System

Abstract

A system is provided.SOLUTION: A system comprising: means for receiving voice input of a patient; means for transmitting received voice data to a server; means for analyzing the voice data and grasping an emotional state and a cognitive state of the patient by the server; means for selecting an optimal rehabilitation technique based on the grasped states; means for transmitting the selected rehabilitation technique to a terminal; and means for presenting rehabilitation contents to the patient and acquiring feedback by the terminal.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Currently, there are two types of symptomatic treatment for dementia patients: drug therapy and non-drug therapy. However, providing rehabilitation methods appropriate for each individual patient in the case of non-drug therapy places a heavy burden on caregivers and family members. Furthermore, conventional systems have difficulty flexibly adjusting rehabilitation content based on the patient's emotional state and physical condition on that day. This makes it difficult to continuously provide the patient with the optimal rehabilitation method, making it difficult to achieve effective care. The present invention aims to solve this problem by providing a system that provides effective rehabilitation methods in real time to slow the progression of dementia in patients. [Means for solving the problem]

[0005] The present invention includes a means for receiving a patient's voice input and transmitting the received voice data to a server. The server analyzes the voice data to grasp the patient's emotional and cognitive states. Based on the grasped state, the server selects an optimal rehabilitation method and transmits the selected rehabilitation method to a terminal. The terminal presents the rehabilitation content to the patient and obtains feedback. Furthermore, by including a server means for analyzing the tone and strength of the voice data to identify emotions, it is possible to select a rehabilitation method according to emotions. Furthermore, by including a means for automatically selecting one of cognitive stimulation therapy, reminiscence therapy, and reality orientation as the rehabilitation method, flexible responses are possible without being limited to a specific rehabilitation method. This makes it possible to provide optimal rehabilitation for each patient while reducing the burden on caregivers and family members.

[0006] "Voice input" refers to information such as words or sounds spoken by the patient, and is data that the system receives and analyzes.

[0007] "Voice data" means voice input converted into digital form for analysis and transmission.

[0008] The "server" is an information processing device that analyzes voice data and selects a rehabilitation method, and plays a central role in the present invention.

[0009] "Analysis" refers to the information processing process that evaluates the patient's emotional and cognitive state based on voice data.

[0010] "Emotional state" refers to the patient's mental and emotional state, as determined by tone and content of voice.

[0011] "Cognitive status" refers to the patient's intellectual state, such as cognitive ability and memory.

[0012] "Rehabilitation techniques" refer to specific methods and approaches to slow the progression of dementia in patients, including cognitive stimulation therapy, reminiscence therapy, and reality orientation.

[0013] "Terminal" refers to a device that receives the patient's voice input and presents the rehabilitation content from the server to the patient.

[0014] "Presentation" refers to the act of the device communicating rehabilitation details to the patient through voice, screen, etc.

[0015] "Feedback" refers to receiving information about the patient's reactions and responses and using this information to reassess their condition.

[0016] "Cognitive stimulation therapy" refers to a rehabilitation method that utilizes the five senses to stimulate the brain and activate cognitive functions.

[0017] "Reminiscence therapy" refers to a rehabilitation technique that aims to activate the brain by eliciting past memories.

[0018] "Reality orientation" is a therapy that helps patients correctly recognize real time and place, and refers to a rehabilitation method that helps patients maintain a sense of reality. [Brief explanation of the drawings]

[0019] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6]FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0020] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0021] First, the terms used in the following description will be explained.

[0022] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0023] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0024] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0025] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0026] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0027] [First embodiment]

[0028] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0029] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0030] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0031] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0032] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0033] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0034] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0035] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0036] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0037] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0038] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0039] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0040] MODE FOR CARRYING OUT THE INVENTION

[0041] This invention is a system for slowing the progression of dementia patients, linking generative AI with a smart speaker to provide optimal rehabilitation methods for each patient. This system analyzes the patient's condition from their voice and adjusts the rehabilitation content in real time to enhance effectiveness.

[0042] System configuration

[0043] 1. Terminal

[0044] The device is a smart speaker that receives voice input from the patient, performs speech recognition at an early stage, and converts the received voice input into text data. The device also has the function of transmitting the received data to a server.

[0045] 2. Server

[0046] The server receives the voice data and converts it into text using advanced voice recognition technology. The server analyzes the tone and strength of the voice data to assess the patient's emotional and cognitive state. Based on the analysis results, the server selects the most appropriate rehabilitation method and transmits the details to the device.

[0047] 3. Rehabilitation methods

[0048] The rehabilitation techniques include cognitive stimulation therapy, reminiscence therapy, and reality orientation. These techniques are automatically selected by the server depending on the patient's condition.

[0049] 4. Feedback

[0050] The terminal presents rehabilitation content to the patient and receives the patient's response again. The response is again sent to the server and used to determine the next rehabilitation content.

[0051] Program processing

[0052] Device behavior

[0053] The device receives the patient's voice input and sends it to the server. For example, the device may ask a simple question while greeting the patient in the morning. For example, the device may ask, "Good morning, how are you feeling today?" to collect the patient's voice.

[0054] Server Operation

[0055] The server analyzes the received voice data and evaluates the patient's emotional and cognitive state. For example, if the patient responds, "I'm a little tired today," the server analyzes the tone and strength of the voice and evaluates it as "feeling tired." Based on the results, it selects a rehabilitation method and sends it to the device.

[0056] Selection of rehabilitation methods

[0057] For example, if the emotional state is judged to be "calm," reminiscence techniques are used to select questions such as "Tell me about the town where you used to live." Conversely, if the emotional state is judged to be "anxious," reassuring reality orientation techniques are used to select questions such as "Do you know what day it is today?"

[0058] Present and get feedback from the device

[0059] The device then verbally communicates the selected rehabilitation content to the patient. If the patient responds, "The town where I used to live was quiet and a very nice place," the voice message is sent back to the server for analysis. This determines the next rehabilitation content.

[0060] Specific examples

[0061] 1. Basic Scenario

[0062] In the morning, the user (a dementia patient) responds to a voice message from the device about their mood that day. The device then sends the voice message to the server, which analyzes it. For example, if the user answers "I didn't have any particular dreams" in response to the question "Good morning. What kind of dreams did you have today?", the server will analyze the response and evaluate the user's emotional state as "Calm." Based on the results, the server will select a reminiscence method and send the next question to the device: "What is your fondest memory from childhood?"

[0063] 2. Utilizing a variety of rehabilitation methods

[0064] If the patient's emotional state is evaluated as "anxious," the server selects reality orientation. The terminal asks, "What are your plans for today?" and the patient answers, "I have an appointment to meet my family today." The server evaluates the patient's sense of security.

[0065] In this way, the system of the present invention can analyze the patient's voice data and provide optimal rehabilitation techniques in real time, thereby effectively stimulating the patient's cognitive function and slowing the progression of dementia.

[0066] The processing flow will be explained below.

[0067] Step 1:

[0068] The terminal receives the patient's voice input, asks the patient questions, and records their responses as audio.

[0069] For example, ask, "Good morning, how are you feeling today?"

[0070] Step 2:

[0071] The device sends the received voice data to the server, where it is converted into text using early-stage voice recognition technology.

[0072] Audio data: "I'm a little tired today."

[0073] Text data: "I'm a little tired today."

[0074] Step 3:

[0075] The server receives the transmitted voice and text data and converts it back into text using advanced voice recognition technology to improve accuracy.

[0076] Audio data: "I'm a little tired today."

[0077] Text data: "I'm a little tired today" (Improved accuracy)

[0078] Step 4:

[0079] The server analyzes the audio and text data to assess the patient's emotional and cognitive state. It identifies the patient's emotional state based on the tone, stress, inflection, and content of the audio.

[0080] Analysis result: "Emotional state: Fatigue"

[0081] Step 5:

[0082] The server selects the optimal rehabilitation method based on the analysis results. For example, if the emotional state is evaluated as "fatigue," it selects a calm and relaxing reminiscence method.

[0083] Selection result: “Reminiscence method”

[0084] Step 6:

[0085] The server generates specific rehabilitation content based on the selected rehabilitation method, such as "Tell us about your favorite hobbies from the past" for reminiscence therapy.

[0086] Generated rehabilitation content: "Tell me about your favorite past hobbies."

[0087] Step 7:

[0088] The server sends the generated rehabilitation content to the terminal. The rehabilitation content is sent as text data, but also includes information for voice synthesis.

[0089] Submitted data: "Tell me about your favorite past hobbies."

[0090] Step 8:

[0091] The device uses voice synthesis technology to ask the patient about the rehabilitation information it has received, and outputs the questions as voice.

[0092] Voice output: "Tell me about your favorite past hobbies."

[0093] Step 9:

[0094] The user (patient) responds to the questions posed by the device. For example, they might reply, "I used to love reading."

[0095] Response: "I used to love reading."

[0096] Step 10:

[0097] The terminal again receives the patient's response as voice data and sends it to the server, where it is also converted into text data.

[0098] Audio data: "I used to love reading."

[0099] Text data: "I used to love reading."

[0100] Step 11:

[0101] The server analyzes the received feedback data and determines the next rehabilitation technique. A loop process is performed to continuously evaluate the patient's condition.

[0102] Analysis results: "Emotional state: calm, rehabilitation method: continued reminiscence therapy"

[0103] This series of steps is repeated to provide optimal rehabilitation for the patient, with the aim of slowing the progression of dementia.

[0104] Example 1

[0105] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0106] Currently, rehabilitation methods for dementia patients are often based on uniform programs, making it difficult to provide individualized training content that takes into account each patient's emotional and cognitive states in real time. Furthermore, there are limited methods for assessing a patient's emotional and cognitive states, making it difficult to select effective rehabilitation methods. As a result, there is a problem in that the progression of dementia cannot be effectively slowed.

[0107] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0108] In this invention, the server includes a means for analyzing the tone and strength of the patient's voice data to identify emotions, a means for grasping the patient's emotional and cognitive states, and a means for automatically selecting one of cognitive stimulation therapy, reminiscence therapy, and reality-focused therapy as a retraining method, thereby enabling the provision of an individual rehabilitation method based on the patient's emotional and cognitive states.

[0109] "Patient" refers to a person who has a progressive disease, such as dementia, and requires rehabilitation.

[0110] "Voice input" refers to the voice data uttered by the patient, which provides basic information for analyzing their emotions and cognitive state.

[0111] A "terminal" is a device that has the function of receiving voice input from a patient and transmitting it to an information processing device. Specifically, this applies to smart speakers.

[0112] "Voice data" refers to digitally recorded speech made by a patient and is used to analyze their emotional and cognitive states.

[0113] "Information processing device" refers to a device that processes received voice data using advanced analytical technology to evaluate a patient's emotional and cognitive state. This includes servers and cloud computing services.

[0114] "Emotional state" refers to the patient's mental state, and is information analyzed from the tone and strength of the voice data.

[0115] "Cognitive status" refers to the patient's cognitive functions, such as intellectual processing ability and memory, and is information evaluated from the content of the voice data and responses.

[0116] "Retraining techniques" refer to the most appropriate rehabilitation methods based on the patient's emotional and cognitive state. These include cognitive stimulation therapy, reminiscence therapy, and reality-oriented therapy.

[0117] "Cognitive stimulation therapy" is a type of training method for maintaining and improving cognitive function in the brain. Specifically, it is carried out using puzzles and calculation problems.

[0118] Reminiscence therapy is a rehabilitation method that stimulates memory and maintains cognitive function by discussing past events.

[0119] "Reality-oriented therapy" is a rehabilitation method that promotes awareness of reality and strengthens cognitive function by having patients talk about real times, places, and everyday events.

[0120] "Feedback" refers to the responses and reactions received by the terminal from the patient, and is data that the information processing device uses to perform further analysis based on that.

[0121] MODE FOR CARRYING OUT THE INVENTION

[0122] This invention is a system for slowing the progression of dementia in patients, linking a generative AI with a voice input device to provide optimal retraining methods for each patient. This system analyzes the emotional and cognitive state of the patient from their voice and adjusts the retraining content in real time to enhance effectiveness.

[0123] System configuration

[0124] 1. Terminal

[0125] The terminal is a voice input device for receiving voice input from the patient. The voice input device receives the patient's voice and converts it into text data through speech recognition at an early stage. Examples of this include smart speakers such as Amazon Alexa and Google Home. The terminal also has the function of transmitting the received data to an information processing device.

[0126] 2. Server (information processing device)

[0127] The server receives the voice data and converts it into text using more advanced speech recognition technology. The server analyzes the tone and intensity of the voice data to assess the patient's emotional and cognitive state. The analysis uses advanced voice analysis tools such as Google Cloud Speech-to-Text and IBM Watson. Based on the analysis results, the server selects the optimal retraining method and sends it to the device.

[0128] 3. Retraining method

[0129] Retraining techniques include cognitive stimulation therapy, reminiscence therapy, and reality-oriented therapy. These methods are automatically selected by the server depending on the patient's condition. Specific questions and instructions are generated using a generative AI model (e.g., GPT-3) built into the server.

[0130] 4. Feedback

[0131] The terminal presents the retraining content to the patient by voice and receives the patient's response again, which is then sent back to the server and used to determine the next retraining content.

[0132] Specific actions

[0133] Device behavior

[0134] The device receives the patient's voice input and sends it to the server. For example, the device may ask a simple question while greeting the patient in the morning. For example, the device may ask, "Good morning, how are you feeling today?" to collect the patient's voice.

[0135] Server Operation

[0136] The server performs advanced analysis of the received voice data to evaluate the patient's emotional and cognitive state. For example, if the patient responds, "I'm a little tired today," the server analyzes the tone and strength of the voice and evaluates the patient's emotional state as "fatigue." Based on the results, it selects a retraining method and sends it to the device.

[0137] Selection of retraining methods

[0138] For example, if the emotional state is determined to be "calm," reminiscence therapy is used to select questions such as "Tell me about the town where you used to live." Conversely, if the emotional state is determined to be "anxious," reality-oriented therapy, which provides reassurance, is used to select questions such as "Do you know what day it is today?" The generative AI model is used to input the following prompt sentences and generate appropriate questions.

[0139] Example prompt sentence:

[0140] "Create questions about how you're feeling today and your past memories to suggest rehabilitation techniques for dementia patients."

[0141] Present and get feedback from the device

[0142] The device then verbally communicates the selected retraining content to the patient. For example, if the device asks, "Tell me about the town where you used to live," and the patient replies, "The town where I used to live was quiet and a very nice place," the device sends the voice back to the server for analysis. This determines the next retraining content.

[0143] In this way, the system of the present invention can analyze the patient's voice data and provide optimal retraining techniques in real time, which can effectively stimulate the patient's cognitive function and slow the progression of dementia.

[0144] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0145] Program processing flow

[0146] Step 1:

[0147] A user speaks into a voice input device (terminal). The voice input device receives the user's voice in real time. The input is the user's voice data, for example, a voice replying "I'm a little tired today" to a question from the terminal such as "Good morning. How are you feeling today?" The output is raw voice data acquired in real time.

[0148] Step 2:

[0149] The device converts the user's voice data into text data using early speech recognition technology. A speech recognition engine is used to convert the input voice data into text data. For example, a speech saying "I'm a little tired today" is converted into text data saying "I'm a little tired today." The output is that text data.

[0150] Step 3:

[0151] The terminal sends text data to the server. The input is the converted text data, which is sent to the server. For example, the text data "I'm a little tired today" is sent to the server via the network. The output is the text data sent to the server.

[0152] Step 4:

[0153] The server performs a detailed analysis of the received text data using advanced analysis techniques. The server combines natural language processing (NLP) techniques and voice tone analysis to evaluate the emotional and cognitive states of the input text data. For example, "Today" indicates "fatigue" as the emotional state. The output is the analysis results: "Emotional state: fatigue" and "Cognitive state."

[0154] Step 5:

[0155] The server selects the optimal retraining method based on the analysis results. The input is data on "emotional state: fatigue" and "cognitive state." For example, in the case of "fatigue," the server selects reality-oriented therapy to promote refreshment. A generative AI model (such as GPT-3) is used to generate specific questions and instructions. For example, it generates the question, "Do you know what day it is today?" The output is the selected retraining method and the generated question.

[0156] Step 6:

[0157] The server sends the selected retraining method and the generated question to the terminal. The input is the retraining method selection result and the generated question, which are sent to the terminal via the Internet. For example, a question such as "Do you know what day it is today?" is sent to the terminal. The output is the question sent to the terminal.

[0158] Step 7:

[0159] The terminal presents the received question to the user by voice. The input is question data received from the server, and the text "Do you know what day it is today?" is converted into voice and presented to the user. The voice output device is used to pose the question to the user. The output is a voice presentation to the user.

[0160] Step 8:

[0161] The user responds to a question from the terminal. The input is a voice question from the terminal, and the user responds to it by voice. For example, the user might reply, "Today is Monday." The output is the user's response voice data.

[0162] Step 9:

[0163] The terminal converts the user's response voice data into text data again and sends it to the server. The voice recognition engine is used again to convert the input voice into text data. For example, the voice saying "Today is Monday" is converted into text data. This is sent to the server. The output is the text data sent to the server.

[0164] Step 10:

[0165] The server analyzes the user's response again and adjusts the next retraining content. The input is the text data "Today is Monday," which the server analyzes to evaluate the user's emotional and cognitive state. Based on the analysis results, a new question or instruction is generated and sent to the device again. For example, a new question is generated: "Do you have any plans with your family today?" The output is the next retraining content.

[0166] Through this process, the system of the present invention analyzes the patient's voice data and provides an optimal retraining method in real time, thereby effectively stimulating the patient's cognitive function and slowing the progression of dementia.

[0167] (Application example 1)

[0168] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0169] It is important to provide appropriate rehabilitation methods for dementia patients in elderly rehabilitation facilities and nursing homes. However, conventional technologies are insufficient in providing optimal rehabilitation methods according to the emotional and cognitive states that vary from patient to patient. In particular, it is difficult to assess a patient's condition in real time and continue to provide appropriate rehabilitation methods. Therefore, there is a need for a system that provides optimal rehabilitation methods for individual patients in real time.

[0170] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0171] In this invention, the server includes: means for receiving voice input from the patient; means for transmitting the received voice data to the server; means for the server to analyze the voice data and grasp the emotional and cognitive states of the patient; means for selecting an optimal rehabilitation method based on the grasped states; means for transmitting the selected rehabilitation method to the terminal; means for the terminal to present rehabilitation content to the patient and obtain feedback; means for converting the patient's voice data into text data and evaluating the emotional state using an emotion analysis model; and means for generating specific prompt sentences based on the evaluation results to provide the rehabilitation method. This makes it possible to evaluate the emotional and cognitive states that differ for each patient in real time and provide the optimal rehabilitation method based on the evaluation.

[0172] A "means for receiving patient voice input" is a device or method for collecting patient-produced speech in real time.

[0173] The "means for transmitting received voice data to a server" refers to a device or method that has the function of transferring voice data to a server via a network.

[0174] "Means for the server to analyze voice data and understand the emotional and cognitive state of the patient" refers to technology that enables the server to process voice data and understand the emotional and cognitive state of the patient.

[0175] "Means for selecting the optimal rehabilitation method based on the grasped condition" refers to a system or algorithm that determines the most appropriate rehabilitation method based on the analysis results.

[0176] The "means for transmitting the selected rehabilitation method to the terminal" refers to a device or method for transmitting the selected rehabilitation method to the client device via a network.

[0177] "Means for a terminal to present rehabilitation content to a patient and obtain feedback" refers to equipment or methods for providing rehabilitation information to a patient and collecting their reactions again.

[0178] "Means for converting a patient's voice data into text data and assessing their emotional state using an emotion analysis model" refers to a technology for converting voice data into text and analyzing the text to assess the user's emotions.

