. The received data is fed back to the user through a speech synthesizer or display. Specifically, text-to-speech technology is used to provide voice feedback such as, "Don't worry." The output is voice or text information in a format that is easy for the user to understand.

[0367] (Application Example 2)

[0368] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0369] Many modern communication devices and systems lack the ability to accurately understand a user's intentions and emotional state and generate appropriate responses accordingly. As a result, users often don't receive the responses they expect and become dissatisfied. Furthermore, current technology struggles to engage in emotionally sensitive dialogue like a human being. There is a need to solve this problem and realize more human-like communication that takes emotions into account.

[0370] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0371] In this invention, the server includes a receiving device that receives user gestures or inputs, an analysis device that analyzes the received data to identify gestures and input content, and an emotion recognition engine that recognizes the user's emotional state based on the analyzed gestures and input content. This enables intent estimation that takes the user's emotional state into account and the generation of appropriate responses in natural language.

[0372] A "receiving device" is a device that detects user gestures and inputs and acquires that data.

[0373] An "analysis device" is a device that analyzes data obtained from a receiving device to identify gestures and distinguish input content.

[0374] An "emotion recognition engine" is an engine that determines the user's emotional state based on analyzed gestures and input content.

[0375] An "estimation device" is a device that infers what a user intends by considering the user's emotional state and analysis results.

[0376] A "generator" is a device that creates appropriate responses in natural language based on the estimated user's intentions and emotional state.

[0377] An "output device" is a device that provides the user with a generated natural language response in either voice or text format.

[0378] The system for realizing this invention comprises multiple devices for processing user gestures and input. The server uses a receiving device to receive gestures and voice input from the user. The received data is analyzed using an analysis device to identify gestures and input content. Furthermore, based on the analyzed data, an emotion recognition engine determines the user's emotional state. Information on the emotional state is used by an estimation device to estimate the user's intentions. Subsequently, a generation device generates a natural language response based on the estimated intentions and emotional state. This response is provided to the user in voice or text through an output device.

[0379] The hardware will consist of image processing sensors and microphones to capture user gestures and facial expressions. A server with high processing power is also required to analyze this data. The software will utilize image recognition algorithms for analysis and emotion recognition engines (e.g., Microsoft Azure Face API) for emotion recognition. Generative AI (e.g., OpenAI GPT) will be used for natural language response generation.

[0380] For example, if a user with a tired expression asks the robot, "Tell me today's news," the server will select news in a calm tone that reflects the user's fatigue, and use a generative AI to form a response such as, "We have some calm news today. For example..." and transmit it to the robot. This allows the user to experience a dialogue that reflects their own emotional state.

[0381] An example of a prompt message would be, "When the user looks tired, adjust the content and tone of the news message to respond."

[0382] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0383] Step 1:

[0384] The server acquires user gestures and voice input via a receiving device. The input data collected includes video data from a camera and audio data from a microphone. This data is stored on the server for subsequent analysis.

[0385] Step 2:

[0386] The server analyzes video and audio data using an analysis device. Image recognition algorithms are used to identify gestures and transcribe audio input. The output obtained here contains information about the user's specific actions and statements.

[0387] Step 3:

[0388] The server uses an emotion recognition engine to determine the user's emotional state from the analyzed gesture and voice input data. This process outputs an emotional state (e.g., "fatigue" or "confusion") based on changes in facial expression and tone of voice.

[0389] Step 4:

[0390] The server uses an estimation device to estimate the user's intentions based on the obtained emotional state and the content of the user's statements. At this stage, the server identifies the user's specific purpose or request from the combination of emotion and content, and outputs that information.

[0391] Step 5:

[0392] The server generates natural language responses based on the intent and emotional state estimated by the generator. It utilizes a generative AI model to create responses with appropriate context and tone. The final output is optimized messaging for the user.

[0393] Step 6:

[0394] The terminal provides the user with a response generated using an output device. The output is either in audio format via speech synthesis or in text format via a display. This allows the user to receive emotionally resonant responses.

[0395] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0396] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0397] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0398] [Third Embodiment]

[0399] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0400] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0401] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0402] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0403] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0404] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0405] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0406] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0407] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0408] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0409] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0410] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0411] This invention provides a communication support system that allows users to communicate their intentions with confidence. This system is implemented by a program that accurately receives user gestures and inputs and generates responses in real time.

[0412] When a user performs a specific gesture, such as a greeting or a question, the device receives this information through its camera or touch interface. The acquired data is sent to a server for analysis. The server uses machine learning algorithms to analyze the data and identify the user's intent from their gestures.

[0413] A crucial part of this system is that the server generates the user's intent in natural language. Based on the estimated intent, the generative AI creates a response that fits the context of a natural conversation. In this process, the generated text is adjusted to be faithful to the user's intended meaning.

[0414] The generated response is sent to the device as audio or text and presented to the user and the other party. For example, if the user makes a waving gesture, the system will output "Hello" as audio or text, helping the recipient communicate smoothly.

[0415] Thus, the system based on the present invention is an embodiment that can be used easily and intuitively by people with disabilities or those who have anxiety about conversation, and promotes social connection.

[0416] The following describes the processing flow.

[0417] Step 1:

[0418] The user performs a gesture or input. This action is initiated using the camera or touch interface.

[0419] Step 2:

[0420] The device receives user gestures and input in real time. The received data is temporarily recorded and converted into a format that can be sent to the server.

[0421] Step 3:

[0422] The device converts the data to a specific format and sends it to the server via the internet.

[0423] Step 4:

[0424] The server analyzes the received data to identify the type of gesture and the input content. This analysis utilizes image recognition algorithms and machine learning models.

[0425] Step 5:

[0426] The server estimates the user's intent based on the analysis results. It evaluates the meaning of gestures and inputs in context and determines the appropriate response.

[0427] Step 6:

[0428] The server generates natural language responses based on estimated intent. Generative AI is used to create contextually appropriate and natural communication.

[0429] Step 7:

[0430] The server sends the generated response data to the terminal. The data is converted to the appropriate format and provided as audio output or text display.

[0431] Step 8:

[0432] The terminal presents the received response data to the user or the other party. The generated content is then transmitted through the speaker or display to constitute actual communication.

[0433] (Example 1)

[0434] Next, we will describe Example 1. 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."