[0179] "Means for generating specific prompt sentences based on the evaluation results and providing rehabilitation techniques" is a mechanism for creating specific questions and instructions based on the results of emotion analysis and using them to advance rehabilitation.

[0180] The following describes an embodiment of the present invention.

[0181] First, the entire system consists of a voice input device (such as a smartphone) held by the patient, a server, and a user terminal for providing feedback.

[0182] Audio input device

[0183] The voice input device is a smartphone, which can collect the patient's speech in real time and convert it into text data. The technologies used include the Python SpeechRecognition library and the Google Speech API.

[0184] server

[0185] The server analyzes the received voice data to understand the patient's emotional and cognitive state. A generative AI model is used for the analysis, and emotion analysis is performed using the Hugging Face Transformers model. Specifically, the voice data is converted into text data, which is then input into an emotion analysis model to evaluate the patient's emotional state. Based on the results of this evaluation, an appropriate rehabilitation method is selected. An appropriate prompt is also generated, and the selected rehabilitation method is sent to the user's device.

[0186] User terminal

[0187] The user device receives the rehabilitation method from the server and presents it to the patient via voice. The user device then collects the feedback again as voice and sends the data to the server. This provides the data needed to determine the next rehabilitation method.

[0188] Program processing natural language explanation

[0189] The server analyzes the patient's voice to understand their emotional and cognitive state and decides on a rehabilitation method based on that. First, the smartphone collects the voice and converts it into text using the Google Speech API. The text data is then sent to the server, where emotion analysis is performed using the Hugging Face Transformers model. Based on the analysis results, a specific rehabilitation method for the target patient is determined and an appropriate prompt is generated. This prompt is then presented to the patient via the user's device.

[0190] Specific examples

[0191] As a specific example, if a patient answers, "I feel a little tired today," the emotion analysis model will determine this as "NEGATIVE," and the server will generate a prompt such as "Is there anything bothering you?" and send it to the user's terminal.

[0192] Prompt Sentence Examples

[0193] "Good morning. How are you feeling today?"

[0194] "Tell me about the town where you used to live."

[0195] "Do you know what day it is?"

[0196] As a result, this system can provide optimal rehabilitation methods in real time based on each patient's emotional and cognitive state.

[0197] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0198] Step 1:

[0199] The user provides voice input. In this case, the patient's voice data is collected using the smartphone's microphone. For example, the patient might say, "I feel a little tired today." This voice data becomes the input.

[0200] Step 2:

[0201] The device converts the input voice into text data. Here, the Google Speech API is used to convert the voice into text. This text data is output and goes to the next analysis step.

[0202] Step 3:

[0203] The terminal sends the converted text data to the server. The text data is transferred to the server via the network. The server receives the text data.

[0204] Step 4:

[0205] The server analyzes the received text data. Specifically, it performs emotion analysis using the Hugging Face Transformers model. This analysis evaluates the patient's emotional state from the input text data. For example, the text "I'm a little tired" is evaluated as "NEGATIVE."

[0206] Step 5:

[0207] The server selects the optimal rehabilitation method based on the results of the emotion analysis and generates a specific prompt. Here, the rehabilitation method is determined based on the results of the emotion analysis model, and an appropriate prompt is generated. A prompt such as "Is there anything that is bothering you?" is generated.

[0208] Step 6:

[0209] The server sends the generated prompt to the terminal. The prompt is sent to the terminal via the network. The sent prompt is received by the terminal.

[0210] Step 7:

[0211] The terminal then presents the received prompt to the user by voice. In this case, the smartphone's text-to-speech function is used to play the prompt aloud. For example, the patient might hear, "Is there anything bothering you?"

[0212] Step 8:

[0213] The user provides vocal feedback on the presented prompt sentence, which is then input into the terminal again, leading to further analysis and the provision of rehabilitation techniques.

[0214] Through these steps, the system can analyze the patient's emotional state in real time and provide optimal rehabilitation methods. It also generates prompts and collects feedback, enabling continuous rehabilitation support.

[0215] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0216] MODE FOR CARRYING OUT THE INVENTION

[0217] This invention is a system that combines generative AI, a smart speaker, and an emotion engine to provide a rehabilitation method to slow the progression of dementia. This system analyzes the patient's condition from their voice and adjusts the rehabilitation content in real time to improve effectiveness.

[0218] System configuration

[0219] 1. Terminal

[0220] The device is a smart speaker that receives voice input from the patient, performs speech recognition at an early stage, and converts the received voice input into text data. The device also has the function of transmitting the received data to a server.

[0221] 2. Server

[0222] The server receives the voice data and converts it into text using advanced voice recognition technology. The server analyzes the tone and strength of the voice data to assess the patient's emotional and cognitive state. Based on the analysis results, it selects the most appropriate rehabilitation method and sends it to the device.

[0223] 3. Emotion Engine

[0224] The emotion engine analyzes the emotional state of the voice data in real time and feeds back the results to the server. The emotion engine has the function of analyzing the emotional state by combining tone, stress, intonation, and other non-verbal elements.

[0225] 4. Rehabilitation methods

[0226] The rehabilitation techniques include cognitive stimulation therapy, reminiscence therapy, and reality orientation. These techniques are automatically selected by the server depending on the patient's condition.

[0227] 5. Feedback

[0228] The terminal presents rehabilitation content to the patient and receives the patient's response again. The response is again sent to the server and used to determine the next rehabilitation content.

[0229] Program processing

[0230] Device behavior

[0231] The device receives the patient's voice input and sends it to the server. For example, the device may ask a simple question while greeting the patient in the morning. For example, the device may ask, "Good morning, how are you feeling today?" to collect the patient's voice.

[0232] Server Operation

[0233] The server analyzes the received voice data and evaluates the patient's emotional and cognitive state. For example, if the patient responds, "I'm a little tired today," the server analyzes the tone and strength of the voice and evaluates it as "feeling tired." Based on the results, it selects a rehabilitation method and sends it to the device.

[0234] Emotion Engine Operation

[0235] The emotion engine analyzes the patient's emotions in real time based on the voice data. For example, if "anxiety" is detected from the tone and intonation, the result is fed back to the server, which then adjusts the rehabilitation program based on that information.

[0236] Selection of rehabilitation methods

[0237] For example, if the emotional state is judged to be "calm," reminiscence techniques are used to select questions such as "Tell me about the town where you used to live." Conversely, if the emotional state is judged to be "anxious," reassuring reality orientation techniques are used to select questions such as "Do you know what day it is today?"

[0238] Present and get feedback from the device

[0239] The device then verbally communicates the selected rehabilitation content to the patient. If the patient responds, "The town where I used to live was quiet and a very nice place," the voice message is sent back to the server for analysis. This determines the next rehabilitation content.

[0240] Specific examples

[0241] 1. Basic Scenario

[0242] In the morning, the user (a dementia patient) responds to a voice message from the device about their mood that day. The device then sends the voice message to the server, which analyzes it. For example, if the user answers "I didn't have any particular dreams" in response to the question "Good morning. What kind of dreams did you have today?", the server will analyze the response and evaluate the user's emotional state as "Calm." Based on the results, the server will select a reminiscence method and send the next question to the device: "What is your fondest memory from childhood?"

[0243] 2. Utilizing the Emotion Engine

[0244] An emotion engine has been added to more accurately grasp the patient's emotional state. For example, if a patient responds, "I'm a little tired today," the emotion engine analyzes the tone and intonation of the response and evaluates it as "fatigue." This result is fed back to the server, which then adjusts the rehabilitation program based on that information.

[0245] 3. Utilizing a variety of rehabilitation methods

[0246] If the patient's emotional state is evaluated as "anxious," the server selects reality orientation. The terminal asks, "What are your plans for today?" and the patient answers, "I have an appointment to meet my family today." The server evaluates the patient's sense of security.

[0247] In this way, the system of the present invention can analyze the patient's voice data and provide optimal rehabilitation techniques in real time, which can effectively stimulate the patient's cognitive function and slow the progression of dementia. The addition of an emotion engine further improves the accuracy and effectiveness of rehabilitation content.

[0248] The processing flow will be explained below.

[0249] Step 1:

[0250] The terminal receives the patient's voice input, asks the patient questions, and records their responses as audio.

[0251] For example, ask, "Good morning, how are you feeling today?"

[0252] Step 2:

[0253] The device sends the received voice data to the server, where it is converted into text using early-stage voice recognition technology.

[0254] Audio data: "I'm a little tired today."

[0255] Text data: "I'm a little tired today."

[0256] Step 3:

[0257] The server receives the transmitted voice and text data and converts it back into text using advanced voice recognition technology to improve accuracy.

[0258] Audio data: "I'm a little tired today."

[0259] Text data: "I'm a little tired today" (Improved accuracy)

[0260] Step 4:

[0261] The server analyzes the audio and text data and uses an emotion engine to assess the patient's emotional and cognitive state. It identifies the patient's emotional state based on the tone, stress, inflection, and content of the audio.

[0262] Analysis result: "Emotional state: Fatigue"

[0263] Step 5:

[0264] The server selects the optimal rehabilitation method based on the analysis results. For example, if the emotional state is evaluated as "fatigue," it selects a calm and relaxing reminiscence method.

[0265] Selection result: “Reminiscence method”

[0266] Step 6:

[0267] The server generates specific rehabilitation content based on the selected rehabilitation method, such as "Tell us about your favorite hobbies from the past" for reminiscence therapy.

[0268] Generated rehabilitation content: "Tell me about your favorite past hobbies."

[0269] Step 7:

[0270] The server sends the generated rehabilitation content to the terminal. The rehabilitation content is sent as text data, but also includes information for voice synthesis.

[0271] Submitted data: "Tell me about your favorite past hobbies."

[0272] Step 8:

[0273] The device uses voice synthesis technology to ask the patient about the rehabilitation information it has received, and outputs the questions as voice.

[0274] Voice output: "Tell me about your favorite past hobbies."

[0275] Step 9:

[0276] The user (patient) responds to the questions posed by the device. For example, they might reply, "I used to love reading."

[0277] Response: "I used to love reading."

[0278] Step 10:

[0279] The terminal again receives the patient's response as voice data and sends it to the server, where it is also converted into text data.

[0280] Audio data: "I used to love reading."

[0281] Text data: "I used to love reading."

[0282] Step 11:

[0283] The server analyzes the received feedback data and determines the next rehabilitation method. The emotion engine runs again and evaluates the patient's condition. A loop process is performed to continuously evaluate the patient's condition.

[0284] Analysis results: "Emotional state: calm, rehabilitation method: continued reminiscence therapy"

[0285] This series of steps is repeated to provide optimal rehabilitation for the patient, with the aim of slowing the progression of dementia. The addition of the emotion engine further improves the accuracy and effectiveness of the rehabilitation content.

[0286] Example 2

[0287] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0288] Currently, rehabilitation systems on the market lack the ability to properly grasp a patient's emotional and cognitive state. As a result, the rehabilitation methods provided are often not optimal for the patient. Furthermore, the lack of a function to obtain real-time feedback and adjust the rehabilitation content limits the effectiveness of rehabilitation. There is a need to solve these problems and provide the most effective rehabilitation methods for patients.

[0289] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for analyzing voice data and grasping the emotional and cognitive states of the patient, a means for an emotion engine to analyze the voice data in real time and provide feedback to the server, and a means for selecting an optimal rehabilitation method. This makes it possible to provide an optimal rehabilitation method in real time based on the patient's condition and maximize its effectiveness.

[0290] "Patient" refers to an individual who is a target of using the rehabilitation system.

[0291] "Voice input" refers to the words or voice data uttered by the patient.

[0292] "Terminal" refers to an electronic device that receives the patient's voice input and performs the necessary processing.

[0293] "Server" refers to a central computer system that analyzes data sent from terminals and performs necessary processing and feedback.

[0294] "Voice data" refers to a digital representation of a patient's voice input.

[0295] "Emotion engine" refers to a software or hardware component for analyzing audio data and assessing emotional state.

[0296] "Emotional state" refers to the psychological state of the patient as judged from the tone and strength of their voice.

[0297] "Cognitive status" refers to the state of a patient's cognitive function.

[0298] "Rehabilitation techniques" refer to therapies provided based on the patient's emotional and cognitive state.

[0299] "Real-time" refers to processing that occurs at a speed close to instantaneous.

[0300] This invention is a system that combines a generative AI model, a smart speaker, and an emotion engine to provide a rehabilitation method to slow the progression of dementia. This system analyzes the patient's condition from their voice and adjusts the rehabilitation content in real time to improve effectiveness.

[0301] System configuration

[0302] 1. Terminal

[0303] The device consists of a smart speaker that receives the patient's voice input. Specifically, the smart speaker picks up the patient's voice using a microphone, performs initial voice recognition, and converts it into text data. The device also has the function of sending this data to a server. For example, the device may ask the patient, "Good morning, how are you feeling today?"

[0304] 2. Server

[0305] The server receives the voice data sent from the device and converts it into text data using more advanced voice recognition technology. One possible technology is the Google Cloud Speech-to-Text API. The server then analyzes the tone and strength of the voice data to evaluate the patient's emotional and cognitive state. For example, the server evaluates the voice data, such as "I'm a little tired today," as "I feel fatigued." Based on the analysis results, the server selects the optimal rehabilitation method and sends it to the device.

[0306] 3. Emotion Engine

[0307] The emotion engine analyzes the emotional state of the voice data in real time and feeds the results back to the server. Specific analysis items include tone, stress, intonation, and other non-verbal elements. For example, if "anxiety" is detected from the intonation of a patient's voice, the result is fed back to the server.

[0308] 4. Rehabilitation methods

[0309] Rehabilitation techniques include cognitive stimulation therapy, reminiscence therapy, and reality orientation. These techniques are automatically selected by the server depending on the patient's condition. For example, if the emotional state is judged to be "calm," reminiscence therapy will be used and questions such as "Tell me about the town where you used to live" will be selected.

[0310] 5. Feedback

[0311] The device presents rehabilitation content to the patient and receives the patient's response again. The received data is sent to the server and used to determine the next rehabilitation content. For example, if the patient responds, "The town where I used to live was quiet and a very nice place," the voice data is sent to the server and analyzed.

[0312] Specific examples

[0313] Basic Scenario

[0314] In the morning, the user (a dementia patient) answers questions posed by the device. For example, to the question, "Good morning. What kind of dreams did you have today?", the user replies, "I didn't have any particular dreams." The server analyzes the voice data and evaluates the user's emotional state as "calm." The server selects the reminiscence method and sends the next question, "What is your best childhood memory?" to the device. The device presents the question to the patient, who responds. The server then analyzes the response again to determine the next rehabilitation plan.

[0315] Utilizing the Emotion Engine

[0316] The emotion engine analyzes the tone and intonation of the patient's voice. For example, if a patient replies, "I'm a little tired today," the emotion engine analyzes the tone and intonation and evaluates it as "feeling tired." This evaluation result is fed back to the server, which then adjusts the rehabilitation program based on that information.

[0317] Utilizing a variety of rehabilitation methods

[0318] If the patient's emotional state is evaluated as "anxious," the server selects reality orientation. The device asks, "What are your plans for today?" and the patient replies, "I have an appointment to meet my family today." The server analyzes this information and evaluates the patient's sense of security. This determines the next rehabilitation content.

[0319] Prompt Sentence Examples

[0320] "To provide rehabilitation techniques for dementia patients, suggest optimal questions based on their emotional and cognitive states. For example, if a patient says, 'I'm a little tired today,' analyze their speech data and generate the appropriate next question."

[0321] In this way, the system of the present invention can analyze the patient's voice data and provide optimal rehabilitation methods in real time, which can effectively stimulate the patient's cognitive function and slow the progression of dementia. The addition of an emotion engine further improves the accuracy and effectiveness of rehabilitation content.

[0322] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0323] Step 1:

[0324] The device receives voice input. The user speaks to the smart speaker. For example, the device asks, "Good morning, how are you feeling today?" and the user replies, "I'm a little tired today." This voice data is input into the device.

[0325] Input: User's voice data: "I'm a little tired today."

[0326] Output: Received audio data

[0327] Step 2:

[0328] The voice data received by the device is converted into text data using an initial voice recognition engine. For example, the voice recognition technology built into a smart speaker can be used to convert the voice into text, such as "I'm a little tired today." This text data is then sent to the server.

[0329] Input: Received audio data

[0330] Output: Text data "I'm a little tired today"

[0331] Step 3:

[0332] The server uses advanced speech recognition technology to convert the voice data sent from the device back into text data, and then uses advanced technology such as the Google Cloud Speech-to-Text API to analyze the details of the voice and generate text data.

[0333] Input: Audio data sent from the device

[0334] Output: Highly refined text data "I'm a little tired today"

[0335] Step 4:

[0336] The server performs detailed analysis of the voice data and evaluates the emotional state using an emotion engine. The server analyzes the tone, strength, and intonation of the voice to identify the emotional state. For example, it may evaluate the voice as "fatigue" based on its low tone and strength.

[0337] Input: Highly refined text data and its speech characteristics

[0338] Output: Emotional state "Feeling tired"

[0339] Step 5:

[0340] The emotion engine feeds back the analyzed emotional state to the server. The emotion engine determines "fatigue" based on the tone and intonation of the voice data and sends the evaluation result to the server.

[0341] Input: Tone and intonation analysis data of speech data

[0342] Output: Emotional state assessment result: "Feeling tired"

[0343] Step 6:

[0344] The server selects the optimal rehabilitation method based on the emotional and cognitive states. Based on the result "feeling tired," the server selects cognitive stimulation therapy as the rehabilitation method and determines its specific content. For example, it selects the question, "Do you know what day it is today?"

[0345] Input: Emotional state assessment result: "Feeling tired"

[0346] Output: The selected rehabilitation technique "Cognitive Stimulation Therapy" and the question "Do you know what day it is today?"

[0347] Step 7:

[0348] The server sends the selected rehabilitation method and its details to the device. As part of the cognitive stimulation therapy, the server sends the question "Do you know what day it is today?" to the device.

[0349] Input: Selected rehabilitation method and specific question

[0350] Output: Instruction data to the terminal

[0351] Step 8:

[0352] The device then presents the received rehabilitation information to the user via voice. For example, the device might ask the user, "Do you know what day it is today?" The user might respond, "Today is Tuesday."

[0353] Input: Rehabilitation details sent from the server

[0354] Output: Audio presentation to the user

[0355] Step 9:

[0356] The user responds to the rehabilitation content. When the user responds, "Today is Tuesday," the terminal receives the voice again. This received data is sent to the server again.

[0357] Input: User's voice reply "Today is Tuesday"

[0358] Output: Response data to the terminal

[0359] Step 10:

[0360] The server analyzes the voice data again and determines the next rehabilitation content based on the results. For example, based on the response "Today is Tuesday," the server selects the next rehabilitation method and sends the next question, "What did you do on Tuesday?" to the terminal.

[0361] Input: User response data received again

[0362] Output: Next rehabilitation technique and specific questions

[0363] This series of processes enables feedback tailored to the user's emotional and cognitive state, providing optimal rehabilitation techniques. This system is expected to slow the progression of dementia in patients.

[0364] (Application example 2)

[0365] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0366] Rehabilitation for dementia patients requires selecting the optimal method to suit each individual patient's condition and providing effective treatment. However, with conventional methods, it has been difficult to accurately grasp the patient's emotional and cognitive state in real time and provide the optimal rehabilitation method based on that. In particular, more advanced technology is required to realize rehabilitation in a virtual environment. The purpose of this project is to solve these issues.

[0367] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0368] In this invention, the server includes means for receiving voice input from the patient, means for transmitting the received voice data to the server, means for analyzing the voice data to grasp the emotional and cognitive states of the patient, means for selecting an optimal rehabilitation method based on the grasped states, means for transmitting the selected rehabilitation method to the terminal, means for the terminal to present rehabilitation content to the patient and obtain feedback, and means for presenting rehabilitation content in a virtual environment. This makes it possible to provide the rehabilitation method optimal for the patient's condition in real time, thereby effectively slowing the progression of dementia.

[0369] Key Word Definitions

[0370] "Means for receiving patient voice input" refers to a device for collecting voice data from a dementia patient, often a microphone or smart speaker.

[0371] "Means for transmitting received voice data to a server" refers to a function for transferring collected voice data to a central computer (server) via a network.

[0372] "Means for analyzing voice data to understand the emotional and cognitive state of a patient" refers to algorithms or analytical software for analyzing received voice data and determining the emotional and cognitive state of a patient.

[0373] "Means for selecting the most appropriate rehabilitation method based on the identified condition" refers to the logic and process for automatically selecting the rehabilitation method that is most appropriate for the patient's current condition based on the results of the analysis.

[0374] "Means for transmitting the selected rehabilitation method to the terminal" refers to a communication means for transmitting information about the rehabilitation method selected by the server to the patient's terminal.

[0375] "Means for the device to present rehabilitation content to the patient and obtain feedback" refers to an interface that presents selected rehabilitation content through audio or a virtual environment and collects the patient's reactions and answers again.

[0376] "Means for presenting rehabilitation content in a virtual environment" refers to software or hardware that uses virtual reality or augmented reality technology to provide patients with visual or auditory rehabilitation content.

[0377] MODE FOR CARRYING OUT THE INVENTION

[0378] To illustrate an embodiment of the present invention, the following system configuration and operation will be described.

[0379] System Configuration

[0380] This system consists of the following main hardware and software:

[0381] Hardware

[0382] Smart glasses or head-mounted displays (HMD)

[0383] microphone

[0384] Cloud Server

[0385] software

[0386] Speech recognition library (e.g., speech_recognition library)

[0387] Sentiment analysis engine (e.g., custom-developed emotion_engine module)

[0388] Rehabilitation method selection logic (e.g., custom-developed rehab_methods module)

[0389] Virtual environment interface (e.g., a custom-developed virtual_environment module)

[0390] A natural language explanation of how to proceed

[0391] 1. Receiving and sending audio data

[0392] The user's voice input is received by a microphone attached to the smart glasses or head-mounted display, and the received voice data is transmitted to a cloud server in real time.

[0393] 2. Analysis of audio data

[0394] The server converts the received voice data into text data using a speech recognition library. It then uses an emotion analysis engine to analyze the tone and strength of the text data to understand the user's emotional and cognitive state. Based on the results of this analysis, the server selects the most appropriate rehabilitation method for the user.

[0395] 3. Selection of rehabilitation method

[0396] Based on the analyzed data, the server uses a rehabilitation method selection logic to determine the appropriate rehabilitation content. The rehabilitation method is automatically selected from cognitive stimulation therapy, reminiscence therapy, and reality orientation.

[0397] 4. Presentation of rehabilitation content

[0398] The selected rehabilitation content is sent from the cloud server to smart glasses or a head-mounted display, and the user can receive the rehabilitation content visually or audibly through virtual reality or augmented reality using a virtual environment interface.

[0399] 5. Get feedback

[0400] If the user provides vocal feedback on the rehabilitation content, the audio is again picked up by the microphone and sent to the cloud server, and this process further adjusts the next rehabilitation content.

[0401] Specific examples

[0402] For example, if a user is asked, "What kind of dream did you have today?" and the user answers, "I didn't have any particular dreams," this voice data is sent to the server, where it is analyzed and the emotional state is evaluated as "calm" based on the tone and intensity of the voice. Based on this evaluation, the server selects a reminiscence method and presents the next question, "What is your best childhood memory?" A nostalgic scene is displayed within the virtual environment.

[0403] Prompt Sentence Examples

[0404] Emotional state: Calm

[0405] Q: Tell me about the town where you used to live.

[0406] In this way, the system can analyze the user's voice data and provide optimal rehabilitation techniques in real time, which can effectively stimulate the user's cognitive functions and slow the progression of dementia.

[0407] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0408] Program processing flow

[0409] Step 1:

[0410] The device receives the user's voice input. The voice data is captured by a microphone and sent to a cloud server in real time. The input is the user's voice, and the output is sent to the server as voice data.

[0411] Step 2:

[0412] The server receives the voice data and converts it into text data using a speech recognition library. The input is voice data and the output is text data. The speech recognition library analyzes the sound wave data and converts it into text based on a language model.

[0413] Step 3:

[0414] The server inputs the converted text data into an emotion analysis engine, which analyzes the emotional and cognitive states. The input is text data, and the output is an evaluation of the emotional and cognitive states. The emotion analysis engine takes into account non-verbal elements such as tone, strength, and intonation during analysis.

[0415] Step 4:

[0416] Based on the analysis results, the server uses the rehabilitation method selection logic to select the optimal rehabilitation method. The input is the evaluation results of the emotional and cognitive states, and the output is the selected rehabilitation method. The rehabilitation method selection logic compares the application conditions of each rehabilitation method with the current evaluation results and selects the most appropriate method.

[0417] Step 5:

[0418] The server sends the selected rehabilitation technique to the terminal. The input is the selected rehabilitation technique, and the output is the rehabilitation technique information sent to the terminal. The selected information is delivered to the terminal using a communication protocol.

[0419] Step 6:

[0420] The device presents rehabilitation content to the user through a virtual environment interface. The input is rehabilitation method information, and the output is a virtual rehabilitation environment that the user perceives visually or audibly. Visual content is displayed and played on a head-mounted display, and audio content is displayed and played on speakers.

[0421] Step 7:

[0422] The user responds to the presented rehabilitation content and takes action, and their reactions are collected again as voice data by the device. The input is the user's voice response, and the output is again sent to the server as voice data.

[0423] Step 8:

[0424] The server analyzes the user's feedback and adjusts the next rehabilitation content. The input is the user's voice data and the rehabilitation method results, and the output is the adjusted rehabilitation method. The server reuses the emotion analysis engine and rehabilitation method selection logic to reselect the optimal rehabilitation method that reflects the feedback.

[0425] Through these steps, the system can adapt flexibly to the user's condition and provide effective rehabilitation.

[0426] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0427] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0428] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0429] [Second embodiment]

[0430] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0431] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0432] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0433] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0434] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0435] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0436] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0437] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0438] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0439] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0440] In the smart glasses 214, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0441] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0442] MODE FOR CARRYING OUT THE INVENTION

[0443] This invention is a system for slowing the progression of dementia patients, linking generative AI with a smart speaker to provide optimal rehabilitation methods for each patient. This system analyzes the patient's condition from their voice and adjusts the rehabilitation content in real time to enhance effectiveness.

[0444] System configuration

[0445] 1. Terminal

[0446] The device is a smart speaker that receives voice input from the patient, performs speech recognition at an early stage, and converts the received voice input into text data. The device also has the function of transmitting the received data to a server.

[0447] 2. Server

[0448] The server receives the voice data and converts it into text using advanced voice recognition technology. The server analyzes the tone and strength of the voice data to assess the patient's emotional and cognitive state. Based on the analysis results, the server selects the most appropriate rehabilitation method and transmits the details to the device.

[0449] 3. Rehabilitation methods

[0450] The rehabilitation techniques include cognitive stimulation therapy, reminiscence therapy, and reality orientation. These techniques are automatically selected by the server depending on the patient's condition.

[0451] 4. Feedback

[0452] The terminal presents rehabilitation content to the patient and receives the patient's response again. The response is again sent to the server and used to determine the next rehabilitation content.

[0453] Program processing

[0454] Device behavior

[0455] The device receives the patient's voice input and sends it to the server. For example, the device may ask a simple question while greeting the patient in the morning. For example, the device may ask, "Good morning, how are you feeling today?" to collect the patient's voice.

[0456] Server Operation

[0457] The server analyzes the received voice data and evaluates the patient's emotional and cognitive state. For example, if the patient responds, "I'm a little tired today," the server analyzes the tone and strength of the voice and evaluates it as "feeling tired." Based on the results, it selects a rehabilitation method and sends it to the device.

[0458] Selection of rehabilitation methods

[0459] For example, if the emotional state is judged to be "calm," reminiscence techniques are used to select questions such as "Tell me about the town where you used to live." Conversely, if the emotional state is judged to be "anxious," reassuring reality orientation techniques are used to select questions such as "Do you know what day it is today?"

[0460] Present and get feedback from the device

[0461] The device then verbally communicates the selected rehabilitation content to the patient. If the patient responds, "The town where I used to live was quiet and a very nice place," the voice message is sent back to the server for analysis. This determines the next rehabilitation content.

[0462] Specific examples

[0463] 1. Basic Scenario

[0464] In the morning, the user (a dementia patient) responds to a voice message from the device about their mood that day. The device then sends the voice message to the server, which analyzes it. For example, if the user answers "I didn't have any particular dreams" in response to the question "Good morning. What kind of dreams did you have today?", the server will analyze the response and evaluate the user's emotional state as "Calm." Based on the results, the server will select a reminiscence method and send the next question to the device: "What is your fondest memory from childhood?"

[0465] 2. Utilizing a variety of rehabilitation methods

[0466] If the patient's emotional state is evaluated as "anxious," the server selects reality orientation. The terminal asks, "What are your plans for today?" and the patient answers, "I have an appointment to meet my family today." The server evaluates the patient's sense of security.

[0467] In this way, the system of the present invention can analyze the patient's voice data and provide optimal rehabilitation techniques in real time, thereby effectively stimulating the patient's cognitive function and slowing the progression of dementia.

[0468] The processing flow will be explained below.

[0469] Step 1:

[0470] The terminal receives the patient's voice input, asks the patient questions, and records their responses as audio.

[0471] For example, ask, "Good morning, how are you feeling today?"

[0472] Step 2:

[0473] The device sends the received voice data to the server, where it is converted into text using early-stage voice recognition technology.

[0474] Audio data: "I'm a little tired today."

[0475] Text data: "I'm a little tired today."

[0476] Step 3:

[0477] The server receives the transmitted voice and text data and converts it back into text using advanced voice recognition technology to improve accuracy.

[0478] Audio data: "I'm a little tired today."

[0479] Text data: "I'm a little tired today" (Improved accuracy)

[0480] Step 4:

[0481] The server analyzes the audio and text data to assess the patient's emotional and cognitive state. It identifies the patient's emotional state based on the tone, stress, inflection, and content of the audio.

[0482] Analysis result: "Emotional state: Fatigue"

[0483] Step 5:

[0484] The server selects the optimal rehabilitation method based on the analysis results. For example, if the emotional state is evaluated as "fatigue," it selects a calm and relaxing reminiscence method.

[0485] Selection result: “Reminiscence method”

[0486] Step 6:

[0487] The server generates specific rehabilitation content based on the selected rehabilitation method, such as "Tell us about your favorite hobbies from the past" for reminiscence therapy.

[0488] Generated rehabilitation content: "Tell me about your favorite past hobbies."

[0489] Step 7:

[0490] The server sends the generated rehabilitation content to the terminal. The rehabilitation content is sent as text data, but also includes information for voice synthesis.

[0491] Submitted data: "Tell me about your favorite past hobbies."

[0492] Step 8:

[0493] The device uses voice synthesis technology to ask the patient about the rehabilitation information it has received, and outputs the questions as voice.

[0494] Voice output: "Tell me about your favorite past hobbies."

[0495] Step 9:

[0496] The user (patient) responds to the questions posed by the device. For example, they might reply, "I used to love reading."

[0497] Response: "I used to love reading."

[0498] Step 10:

[0499] The terminal again receives the patient's response as voice data and sends it to the server, where it is also converted into text data.

[0500] Audio data: "I used to love reading."

[0501] Text data: "I used to love reading."

[0502] Step 11:

[0503] The server analyzes the received feedback data and determines the next rehabilitation technique. A loop process is performed to continuously evaluate the patient's condition.

[0504] Analysis results: "Emotional state: calm, rehabilitation method: continued reminiscence therapy"

[0505] This series of steps is repeated to provide optimal rehabilitation for the patient, with the aim of slowing the progression of dementia.

[0506] Example 1

[0507] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0508] Currently, rehabilitation methods for dementia patients are often based on uniform programs, making it difficult to provide individualized training content that takes into account each patient's emotional and cognitive states in real time. Furthermore, there are limited methods for assessing a patient's emotional and cognitive states, making it difficult to select effective rehabilitation methods. As a result, there is a problem in that the progression of dementia cannot be effectively slowed.

[0509] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0510] In this invention, the server includes a means for analyzing the tone and strength of the patient's voice data to identify emotions, a means for grasping the patient's emotional and cognitive states, and a means for automatically selecting one of cognitive stimulation therapy, reminiscence therapy, and reality-focused therapy as a retraining method, thereby enabling the provision of an individual rehabilitation method based on the patient's emotional and cognitive states.

[0511] "Patient" refers to a person who has a progressive disease, such as dementia, and requires rehabilitation.

[0512] "Voice input" refers to the voice data uttered by the patient, which provides basic information for analyzing their emotions and cognitive state.

[0513] A "terminal" is a device that has the function of receiving voice input from a patient and transmitting it to an information processing device. Specifically, this applies to smart speakers.

[0514] "Voice data" refers to digitally recorded speech made by a patient and is used to analyze their emotional and cognitive states.

[0515] "Information processing device" refers to a device that processes received voice data using advanced analytical technology to evaluate a patient's emotional and cognitive state. This includes servers and cloud computing services.

[0516] "Emotional state" refers to the patient's mental state, and is information analyzed from the tone and strength of the voice data.

[0517] "Cognitive status" refers to the patient's cognitive functions, such as intellectual processing ability and memory, and is information evaluated from the content of the voice data and responses.

[0518] "Retraining techniques" refer to the most appropriate rehabilitation methods based on the patient's emotional and cognitive state. These include cognitive stimulation therapy, reminiscence therapy, and reality-oriented therapy.

[0519] "Cognitive stimulation therapy" is a type of training method for maintaining and improving cognitive function in the brain. Specifically, it is carried out using puzzles and calculation problems.

[0520] Reminiscence therapy is a rehabilitation method that stimulates memory and maintains cognitive function by discussing past events.

[0521] "Reality-oriented therapy" is a rehabilitation method that promotes awareness of reality and strengthens cognitive function by having patients talk about real times, places, and everyday events.

[0522] "Feedback" refers to the responses and reactions received by the terminal from the patient, and is data that the information processing device uses to perform further analysis based on that.

[0523] MODE FOR CARRYING OUT THE INVENTION

[0524] This invention is a system for slowing the progression of dementia in patients, linking a generative AI with a voice input device to provide optimal retraining methods for each patient. This system analyzes the emotional and cognitive state of the patient from their voice and adjusts the retraining content in real time to enhance effectiveness.

[0525] System configuration

[0526] 1. Terminal

[0527] The terminal is a voice input device for receiving voice input from the patient. The voice input device receives the patient's voice and converts it into text data through speech recognition at an early stage. Examples of this include smart speakers such as Amazon Alexa and Google Home. The terminal also has the function of transmitting the received data to an information processing device.

[0528] 2. Server (information processing device)

[0529] The server receives the voice data and converts it into text using more advanced speech recognition technology. The server analyzes the tone and intensity of the voice data to assess the patient's emotional and cognitive state. The analysis uses advanced voice analysis tools such as Google Cloud Speech-to-Text and IBM Watson. Based on the analysis results, the server selects the optimal retraining method and sends it to the device.

[0530] 3. Retraining method

[0531] Retraining techniques include cognitive stimulation therapy, reminiscence therapy, and reality-oriented therapy. These methods are automatically selected by the server depending on the patient's condition. Specific questions and instructions are generated using a generative AI model (e.g., GPT-3) built into the server.

[0532] 4. Feedback

[0533] The terminal presents the retraining content to the patient by voice and receives the patient's response again, which is then sent back to the server and used to determine the next retraining content.

[0534] Specific actions

[0535] Device behavior

[0536] The device receives the patient's voice input and sends it to the server. For example, the device may ask a simple question while greeting the patient in the morning. For example, the device may ask, "Good morning, how are you feeling today?" to collect the patient's voice.

[0537] Server Operation

[0538] The server performs advanced analysis of the received voice data to evaluate the patient's emotional and cognitive state. For example, if the patient responds, "I'm a little tired today," the server analyzes the tone and strength of the voice and evaluates the patient's emotional state as "fatigue." Based on the results, it selects a retraining method and sends it to the device.

[0539] Selection of retraining methods

[0540] For example, if the emotional state is determined to be "calm," reminiscence therapy is used to select questions such as "Tell me about the town where you used to live." Conversely, if the emotional state is determined to be "anxious," reality-oriented therapy, which provides reassurance, is used to select questions such as "Do you know what day it is today?" The generative AI model is used to input the following prompt sentences and generate appropriate questions.

[0541] Example prompt sentence:

[0542] "Create questions about how you're feeling today and your past memories to suggest rehabilitation techniques for dementia patients."

[0543] Present and get feedback from the device

[0544] The device then verbally communicates the selected retraining content to the patient. For example, if the device asks, "Tell me about the town where you used to live," and the patient replies, "The town where I used to live was quiet and a very nice place," the device sends the voice back to the server for analysis. This determines the next retraining content.

[0545] In this way, the system of the present invention can analyze the patient's voice data and provide optimal retraining techniques in real time, which can effectively stimulate the patient's cognitive function and slow the progression of dementia.

[0546] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0547] Program processing flow

[0548] Step 1:

[0549] A user speaks into a voice input device (terminal). The voice input device receives the user's voice in real time. The input is the user's voice data, for example, a voice replying "I'm a little tired today" to a question from the terminal such as "Good morning. How are you feeling today?" The output is raw voice data acquired in real time.

[0550] Step 2:

[0551] The device converts the user's voice data into text data using early speech recognition technology. A speech recognition engine is used to convert the input voice data into text data. For example, a speech saying "I'm a little tired today" is converted into text data saying "I'm a little tired today." The output is that text data.

[0552] Step 3:

[0553] The terminal sends text data to the server. The input is the converted text data, which is sent to the server. For example, the text data "I'm a little tired today" is sent to the server via the network. The output is the text data sent to the server.

[0554] Step 4:

[0555] The server performs a detailed analysis of the received text data using advanced analysis techniques. The server combines natural language processing (NLP) techniques and voice tone analysis to evaluate the emotional and cognitive states of the input text data. For example, "Today" indicates "fatigue" as the emotional state. The output is the analysis results: "Emotional state: fatigue" and "Cognitive state."

[0556] Step 5:

[0557] The server selects the optimal retraining method based on the analysis results. The input is data on "emotional state: fatigue" and "cognitive state." For example, in the case of "fatigue," the server selects reality-oriented therapy to promote refreshment. A generative AI model (such as GPT-3) is used to generate specific questions and instructions. For example, it generates the question, "Do you know what day it is today?" The output is the selected retraining method and the generated question.

[0558] Step 6:

[0559] The server sends the selected retraining method and the generated question to the terminal. The input is the retraining method selection result and the generated question, which are sent to the terminal via the Internet. For example, a question such as "Do you know what day it is today?" is sent to the terminal. The output is the question sent to the terminal.

[0560] Step 7:

[0561] The terminal presents the received question to the user by voice. The input is question data received from the server, and the text "Do you know what day it is today?" is converted into voice and presented to the user. The voice output device is used to pose the question to the user. The output is a voice presentation to the user.

[0562] Step 8:

[0563] The user responds to a question from the terminal. The input is a voice question from the terminal, and the user responds to it by voice. For example, the user might reply, "Today is Monday." The output is the user's response voice data.