[0435] Conventional communication support technologies have struggled to accurately recognize users' intentions and generate appropriate responses based on them. Therefore, a key challenge has been mitigating communication barriers, particularly faced by people with disabilities and those who experience anxiety in conversation.

[0436] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0437] In this invention, the server includes an imaging device means for receiving user actions, a computing device means for analyzing the received data and identifying the actions, and an evaluation device means for estimating the user's intent based on the analyzed actions. This makes it possible to accurately recognize the user's intent and generate and present a rapid and appropriate natural language response.

[0438] An "imaging device" is a device equipped with optical sensors or contact surfaces for acquiring user movements and gestures.

[0439] A "processing unit" is a device that has information processing capabilities to analyze received data and identify its contents.

[0440] An "evaluation device" is a device that performs processing to estimate the user's intent based on the analyzed data.

[0441] A "response generation device" is a device that has the function of generating a response in natural language based on an estimated intention.

[0442] A "presentation device" is a device that provides a generated response to the user or other party visually or audibly.

[0443] A "learning model" refers to an algorithm that is trained using a large amount of data and generates an output that is appropriate to the given input information.

[0444] This invention provides a communication support system that recognizes user gestures and generates and provides natural language responses based on those gestures. The system mainly consists of a terminal operated by the user and a server that supports it.

[0445] Terminal role

[0446] The device is equipped with a camera as an imaging device and a touch interface as a contact surface to capture user movements and gestures. For example, if a user performs a "wave" motion, that motion is received by the device.

[0447] Server Role

[0448] Data received by the terminal is sent to the server. Here, the server functions as a computing device, analyzing the received data to determine its operation. Computer vision technology and machine learning algorithms are used for this purpose. During the analysis process, the server also functions as an evaluation device, accurately estimating the user's intent.

[0449] Next, based on the estimated intent, the server uses a generative AI model to generate a natural language response as a response generator. For example, if the wavering motion is recognized as the greeting "hello," the server will create a response saying "hello."

[0450] In addition, the server sends the generated response to the presentation device, which then provides it to the user or recipient as voice or text. Voice responses are produced using speech synthesis technology, while text responses are displayed on the screen.

[0451] Specific example

[0452] The user makes a "thumbs up" gesture. The device receives this action and sends data to the server. The server recognizes the action as an "OK" intent and uses a generative AI model to generate a "thumbs up" response. Finally, the device either plays this response aloud or displays it as text.

[0453] Example of a prompt

[0454] "The user has detected a waving gesture. Please generate an appropriate greeting based on this gesture."

[0455] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0456] Step 1:

[0457] The user performs a specific gesture, and the device receives this action. The device records this action using imaging devices such as a camera or touch interface. User action information is acquired as input, and image data of the action or contact location data is generated as output.

[0458] Step 2:

[0459] The terminal transmits acquired operational data to a server via the internet. The input is operational data stored within the terminal, and the output is the transmission of that data to the server. Specifically, data transmission is performed using a wireless communication module.

[0460] Step 3:

[0461] The server analyzes the received motion data. The input is motion data from the terminal, and the output is the result of gesture identification. The server uses computer vision technology and machine learning algorithms to analyze the data and identify gestures. Specifically, feature extraction is performed using an image analysis program.

[0462] Step 4:

[0463] The server estimates the user's intent based on the analyzed actions. The input is the result of identifying the gesture, and the output is the estimated intent. The server utilizes a learning model to evaluate intent based on past data. Specifically, an intent estimation algorithm is applied.

[0464] Step 5:

[0465] The server generates natural language responses using a generative AI model based on estimated intent. The input is the estimated intent, and the output is the generated natural language response. Specifically, the generative AI model creates appropriate prompts.

[0466] Step 6:

[0467] The server sends the generated response to the terminal. The input is the generated natural language response, and the output is the transmission of the response to the terminal. Specifically, data transfer takes place through a communication protocol.

[0468] Step 7:

[0469] The terminal presents the received response to the user or the other party in audio or text format. Input is a natural language response from the server, and output is audio playback or display. Specifically, this involves audio output via the speaker or text display on the screen.

[0470] (Application Example 1)

[0471] Next, we will explain Application Example 1. In the following explanation, 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."

[0472] Currently, many elderly people and individuals requiring care face difficulties in communicating smoothly with others in their daily lives. Communication is a particularly significant challenge for those with physical limitations or anxieties about communication. Therefore, there is a need to improve their quality of life and provide them with means to express themselves with confidence.

[0473] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0474] In this invention, the server includes receiving means for receiving user actions, analysis means for analyzing the received action data and identifying the classification and content of the actions, and estimation means for estimating the user's intentions based on the analyzed actions. This enables elderly people and caregivers to communicate quickly and accurately through actions.

[0475] "User actions" refer to physical movements and gestures performed by the user, and these actions are what the system recognizes.

[0476] A "receiving means" is a device or interface for capturing user actions, and has the function of acquiring data using a video acquisition device or a contact interface.

[0477] "Analysis means" refers to means for analyzing motion data captured by the receiving means and classifying and identifying the intent behind the motion.

[0478] "Estimation means" are methods used to determine and identify the intentions indicated by the analyzed user actions.

[0479] A "generative means" is a means for forming and creating a response in natural language based on the intention identified by the estimation means.

[0480] A "presentation means" is a means that transmits the response formed by the generation means to the user or recipient in a physical or digital form.

[0481] A "video acquisition device" is a device that records a user's movements as digital data, and cameras are an example of such devices.

[0482] A "contact interface" is an interface designed to capture user touch, and touchscreens are an example of this.

[0483] A "learning model" refers to an algorithm or program that has been trained to predict and generate user intent and appropriate responses based on data.

[0484] To realize this invention, the following system configuration and program are required. The system mainly consists of a server, a terminal, and a user interface, and each component works in cooperation with each other.

[0485] The server utilizes computing resources in the cloud to process data for analyzing user actions. Specifically, the server first receives video and contact data received via video acquisition devices and contact interfaces. Next, it analyzes the received data using a learning model (e.g., TensorFlow) to classify and identify user actions. Based on the analysis results, it estimates the user's intent and generates a natural language response using a generation AI model (e.g., GPT). The generated response is then adjusted to provide the user with the necessary information clearly and quickly.

[0486] The terminal functions as an interface between the user and the system, presenting natural language responses sent from the server to the user in voice or text. This allows the user to engage in meaningful communication through their own actions.