[0564] Step 9:

[0565] The terminal converts the user's response voice data into text data again and sends it to the server. The voice recognition engine is used again to convert the input voice into text data. For example, the voice saying "Today is Monday" is converted into text data. This is sent to the server. The output is the text data sent to the server.

[0566] Step 10:

[0567] The server analyzes the user's response again and adjusts the next retraining content. The input is the text data "Today is Monday," which the server analyzes to evaluate the user's emotional and cognitive state. Based on the analysis results, a new question or instruction is generated and sent to the device again. For example, a new question is generated: "Do you have any plans with your family today?" The output is the next retraining content.

[0568] Through this process, the system of the present invention analyzes the patient's voice data and provides an optimal retraining method in real time, thereby effectively stimulating the patient's cognitive function and slowing the progression of dementia.

[0569] (Application example 1)

[0570] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0571] It is important to provide appropriate rehabilitation methods for dementia patients in elderly rehabilitation facilities and nursing homes. However, conventional technologies are insufficient in providing optimal rehabilitation methods according to the emotional and cognitive states that vary from patient to patient. In particular, it is difficult to assess a patient's condition in real time and continue to provide appropriate rehabilitation methods. Therefore, there is a need for a system that provides optimal rehabilitation methods for individual patients in real time.

[0572] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0573] In this invention, the server includes: means for receiving voice input from the patient; means for transmitting the received voice data to the server; means for the server to analyze the voice data and grasp the emotional and cognitive states of the patient; means for selecting an optimal rehabilitation method based on the grasped states; means for transmitting the selected rehabilitation method to the terminal; means for the terminal to present rehabilitation content to the patient and obtain feedback; means for converting the patient's voice data into text data and evaluating the emotional state using an emotion analysis model; and means for generating specific prompt sentences based on the evaluation results to provide the rehabilitation method. This makes it possible to evaluate the emotional and cognitive states that differ for each patient in real time and provide the optimal rehabilitation method based on the evaluation.

[0574] A "means for receiving patient voice input" is a device or method for collecting patient-produced speech in real time.

[0575] The "means for transmitting received voice data to a server" refers to a device or method that has the function of transferring voice data to a server via a network.

[0576] "Means for the server to analyze voice data and understand the emotional and cognitive state of the patient" refers to technology that enables the server to process voice data and understand the emotional and cognitive state of the patient.

[0577] "Means for selecting the optimal rehabilitation method based on the grasped condition" refers to a system or algorithm that determines the most appropriate rehabilitation method based on the analysis results.

[0578] The "means for transmitting the selected rehabilitation method to the terminal" refers to a device or method for transmitting the selected rehabilitation method to the client device via a network.

[0579] "Means for a terminal to present rehabilitation content to a patient and obtain feedback" refers to equipment or methods for providing rehabilitation information to a patient and collecting their reactions again.

[0580] "Means for converting a patient's voice data into text data and assessing their emotional state using an emotion analysis model" refers to a technology for converting voice data into text and analyzing the text to assess the user's emotions.

[0581] "Means for generating specific prompt sentences based on the evaluation results and providing rehabilitation techniques" is a mechanism for creating specific questions and instructions based on the results of emotion analysis and using them to advance rehabilitation.

[0582] The following describes an embodiment of the present invention.

[0583] First, the entire system consists of a voice input device (such as a smartphone) held by the patient, a server, and a user terminal for providing feedback.

[0584] Audio input device

[0585] The voice input device is a smartphone, which can collect the patient's speech in real time and convert it into text data. The technologies used include the Python SpeechRecognition library and the Google Speech API.

[0586] server

[0587] The server analyzes the received voice data to understand the patient's emotional and cognitive state. A generative AI model is used for the analysis, and emotion analysis is performed using the Hugging Face Transformers model. Specifically, the voice data is converted into text data, which is then input into an emotion analysis model to evaluate the patient's emotional state. Based on the results of this evaluation, an appropriate rehabilitation method is selected. An appropriate prompt is also generated, and the selected rehabilitation method is sent to the user's device.

[0588] User terminal

[0589] The user device receives the rehabilitation method from the server and presents it to the patient via voice. The user device then collects the feedback again as voice and sends the data to the server. This provides the data needed to determine the next rehabilitation method.

[0590] Program processing natural language explanation

[0591] The server analyzes the patient's voice to understand their emotional and cognitive state and decides on a rehabilitation method based on that. First, the smartphone collects the voice and converts it into text using the Google Speech API. The text data is then sent to the server, where emotion analysis is performed using the Hugging Face Transformers model. Based on the analysis results, a specific rehabilitation method for the target patient is determined and an appropriate prompt is generated. This prompt is then presented to the patient via the user's device.

[0592] Specific examples

[0593] As a specific example, if a patient answers, "I feel a little tired today," the emotion analysis model will determine this as "NEGATIVE," and the server will generate a prompt such as "Is there anything bothering you?" and send it to the user's terminal.

[0594] Prompt Sentence Examples

[0595] "Good morning. How are you feeling today?"

[0596] "Tell me about the town where you used to live."

[0597] "Do you know what day it is?"

[0598] As a result, this system can provide optimal rehabilitation methods in real time based on each patient's emotional and cognitive state.

[0599] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0600] Step 1:

[0601] The user provides voice input. In this case, the patient's voice data is collected using the smartphone's microphone. For example, the patient might say, "I feel a little tired today." This voice data becomes the input.

[0602] Step 2:

[0603] The device converts the input voice into text data. Here, the Google Speech API is used to convert the voice into text. This text data is output and goes to the next analysis step.

[0604] Step 3:

[0605] The terminal sends the converted text data to the server. The text data is transferred to the server via the network. The server receives the text data.

[0606] Step 4:

[0607] The server analyzes the received text data. Specifically, it performs emotion analysis using the Hugging Face Transformers model. This analysis evaluates the patient's emotional state from the input text data. For example, the text "I'm a little tired" is evaluated as "NEGATIVE."

[0608] Step 5:

[0609] The server selects the optimal rehabilitation method based on the results of the emotion analysis and generates a specific prompt. Here, the rehabilitation method is determined based on the results of the emotion analysis model, and an appropriate prompt is generated. A prompt such as "Is there anything that is bothering you?" is generated.

[0610] Step 6:

[0611] The server sends the generated prompt to the terminal. The prompt is sent to the terminal via the network. The sent prompt is received by the terminal.

[0612] Step 7:

[0613] The terminal then presents the received prompt to the user by voice. In this case, the smartphone's text-to-speech function is used to play the prompt aloud. For example, the patient might hear, "Is there anything bothering you?"

[0614] Step 8:

[0615] The user provides vocal feedback on the presented prompt sentence, which is then input into the terminal again, leading to further analysis and the provision of rehabilitation techniques.

[0616] Through these steps, the system can analyze the patient's emotional state in real time and provide optimal rehabilitation methods. It also generates prompts and collects feedback, enabling continuous rehabilitation support.

[0617] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0618] MODE FOR CARRYING OUT THE INVENTION

[0619] This invention is a system that combines generative AI, a smart speaker, and an emotion engine to provide a rehabilitation method to slow the progression of dementia. This system analyzes the patient's condition from their voice and adjusts the rehabilitation content in real time to improve effectiveness.

[0620] System configuration

[0621] 1. Terminal

[0622] The device is a smart speaker that receives voice input from the patient, performs speech recognition at an early stage, and converts the received voice input into text data. The device also has the function of transmitting the received data to a server.

[0623] 2. Server

[0624] The server receives the voice data and converts it into text using advanced voice recognition technology. The server analyzes the tone and strength of the voice data to assess the patient's emotional and cognitive state. Based on the analysis results, it selects the most appropriate rehabilitation method and sends it to the device.

[0625] 3. Emotion Engine

[0626] The emotion engine analyzes the emotional state of the voice data in real time and feeds back the results to the server. The emotion engine has the function of analyzing the emotional state by combining tone, stress, intonation, and other non-verbal elements.

[0627] 4. Rehabilitation methods

[0628] The rehabilitation techniques include cognitive stimulation therapy, reminiscence therapy, and reality orientation. These techniques are automatically selected by the server depending on the patient's condition.

[0629] 5. Feedback

[0630] The terminal presents rehabilitation content to the patient and receives the patient's response again. The response is again sent to the server and used to determine the next rehabilitation content.

[0631] Program processing

[0632] Device behavior

[0633] The device receives the patient's voice input and sends it to the server. For example, the device may ask a simple question while greeting the patient in the morning. For example, the device may ask, "Good morning, how are you feeling today?" to collect the patient's voice.

[0634] Server Operation

[0635] The server analyzes the received voice data and evaluates the patient's emotional and cognitive state. For example, if the patient responds, "I'm a little tired today," the server analyzes the tone and strength of the voice and evaluates it as "feeling tired." Based on the results, it selects a rehabilitation method and sends it to the device.

[0636] Emotion Engine Operation

[0637] The emotion engine analyzes the patient's emotions in real time based on the voice data. For example, if "anxiety" is detected from the tone and intonation, the result is fed back to the server, which then adjusts the rehabilitation program based on that information.

[0638] Selection of rehabilitation methods

[0639] For example, if the emotional state is judged to be "calm," reminiscence techniques are used to select questions such as "Tell me about the town where you used to live." Conversely, if the emotional state is judged to be "anxious," reassuring reality orientation techniques are used to select questions such as "Do you know what day it is today?"

[0640] Present and get feedback from the device

[0641] The device then verbally communicates the selected rehabilitation content to the patient. If the patient responds, "The town where I used to live was quiet and a very nice place," the voice message is sent back to the server for analysis. This determines the next rehabilitation content.

[0642] Specific examples

[0643] 1. Basic Scenario

[0644] In the morning, the user (a dementia patient) responds to a voice message from the device about their mood that day. The device then sends the voice message to the server, which analyzes it. For example, if the user answers "I didn't have any particular dreams" in response to the question "Good morning. What kind of dreams did you have today?", the server will analyze the response and evaluate the user's emotional state as "Calm." Based on the results, the server will select a reminiscence method and send the next question to the device: "What is your fondest memory from childhood?"

[0645] 2. Utilizing the Emotion Engine

[0646] An emotion engine has been added to more accurately grasp the patient's emotional state. For example, if a patient responds, "I'm a little tired today," the emotion engine analyzes the tone and intonation of the response and evaluates it as "fatigue." This result is fed back to the server, which then adjusts the rehabilitation program based on that information.

[0647] 3. Utilizing a variety of rehabilitation methods

[0648] If the patient's emotional state is evaluated as "anxious," the server selects reality orientation. The terminal asks, "What are your plans for today?" and the patient answers, "I have an appointment to meet my family today." The server evaluates the patient's sense of security.

[0649] In this way, the system of the present invention can analyze the patient's voice data and provide optimal rehabilitation techniques in real time, which can effectively stimulate the patient's cognitive function and slow the progression of dementia. The addition of an emotion engine further improves the accuracy and effectiveness of rehabilitation content.

[0650] The processing flow will be explained below.

[0651] Step 1:

[0652] The terminal receives the patient's voice input, asks the patient questions, and records their responses as audio.

[0653] For example, ask, "Good morning, how are you feeling today?"

[0654] Step 2:

[0655] The device sends the received voice data to the server, where it is converted into text using early-stage voice recognition technology.

[0656] Audio data: "I'm a little tired today."

[0657] Text data: "I'm a little tired today."

[0658] Step 3:

[0659] The server receives the transmitted voice and text data and converts it back into text using advanced voice recognition technology to improve accuracy.

[0660] Audio data: "I'm a little tired today."

[0661] Text data: "I'm a little tired today" (Improved accuracy)

[0662] Step 4:

[0663] The server analyzes the audio and text data and uses an emotion engine to assess the patient's emotional and cognitive state. It identifies the patient's emotional state based on the tone, stress, inflection, and content of the audio.

[0664] Analysis result: "Emotional state: Fatigue"

[0665] Step 5:

[0666] The server selects the optimal rehabilitation method based on the analysis results. For example, if the emotional state is evaluated as "fatigue," it selects a calm and relaxing reminiscence method.

[0667] Selection result: “Reminiscence method”

[0668] Step 6:

[0669] The server generates specific rehabilitation content based on the selected rehabilitation method, such as "Tell us about your favorite hobbies from the past" for reminiscence therapy.

[0670] Generated rehabilitation content: "Tell me about your favorite past hobbies."

[0671] Step 7:

[0672] The server sends the generated rehabilitation content to the terminal. The rehabilitation content is sent as text data, but also includes information for voice synthesis.

[0673] Submitted data: "Tell me about your favorite past hobbies."

[0674] Step 8:

[0675] The device uses voice synthesis technology to ask the patient about the rehabilitation information it has received, and outputs the questions as voice.

[0676] Voice output: "Tell me about your favorite past hobbies."

[0677] Step 9:

[0678] The user (patient) responds to the questions posed by the device. For example, they might reply, "I used to love reading."

[0679] Response: "I used to love reading."

[0680] Step 10:

[0681] The terminal again receives the patient's response as voice data and sends it to the server, where it is also converted into text data.

[0682] Audio data: "I used to love reading."

[0683] Text data: "I used to love reading."

[0684] Step 11:

[0685] The server analyzes the received feedback data and determines the next rehabilitation method. The emotion engine runs again and evaluates the patient's condition. A loop process is performed to continuously evaluate the patient's condition.

[0686] Analysis results: "Emotional state: calm, rehabilitation method: continued reminiscence therapy"

[0687] This series of steps is repeated to provide optimal rehabilitation for the patient, with the aim of slowing the progression of dementia. The addition of the emotion engine further improves the accuracy and effectiveness of the rehabilitation content.

[0688] Example 2

[0689] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0690] Currently, rehabilitation systems on the market lack the ability to properly grasp a patient's emotional and cognitive state. As a result, the rehabilitation methods provided are often not optimal for the patient. Furthermore, the lack of a function to obtain real-time feedback and adjust the rehabilitation content limits the effectiveness of rehabilitation. There is a need to solve these problems and provide the most effective rehabilitation methods for patients.

[0691] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for analyzing voice data and grasping the emotional and cognitive states of the patient, a means for an emotion engine to analyze the voice data in real time and provide feedback to the server, and a means for selecting an optimal rehabilitation method. This makes it possible to provide an optimal rehabilitation method in real time based on the patient's condition and maximize its effectiveness.

[0692] "Patient" refers to an individual who is a target of using the rehabilitation system.

[0693] "Voice input" refers to the words or voice data uttered by the patient.

[0694] "Terminal" refers to an electronic device that receives the patient's voice input and performs the necessary processing.

[0695] "Server" refers to a central computer system that analyzes data sent from terminals and performs necessary processing and feedback.

[0696] "Voice data" refers to a digital representation of a patient's voice input.

[0697] "Emotion engine" refers to a software or hardware component for analyzing audio data and assessing emotional state.

[0698] "Emotional state" refers to the psychological state of the patient as judged from the tone and strength of their voice.

[0699] "Cognitive status" refers to the state of a patient's cognitive function.

[0700] "Rehabilitation techniques" refer to therapies provided based on the patient's emotional and cognitive state.

[0701] "Real-time" refers to processing that occurs at a speed close to instantaneous.

[0702] This invention is a system that combines a generative AI model, a smart speaker, and an emotion engine to provide a rehabilitation method to slow the progression of dementia. This system analyzes the patient's condition from their voice and adjusts the rehabilitation content in real time to improve effectiveness.

[0703] System configuration

[0704] 1. Terminal

[0705] The device consists of a smart speaker that receives the patient's voice input. Specifically, the smart speaker picks up the patient's voice using a microphone, performs initial voice recognition, and converts it into text data. The device also has the function of sending this data to a server. For example, the device may ask the patient, "Good morning, how are you feeling today?"

[0706] 2. Server

[0707] The server receives the voice data sent from the device and converts it into text data using more advanced voice recognition technology. One possible technology is the Google Cloud Speech-to-Text API. The server then analyzes the tone and strength of the voice data to evaluate the patient's emotional and cognitive state. For example, the server evaluates the voice data, such as "I'm a little tired today," as "I feel fatigued." Based on the analysis results, the server selects the optimal rehabilitation method and sends it to the device.

[0708] 3. Emotion Engine

[0709] The emotion engine analyzes the emotional state of the voice data in real time and feeds the results back to the server. Specific analysis items include tone, stress, intonation, and other non-verbal elements. For example, if "anxiety" is detected from the intonation of a patient's voice, the result is fed back to the server.

[0710] 4. Rehabilitation methods

[0711] Rehabilitation techniques include cognitive stimulation therapy, reminiscence therapy, and reality orientation. These techniques are automatically selected by the server depending on the patient's condition. For example, if the emotional state is judged to be "calm," reminiscence therapy will be used and questions such as "Tell me about the town where you used to live" will be selected.

[0712] 5. Feedback

[0713] The device presents rehabilitation content to the patient and receives the patient's response again. The received data is sent to the server and used to determine the next rehabilitation content. For example, if the patient responds, "The town where I used to live was quiet and a very nice place," the voice data is sent to the server and analyzed.

[0714] Specific examples

[0715] Basic Scenario

[0716] In the morning, the user (a dementia patient) answers questions posed by the device. For example, to the question, "Good morning. What kind of dreams did you have today?", the user replies, "I didn't have any particular dreams." The server analyzes the voice data and evaluates the user's emotional state as "calm." The server selects the reminiscence method and sends the next question, "What is your best childhood memory?" to the device. The device presents the question to the patient, who responds. The server then analyzes the response again to determine the next rehabilitation plan.

[0717] Utilizing the Emotion Engine

[0718] The emotion engine analyzes the tone and intonation of the patient's voice. For example, if a patient replies, "I'm a little tired today," the emotion engine analyzes the tone and intonation and evaluates it as "feeling tired." This evaluation result is fed back to the server, which then adjusts the rehabilitation program based on that information.

[0719] Utilizing a variety of rehabilitation methods

[0720] If the patient's emotional state is evaluated as "anxious," the server selects reality orientation. The device asks, "What are your plans for today?" and the patient replies, "I have an appointment to meet my family today." The server analyzes this information and evaluates the patient's sense of security. This determines the next rehabilitation content.

[0721] Prompt Sentence Examples

[0722] "To provide rehabilitation techniques for dementia patients, suggest optimal questions based on their emotional and cognitive states. For example, if a patient says, 'I'm a little tired today,' analyze their speech data and generate the appropriate next question."

[0723] In this way, the system of the present invention can analyze the patient's voice data and provide optimal rehabilitation methods in real time, which can effectively stimulate the patient's cognitive function and slow the progression of dementia. The addition of an emotion engine further improves the accuracy and effectiveness of rehabilitation content.

[0724] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0725] Step 1:

[0726] The device receives voice input. The user speaks to the smart speaker. For example, the device asks, "Good morning, how are you feeling today?" and the user replies, "I'm a little tired today." This voice data is input into the device.

[0727] Input: User's voice data: "I'm a little tired today."

[0728] Output: Received audio data

[0729] Step 2:

[0730] The voice data received by the device is converted into text data using an initial voice recognition engine. For example, the voice recognition technology built into a smart speaker can be used to convert the voice into text, such as "I'm a little tired today." This text data is then sent to the server.

[0731] Input: Received audio data

[0732] Output: Text data "I'm a little tired today"

[0733] Step 3:

[0734] The server uses advanced speech recognition technology to convert the voice data sent from the device back into text data, and then uses advanced technology such as the Google Cloud Speech-to-Text API to analyze the details of the voice and generate text data.

[0735] Input: Audio data sent from the device

[0736] Output: Highly refined text data "I'm a little tired today"

[0737] Step 4:

[0738] The server performs detailed analysis of the voice data and evaluates the emotional state using an emotion engine. The server analyzes the tone, strength, and intonation of the voice to identify the emotional state. For example, it may evaluate the voice as "fatigue" based on its low tone and strength.