[0487] Users primarily use smart glasses or smartphones to perform actions. For example, if a user waves their hand, this is sent to the cloud, interpreted as an intention to "want water," and a generated voice message saying "I want water" is output from the device.

[0488] As a concrete example, the following prompt statement is possible:

[0489] Prompt: "Please describe the process of interpreting intent and generating a response when a user makes a hand-waving gesture."

[0490] Such systems can remove communication barriers and provide a richer communication experience for the elderly and those who have difficulty conversing.

[0491] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0492] Step 1:

[0493] The user performs actions using smart glasses or a smartphone. This allows video acquisition devices and contact interfaces to capture the user's motion data. The input is the user's physical movements, and the output is motion data based on these movements. A concrete example of such an action is the user waving their hand.

[0494] Step 2:

[0495] The server receives behavioral data sent from the terminal. Based on the received data, it performs analysis using a machine learning model (e.g., TensorFlow). In this step, the input is behavioral data, and the output is the classification or identification of the behavior's content. Data processing includes executing a machine learning algorithm using the features of the behavior.

[0496] Step 3:

[0497] The server estimates the user's intent from the analyzed results. This estimation process uses the data obtained through analysis as input to perform data calculations that predict what intent it represents. The output is data representing the user's intent. Specifically, this involves determining that "this action indicates the intention 'I want water'."

[0498] Step 4:

[0499] The server generates natural language responses using a generative AI model (e.g., GPT) based on the estimated intent. Here, user intent data is input, and a response in natural language format is output. The generated response is then adapted to human language appropriate to the specific context. A concrete example of this process is when the response to the intent "I want water" is "I want water."

[0500] Step 5:

[0501] The device presents the user with the response received from the server. The response is displayed as audio or text. The input is natural language response data, and the output is the actual notification or voice message provided to the user. A specific example of this operation is using the speech synthesis function of smart glasses to say "I want water."

[0502] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0503] This invention provides a system that effectively supports communication by receiving and analyzing user gestures and inputs. Furthermore, by combining it with an emotion engine, it enables the generation of responses that take the user's emotions into consideration.

[0504] When a user performs a specific gesture or inputs via the touch interface during use, the device receives this information. The device then sends this data to a server where it is analyzed. During the analysis process, the gesture or input content is first identified, and then the user's emotional state is recognized using an emotion engine.

[0505] The server estimates the user's intent while considering the emotion recognition results. It then uses generative AI to create a natural language response, employing a tone and expression that matches each user's emotions. For example, if the server recognizes that the user is confused, it adjusts the response to provide a gentler and more specific explanation.

[0506] The device receives the final generated response and presents it to the user or the other party in voice or text format. By utilizing speech synthesis technology and display functions, communication support optimized for the user's current emotional state is achieved.

[0507] As a concrete example, imagine a user making a questioning gesture while showing a "troubled" expression in front of the camera. In this case, the system recognizes the user's emotion as "confusion" from the expression and generates a response that includes kindness and encouragement, such as "Are you okay?". This response is then fed back to the user via voice, resulting in a more human-like conversation.

[0508] Thus, by integrating an emotion engine, the present invention provides high-quality communication support that takes into account not only normal input analysis but also the user's emotions. This enables a more reassuring conversational experience for the user.

[0509] The following describes the processing flow.

[0510] Step 1:

[0511] The user makes gestures or touch inputs. For example, they might express their intentions by waving their hand or making a confused facial expression.

[0512] Step 2:

[0513] The device receives user gestures and facial expressions in real time via the camera and touch interface. The received data is temporarily stored as image and signal data.

[0514] Step 3:

[0515] The terminal sends the received data to the server for analysis. The data is transferred quickly using a secure protocol.

[0516] Step 4:

[0517] The server begins analyzing the received data. First, it identifies the type of gesture or input, and then uses the emotion engine to recognize the user's emotional state from their facial expressions.

[0518] Step 5:

[0519] The server estimates the user's intent based on the analysis results. The recognized emotional state is used to understand the intent and adjust the response accordingly.

[0520] Step 6:

[0521] The server uses generative AI to construct natural language responses based on estimated intentions and emotional states. The responses are appropriately adjusted in tone and content according to the recognized emotions.

[0522] Step 7:

[0523] The server sends the generated response to the terminal. The response data is prepared as audio or text.

[0524] Step 8:

[0525] The device presents the received response to the user or the other party through a speaker or display. In the case of voice output, natural and emotionally sensitive expressions are reproduced using synthesized speech technology.

[0526] (Example 2)

[0527] Next, we will describe Example 2. 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."

[0528] Conventional communication support systems simply receive and analyze user gestures and input, making it impossible to generate responses that take user emotions into account. This makes natural and user-friendly communication difficult, and often results in inadequate responses, especially in interactions involving emotional nuances.

[0529] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0530] In this invention, the server includes a receiving means, an analysis means, and an emotion recognition means. This makes it possible to generate a response that takes into account the user's emotional state in addition to their gestures and input content.

[0531] "Receiving means" refers to the function of receiving gestures and touch inputs from the user through sensors and interfaces.

[0532] "Analysis means" refers to a function that analyzes received data to identify gestures and input content.

[0533] "Emotion recognition means" refers to a function that recognizes the user's emotional state based on analyzed data.

[0534] "Estimation method" refers to a function that estimates the user's intentions from the recognized emotional state and the analyzed input content.

[0535] "Generation method" refers to a function that generates responses in natural language based on the estimated user's intentions and emotional state.

[0536] "Output means" refers to a function that provides the generated response to the user or the other party in text or voice.

[0537] The communication support system of the present invention aims to receive and analyze user gestures and inputs to generate appropriate responses. This system mainly consists of two components: a terminal and a server.

[0538] When a user performs a specific gesture or inputs via a touch interface, that input is received by the terminal. Hardware such as image sensors and touch displays are utilized here. The terminal processes this input data in real time and transfers it to a server. The data is transmitted via a secure protocol over a communication network.

[0539] To analyze the received data, the server first applies pattern recognition algorithms to identify gestures and input content. Open-source libraries such as OpenCV can be used for this purpose. Based on the analyzed data, the server utilizes an emotion recognition engine to recognize the user's emotional state from text and facial expression data. Deep learning models are suitable for emotion recognition. For example, emotion classification can be performed using a model based on TensorFlow.