[0739] Input: Highly refined text data and its speech characteristics

[0740] Output: Emotional state "Feeling tired"

[0741] Step 5:

[0742] The emotion engine feeds back the analyzed emotional state to the server. The emotion engine determines "fatigue" based on the tone and intonation of the voice data and sends the evaluation result to the server.

[0743] Input: Tone and intonation analysis data of speech data

[0744] Output: Emotional state assessment result: "Feeling tired"

[0745] Step 6:

[0746] The server selects the optimal rehabilitation method based on the emotional and cognitive states. Based on the result "feeling tired," the server selects cognitive stimulation therapy as the rehabilitation method and determines its specific content. For example, it selects the question, "Do you know what day it is today?"

[0747] Input: Emotional state assessment result: "Feeling tired"

[0748] Output: The selected rehabilitation technique "Cognitive Stimulation Therapy" and the question "Do you know what day it is today?"

[0749] Step 7:

[0750] The server sends the selected rehabilitation method and its details to the device. As part of the cognitive stimulation therapy, the server sends the question "Do you know what day it is today?" to the device.

[0751] Input: Selected rehabilitation method and specific question

[0752] Output: Instruction data to the terminal

[0753] Step 8:

[0754] The device then presents the received rehabilitation information to the user via voice. For example, the device might ask the user, "Do you know what day it is today?" The user might respond, "Today is Tuesday."

[0755] Input: Rehabilitation details sent from the server

[0756] Output: Audio presentation to the user

[0757] Step 9:

[0758] The user responds to the rehabilitation content. When the user responds, "Today is Tuesday," the terminal receives the voice again. This received data is sent to the server again.

[0759] Input: User's voice reply "Today is Tuesday"

[0760] Output: Response data to the terminal

[0761] Step 10:

[0762] The server analyzes the voice data again and determines the next rehabilitation content based on the results. For example, based on the response "Today is Tuesday," the server selects the next rehabilitation method and sends the next question, "What did you do on Tuesday?" to the terminal.

[0763] Input: User response data received again

[0764] Output: Next rehabilitation technique and specific questions

[0765] This series of processes enables feedback tailored to the user's emotional and cognitive state, providing optimal rehabilitation techniques. This system is expected to slow the progression of dementia in patients.

[0766] (Application example 2)

[0767] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0768] Rehabilitation for dementia patients requires selecting the optimal method to suit each individual patient's condition and providing effective treatment. However, with conventional methods, it has been difficult to accurately grasp the patient's emotional and cognitive state in real time and provide the optimal rehabilitation method based on that. In particular, more advanced technology is required to realize rehabilitation in a virtual environment. The purpose of this project is to solve these issues.

[0769] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0770] In this invention, the server includes means for receiving voice input from the patient, means for transmitting the received voice data to the server, means for analyzing the voice data to grasp the emotional and cognitive states of the patient, means for selecting an optimal rehabilitation method based on the grasped states, means for transmitting the selected rehabilitation method to the terminal, means for the terminal to present rehabilitation content to the patient and obtain feedback, and means for presenting rehabilitation content in a virtual environment. This makes it possible to provide the rehabilitation method optimal for the patient's condition in real time, thereby effectively slowing the progression of dementia.

[0771] Key Word Definitions

[0772] "Means for receiving patient voice input" refers to a device for collecting voice data from a dementia patient, often a microphone or smart speaker.

[0773] "Means for transmitting received voice data to a server" refers to a function for transferring collected voice data to a central computer (server) via a network.

[0774] "Means for analyzing voice data to understand the emotional and cognitive state of a patient" refers to algorithms or analytical software for analyzing received voice data and determining the emotional and cognitive state of a patient.

[0775] "Means for selecting the most appropriate rehabilitation method based on the identified condition" refers to the logic and process for automatically selecting the rehabilitation method that is most appropriate for the patient's current condition based on the results of the analysis.

[0776] "Means for transmitting the selected rehabilitation method to the terminal" refers to a communication means for transmitting information about the rehabilitation method selected by the server to the patient's terminal.

[0777] "Means for the device to present rehabilitation content to the patient and obtain feedback" refers to an interface that presents selected rehabilitation content through audio or a virtual environment and collects the patient's reactions and answers again.

[0778] "Means for presenting rehabilitation content in a virtual environment" refers to software or hardware that uses virtual reality or augmented reality technology to provide patients with visual or auditory rehabilitation content.

[0779] MODE FOR CARRYING OUT THE INVENTION

[0780] To illustrate an embodiment of the present invention, the following system configuration and operation will be described.

[0781] System Configuration

[0782] This system consists of the following main hardware and software:

[0783] Hardware

[0784] Smart glasses or head-mounted displays (HMD)

[0785] microphone

[0786] Cloud Server

[0787] software

[0788] Speech recognition library (e.g., speech_recognition library)

[0789] Sentiment analysis engine (e.g., custom-developed emotion_engine module)

[0790] Rehabilitation method selection logic (e.g., custom-developed rehab_methods module)

[0791] Virtual environment interface (e.g., a custom-developed virtual_environment module)

[0792] A natural language explanation of how to proceed

[0793] 1. Receiving and sending audio data

[0794] The user's voice input is received by a microphone attached to the smart glasses or head-mounted display, and the received voice data is transmitted to a cloud server in real time.

[0795] 2. Analysis of audio data

[0796] The server converts the received voice data into text data using a speech recognition library. It then uses an emotion analysis engine to analyze the tone and strength of the text data to understand the user's emotional and cognitive state. Based on the results of this analysis, the server selects the most appropriate rehabilitation method for the user.

[0797] 3. Selection of rehabilitation method

[0798] Based on the analyzed data, the server uses a rehabilitation method selection logic to determine the appropriate rehabilitation content. The rehabilitation method is automatically selected from cognitive stimulation therapy, reminiscence therapy, and reality orientation.

[0799] 4. Presentation of rehabilitation content

[0800] The selected rehabilitation content is sent from the cloud server to smart glasses or a head-mounted display, and the user can receive the rehabilitation content visually or audibly through virtual reality or augmented reality using a virtual environment interface.

[0801] 5. Get feedback

[0802] If the user provides vocal feedback on the rehabilitation content, the audio is again picked up by the microphone and sent to the cloud server, and this process further adjusts the next rehabilitation content.

[0803] Specific examples

[0804] For example, if a user is asked, "What kind of dream did you have today?" and the user answers, "I didn't have any particular dreams," this voice data is sent to the server, where it is analyzed and the emotional state is evaluated as "calm" based on the tone and intensity of the voice. Based on this evaluation, the server selects a reminiscence method and presents the next question, "What is your best childhood memory?" A nostalgic scene is displayed within the virtual environment.

[0805] Prompt Sentence Examples

[0806] Emotional state: Calm

[0807] Q: Tell me about the town where you used to live.

[0808] In this way, the system can analyze the user's voice data and provide optimal rehabilitation techniques in real time, which can effectively stimulate the user's cognitive functions and slow the progression of dementia.

[0809] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0810] Program processing flow

[0811] Step 1:

[0812] The device receives the user's voice input. The voice data is captured by a microphone and sent to a cloud server in real time. The input is the user's voice, and the output is sent to the server as voice data.

[0813] Step 2:

[0814] The server receives the voice data and converts it into text data using a speech recognition library. The input is voice data and the output is text data. The speech recognition library analyzes the sound wave data and converts it into text based on a language model.

[0815] Step 3:

[0816] The server inputs the converted text data into an emotion analysis engine, which analyzes the emotional and cognitive states. The input is text data, and the output is an evaluation of the emotional and cognitive states. The emotion analysis engine takes into account non-verbal elements such as tone, strength, and intonation during analysis.

[0817] Step 4:

[0818] Based on the analysis results, the server uses the rehabilitation method selection logic to select the optimal rehabilitation method. The input is the evaluation results of the emotional and cognitive states, and the output is the selected rehabilitation method. The rehabilitation method selection logic compares the application conditions of each rehabilitation method with the current evaluation results and selects the most appropriate method.

[0819] Step 5:

[0820] The server sends the selected rehabilitation technique to the terminal. The input is the selected rehabilitation technique, and the output is the rehabilitation technique information sent to the terminal. The selected information is delivered to the terminal using a communication protocol.

[0821] Step 6:

[0822] The device presents rehabilitation content to the user through a virtual environment interface. The input is rehabilitation method information, and the output is a virtual rehabilitation environment that the user perceives visually or audibly. Visual content is displayed and played on a head-mounted display, and audio content is displayed and played on speakers.

[0823] Step 7:

[0824] The user responds to the presented rehabilitation content and takes action, and their reactions are collected again as voice data by the device. The input is the user's voice response, and the output is again sent to the server as voice data.

[0825] Step 8:

[0826] The server analyzes the user's feedback and adjusts the next rehabilitation content. The input is the user's voice data and the rehabilitation method results, and the output is the adjusted rehabilitation method. The server reuses the emotion analysis engine and rehabilitation method selection logic to reselect the optimal rehabilitation method that reflects the feedback.

[0827] Through these steps, the system can adapt flexibly to the user's condition and provide effective rehabilitation.

[0828] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0829] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0830] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0831] [Third embodiment]

[0832] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0833] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0834] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0835] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0836] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0837] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0838] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0839] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0840] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0841] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0842] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0843] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0844] MODE FOR CARRYING OUT THE INVENTION

[0845] This invention is a system for slowing the progression of dementia patients, linking generative AI with a smart speaker to provide optimal rehabilitation methods for each patient. This system analyzes the patient's condition from their voice and adjusts the rehabilitation content in real time to enhance effectiveness.

[0846] System configuration

[0847] 1. Terminal

[0848] The device is a smart speaker that receives voice input from the patient, performs speech recognition at an early stage, and converts the received voice input into text data. The device also has the function of transmitting the received data to a server.

[0849] 2. Server

[0850] The server receives the voice data and converts it into text using advanced voice recognition technology. The server analyzes the tone and strength of the voice data to assess the patient's emotional and cognitive state. Based on the analysis results, the server selects the most appropriate rehabilitation method and transmits the details to the device.

[0851] 3. Rehabilitation methods

[0852] The rehabilitation techniques include cognitive stimulation therapy, reminiscence therapy, and reality orientation. These techniques are automatically selected by the server depending on the patient's condition.

[0853] 4. Feedback

[0854] The terminal presents rehabilitation content to the patient and receives the patient's response again. The response is again sent to the server and used to determine the next rehabilitation content.

[0855] Program processing

[0856] Device behavior

[0857] The device receives the patient's voice input and sends it to the server. For example, the device may ask a simple question while greeting the patient in the morning. For example, the device may ask, "Good morning, how are you feeling today?" to collect the patient's voice.

[0858] Server Operation

[0859] The server analyzes the received voice data and evaluates the patient's emotional and cognitive state. For example, if the patient responds, "I'm a little tired today," the server analyzes the tone and strength of the voice and evaluates it as "feeling tired." Based on the results, it selects a rehabilitation method and sends it to the device.

[0860] Selection of rehabilitation methods

[0861] For example, if the emotional state is judged to be "calm," reminiscence techniques are used to select questions such as "Tell me about the town where you used to live." Conversely, if the emotional state is judged to be "anxious," reassuring reality orientation techniques are used to select questions such as "Do you know what day it is today?"

[0862] Present and get feedback from the device

[0863] The device then verbally communicates the selected rehabilitation content to the patient. If the patient responds, "The town where I used to live was quiet and a very nice place," the voice message is sent back to the server for analysis. This determines the next rehabilitation content.

[0864] Specific examples

[0865] 1. Basic Scenario

[0866] In the morning, the user (a dementia patient) responds to a voice message from the device about their mood that day. The device then sends the voice message to the server, which analyzes it. For example, if the user answers "I didn't have any particular dreams" in response to the question "Good morning. What kind of dreams did you have today?", the server will analyze the response and evaluate the user's emotional state as "Calm." Based on the results, the server will select a reminiscence method and send the next question to the device: "What is your fondest memory from childhood?"

[0867] 2. Utilizing a variety of rehabilitation methods

[0868] If the patient's emotional state is evaluated as "anxious," the server selects reality orientation. The terminal asks, "What are your plans for today?" and the patient answers, "I have an appointment to meet my family today." The server evaluates the patient's sense of security.

[0869] In this way, the system of the present invention can analyze the patient's voice data and provide optimal rehabilitation techniques in real time, thereby effectively stimulating the patient's cognitive function and slowing the progression of dementia.

[0870] The processing flow will be explained below.

[0871] Step 1:

[0872] The terminal receives the patient's voice input, asks the patient questions, and records their responses as audio.

[0873] For example, ask, "Good morning, how are you feeling today?"

[0874] Step 2:

[0875] The device sends the received voice data to the server, where it is converted into text using early-stage voice recognition technology.

[0876] Audio data: "I'm a little tired today."

[0877] Text data: "I'm a little tired today."

[0878] Step 3:

[0879] The server receives the transmitted voice and text data and converts it back into text using advanced voice recognition technology to improve accuracy.

[0880] Audio data: "I'm a little tired today."

[0881] Text data: "I'm a little tired today" (Improved accuracy)

[0882] Step 4:

[0883] The server analyzes the audio and text data to assess the patient's emotional and cognitive state. It identifies the patient's emotional state based on the tone, stress, inflection, and content of the audio.

[0884] Analysis result: "Emotional state: Fatigue"

[0885] Step 5:

[0886] The server selects the optimal rehabilitation method based on the analysis results. For example, if the emotional state is evaluated as "fatigue," it selects a calm and relaxing reminiscence method.

[0887] Selection result: “Reminiscence method”

[0888] Step 6:

[0889] The server generates specific rehabilitation content based on the selected rehabilitation method, such as "Tell us about your favorite hobbies from the past" for reminiscence therapy.

[0890] Generated rehabilitation content: "Tell me about your favorite past hobbies."

[0891] Step 7:

[0892] The server sends the generated rehabilitation content to the terminal. The rehabilitation content is sent as text data, but also includes information for voice synthesis.

[0893] Submitted data: "Tell me about your favorite past hobbies."

[0894] Step 8:

[0895] The device uses voice synthesis technology to ask the patient about the rehabilitation information it has received, and outputs the questions as voice.

[0896] Voice output: "Tell me about your favorite past hobbies."

[0897] Step 9:

[0898] The user (patient) responds to the questions posed by the device. For example, they might reply, "I used to love reading."

[0899] Response: "I used to love reading."

[0900] Step 10:

[0901] The terminal again receives the patient's response as voice data and sends it to the server, where it is also converted into text data.

[0902] Audio data: "I used to love reading."

[0903] Text data: "I used to love reading."

[0904] Step 11:

[0905] The server analyzes the received feedback data and determines the next rehabilitation technique. A loop process is performed to continuously evaluate the patient's condition.

[0906] Analysis results: "Emotional state: calm, rehabilitation method: continued reminiscence therapy"

[0907] This series of steps is repeated to provide optimal rehabilitation for the patient, with the aim of slowing the progression of dementia.

[0908] Example 1

[0909] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0910] Currently, rehabilitation methods for dementia patients are often based on uniform programs, making it difficult to provide individualized training content that takes into account each patient's emotional and cognitive states in real time. Furthermore, there are limited methods for assessing a patient's emotional and cognitive states, making it difficult to select effective rehabilitation methods. As a result, there is a problem in that the progression of dementia cannot be effectively slowed.

[0911] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0912] In this invention, the server includes a means for analyzing the tone and strength of the patient's voice data to identify emotions, a means for grasping the patient's emotional and cognitive states, and a means for automatically selecting one of cognitive stimulation therapy, reminiscence therapy, and reality-focused therapy as a retraining method, thereby enabling the provision of an individual rehabilitation method based on the patient's emotional and cognitive states.

[0913] "Patient" refers to a person who has a progressive disease, such as dementia, and requires rehabilitation.

[0914] "Voice input" refers to the voice data uttered by the patient, which provides basic information for analyzing their emotions and cognitive state.

[0915] A "terminal" is a device that has the function of receiving voice input from a patient and transmitting it to an information processing device. Specifically, this applies to smart speakers.

[0916] "Voice data" refers to digitally recorded speech made by a patient and is used to analyze their emotional and cognitive states.

[0917] "Information processing device" refers to a device that processes received voice data using advanced analytical technology to evaluate a patient's emotional and cognitive state. This includes servers and cloud computing services.

[0918] "Emotional state" refers to the patient's mental state, and is information analyzed from the tone and strength of the voice data.

[0919] "Cognitive status" refers to the patient's cognitive functions, such as intellectual processing ability and memory, and is information evaluated from the content of the voice data and responses.

[0920] "Retraining techniques" refer to the most appropriate rehabilitation methods based on the patient's emotional and cognitive state. These include cognitive stimulation therapy, reminiscence therapy, and reality-oriented therapy.

[0921] "Cognitive stimulation therapy" is a type of training method for maintaining and improving cognitive function in the brain. Specifically, it is carried out using puzzles and calculation problems.

[0922] Reminiscence therapy is a rehabilitation method that stimulates memory and maintains cognitive function by discussing past events.

[0923] "Reality-oriented therapy" is a rehabilitation method that promotes awareness of reality and strengthens cognitive function by having patients talk about real times, places, and everyday events.

[0924] "Feedback" refers to the responses and reactions received by the terminal from the patient, and is data that the information processing device uses to perform further analysis based on that.

[0925] MODE FOR CARRYING OUT THE INVENTION

[0926] This invention is a system for slowing the progression of dementia in patients, linking a generative AI with a voice input device to provide optimal retraining methods for each patient. This system analyzes the emotional and cognitive state of the patient from their voice and adjusts the retraining content in real time to enhance effectiveness.

[0927] System configuration

[0928] 1. Terminal

[0929] The terminal is a voice input device for receiving voice input from the patient. The voice input device receives the patient's voice and converts it into text data through speech recognition at an early stage. Examples of this include smart speakers such as Amazon Alexa and Google Home. The terminal also has the function of transmitting the received data to an information processing device.

[0930] 2. Server (information processing device)

[0931] The server receives the voice data and converts it into text using more advanced speech recognition technology. The server analyzes the tone and intensity of the voice data to assess the patient's emotional and cognitive state. The analysis uses advanced voice analysis tools such as Google Cloud Speech-to-Text and IBM Watson. Based on the analysis results, the server selects the optimal retraining method and sends it to the device.

[0932] 3. Retraining method

[0933] Retraining techniques include cognitive stimulation therapy, reminiscence therapy, and reality-oriented therapy. These methods are automatically selected by the server depending on the patient's condition. Specific questions and instructions are generated using a generative AI model (e.g., GPT-3) built into the server.

[0934] 4. Feedback

[0935] The terminal presents the retraining content to the patient by voice and receives the patient's response again, which is then sent back to the server and used to determine the next retraining content.

[0936] Specific actions

[0937] Device behavior

[0938] The device receives the patient's voice input and sends it to the server. For example, the device may ask a simple question while greeting the patient in the morning. For example, the device may ask, "Good morning, how are you feeling today?" to collect the patient's voice.

[0939] Server Operation

[0940] The server performs advanced analysis of the received voice data to evaluate the patient's emotional and cognitive state. For example, if the patient responds, "I'm a little tired today," the server analyzes the tone and strength of the voice and evaluates the patient's emotional state as "fatigue." Based on the results, it selects a retraining method and sends it to the device.

[0941] Selection of retraining methods

[0942] For example, if the emotional state is determined to be "calm," reminiscence therapy is used to select questions such as "Tell me about the town where you used to live." Conversely, if the emotional state is determined to be "anxious," reality-oriented therapy, which provides reassurance, is used to select questions such as "Do you know what day it is today?" The generative AI model is used to input the following prompt sentences and generate appropriate questions.