[0540] Next, the server utilizes a generative AI model to generate natural language responses that are optimal for the user's intent and emotional state. By using prompts, flexible and user-friendly responses are achieved. For example, by using a prompt such as "This user is confused. How do you support them?", the generative AI will generate a human-like response such as "Don't worry, I'll explain in detail."

[0541] Finally, the device receives this generated response and presents it to the user in either audio or text format. Text-to-speech technology is used for speech synthesis, and it is desirable to utilize techniques such as WaveNet to enhance the naturalness of the speech. When displaying text via a screen, adopting a clear and intuitive UI design can further improve the user experience.

[0542] By implementing this system, users can receive flexible and considerate responses that are tailored to their emotions, resulting in a more natural communication experience.

[0543] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0544] Step 1:

[0545] The user performs gestures and touch inputs. For example, they might tap the smartphone screen or wave their hand in front of the camera. This is the initial input for the system. The device receives this input using its built-in image sensor and touch interface. The data received at this time includes the video of the gesture and the touch coordinates.

[0546] Step 2:

[0547] The terminal sends the user's input data to the server. During this process, pre-processing is performed on the terminal, and the data is compressed or encrypted before transmission. The server decompresses or decrypts the received data and prepares it for analysis. The input consists of gesture images and coordinate data, while the output is data formatted for analysis.

[0548] Step 3:

[0549] The server performs pattern recognition on the received data to identify gestures and input content. Specifically, it uses the OpenCV library to perform image processing and classify gestures. In this process, the input is video data, and the output is the label of the identified gesture and the type of touch.

[0550] Step 4:

[0551] The server performs emotion recognition on identified gestures. Here, a deep learning model is used to estimate emotions from the user's facial expressions and voice. A pre-trained emotion classification model using TensorFlow is used. The input is analyzed data, and the output is an emotion label (e.g., joy, sadness, confusion).

[0552] Step 5:

[0553] The server uses a generative AI model to generate prompts based on the user's emotional state and gestures. For example, it might generate a prompt such as, "This user is confused. How do you support them?" Furthermore, this prompt is used as input to generate a natural language response. The output is a customized text response tailored to the user's state.

[0554] Step 6:

[0555] The terminal receives a response from the server. This response includes contextual tone and expression. The received data is fed back to the user through a speech synthesizer or display. Specifically, text-to-speech technology is used to provide voice feedback such as, "Don't worry." The output is voice or text information in a format that is easy for the user to understand.

[0556] (Application Example 2)

[0557] Next, we will explain application example 2. In the following explanation, 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."

[0558] Many modern communication devices and systems lack the ability to accurately understand a user's intentions and emotional state and generate appropriate responses accordingly. As a result, users often don't receive the responses they expect and become dissatisfied. Furthermore, current technology struggles to engage in emotionally sensitive dialogue like a human being. There is a need to solve this problem and realize more human-like communication that takes emotions into account.

[0559] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0560] In this invention, the server includes a receiving device that receives user gestures or inputs, an analysis device that analyzes the received data to identify gestures and input content, and an emotion recognition engine that recognizes the user's emotional state based on the analyzed gestures and input content. This enables intent estimation that takes the user's emotional state into account and the generation of appropriate responses in natural language.

[0561] A "receiving device" is a device that detects user gestures and inputs and acquires that data.

[0562] An "analysis device" is a device that analyzes data obtained from a receiving device to identify gestures and distinguish input content.

[0563] An "emotion recognition engine" is an engine that determines the user's emotional state based on analyzed gestures and input content.

[0564] An "estimation device" is a device that infers what a user intends by considering the user's emotional state and analysis results.

[0565] A "generator" is a device that creates appropriate responses in natural language based on the estimated user's intentions and emotional state.

[0566] An "output device" is a device that provides the user with a generated natural language response in either voice or text format.

[0567] The system for realizing this invention comprises multiple devices for processing user gestures and input. The server uses a receiving device to receive gestures and voice input from the user. The received data is analyzed using an analysis device to identify gestures and input content. Furthermore, based on the analyzed data, an emotion recognition engine determines the user's emotional state. Information on the emotional state is used by an estimation device to estimate the user's intentions. Subsequently, a generation device generates a natural language response based on the estimated intentions and emotional state. This response is provided to the user in voice or text through an output device.

[0568] The hardware will consist of image processing sensors and microphones to capture user gestures and facial expressions. A server with high processing power is also required to analyze this data. The software will utilize image recognition algorithms for analysis and emotion recognition engines (e.g., Microsoft Azure Face API) for emotion recognition. Generative AI (e.g., OpenAI GPT) will be used for natural language response generation.

[0569] For example, if a user with a tired expression asks the robot, "Tell me today's news," the server will select news in a calm tone that reflects the user's fatigue, and use a generative AI to form a response such as, "We have some calm news today. For example..." and transmit it to the robot. This allows the user to experience a dialogue that reflects their own emotional state.

[0570] An example of a prompt message would be, "When the user looks tired, adjust the content and tone of the news message to respond."

[0571] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0572] Step 1:

[0573] The server acquires user gestures and voice input via a receiving device. The input data collected includes video data from a camera and audio data from a microphone. This data is stored on the server for subsequent analysis.

[0574] Step 2:

[0575] The server analyzes video and audio data using an analysis device. Image recognition algorithms are used to identify gestures and transcribe audio input. The output obtained here contains information about the user's specific actions and statements.

[0576] Step 3:

[0577] The server uses an emotion recognition engine to determine the user's emotional state from the analyzed gesture and voice input data. This process outputs an emotional state (e.g., "fatigue" or "confusion") based on changes in facial expression and tone of voice.

[0578] Step 4:

[0579] The server uses an estimation device to estimate the user's intentions based on the obtained emotional state and the content of the user's statements. At this stage, the server identifies the user's specific purpose or request from the combination of emotion and content, and outputs that information.

[0580] Step 5:

[0581] The server generates natural language responses based on the intent and emotional state estimated by the generator. It utilizes a generative AI model to create responses with appropriate context and tone. The final output is optimized messaging for the user.

[0582] Step 6:

[0583] The terminal provides the user with a response generated using an output device. The output is either in audio format via speech synthesis or in text format via a display. This allows the user to receive emotionally resonant responses.

[0584] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0585] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0586] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0587] [Fourth Embodiment]

[0588] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0589] As shown in Figure 7, the 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.