[0943] Example prompt sentence:

[0944] "Create questions about how you're feeling today and your past memories to suggest rehabilitation techniques for dementia patients."

[0945] Present and get feedback from the device

[0946] The device then verbally communicates the selected retraining content to the patient. For example, if the device asks, "Tell me about the town where you used to live," and the patient replies, "The town where I used to live was quiet and a very nice place," the device sends the voice back to the server for analysis. This determines the next retraining content.

[0947] In this way, the system of the present invention can analyze the patient's voice data and provide optimal retraining techniques in real time, which can effectively stimulate the patient's cognitive function and slow the progression of dementia.

[0948] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0949] Program processing flow

[0950] Step 1:

[0951] A user speaks into a voice input device (terminal). The voice input device receives the user's voice in real time. The input is the user's voice data, for example, a voice replying "I'm a little tired today" to a question from the terminal such as "Good morning. How are you feeling today?" The output is raw voice data acquired in real time.

[0952] Step 2:

[0953] The device converts the user's voice data into text data using early speech recognition technology. A speech recognition engine is used to convert the input voice data into text data. For example, a speech saying "I'm a little tired today" is converted into text data saying "I'm a little tired today." The output is that text data.

[0954] Step 3:

[0955] The terminal sends text data to the server. The input is the converted text data, which is sent to the server. For example, the text data "I'm a little tired today" is sent to the server via the network. The output is the text data sent to the server.

[0956] Step 4:

[0957] The server performs a detailed analysis of the received text data using advanced analysis techniques. The server combines natural language processing (NLP) techniques and voice tone analysis to evaluate the emotional and cognitive states of the input text data. For example, "Today" indicates "fatigue" as the emotional state. The output is the analysis results: "Emotional state: fatigue" and "Cognitive state."

[0958] Step 5:

[0959] The server selects the optimal retraining method based on the analysis results. The input is data on "emotional state: fatigue" and "cognitive state." For example, in the case of "fatigue," the server selects reality-oriented therapy to promote refreshment. A generative AI model (such as GPT-3) is used to generate specific questions and instructions. For example, it generates the question, "Do you know what day it is today?" The output is the selected retraining method and the generated question.

[0960] Step 6:

[0961] The server sends the selected retraining method and the generated question to the terminal. The input is the retraining method selection result and the generated question, which are sent to the terminal via the Internet. For example, a question such as "Do you know what day it is today?" is sent to the terminal. The output is the question sent to the terminal.

[0962] Step 7:

[0963] The terminal presents the received question to the user by voice. The input is question data received from the server, and the text "Do you know what day it is today?" is converted into voice and presented to the user. The voice output device is used to pose the question to the user. The output is a voice presentation to the user.

[0964] Step 8:

[0965] The user responds to a question from the terminal. The input is a voice question from the terminal, and the user responds to it by voice. For example, the user might reply, "Today is Monday." The output is the user's response voice data.

[0966] Step 9:

[0967] The terminal converts the user's response voice data into text data again and sends it to the server. The voice recognition engine is used again to convert the input voice into text data. For example, the voice saying "Today is Monday" is converted into text data. This is sent to the server. The output is the text data sent to the server.

[0968] Step 10:

[0969] The server analyzes the user's response again and adjusts the next retraining content. The input is the text data "Today is Monday," which the server analyzes to evaluate the user's emotional and cognitive state. Based on the analysis results, a new question or instruction is generated and sent to the device again. For example, a new question is generated: "Do you have any plans with your family today?" The output is the next retraining content.

[0970] Through this process, the system of the present invention analyzes the patient's voice data and provides an optimal retraining method in real time, thereby effectively stimulating the patient's cognitive function and slowing the progression of dementia.

[0971] (Application example 1)

[0972] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0973] It is important to provide appropriate rehabilitation methods for dementia patients in elderly rehabilitation facilities and nursing homes. However, conventional technologies are insufficient in providing optimal rehabilitation methods according to the emotional and cognitive states that vary from patient to patient. In particular, it is difficult to assess a patient's condition in real time and continue to provide appropriate rehabilitation methods. Therefore, there is a need for a system that provides optimal rehabilitation methods for individual patients in real time.

[0974] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0975] In this invention, the server includes: means for receiving voice input from the patient; means for transmitting the received voice data to the server; means for the server to analyze the voice data and grasp the emotional and cognitive states of the patient; means for selecting an optimal rehabilitation method based on the grasped states; means for transmitting the selected rehabilitation method to the terminal; means for the terminal to present rehabilitation content to the patient and obtain feedback; means for converting the patient's voice data into text data and evaluating the emotional state using an emotion analysis model; and means for generating specific prompt sentences based on the evaluation results to provide the rehabilitation method. This makes it possible to evaluate the emotional and cognitive states that differ for each patient in real time and provide the optimal rehabilitation method based on the evaluation.

[0976] A "means for receiving patient voice input" is a device or method for collecting patient-produced speech in real time.

[0977] The "means for transmitting received voice data to a server" refers to a device or method that has the function of transferring voice data to a server via a network.

[0978] "Means for the server to analyze voice data and understand the emotional and cognitive state of the patient" refers to technology that enables the server to process voice data and understand the emotional and cognitive state of the patient.

[0979] "Means for selecting the optimal rehabilitation method based on the grasped condition" refers to a system or algorithm that determines the most appropriate rehabilitation method based on the analysis results.

[0980] The "means for transmitting the selected rehabilitation method to the terminal" refers to a device or method for transmitting the selected rehabilitation method to the client device via a network.

[0981] "Means for a terminal to present rehabilitation content to a patient and obtain feedback" refers to equipment or methods for providing rehabilitation information to a patient and collecting their reactions again.

[0982] "Means for converting a patient's voice data into text data and assessing their emotional state using an emotion analysis model" refers to a technology for converting voice data into text and analyzing the text to assess the user's emotions.

[0983] "Means for generating specific prompt sentences based on the evaluation results and providing rehabilitation techniques" is a mechanism for creating specific questions and instructions based on the results of emotion analysis and using them to advance rehabilitation.

[0984] The following describes an embodiment of the present invention.

[0985] First, the entire system consists of a voice input device (such as a smartphone) held by the patient, a server, and a user terminal for providing feedback.

[0986] Audio input device

[0987] The voice input device is a smartphone, which can collect the patient's speech in real time and convert it into text data. The technologies used include the Python SpeechRecognition library and the Google Speech API.

[0988] server

[0989] The server analyzes the received voice data to understand the patient's emotional and cognitive state. A generative AI model is used for the analysis, and emotion analysis is performed using the Hugging Face Transformers model. Specifically, the voice data is converted into text data, which is then input into an emotion analysis model to evaluate the patient's emotional state. Based on the results of this evaluation, an appropriate rehabilitation method is selected. An appropriate prompt is also generated, and the selected rehabilitation method is sent to the user's device.

[0990] User terminal

[0991] The user device receives the rehabilitation method from the server and presents it to the patient via voice. The user device then collects the feedback again as voice and sends the data to the server. This provides the data needed to determine the next rehabilitation method.

[0992] Program processing natural language explanation

[0993] The server analyzes the patient's voice to understand their emotional and cognitive state and decides on a rehabilitation method based on that. First, the smartphone collects the voice and converts it into text using the Google Speech API. The text data is then sent to the server, where emotion analysis is performed using the Hugging Face Transformers model. Based on the analysis results, a specific rehabilitation method for the target patient is determined and an appropriate prompt is generated. This prompt is then presented to the patient via the user's device.

[0994] Specific examples

[0995] As a specific example, if a patient answers, "I feel a little tired today," the emotion analysis model will determine this as "NEGATIVE," and the server will generate a prompt such as "Is there anything bothering you?" and send it to the user's terminal.

[0996] Prompt Sentence Examples

[0997] "Good morning. How are you feeling today?"

[0998] "Tell me about the town where you used to live."

[0999] "Do you know what day it is?"

[1000] As a result, this system can provide optimal rehabilitation methods in real time based on each patient's emotional and cognitive state.

[1001] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1002] Step 1:

[1003] The user provides voice input. In this case, the patient's voice data is collected using the smartphone's microphone. For example, the patient might say, "I feel a little tired today." This voice data becomes the input.

[1004] Step 2:

[1005] The device converts the input voice into text data. Here, the Google Speech API is used to convert the voice into text. This text data is output and goes to the next analysis step.

[1006] Step 3:

[1007] The terminal sends the converted text data to the server. The text data is transferred to the server via the network. The server receives the text data.

[1008] Step 4:

[1009] The server analyzes the received text data. Specifically, it performs emotion analysis using the Hugging Face Transformers model. This analysis evaluates the patient's emotional state from the input text data. For example, the text "I'm a little tired" is evaluated as "NEGATIVE."

[1010] Step 5:

[1011] The server selects the optimal rehabilitation method based on the results of the emotion analysis and generates a specific prompt. Here, the rehabilitation method is determined based on the results of the emotion analysis model, and an appropriate prompt is generated. A prompt such as "Is there anything that is bothering you?" is generated.

[1012] Step 6:

[1013] The server sends the generated prompt to the terminal. The prompt is sent to the terminal via the network. The sent prompt is received by the terminal.

[1014] Step 7:

[1015] The terminal then presents the received prompt to the user by voice. In this case, the smartphone's text-to-speech function is used to play the prompt aloud. For example, the patient might hear, "Is there anything bothering you?"

[1016] Step 8:

[1017] The user provides vocal feedback on the presented prompt sentence, which is then input into the terminal again, leading to further analysis and the provision of rehabilitation techniques.

[1018] Through these steps, the system can analyze the patient's emotional state in real time and provide optimal rehabilitation methods. It also generates prompts and collects feedback, enabling continuous rehabilitation support.

[1019] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1020] MODE FOR CARRYING OUT THE INVENTION

[1021] This invention is a system that combines generative AI, a smart speaker, and an emotion engine to provide a rehabilitation method to slow the progression of dementia. This system analyzes the patient's condition from their voice and adjusts the rehabilitation content in real time to improve effectiveness.

[1022] System configuration

[1023] 1. Terminal

[1024] The device is a smart speaker that receives voice input from the patient, performs speech recognition at an early stage, and converts the received voice input into text data. The device also has the function of transmitting the received data to a server.

[1025] 2. Server

[1026] The server receives the voice data and converts it into text using advanced voice recognition technology. The server analyzes the tone and strength of the voice data to assess the patient's emotional and cognitive state. Based on the analysis results, it selects the most appropriate rehabilitation method and sends it to the device.

[1027] 3. Emotion Engine

[1028] The emotion engine analyzes the emotional state of the voice data in real time and feeds back the results to the server. The emotion engine has the function of analyzing the emotional state by combining tone, stress, intonation, and other non-verbal elements.

[1029] 4. Rehabilitation methods

[1030] The rehabilitation techniques include cognitive stimulation therapy, reminiscence therapy, and reality orientation. These techniques are automatically selected by the server depending on the patient's condition.

[1031] 5. Feedback

[1032] The terminal presents rehabilitation content to the patient and receives the patient's response again. The response is again sent to the server and used to determine the next rehabilitation content.

[1033] Program processing

[1034] Device behavior

[1035] The device receives the patient's voice input and sends it to the server. For example, the device may ask a simple question while greeting the patient in the morning. For example, the device may ask, "Good morning, how are you feeling today?" to collect the patient's voice.

[1036] Server Operation

[1037] The server analyzes the received voice data and evaluates the patient's emotional and cognitive state. For example, if the patient responds, "I'm a little tired today," the server analyzes the tone and strength of the voice and evaluates it as "feeling tired." Based on the results, it selects a rehabilitation method and sends it to the device.

[1038] Emotion Engine Operation

[1039] The emotion engine analyzes the patient's emotions in real time based on the voice data. For example, if "anxiety" is detected from the tone and intonation, the result is fed back to the server, which then adjusts the rehabilitation program based on that information.

[1040] Selection of rehabilitation methods

[1041] For example, if the emotional state is judged to be "calm," reminiscence techniques are used to select questions such as "Tell me about the town where you used to live." Conversely, if the emotional state is judged to be "anxious," reassuring reality orientation techniques are used to select questions such as "Do you know what day it is today?"

[1042] Present and get feedback from the device

[1043] The device then verbally communicates the selected rehabilitation content to the patient. If the patient responds, "The town where I used to live was quiet and a very nice place," the voice message is sent back to the server for analysis. This determines the next rehabilitation content.

[1044] Specific examples

[1045] 1. Basic Scenario

[1046] In the morning, the user (a dementia patient) responds to a voice message from the device about their mood that day. The device then sends the voice message to the server, which analyzes it. For example, if the user answers "I didn't have any particular dreams" in response to the question "Good morning. What kind of dreams did you have today?", the server will analyze the response and evaluate the user's emotional state as "Calm." Based on the results, the server will select a reminiscence method and send the next question to the device: "What is your fondest memory from childhood?"

[1047] 2. Utilizing the Emotion Engine

[1048] An emotion engine has been added to more accurately grasp the patient's emotional state. For example, if a patient responds, "I'm a little tired today," the emotion engine analyzes the tone and intonation of the response and evaluates it as "fatigue." This result is fed back to the server, which then adjusts the rehabilitation program based on that information.

[1049] 3. Utilizing a variety of rehabilitation methods

[1050] If the patient's emotional state is evaluated as "anxious," the server selects reality orientation. The terminal asks, "What are your plans for today?" and the patient answers, "I have an appointment to meet my family today." The server evaluates the patient's sense of security.

[1051] In this way, the system of the present invention can analyze the patient's voice data and provide optimal rehabilitation techniques in real time, which can effectively stimulate the patient's cognitive function and slow the progression of dementia. The addition of an emotion engine further improves the accuracy and effectiveness of rehabilitation content.

[1052] The processing flow will be explained below.

[1053] Step 1:

[1054] The terminal receives the patient's voice input, asks the patient questions, and records their responses as audio.

[1055] For example, ask, "Good morning, how are you feeling today?"

[1056] Step 2:

[1057] The device sends the received voice data to the server, where it is converted into text using early-stage voice recognition technology.

[1058] Audio data: "I'm a little tired today."

[1059] Text data: "I'm a little tired today."

[1060] Step 3:

[1061] The server receives the transmitted voice and text data and converts it back into text using advanced voice recognition technology to improve accuracy.

[1062] Audio data: "I'm a little tired today."

[1063] Text data: "I'm a little tired today" (Improved accuracy)

[1064] Step 4:

[1065] The server analyzes the audio and text data and uses an emotion engine to assess the patient's emotional and cognitive state. It identifies the patient's emotional state based on the tone, stress, inflection, and content of the audio.

[1066] Analysis result: "Emotional state: Fatigue"

[1067] Step 5:

[1068] The server selects the optimal rehabilitation method based on the analysis results. For example, if the emotional state is evaluated as "fatigue," it selects a calm and relaxing reminiscence method.

[1069] Selection result: “Reminiscence method”

[1070] Step 6:

[1071] The server generates specific rehabilitation content based on the selected rehabilitation method, such as "Tell us about your favorite hobbies from the past" for reminiscence therapy.

[1072] Generated rehabilitation content: "Tell me about your favorite past hobbies."

[1073] Step 7:

[1074] The server sends the generated rehabilitation content to the terminal. The rehabilitation content is sent as text data, but also includes information for voice synthesis.

[1075] Submitted data: "Tell me about your favorite past hobbies."

[1076] Step 8:

[1077] The device uses voice synthesis technology to ask the patient about the rehabilitation information it has received, and outputs the questions as voice.

[1078] Voice output: "Tell me about your favorite past hobbies."

[1079] Step 9:

[1080] The user (patient) responds to the questions posed by the device. For example, they might reply, "I used to love reading."

[1081] Response: "I used to love reading."

[1082] Step 10:

[1083] The terminal again receives the patient's response as voice data and sends it to the server, where it is also converted into text data.

[1084] Audio data: "I used to love reading."

[1085] Text data: "I used to love reading."

[1086] Step 11:

[1087] The server analyzes the received feedback data and determines the next rehabilitation method. The emotion engine runs again and evaluates the patient's condition. A loop process is performed to continuously evaluate the patient's condition.

[1088] Analysis results: "Emotional state: calm, rehabilitation method: continued reminiscence therapy"

[1089] This series of steps is repeated to provide optimal rehabilitation for the patient, with the aim of slowing the progression of dementia. The addition of the emotion engine further improves the accuracy and effectiveness of the rehabilitation content.

[1090] Example 2

[1091] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1092] Currently, rehabilitation systems on the market lack the ability to properly grasp a patient's emotional and cognitive state. As a result, the rehabilitation methods provided are often not optimal for the patient. Furthermore, the lack of a function to obtain real-time feedback and adjust the rehabilitation content limits the effectiveness of rehabilitation. There is a need to solve these problems and provide the most effective rehabilitation methods for patients.

[1093] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for analyzing voice data and grasping the emotional and cognitive states of the patient, a means for an emotion engine to analyze the voice data in real time and provide feedback to the server, and a means for selecting an optimal rehabilitation method. This makes it possible to provide an optimal rehabilitation method in real time based on the patient's condition and maximize its effectiveness.

[1094] "Patient" refers to an individual who is a target of using the rehabilitation system.

[1095] "Voice input" refers to the words or voice data uttered by the patient.

[1096] "Terminal" refers to an electronic device that receives the patient's voice input and performs the necessary processing.

[1097] "Server" refers to a central computer system that analyzes data sent from terminals and performs necessary processing and feedback.

[1098] "Voice data" refers to a digital representation of a patient's voice input.

[1099] "Emotion engine" refers to a software or hardware component for analyzing audio data and assessing emotional state.

[1100] "Emotional state" refers to the psychological state of the patient as judged from the tone and strength of their voice.

[1101] "Cognitive status" refers to the state of a patient's cognitive function.

[1102] "Rehabilitation techniques" refer to therapies provided based on the patient's emotional and cognitive state.

[1103] "Real-time" refers to processing that occurs at a speed close to instantaneous.

[1104] This invention is a system that combines a generative AI model, a smart speaker, and an emotion engine to provide a rehabilitation method to slow the progression of dementia. This system analyzes the patient's condition from their voice and adjusts the rehabilitation content in real time to improve effectiveness.

[1105] System configuration

[1106] 1. Terminal

[1107] The device consists of a smart speaker that receives the patient's voice input. Specifically, the smart speaker picks up the patient's voice using a microphone, performs initial voice recognition, and converts it into text data. The device also has the function of sending this data to a server. For example, the device may ask the patient, "Good morning, how are you feeling today?"

[1108] 2. Server

[1109] The server receives the voice data sent from the device and converts it into text data using more advanced voice recognition technology. One possible technology is the Google Cloud Speech-to-Text API. The server then analyzes the tone and strength of the voice data to evaluate the patient's emotional and cognitive state. For example, the server evaluates the voice data, such as "I'm a little tired today," as "I feel fatigued." Based on the analysis results, the server selects the optimal rehabilitation method and sends it to the device.

[1110] 3. Emotion Engine

[1111] The emotion engine analyzes the emotional state of the voice data in real time and feeds the results back to the server. Specific analysis items include tone, stress, intonation, and other non-verbal elements. For example, if "anxiety" is detected from the intonation of a patient's voice, the result is fed back to the server.

[1112] 4. Rehabilitation methods

[1113] Rehabilitation techniques include cognitive stimulation therapy, reminiscence therapy, and reality orientation. These techniques are automatically selected by the server depending on the patient's condition. For example, if the emotional state is judged to be "calm," reminiscence therapy will be used and questions such as "Tell me about the town where you used to live" will be selected.

[1114] 5. Feedback

[1115] The device presents rehabilitation content to the patient and receives the patient's response again. The received data is sent to the server and used to determine the next rehabilitation content. For example, if the patient responds, "The town where I used to live was quiet and a very nice place," the voice data is sent to the server and analyzed.