[0590] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0591] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0592] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0593] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0594] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0595] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0596] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0597] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0598] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0599] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0600] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0601] This invention provides a communication support system that allows users to communicate their intentions with confidence. This system is implemented by a program that accurately receives user gestures and inputs and generates responses in real time.

[0602] When a user performs a specific gesture, such as a greeting or a question, the device receives this information through its camera or touch interface. The acquired data is sent to a server for analysis. The server uses machine learning algorithms to analyze the data and identify the user's intent from their gestures.

[0603] A crucial part of this system is that the server generates the user's intent in natural language. Based on the estimated intent, the generative AI creates a response that fits the context of a natural conversation. In this process, the generated text is adjusted to be faithful to the user's intended meaning.

[0604] The generated response is sent to the device as audio or text and presented to the user and the other party. For example, if the user makes a waving gesture, the system will output "Hello" as audio or text, helping the recipient communicate smoothly.

[0605] Thus, the system based on the present invention is an embodiment that can be used easily and intuitively by people with disabilities or those who have anxiety about conversation, and promotes social connection.

[0606] The following describes the processing flow.

[0607] Step 1:

[0608] The user performs a gesture or input. This action is initiated using the camera or touch interface.

[0609] Step 2:

[0610] The device receives user gestures and input in real time. The received data is temporarily recorded and converted into a format that can be sent to the server.

[0611] Step 3:

[0612] The device converts the data to a specific format and sends it to the server via the internet.

[0613] Step 4:

[0614] The server analyzes the received data to identify the type of gesture and the input content. This analysis utilizes image recognition algorithms and machine learning models.

[0615] Step 5:

[0616] The server estimates the user's intent based on the analysis results. It evaluates the meaning of gestures and inputs in context and determines the appropriate response.

[0617] Step 6:

[0618] The server generates natural language responses based on estimated intent. Generative AI is used to create contextually appropriate and natural communication.

[0619] Step 7:

[0620] The server sends the generated response data to the terminal. The data is converted to the appropriate format and provided as audio output or text display.

[0621] Step 8:

[0622] The terminal presents the received response data to the user or the other party. The generated content is then transmitted through the speaker or display to constitute actual communication.

[0623] (Example 1)

[0624] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0625] Conventional communication support technologies have struggled to accurately recognize users' intentions and generate appropriate responses based on them. Therefore, a key challenge has been mitigating communication barriers, particularly faced by people with disabilities and those who experience anxiety in conversation.

[0626] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0627] In this invention, the server includes an imaging device means for receiving user actions, a computing device means for analyzing the received data and identifying the actions, and an evaluation device means for estimating the user's intent based on the analyzed actions. This makes it possible to accurately recognize the user's intent and generate and present a rapid and appropriate natural language response.

[0628] An "imaging device" is a device equipped with optical sensors or contact surfaces for acquiring user movements and gestures.

[0629] A "processing unit" is a device that has information processing capabilities to analyze received data and identify its contents.

[0630] An "evaluation device" is a device that performs processing to estimate the user's intent based on the analyzed data.

[0631] A "response generation device" is a device that has the function of generating a response in natural language based on an estimated intention.

[0632] A "presentation device" is a device that provides a generated response to the user or other party visually or audibly.

[0633] A "learning model" refers to an algorithm that is trained using a large amount of data and generates an output that is appropriate to the given input information.

[0634] This invention provides a communication support system that recognizes user gestures and generates and provides natural language responses based on those gestures. The system mainly consists of a terminal operated by the user and a server that supports it.

[0635] Terminal role

[0636] The device is equipped with a camera as an imaging device and a touch interface as a contact surface to capture user movements and gestures. For example, if a user performs a "wave" motion, that motion is received by the device.

[0637] Server Role

[0638] Data received by the terminal is sent to the server. Here, the server functions as a computing device, analyzing the received data to determine its operation. Computer vision technology and machine learning algorithms are used for this purpose. During the analysis process, the server also functions as an evaluation device, accurately estimating the user's intent.

[0639] Next, based on the estimated intent, the server uses a generative AI model to generate a natural language response as a response generator. For example, if the wavering motion is recognized as the greeting "hello," the server will create a response saying "hello."

[0640] In addition, the server sends the generated response to the presentation device, which then provides it to the user or recipient as voice or text. Voice responses are produced using speech synthesis technology, while text responses are displayed on the screen.

[0641] Specific example

[0642] The user makes a "thumbs up" gesture. The device receives this action and sends data to the server. The server recognizes the action as an "OK" intent and uses a generative AI model to generate a "thumbs up" response. Finally, the device either plays this response aloud or displays it as text.

[0643] Example of a prompt

[0644] "The user has detected a waving gesture. Please generate an appropriate greeting based on this gesture."

[0645] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0646] Step 1:

[0647] The user performs a specific gesture, and the device receives this action. The device records this action using imaging devices such as a camera or touch interface. User action information is acquired as input, and image data of the action or contact location data is generated as output.

[0648] Step 2:

[0649] The terminal transmits acquired operational data to a server via the internet. The input is operational data stored within the terminal, and the output is the transmission of that data to the server. Specifically, data transmission is performed using a wireless communication module.

[0650] Step 3:

[0651] The server analyzes the received motion data. The input is motion data from the terminal, and the output is the result of gesture identification. The server uses computer vision technology and machine learning algorithms to analyze the data and identify gestures. Specifically, feature extraction is performed using an image analysis program.

[0652] Step 4:

[0653] The server estimates the user's intent based on the analyzed actions. The input is the result of identifying the gesture, and the output is the estimated intent. The server utilizes a learning model to evaluate intent based on past data. Specifically, an intent estimation algorithm is applied.

[0654] Step 5:

[0655] The server generates natural language responses using a generative AI model based on estimated intent. The input is the estimated intent, and the output is the generated natural language response. Specifically, the generative AI model creates appropriate prompts.

[0656] Step 6:

[0657] The server sends the generated response to the terminal. The input is the generated natural language response, and the output is the transmission of the response to the terminal. Specifically, data transfer takes place through a communication protocol.

[0658] Step 7:

[0659] The terminal presents the received response to the user or the other party in audio or text format. Input is a natural language response from the server, and output is audio playback or display. Specifically, this involves audio output via the speaker or text display on the screen.

[0660] (Application Example 1)

[0661] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0662] Currently, many elderly people and individuals requiring care face difficulties in communicating smoothly with others in their daily lives. Communication is a particularly significant challenge for those with physical limitations or anxieties about communication. Therefore, there is a need to improve their quality of life and provide them with means to express themselves with confidence.

[0663] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0664] In this invention, the server includes receiving means for receiving user actions, analysis means for analyzing the received action data and identifying the classification and content of the actions, and estimation means for estimating the user's intentions based on the analyzed actions. This enables elderly people and caregivers to communicate quickly and accurately through actions.

[0665] "User actions" refer to physical movements and gestures performed by the user, and these actions are what the system recognizes.

[0666] A "receiving means" is a device or interface for capturing user actions, and has the function of acquiring data using a video acquisition device or a contact interface.

[0667] "Analysis means" refers to means for analyzing motion data captured by the receiving means and classifying and identifying the intent behind the motion.

[0668] "Estimation means" are methods used to determine and identify the intentions indicated by the analyzed user actions.

[0669] A "generative means" is a means for forming and creating a response in natural language based on the intention identified by the estimation means.

[0670] A "presentation means" is a means that transmits the response formed by the generation means to the user or recipient in a physical or digital form.

[0671] A "video acquisition device" is a device that records a user's movements as digital data, and cameras are an example of such devices.

[0672] A "contact interface" is an interface designed to capture user touch, and touchscreens are an example of this.

[0673] A "learning model" refers to an algorithm or program that has been trained to predict and generate user intent and appropriate responses based on data.

[0674] To realize this invention, the following system configuration and program are required. The system mainly consists of a server, a terminal, and a user interface, and each component works in cooperation with each other.

[0675] The server utilizes computing resources in the cloud to process data for analyzing user actions. Specifically, the server first receives video and contact data received via video acquisition devices and contact interfaces. Next, it analyzes the received data using a learning model (e.g., TensorFlow) to classify and identify user actions. Based on the analysis results, it estimates the user's intent and generates a natural language response using a generation AI model (e.g., GPT). The generated response is then adjusted to provide the user with the necessary information clearly and quickly.

[0676] The terminal functions as an interface between the user and the system, presenting natural language responses sent from the server to the user in voice or text. This allows the user to engage in meaningful communication through their own actions.

[0677] Users primarily use smart glasses or smartphones to perform actions. For example, if a user waves their hand, this is sent to the cloud, interpreted as an intention to "want water," and a generated voice message saying "I want water" is output from the device.

[0678] As a concrete example, the following prompt statement is possible:

[0679] Prompt: "Please describe the process of interpreting intent and generating a response when a user makes a hand-waving gesture."

[0680] Such systems can remove communication barriers and provide a richer communication experience for the elderly and those who have difficulty conversing.

[0681] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0682] Step 1:

[0683] The user performs actions using smart glasses or a smartphone. This allows video acquisition devices and contact interfaces to capture the user's motion data. The input is the user's physical movements, and the output is motion data based on these movements. A concrete example of such an action is the user waving their hand.

[0684] Step 2:

[0685] The server receives behavioral data sent from the terminal. Based on the received data, it performs analysis using a machine learning model (e.g., TensorFlow). In this step, the input is behavioral data, and the output is the classification or identification of the behavior's content. Data processing includes executing a machine learning algorithm using the features of the behavior.

[0686] Step 3:

[0687] The server estimates the user's intent from the analyzed results. This estimation process uses the data obtained through analysis as input to perform data calculations that predict what intent it represents. The output is data representing the user's intent. Specifically, this involves determining that "this action indicates the intention 'I want water'."

[0688] Step 4:

[0689] The server generates natural language responses using a generative AI model (e.g., GPT) based on the estimated intent. Here, user intent data is input, and a response in natural language format is output. The generated response is then adapted to human language appropriate to the specific context. A concrete example of this process is when the response to the intent "I want water" is "I want water."

[0690] Step 5:

[0691] The device presents the user with the response received from the server. The response is displayed as audio or text. The input is natural language response data, and the output is the actual notification or voice message provided to the user. A specific example of this operation is using the speech synthesis function of smart glasses to say "I want water."

[0692] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0693] This invention provides a system that effectively supports communication by receiving and analyzing user gestures and inputs. Furthermore, by combining it with an emotion engine, it enables the generation of responses that take the user's emotions into consideration.

[0694] When a user performs a specific gesture or inputs via the touch interface during use, the device receives this information. The device then sends this data to a server where it is analyzed. During the analysis process, the gesture or input content is first identified, and then the user's emotional state is recognized using an emotion engine.

[0695] The server estimates the user's intent while considering the emotion recognition results. It then uses generative AI to create a natural language response, employing a tone and expression that matches each user's emotions. For example, if the server recognizes that the user is confused, it adjusts the response to provide a gentler and more specific explanation.

[0696] The device receives the final generated response and presents it to the user or the other party in voice or text format. By utilizing speech synthesis technology and display functions, communication support optimized for the user's current emotional state is achieved.

[0697] As a concrete example, imagine a user making a questioning gesture while showing a "troubled" expression in front of the camera. In this case, the system recognizes the user's emotion as "confusion" from the expression and generates a response that includes kindness and encouragement, such as "Are you okay?". This response is then fed back to the user via voice, resulting in a more human-like conversation.

[0698] Thus, by integrating an emotion engine, the present invention provides high-quality communication support that takes into account not only normal input analysis but also the user's emotions. This enables a more reassuring conversational experience for the user.

[0699] The following describes the processing flow.

[0700] Step 1:

[0701] The user makes gestures or touch inputs. For example, they might express their intentions by waving their hand or making a confused facial expression.

[0702] Step 2:

[0703] The device receives user gestures and facial expressions in real time via the camera and touch interface. The received data is temporarily stored as image and signal data.

[0704] Step 3:

[0705] The terminal sends the received data to the server for analysis. The data is transferred quickly using a secure protocol.

[0706] Step 4:

[0707] The server begins analyzing the received data. First, it identifies the type of gesture or input, and then uses the emotion engine to recognize the user's emotional state from their facial expressions.

[0708] Step 5:

[0709] The server estimates the user's intent based on the analysis results. The recognized emotional state is used to understand the intent and adjust the response accordingly.

[0710] Step 6:

[0711] The server uses generative AI to construct natural language responses based on estimated intentions and emotional states. The responses are appropriately adjusted in tone and content according to the recognized emotions.

[0712] Step 7:

[0713] The server sends the generated response to the terminal. The response data is prepared as audio or text.