[1116] Specific examples

[1117] Basic Scenario

[1118] In the morning, the user (a dementia patient) answers questions posed by the device. For example, to the question, "Good morning. What kind of dreams did you have today?", the user replies, "I didn't have any particular dreams." The server analyzes the voice data and evaluates the user's emotional state as "calm." The server selects the reminiscence method and sends the next question, "What is your best childhood memory?" to the device. The device presents the question to the patient, who responds. The server then analyzes the response again to determine the next rehabilitation plan.

[1119] Utilizing the Emotion Engine

[1120] The emotion engine analyzes the tone and intonation of the patient's voice. For example, if a patient replies, "I'm a little tired today," the emotion engine analyzes the tone and intonation and evaluates it as "feeling tired." This evaluation result is fed back to the server, which then adjusts the rehabilitation program based on that information.

[1121] Utilizing a variety of rehabilitation methods

[1122] If the patient's emotional state is evaluated as "anxious," the server selects reality orientation. The device asks, "What are your plans for today?" and the patient replies, "I have an appointment to meet my family today." The server analyzes this information and evaluates the patient's sense of security. This determines the next rehabilitation content.

[1123] Prompt Sentence Examples

[1124] "To provide rehabilitation techniques for dementia patients, suggest optimal questions based on their emotional and cognitive states. For example, if a patient says, 'I'm a little tired today,' analyze their speech data and generate the appropriate next question."

[1125] In this way, the system of the present invention can analyze the patient's voice data and provide optimal rehabilitation methods in real time, which can effectively stimulate the patient's cognitive function and slow the progression of dementia. The addition of an emotion engine further improves the accuracy and effectiveness of rehabilitation content.

[1126] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1127] Step 1:

[1128] The device receives voice input. The user speaks to the smart speaker. For example, the device asks, "Good morning, how are you feeling today?" and the user replies, "I'm a little tired today." This voice data is input into the device.

[1129] Input: User's voice data: "I'm a little tired today."

[1130] Output: Received audio data

[1131] Step 2:

[1132] The voice data received by the device is converted into text data using an initial voice recognition engine. For example, the voice recognition technology built into a smart speaker can be used to convert the voice into text, such as "I'm a little tired today." This text data is then sent to the server.

[1133] Input: Received audio data

[1134] Output: Text data "I'm a little tired today"

[1135] Step 3:

[1136] The server uses advanced speech recognition technology to convert the voice data sent from the device back into text data, and then uses advanced technology such as the Google Cloud Speech-to-Text API to analyze the details of the voice and generate text data.

[1137] Input: Audio data sent from the device

[1138] Output: Highly refined text data "I'm a little tired today"

[1139] Step 4:

[1140] The server performs detailed analysis of the voice data and evaluates the emotional state using an emotion engine. The server analyzes the tone, strength, and intonation of the voice to identify the emotional state. For example, it may evaluate the voice as "fatigue" based on its low tone and strength.

[1141] Input: Highly refined text data and its speech characteristics

[1142] Output: Emotional state "Feeling tired"

[1143] Step 5:

[1144] The emotion engine feeds back the analyzed emotional state to the server. The emotion engine determines "fatigue" based on the tone and intonation of the voice data and sends the evaluation result to the server.

[1145] Input: Tone and intonation analysis data of speech data

[1146] Output: Emotional state assessment result: "Feeling tired"

[1147] Step 6:

[1148] The server selects the optimal rehabilitation method based on the emotional and cognitive states. Based on the result "feeling tired," the server selects cognitive stimulation therapy as the rehabilitation method and determines its specific content. For example, it selects the question, "Do you know what day it is today?"

[1149] Input: Emotional state assessment result: "Feeling tired"

[1150] Output: The selected rehabilitation technique "Cognitive Stimulation Therapy" and the question "Do you know what day it is today?"

[1151] Step 7:

[1152] The server sends the selected rehabilitation method and its details to the device. As part of the cognitive stimulation therapy, the server sends the question "Do you know what day it is today?" to the device.

[1153] Input: Selected rehabilitation method and specific question

[1154] Output: Instruction data to the terminal

[1155] Step 8:

[1156] The device then presents the received rehabilitation information to the user via voice. For example, the device might ask the user, "Do you know what day it is today?" The user might respond, "Today is Tuesday."

[1157] Input: Rehabilitation details sent from the server

[1158] Output: Audio presentation to the user

[1159] Step 9:

[1160] The user responds to the rehabilitation content. When the user responds, "Today is Tuesday," the terminal receives the voice again. This received data is sent to the server again.

[1161] Input: User's voice reply "Today is Tuesday"

[1162] Output: Response data to the terminal

[1163] Step 10:

[1164] The server analyzes the voice data again and determines the next rehabilitation content based on the results. For example, based on the response "Today is Tuesday," the server selects the next rehabilitation method and sends the next question, "What did you do on Tuesday?" to the terminal.

[1165] Input: User response data received again

[1166] Output: Next rehabilitation technique and specific questions

[1167] This series of processes enables feedback tailored to the user's emotional and cognitive state, providing optimal rehabilitation techniques. This system is expected to slow the progression of dementia in patients.

[1168] (Application example 2)

[1169] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1170] Rehabilitation for dementia patients requires selecting the optimal method to suit each individual patient's condition and providing effective treatment. However, with conventional methods, it has been difficult to accurately grasp the patient's emotional and cognitive state in real time and provide the optimal rehabilitation method based on that. In particular, more advanced technology is required to realize rehabilitation in a virtual environment. The purpose of this project is to solve these issues.

[1171] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1172] In this invention, the server includes means for receiving voice input from the patient, means for transmitting the received voice data to the server, means for analyzing the voice data to grasp the emotional and cognitive states of the patient, means for selecting an optimal rehabilitation method based on the grasped states, means for transmitting the selected rehabilitation method to the terminal, means for the terminal to present rehabilitation content to the patient and obtain feedback, and means for presenting rehabilitation content in a virtual environment. This makes it possible to provide the rehabilitation method optimal for the patient's condition in real time, thereby effectively slowing the progression of dementia.

[1173] Key Word Definitions

[1174] "Means for receiving patient voice input" refers to a device for collecting voice data from a dementia patient, often a microphone or smart speaker.

[1175] "Means for transmitting received voice data to a server" refers to a function for transferring collected voice data to a central computer (server) via a network.

[1176] "Means for analyzing voice data to understand the emotional and cognitive state of a patient" refers to algorithms or analytical software for analyzing received voice data and determining the emotional and cognitive state of a patient.

[1177] "Means for selecting the most appropriate rehabilitation method based on the identified condition" refers to the logic and process for automatically selecting the rehabilitation method that is most appropriate for the patient's current condition based on the results of the analysis.

[1178] "Means for transmitting the selected rehabilitation method to the terminal" refers to a communication means for transmitting information about the rehabilitation method selected by the server to the patient's terminal.

[1179] "Means for the device to present rehabilitation content to the patient and obtain feedback" refers to an interface that presents selected rehabilitation content through audio or a virtual environment and collects the patient's reactions and answers again.

[1180] "Means for presenting rehabilitation content in a virtual environment" refers to software or hardware that uses virtual reality or augmented reality technology to provide patients with visual or auditory rehabilitation content.

[1181] MODE FOR CARRYING OUT THE INVENTION

[1182] To illustrate an embodiment of the present invention, the following system configuration and operation will be described.

[1183] System Configuration

[1184] This system consists of the following main hardware and software:

[1185] Hardware

[1186] Smart glasses or head-mounted displays (HMD)

[1187] microphone

[1188] Cloud Server

[1189] software

[1190] Speech recognition library (e.g., speech_recognition library)

[1191] Sentiment analysis engine (e.g., custom-developed emotion_engine module)

[1192] Rehabilitation method selection logic (e.g., custom-developed rehab_methods module)

[1193] Virtual environment interface (e.g., a custom-developed virtual_environment module)

[1194] A natural language explanation of how to proceed

[1195] 1. Receiving and sending audio data

[1196] The user's voice input is received by a microphone attached to the smart glasses or head-mounted display, and the received voice data is transmitted to a cloud server in real time.

[1197] 2. Analysis of audio data

[1198] The server converts the received voice data into text data using a speech recognition library. It then uses an emotion analysis engine to analyze the tone and strength of the text data to understand the user's emotional and cognitive state. Based on the results of this analysis, the server selects the most appropriate rehabilitation method for the user.

[1199] 3. Selection of rehabilitation method

[1200] Based on the analyzed data, the server uses a rehabilitation method selection logic to determine the appropriate rehabilitation content. The rehabilitation method is automatically selected from cognitive stimulation therapy, reminiscence therapy, and reality orientation.

[1201] 4. Presentation of rehabilitation content

[1202] The selected rehabilitation content is sent from the cloud server to smart glasses or a head-mounted display, and the user can receive the rehabilitation content visually or audibly through virtual reality or augmented reality using a virtual environment interface.

[1203] 5. Get feedback

[1204] If the user provides vocal feedback on the rehabilitation content, the audio is again picked up by the microphone and sent to the cloud server, and this process further adjusts the next rehabilitation content.

[1205] Specific examples

[1206] For example, if a user is asked, "What kind of dream did you have today?" and the user answers, "I didn't have any particular dreams," this voice data is sent to the server, where it is analyzed and the emotional state is evaluated as "calm" based on the tone and intensity of the voice. Based on this evaluation, the server selects a reminiscence method and presents the next question, "What is your best childhood memory?" A nostalgic scene is displayed within the virtual environment.

[1207] Prompt Sentence Examples

[1208] Emotional state: Calm

[1209] Q: Tell me about the town where you used to live.

[1210] In this way, the system can analyze the user's voice data and provide optimal rehabilitation techniques in real time, which can effectively stimulate the user's cognitive functions and slow the progression of dementia.

[1211] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1212] Program processing flow

[1213] Step 1:

[1214] The device receives the user's voice input. The voice data is captured by a microphone and sent to a cloud server in real time. The input is the user's voice, and the output is sent to the server as voice data.

[1215] Step 2:

[1216] The server receives the voice data and converts it into text data using a speech recognition library. The input is voice data and the output is text data. The speech recognition library analyzes the sound wave data and converts it into text based on a language model.

[1217] Step 3:

[1218] The server inputs the converted text data into an emotion analysis engine, which analyzes the emotional and cognitive states. The input is text data, and the output is an evaluation of the emotional and cognitive states. The emotion analysis engine takes into account non-verbal elements such as tone, strength, and intonation during analysis.

[1219] Step 4:

[1220] Based on the analysis results, the server uses the rehabilitation method selection logic to select the optimal rehabilitation method. The input is the evaluation results of the emotional and cognitive states, and the output is the selected rehabilitation method. The rehabilitation method selection logic compares the application conditions of each rehabilitation method with the current evaluation results and selects the most appropriate method.

[1221] Step 5:

[1222] The server sends the selected rehabilitation technique to the terminal. The input is the selected rehabilitation technique, and the output is the rehabilitation technique information sent to the terminal. The selected information is delivered to the terminal using a communication protocol.

[1223] Step 6:

[1224] The device presents rehabilitation content to the user through a virtual environment interface. The input is rehabilitation method information, and the output is a virtual rehabilitation environment that the user perceives visually or audibly. Visual content is displayed and played on a head-mounted display, and audio content is displayed and played on speakers.

[1225] Step 7:

[1226] The user responds to the presented rehabilitation content and takes action, and their reactions are collected again as voice data by the device. The input is the user's voice response, and the output is again sent to the server as voice data.

[1227] Step 8:

[1228] The server analyzes the user's feedback and adjusts the next rehabilitation content. The input is the user's voice data and the rehabilitation method results, and the output is the adjusted rehabilitation method. The server reuses the emotion analysis engine and rehabilitation method selection logic to reselect the optimal rehabilitation method that reflects the feedback.

[1229] Through these steps, the system can adapt flexibly to the user's condition and provide effective rehabilitation.

[1230] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1231] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1232] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1233] [Fourth embodiment]

[1234] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1235] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1236] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1237] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1238] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1239] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1240] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1241] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1242] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1243] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1244] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1245] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1246] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1247] MODE FOR CARRYING OUT THE INVENTION

[1248] This invention is a system for slowing the progression of dementia patients, linking generative AI with a smart speaker to provide optimal rehabilitation methods for each patient. This system analyzes the patient's condition from their voice and adjusts the rehabilitation content in real time to enhance effectiveness.

[1249] System configuration

[1250] 1. Terminal

[1251] The device is a smart speaker that receives voice input from the patient, performs speech recognition at an early stage, and converts the received voice input into text data. The device also has the function of transmitting the received data to a server.

[1252] 2. Server

[1253] The server receives the voice data and converts it into text using advanced voice recognition technology. The server analyzes the tone and strength of the voice data to assess the patient's emotional and cognitive state. Based on the analysis results, the server selects the most appropriate rehabilitation method and transmits the details to the device.

[1254] 3. Rehabilitation methods

[1255] The rehabilitation techniques include cognitive stimulation therapy, reminiscence therapy, and reality orientation. These techniques are automatically selected by the server depending on the patient's condition.

[1256] 4. Feedback

[1257] The terminal presents rehabilitation content to the patient and receives the patient's response again. The response is again sent to the server and used to determine the next rehabilitation content.

[1258] Program processing

[1259] Device behavior

[1260] The device receives the patient's voice input and sends it to the server. For example, the device may ask a simple question while greeting the patient in the morning. For example, the device may ask, "Good morning, how are you feeling today?" to collect the patient's voice.

[1261] Server Operation

[1262] The server analyzes the received voice data and evaluates the patient's emotional and cognitive state. For example, if the patient responds, "I'm a little tired today," the server analyzes the tone and strength of the voice and evaluates it as "feeling tired." Based on the results, it selects a rehabilitation method and sends it to the device.

[1263] Selection of rehabilitation methods

[1264] For example, if the emotional state is judged to be "calm," reminiscence techniques are used to select questions such as "Tell me about the town where you used to live." Conversely, if the emotional state is judged to be "anxious," reassuring reality orientation techniques are used to select questions such as "Do you know what day it is today?"

[1265] Present and get feedback from the device

[1266] The device then verbally communicates the selected rehabilitation content to the patient. If the patient responds, "The town where I used to live was quiet and a very nice place," the voice message is sent back to the server for analysis. This determines the next rehabilitation content.

[1267] Specific examples

[1268] 1. Basic Scenario

[1269] In the morning, the user (a dementia patient) responds to a voice message from the device about their mood that day. The device then sends the voice message to the server, which analyzes it. For example, if the user answers "I didn't have any particular dreams" in response to the question "Good morning. What kind of dreams did you have today?", the server will analyze the response and evaluate the user's emotional state as "Calm." Based on the results, the server will select a reminiscence method and send the next question to the device: "What is your fondest memory from childhood?"

[1270] 2. Utilizing a variety of rehabilitation methods

[1271] If the patient's emotional state is evaluated as "anxious," the server selects reality orientation. The terminal asks, "What are your plans for today?" and the patient answers, "I have an appointment to meet my family today." The server evaluates the patient's sense of security.

[1272] In this way, the system of the present invention can analyze the patient's voice data and provide optimal rehabilitation techniques in real time, thereby effectively stimulating the patient's cognitive function and slowing the progression of dementia.

[1273] The processing flow will be explained below.

[1274] Step 1:

[1275] The terminal receives the patient's voice input, asks the patient questions, and records their responses as audio.

[1276] For example, ask, "Good morning, how are you feeling today?"

[1277] Step 2:

[1278] The device sends the received voice data to the server, where it is converted into text using early-stage voice recognition technology.

[1279] Audio data: "I'm a little tired today."

[1280] Text data: "I'm a little tired today."

[1281] Step 3:

[1282] The server receives the transmitted voice and text data and converts it back into text using advanced voice recognition technology to improve accuracy.

[1283] Audio data: "I'm a little tired today."

[1284] Text data: "I'm a little tired today" (Improved accuracy)

[1285] Step 4:

[1286] The server analyzes the audio and text data to assess the patient's emotional and cognitive state. It identifies the patient's emotional state based on the tone, stress, inflection, and content of the audio.

[1287] Analysis result: "Emotional state: Fatigue"

[1288] Step 5:

[1289] The server selects the optimal rehabilitation method based on the analysis results. For example, if the emotional state is evaluated as "fatigue," it selects a calm and relaxing reminiscence method.

[1290] Selection result: “Reminiscence method”

[1291] Step 6:

[1292] The server generates specific rehabilitation content based on the selected rehabilitation method, such as "Tell us about your favorite hobbies from the past" for reminiscence therapy.

[1293] Generated rehabilitation content: "Tell me about your favorite past hobbies."

[1294] Step 7:

[1295] The server sends the generated rehabilitation content to the terminal. The rehabilitation content is sent as text data, but also includes information for voice synthesis.

[1296] Submitted data: "Tell me about your favorite past hobbies."

[1297] Step 8:

[1298] The device uses voice synthesis technology to ask the patient about the rehabilitation information it has received, and outputs the questions as voice.

[1299] Voice output: "Tell me about your favorite past hobbies."

[1300] Step 9:

[1301] The user (patient) responds to the questions posed by the device. For example, they might reply, "I used to love reading."

[1302] Response: "I used to love reading."

[1303] Step 10:

[1304] The terminal again receives the patient's response as voice data and sends it to the server, where it is also converted into text data.

[1305] Audio data: "I used to love reading."

[1306] Text data: "I used to love reading."

[1307] Step 11:

[1308] The server analyzes the received feedback data and determines the next rehabilitation technique. A loop process is performed to continuously evaluate the patient's condition.

[1309] Analysis results: "Emotional state: calm, rehabilitation method: continued reminiscence therapy"

[1310] This series of steps is repeated to provide optimal rehabilitation for the patient, with the aim of slowing the progression of dementia.

[1311] Example 1

[1312] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1313] Currently, rehabilitation methods for dementia patients are often based on uniform programs, making it difficult to provide individualized training content that takes into account each patient's emotional and cognitive states in real time. Furthermore, there are limited methods for assessing a patient's emotional and cognitive states, making it difficult to select effective rehabilitation methods. As a result, there is a problem in that the progression of dementia cannot be effectively slowed.

[1314] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1315] In this invention, the server includes a means for analyzing the tone and strength of the patient's voice data to identify emotions, a means for grasping the patient's emotional and cognitive states, and a means for automatically selecting one of cognitive stimulation therapy, reminiscence therapy, and reality-focused therapy as a retraining method, thereby enabling the provision of an individual rehabilitation method based on the patient's emotional and cognitive states.

[1316] "Patient" refers to a person who has a progressive disease, such as dementia, and requires rehabilitation.

[1317] "Voice input" refers to the voice data uttered by the patient, which provides basic information for analyzing their emotions and cognitive state.

[1318] A "terminal" is a device that has the function of receiving voice input from a patient and transmitting it to an information processing device. Specifically, this applies to smart speakers.

[1319] "Voice data" refers to digitally recorded speech made by a patient and is used to analyze their emotional and cognitive states.

[1320] "Information processing device" refers to a device that processes received voice data using advanced analytical technology to evaluate a patient's emotional and cognitive state. This includes servers and cloud computing services.

[1321] "Emotional state" refers to the patient's mental state, and is information analyzed from the tone and strength of the voice data.

[1322] "Cognitive status" refers to the patient's cognitive functions, such as intellectual processing ability and memory, and is information evaluated from the content of the voice data and responses.

[1323] "Retraining techniques" refer to the most appropriate rehabilitation methods based on the patient's emotional and cognitive state. These include cognitive stimulation therapy, reminiscence therapy, and reality-oriented therapy.

[1324] "Cognitive stimulation therapy" is a type of training method for maintaining and improving cognitive function in the brain. Specifically, it is carried out using puzzles and calculation problems.