[0714] Step 8:

[0715] The device presents the received response to the user or the other party through a speaker or display. In the case of voice output, natural and emotionally sensitive expressions are reproduced using synthesized speech technology.

[0716] (Example 2)

[0717] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0718] Conventional communication support systems simply receive and analyze user gestures and input, making it impossible to generate responses that take user emotions into account. This makes natural and user-friendly communication difficult, and often results in inadequate responses, especially in interactions involving emotional nuances.

[0719] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0720] In this invention, the server includes a receiving means, an analysis means, and an emotion recognition means. This makes it possible to generate a response that takes into account the user's emotional state in addition to their gestures and input content.

[0721] "Receiving means" refers to the function of receiving gestures and touch inputs from the user through sensors and interfaces.

[0722] "Analysis means" refers to a function that analyzes received data to identify gestures and input content.

[0723] "Emotion recognition means" refers to a function that recognizes the user's emotional state based on analyzed data.

[0724] "Estimation method" refers to a function that estimates the user's intentions from the recognized emotional state and the analyzed input content.

[0725] "Generation method" refers to a function that generates responses in natural language based on the estimated user's intentions and emotional state.

[0726] "Output means" refers to a function that provides the generated response to the user or the other party in text or voice.

[0727] The communication support system of the present invention aims to receive and analyze user gestures and inputs to generate appropriate responses. This system mainly consists of two components: a terminal and a server.

[0728] When a user performs a specific gesture or inputs via a touch interface, that input is received by the terminal. Hardware such as image sensors and touch displays are utilized here. The terminal processes this input data in real time and transfers it to a server. The data is transmitted via a secure protocol over a communication network.

[0729] To analyze the received data, the server first applies pattern recognition algorithms to identify gestures and input content. Open-source libraries such as OpenCV can be used for this purpose. Based on the analyzed data, the server utilizes an emotion recognition engine to recognize the user's emotional state from text and facial expression data. Deep learning models are suitable for emotion recognition. For example, emotion classification can be performed using a model based on TensorFlow.

[0730] Next, the server utilizes a generative AI model to generate natural language responses that are optimal for the user's intent and emotional state. By using prompts, flexible and user-friendly responses are achieved. For example, by using a prompt such as "This user is confused. How do you support them?", the generative AI will generate a human-like response such as "Don't worry, I'll explain in detail."

[0731] Finally, the device receives this generated response and presents it to the user in either audio or text format. Text-to-speech technology is used for speech synthesis, and it is desirable to utilize techniques such as WaveNet to enhance the naturalness of the speech. When displaying text via a screen, adopting a clear and intuitive UI design can further improve the user experience.

[0732] By implementing this system, users can receive flexible and considerate responses that are tailored to their emotions, resulting in a more natural communication experience.

[0733] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0734] Step 1:

[0735] The user performs gestures and touch inputs. For example, they might tap the smartphone screen or wave their hand in front of the camera. This is the initial input for the system. The device receives this input using its built-in image sensor and touch interface. The data received at this time includes the video of the gesture and the touch coordinates.

[0736] Step 2:

[0737] The terminal sends the user's input data to the server. During this process, pre-processing is performed on the terminal, and the data is compressed or encrypted before transmission. The server decompresses or decrypts the received data and prepares it for analysis. The input consists of gesture images and coordinate data, while the output is data formatted for analysis.

[0738] Step 3:

[0739] The server performs pattern recognition on the received data to identify gestures and input content. Specifically, it uses the OpenCV library to perform image processing and classify gestures. In this process, the input is video data, and the output is the label of the identified gesture and the type of touch.

[0740] Step 4:

[0741] The server performs emotion recognition on identified gestures. Here, a deep learning model is used to estimate emotions from the user's facial expressions and voice. A pre-trained emotion classification model using TensorFlow is used. The input is analyzed data, and the output is an emotion label (e.g., joy, sadness, confusion).

[0742] Step 5:

[0743] The server uses a generative AI model to generate prompts based on the user's emotional state and gestures. For example, it might generate a prompt such as, "This user is confused. How do you support them?" Furthermore, this prompt is used as input to generate a natural language response. The output is a customized text response tailored to the user's state.

[0744] Step 6:

[0745] The terminal receives a response from the server. This response includes contextual tone and expression. The received data is fed back to the user through a speech synthesizer or display. Specifically, text-to-speech technology is used to provide voice feedback such as, "Don't worry." The output is voice or text information in a format that is easy for the user to understand.

[0746] (Application Example 2)

[0747] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0748] Many modern communication devices and systems lack the ability to accurately understand a user's intentions and emotional state and generate appropriate responses accordingly. As a result, users often don't receive the responses they expect and become dissatisfied. Furthermore, current technology struggles to engage in emotionally sensitive dialogue like a human being. There is a need to solve this problem and realize more human-like communication that takes emotions into account.

[0749] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0750] In this invention, the server includes a receiving device that receives user gestures or inputs, an analysis device that analyzes the received data to identify gestures and input content, and an emotion recognition engine that recognizes the user's emotional state based on the analyzed gestures and input content. This enables intent estimation that takes the user's emotional state into account and the generation of appropriate responses in natural language.

[0751] A "receiving device" is a device that detects user gestures and inputs and acquires that data.

[0752] An "analysis device" is a device that analyzes data obtained from a receiving device to identify gestures and distinguish input content.

[0753] An "emotion recognition engine" is an engine that determines the user's emotional state based on analyzed gestures and input content.

[0754] An "estimation device" is a device that infers what a user intends by considering the user's emotional state and analysis results.

[0755] A "generator" is a device that creates appropriate responses in natural language based on the estimated user's intentions and emotional state.

[0756] An "output device" is a device that provides the user with a generated natural language response in either voice or text format.

[0757] The system for realizing this invention comprises multiple devices for processing user gestures and input. The server uses a receiving device to receive gestures and voice input from the user. The received data is analyzed using an analysis device to identify gestures and input content. Furthermore, based on the analyzed data, an emotion recognition engine determines the user's emotional state. Information on the emotional state is used by an estimation device to estimate the user's intentions. Subsequently, a generation device generates a natural language response based on the estimated intentions and emotional state. This response is provided to the user in voice or text through an output device.

[0758] The hardware will consist of image processing sensors and microphones to capture user gestures and facial expressions. A server with high processing power is also required to analyze this data. The software will utilize image recognition algorithms for analysis and emotion recognition engines (e.g., Microsoft Azure Face API) for emotion recognition. Generative AI (e.g., OpenAI GPT) will be used for natural language response generation.

[0759] For example, if a user with a tired expression asks the robot, "Tell me today's news," the server will select news in a calm tone that reflects the user's fatigue, and use a generative AI to form a response such as, "We have some calm news today. For example..." and transmit it to the robot. This allows the user to experience a dialogue that reflects their own emotional state.

[0760] An example of a prompt message would be, "When the user looks tired, adjust the content and tone of the news message to respond."

[0761] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0762] Step 1:

[0763] The server acquires user gestures and voice input via a receiving device. The input data collected includes video data from a camera and audio data from a microphone. This data is stored on the server for subsequent analysis.

[0764] Step 2:

[0765] The server analyzes video and audio data using an analysis device. Image recognition algorithms are used to identify gestures and transcribe audio input. The output obtained here contains information about the user's specific actions and statements.

[0766] Step 3:

[0767] The server uses an emotion recognition engine to determine the user's emotional state from the analyzed gesture and voice input data. This process outputs an emotional state (e.g., "fatigue" or "confusion") based on changes in facial expression and tone of voice.

[0768] Step 4:

[0769] The server uses an estimation device to estimate the user's intentions based on the obtained emotional state and the content of the user's statements. At this stage, the server identifies the user's specific purpose or request from the combination of emotion and content, and outputs that information.

[0770] Step 5:

[0771] The server generates natural language responses based on the intent and emotional state estimated by the generator. It utilizes a generative AI model to create responses with appropriate context and tone. The final output is optimized messaging for the user.

[0772] Step 6:

[0773] The terminal provides the user with a response generated using an output device. The output is either in audio format via speech synthesis or in text format via a display. This allows the user to receive emotionally resonant responses.

[0774] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0775] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0776] 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 this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0777] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0778] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0779] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0780] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0781] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0782] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is 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 the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0783] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0784] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0785] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0786] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0787] 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.

[0788] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0789] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0790] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0791] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0792] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0793] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0794] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0795] The following is further disclosed regarding the embodiments described above.

[0796] (Claim 1)

[0797] A receiving means for receiving user gestures and input,

[0798] An analysis means that analyzes the received data to identify gestures and input content,

[0799] An estimation means for estimating the user's intent based on analyzed gestures and input content,

[0800] A generation means for generating a natural language response based on an estimated intention,

[0801] An output means for providing the generated response to the user or the other party,

[0802] A communication support system that includes this.

[0803] (Claim 2)

[0804] The communication support system according to claim 1, characterized in that the receiving means uses an image sensor or a touch interface.

[0805] (Claim 3)

[0806] The communication support system according to claim 1, characterized in that the generation means generates a natural language response using a machine learning model.

[0807] "Example 1"

[0808] (Claim 1)

[0809] An imaging device that receives user actions,

[0810] A computing device that analyzes the received data and identifies the operation,

[0811] An evaluation device that estimates the user's intent based on the analyzed actions,

[0812] A response generation device that generates natural language responses based on estimated intentions,

[0813] A presentation device that provides the generated response to the user or the other party,

[0814] A system that includes this.

[0815] (Claim 2)

[0816] The system according to claim 1, characterized in that the imaging device uses an optical sensor or a contact surface.

[0817] (Claim 3)

[0818] The system according to claim 1, characterized in that the response generation device generates a natural language response using a learning model.

[0819] "Application Example 1"

[0820] (Claim 1)

[0821] A receiving means for receiving user actions,

[0822] An analysis means that analyzes the received motion data and identifies the classification and content of the motion,

[0823] An estimation means for estimating the user's intent based on the analyzed behavior,

[0824] A generation means for generating a natural language response based on an estimated intention,

[0825] A means of providing the generated response to the user or other party,

[0826] A means to communicate specific intentions to caregivers using user actions, and to provide additional functions to support communication.

[0827] A system that includes this.

[0828] (Claim 2)

[0829] The system according to claim 1, characterized in that the receiving means uses an image acquisition device or a contact interface.

[0830] (Claim 3)

[0831] The system according to claim 1, characterized in that the generation means generates a natural language response using a learning model.

[0832] "Example 2 of combining an emotion engine"

[0833] (Claim 1)

[0834] A receiving means for receiving user gestures and input,

[0835] An analysis means that analyzes the received data and identifies gestures and input content,

[0836] An emotion recognition means for recognizing emotional states during the analysis process,

[0837] An estimation means for estimating the user's intent based on recognized emotional states and identified content,

[0838] A generation means for generating natural language responses based on estimated intentions and emotional states,

[0839] An output means for providing the generated response to the user or the other party,

[0840] A system that includes this.

[0841] (Claim 2)

[0842] The system according to claim 1, characterized in that the receiving means uses an image sensor or a touch interface.

[0843] (Claim 3)

[0844] The system according to claim 1, characterized in that the generation means generates a natural language response using a machine learning model.

[0845] "Application example 2 when combining with an emotional engine"

[0846] (Claim 1)

[0847] A receiving device that receives user gestures or input,

[0848] An analysis device that analyzes received data to identify gestures and identify input content,

[0849] An emotion recognition engine that recognizes the user's emotional state based on analyzed gestures and input content,

[0850] An estimation device that estimates the user's intentions by taking into account the recognized emotional state,

[0851] A generator that generates responses in natural language based on estimated intentions and emotional states,

[0852] An output device that outputs the generated response via an information processing device,

[0853] A communication support system that includes this.

[0854] (Claim 2)

[0855] The system according to claim 1, characterized in that the receiving device uses an image processing sensor or a tactile interface.

[0856] (Claim 3)

[0857] The system according to claim 1, characterized in that the generation device generates natural language responses using an artificial intelligence model. [Explanation of Symbols]

[0858] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A receiving means for receiving user actions, An analysis means that analyzes the received motion data and identifies the classification and content of the motion, An estimation means for estimating the user's intent based on the analyzed behavior, A generation means for generating a natural language response based on an estimated intention, A means of providing the generated response to the user or other party, A means to communicate specific intentions to caregivers using user actions, and to provide additional functions to support communication. A system that includes this.

2. The system according to claim 1, characterized in that the receiving means uses an image acquisition device or a contact interface.

3. The system according to claim 1, characterized in that the generation means generates a natural language response using a learning model.

Images