[1325] Reminiscence therapy is a rehabilitation method that stimulates memory and maintains cognitive function by discussing past events.

[1326] "Reality-oriented therapy" is a rehabilitation method that promotes awareness of reality and strengthens cognitive function by having patients talk about real times, places, and everyday events.

[1327] "Feedback" refers to the responses and reactions received by the terminal from the patient, and is data that the information processing device uses to perform further analysis based on that.

[1328] MODE FOR CARRYING OUT THE INVENTION

[1329] This invention is a system for slowing the progression of dementia in patients, linking a generative AI with a voice input device to provide optimal retraining methods for each patient. This system analyzes the emotional and cognitive state of the patient from their voice and adjusts the retraining content in real time to enhance effectiveness.

[1330] System configuration

[1331] 1. Terminal

[1332] The terminal is a voice input device for receiving voice input from the patient. The voice input device receives the patient's voice and converts it into text data through speech recognition at an early stage. Examples of this include smart speakers such as Amazon Alexa and Google Home. The terminal also has the function of transmitting the received data to an information processing device.

[1333] 2. Server (information processing device)

[1334] The server receives the voice data and converts it into text using more advanced speech recognition technology. The server analyzes the tone and intensity of the voice data to assess the patient's emotional and cognitive state. The analysis uses advanced voice analysis tools such as Google Cloud Speech-to-Text and IBM Watson. Based on the analysis results, the server selects the optimal retraining method and sends it to the device.

[1335] 3. Retraining method

[1336] Retraining techniques include cognitive stimulation therapy, reminiscence therapy, and reality-oriented therapy. These methods are automatically selected by the server depending on the patient's condition. Specific questions and instructions are generated using a generative AI model (e.g., GPT-3) built into the server.

[1337] 4. Feedback

[1338] The terminal presents the retraining content to the patient by voice and receives the patient's response again, which is then sent back to the server and used to determine the next retraining content.

[1339] Specific actions

[1340] Device behavior

[1341] The device receives the patient's voice input and sends it to the server. For example, the device may ask a simple question while greeting the patient in the morning. For example, the device may ask, "Good morning, how are you feeling today?" to collect the patient's voice.

[1342] Server Operation

[1343] The server performs advanced analysis of the received voice data to evaluate the patient's emotional and cognitive state. For example, if the patient responds, "I'm a little tired today," the server analyzes the tone and strength of the voice and evaluates the patient's emotional state as "fatigue." Based on the results, it selects a retraining method and sends it to the device.

[1344] Selection of retraining methods

[1345] For example, if the emotional state is determined to be "calm," reminiscence therapy is used to select questions such as "Tell me about the town where you used to live." Conversely, if the emotional state is determined to be "anxious," reality-oriented therapy, which provides reassurance, is used to select questions such as "Do you know what day it is today?" The generative AI model is used to input the following prompt sentences and generate appropriate questions.

[1346] Example prompt sentence:

[1347] "Create questions about how you're feeling today and your past memories to suggest rehabilitation techniques for dementia patients."

[1348] Present and get feedback from the device

[1349] The device then verbally communicates the selected retraining content to the patient. For example, if the device asks, "Tell me about the town where you used to live," and the patient replies, "The town where I used to live was quiet and a very nice place," the device sends the voice back to the server for analysis. This determines the next retraining content.

[1350] In this way, the system of the present invention can analyze the patient's voice data and provide optimal retraining techniques in real time, which can effectively stimulate the patient's cognitive function and slow the progression of dementia.

[1351] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1352] Program processing flow

[1353] Step 1:

[1354] A user speaks into a voice input device (terminal). The voice input device receives the user's voice in real time. The input is the user's voice data, for example, a voice replying "I'm a little tired today" to a question from the terminal such as "Good morning. How are you feeling today?" The output is raw voice data acquired in real time.

[1355] Step 2:

[1356] The device converts the user's voice data into text data using early speech recognition technology. A speech recognition engine is used to convert the input voice data into text data. For example, a speech saying "I'm a little tired today" is converted into text data saying "I'm a little tired today." The output is that text data.

[1357] Step 3:

[1358] The terminal sends text data to the server. The input is the converted text data, which is sent to the server. For example, the text data "I'm a little tired today" is sent to the server via the network. The output is the text data sent to the server.

[1359] Step 4:

[1360] The server performs a detailed analysis of the received text data using advanced analysis techniques. The server combines natural language processing (NLP) techniques and voice tone analysis to evaluate the emotional and cognitive states of the input text data. For example, "Today" indicates "fatigue" as the emotional state. The output is the analysis results: "Emotional state: fatigue" and "Cognitive state."

[1361] Step 5:

[1362] The server selects the optimal retraining method based on the analysis results. The input is data on "emotional state: fatigue" and "cognitive state." For example, in the case of "fatigue," the server selects reality-oriented therapy to promote refreshment. A generative AI model (such as GPT-3) is used to generate specific questions and instructions. For example, it generates the question, "Do you know what day it is today?" The output is the selected retraining method and the generated question.

[1363] Step 6:

[1364] The server sends the selected retraining method and the generated question to the terminal. The input is the retraining method selection result and the generated question, which are sent to the terminal via the Internet. For example, a question such as "Do you know what day it is today?" is sent to the terminal. The output is the question sent to the terminal.

[1365] Step 7:

[1366] The terminal presents the received question to the user by voice. The input is question data received from the server, and the text "Do you know what day it is today?" is converted into voice and presented to the user. The voice output device is used to pose the question to the user. The output is a voice presentation to the user.

[1367] Step 8:

[1368] The user responds to a question from the terminal. The input is a voice question from the terminal, and the user responds to it by voice. For example, the user might reply, "Today is Monday." The output is the user's response voice data.

[1369] Step 9:

[1370] The terminal converts the user's response voice data into text data again and sends it to the server. The voice recognition engine is used again to convert the input voice into text data. For example, the voice saying "Today is Monday" is converted into text data. This is sent to the server. The output is the text data sent to the server.

[1371] Step 10:

[1372] The server analyzes the user's response again and adjusts the next retraining content. The input is the text data "Today is Monday," which the server analyzes to evaluate the user's emotional and cognitive state. Based on the analysis results, a new question or instruction is generated and sent to the device again. For example, a new question is generated: "Do you have any plans with your family today?" The output is the next retraining content.

[1373] Through this process, the system of the present invention analyzes the patient's voice data and provides an optimal retraining method in real time, thereby effectively stimulating the patient's cognitive function and slowing the progression of dementia.

[1374] (Application example 1)

[1375] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1376] It is important to provide appropriate rehabilitation methods for dementia patients in elderly rehabilitation facilities and nursing homes. However, conventional technologies are insufficient in providing optimal rehabilitation methods according to the emotional and cognitive states that vary from patient to patient. In particular, it is difficult to assess a patient's condition in real time and continue to provide appropriate rehabilitation methods. Therefore, there is a need for a system that provides optimal rehabilitation methods for individual patients in real time.

[1377] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1378] In this invention, the server includes: means for receiving voice input from the patient; means for transmitting the received voice data to the server; means for the server to analyze the voice data and grasp the emotional and cognitive states of the patient; means for selecting an optimal rehabilitation method based on the grasped states; means for transmitting the selected rehabilitation method to the terminal; means for the terminal to present rehabilitation content to the patient and obtain feedback; means for converting the patient's voice data into text data and evaluating the emotional state using an emotion analysis model; and means for generating specific prompt sentences based on the evaluation results to provide the rehabilitation method. This makes it possible to evaluate the emotional and cognitive states that differ for each patient in real time and provide the optimal rehabilitation method based on the evaluation.

[1379] A "means for receiving patient voice input" is a device or method for collecting patient-produced speech in real time.

[1380] The "means for transmitting received voice data to a server" refers to a device or method that has the function of transferring voice data to a server via a network.

[1381] "Means for the server to analyze voice data and understand the emotional and cognitive state of the patient" refers to technology that enables the server to process voice data and understand the emotional and cognitive state of the patient.

[1382] "Means for selecting the optimal rehabilitation method based on the grasped condition" refers to a system or algorithm that determines the most appropriate rehabilitation method based on the analysis results.

[1383] The "means for transmitting the selected rehabilitation method to the terminal" refers to a device or method for transmitting the selected rehabilitation method to the client device via a network.

[1384] "Means for a terminal to present rehabilitation content to a patient and obtain feedback" refers to equipment or methods for providing rehabilitation information to a patient and collecting their reactions again.

[1385] "Means for converting a patient's voice data into text data and assessing their emotional state using an emotion analysis model" refers to a technology for converting voice data into text and analyzing the text to assess the user's emotions.

[1386] "Means for generating specific prompt sentences based on the evaluation results and providing rehabilitation techniques" is a mechanism for creating specific questions and instructions based on the results of emotion analysis and using them to advance rehabilitation.

[1387] The following describes an embodiment of the present invention.

[1388] First, the entire system consists of a voice input device (such as a smartphone) held by the patient, a server, and a user terminal for providing feedback.

[1389] Audio input device

[1390] The voice input device is a smartphone, which can collect the patient's speech in real time and convert it into text data. The technologies used include the Python SpeechRecognition library and the Google Speech API.

[1391] server

[1392] The server analyzes the received voice data to understand the patient's emotional and cognitive state. A generative AI model is used for the analysis, and emotion analysis is performed using the Hugging Face Transformers model. Specifically, the voice data is converted into text data, which is then input into an emotion analysis model to evaluate the patient's emotional state. Based on the results of this evaluation, an appropriate rehabilitation method is selected. An appropriate prompt is also generated, and the selected rehabilitation method is sent to the user's device.

[1393] User terminal

[1394] The user device receives the rehabilitation method from the server and presents it to the patient via voice. The user device then collects the feedback again as voice and sends the data to the server. This provides the data needed to determine the next rehabilitation method.

[1395] Program processing natural language explanation

[1396] The server analyzes the patient's voice to understand their emotional and cognitive state and decides on a rehabilitation method based on that. First, the smartphone collects the voice and converts it into text using the Google Speech API. The text data is then sent to the server, where emotion analysis is performed using the Hugging Face Transformers model. Based on the analysis results, a specific rehabilitation method for the target patient is determined and an appropriate prompt is generated. This prompt is then presented to the patient via the user's device.

[1397] Specific examples

[1398] As a specific example, if a patient answers, "I feel a little tired today," the emotion analysis model will determine this as "NEGATIVE," and the server will generate a prompt such as "Is there anything bothering you?" and send it to the user's terminal.

[1399] Prompt Sentence Examples

[1400] "Good morning. How are you feeling today?"

[1401] "Tell me about the town where you used to live."

[1402] "Do you know what day it is?"

[1403] As a result, this system can provide optimal rehabilitation methods in real time based on each patient's emotional and cognitive state.

[1404] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1405] Step 1:

[1406] The user provides voice input. In this case, the patient's voice data is collected using the smartphone's microphone. For example, the patient might say, "I feel a little tired today." This voice data becomes the input.

[1407] Step 2:

[1408] The device converts the input voice into text data. Here, the Google Speech API is used to convert the voice into text. This text data is output and goes to the next analysis step.

[1409] Step 3:

[1410] The terminal sends the converted text data to the server. The text data is transferred to the server via the network. The server receives the text data.

[1411] Step 4:

[1412] The server analyzes the received text data. Specifically, it performs emotion analysis using the Hugging Face Transformers model. This analysis evaluates the patient's emotional state from the input text data. For example, the text "I'm a little tired" is evaluated as "NEGATIVE."

[1413] Step 5:

[1414] The server selects the optimal rehabilitation method based on the results of the emotion analysis and generates a specific prompt. Here, the rehabilitation method is determined based on the results of the emotion analysis model, and an appropriate prompt is generated. A prompt such as "Is there anything that is bothering you?" is generated.

[1415] Step 6:

[1416] The server sends the generated prompt to the terminal. The prompt is sent to the terminal via the network. The sent prompt is received by the terminal.

[1417] Step 7:

[1418] The terminal then presents the received prompt to the user by voice. In this case, the smartphone's text-to-speech function is used to play the prompt aloud. For example, the patient might hear, "Is there anything bothering you?"

[1419] Step 8:

[1420] The user provides vocal feedback on the presented prompt sentence, which is then input into the terminal again, leading to further analysis and the provision of rehabilitation techniques.

[1421] Through these steps, the system can analyze the patient's emotional state in real time and provide optimal rehabilitation methods. It also generates prompts and collects feedback, enabling continuous rehabilitation support.

[1422] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1423] MODE FOR CARRYING OUT THE INVENTION

[1424] This invention is a system that combines generative AI, a smart speaker, and an emotion engine to provide a rehabilitation method to slow the progression of dementia. This system analyzes the patient's condition from their voice and adjusts the rehabilitation content in real time to improve effectiveness.

[1425] System configuration

[1426] 1. Terminal

[1427] The device is a smart speaker that receives voice input from the patient, performs speech recognition at an early stage, and converts the received voice input into text data. The device also has the function of transmitting the received data to a server.

[1428] 2. Server

[1429] The server receives the voice data and converts it into text using advanced voice recognition technology. The server analyzes the tone and strength of the voice data to assess the patient's emotional and cognitive state. Based on the analysis results, it selects the most appropriate rehabilitation method and sends it to the device.

[1430] 3. Emotion Engine

[1431] The emotion engine analyzes the emotional state of the voice data in real time and feeds back the results to the server. The emotion engine has the function of analyzing the emotional state by combining tone, stress, intonation, and other non-verbal elements.

[1432] 4. Rehabilitation methods

[1433] The rehabilitation techniques include cognitive stimulation therapy, reminiscence therapy, and reality orientation. These techniques are automatically selected by the server depending on the patient's condition.

[1434] 5. Feedback

[1435] The terminal presents rehabilitation content to the patient and receives the patient's response again. The response is again sent to the server and used to determine the next rehabilitation content.

[1436] Program processing

[1437] Device behavior

[1438] The device receives the patient's voice input and sends it to the server. For example, the device may ask a simple question while greeting the patient in the morning. For example, the device may ask, "Good morning, how are you feeling today?" to collect the patient's voice.

[1439] Server Operation

[1440] The server analyzes the received voice data and evaluates the patient's emotional and cognitive state. For example, if the patient responds, "I'm a little tired today," the server analyzes the tone and strength of the voice and evaluates it as "feeling tired." Based on the results, it selects a rehabilitation method and sends it to the device.

[1441] Emotion Engine Operation

[1442] The emotion engine analyzes the patient's emotions in real time based on the voice data. For example, if "anxiety" is detected from the tone and intonation, the result is fed back to the server, which then adjusts the rehabilitation program based on that information.

[1443] Selection of rehabilitation methods

[1444] For example, if the emotional state is judged to be "calm," reminiscence techniques are used to select questions such as "Tell me about the town where you used to live." Conversely, if the emotional state is judged to be "anxious," reassuring reality orientation techniques are used to select questions such as "Do you know what day it is today?"

[1445] Present and get feedback from the device

[1446] The device then verbally communicates the selected rehabilitation content to the patient. If the patient responds, "The town where I used to live was quiet and a very nice place," the voice message is sent back to the server for analysis. This determines the next rehabilitation content.

[1447] Specific examples

[1448] 1. Basic Scenario

[1449] In the morning, the user (a dementia patient) responds to a voice message from the device about their mood that day. The device then sends the voice message to the server, which analyzes it. For example, if the user answers "I didn't have any particular dreams" in response to the question "Good morning. What kind of dreams did you have today?", the server will analyze the response and evaluate the user's emotional state as "Calm." Based on the results, the server will select a reminiscence method and send the next question to the device: "What is your fondest memory from childhood?"

[1450] 2. Utilizing the Emotion Engine

[1451] An emotion engine has been added to more accurately grasp the patient's emotional state. For example, if a patient responds, "I'm a little tired today," the emotion engine analyzes the tone and intonation of the response and evaluates it as "fatigue." This result is fed back to the server, which then adjusts the rehabilitation program based on that information.

[1452] 3. Utilizing a variety of rehabilitation methods

[1453] If the patient's emotional state is evaluated as "anxious," the server selects reality orientation. The terminal asks, "What are your plans for today?" and the patient answers, "I have an appointment to meet my family today." The server evaluates the patient's sense of security.

[1454] In this way, the system of the present invention can analyze the patient's voice data and provide optimal rehabilitation techniques in real time, which can effectively stimulate the patient's cognitive function and slow the progression of dementia. The addition of an emotion engine further improves the accuracy and effectiveness of rehabilitation content.

[1455] The processing flow will be explained below.

[1456] Step 1:

[1457] The terminal receives the patient's voice input, asks the patient questions, and records their responses as audio.

[1458] For example, ask, "Good morning, how are you feeling today?"

[1459] Step 2:

[1460] The device sends the received voice data to the server, where it is converted into text using early-stage voice recognition technology.

[1461] Audio data: "I'm a little tired today."

[1462] Text data: "I'm a little tired today."

[1463] Step 3:

[1464] The server receives the transmitted voice and text data and converts it back into text using advanced voice recognition technology to improve accuracy.

[1465] Audio data: "I'm a little tired today."

[1466] Text data: "I'm a little tired today" (Improved accuracy)

[1467] Step 4:

[1468] The server analyzes the audio and text data and uses an emotion engine to assess the patient's emotional and cognitive state. It identifies the patient's emotional state based on the tone, stress, inflection, and content of the audio.

[1469] Analysis result: "Emotional state: Fatigue"

[1470] Step 5:

[1471] The server selects the optimal rehabilitation method based on the analysis results. For example, if the emotional state is evaluated as "fatigue," it selects a calm and relaxing reminiscence method.

[1472] Selection result: “Reminiscence method”

[1473] Step 6:

[1474] The server generates specific rehabilitation content based on the selected rehabilitation method, such as "Tell us about your favorite hobbies from the past" for reminiscence therapy.

[1475] Generated rehabilitation content: "Tell me about your favorite past hobbies."

[1476] Step 7:

[1477] The server sends the generated rehabilitation content to the terminal. The rehabilitation content is sent as text data, but also includes information for voice synthesis.

[1478] Submitted data: "Tell me about your favorite past hobbies."

[1479] Step 8:

[1480] The device uses voice synthesis technology to ask the patient about the rehabilitation information it has received, and outputs the questions as voice.

[1481] Voice output: "Tell me about your favorite past hobbies."

[1482] Step 9:

[1483] The user (patient) responds to the questions posed by the device. For example, they might reply, "I used to love reading."

[1484] Response: "I used to love reading."

[1485] Step 10:

[1486] The terminal again receives the patient's response as voice data and sends it to the server, where it is also converted into text data.

[1487] Audio data: "I used to love reading."

[1488] Text data: "I used to love reading."

[1489] Step 11:

[1490] The server analyzes the received feedback data and determines the next rehabilitation method. The emotion engine runs again and evaluates the patient's condition. A loop process is performed to continuously evaluate the patient's condition.

[1491] Analysis results: "Emotional state: calm, rehabilitation method: continued reminiscence therapy"

[1492] This series of steps is repeated to provide optimal rehabilitation for the patient, with the aim of slowing the progression of dementia. The addition of the emotion engine further improves the accuracy and effectiveness of the rehabilitation content.

[1493] Example 2

[1494] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1495] Currently, rehabilitation systems on the market lack the ability to properly grasp a patient's emotional and cognitive state. As a result, the rehabilitation methods provided are often not optimal for the patient. Furthermore, the lack of a function to obtain real-time feedback and adjust the rehabilitation content limits the effectiveness of rehabilitation. There is a need to solve these problems and provide the most effective rehabilitation methods for patients.

[1496] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for analyzing voice data and grasping the emotional and cognitive states of the patient, a means for an emotion engine to analyze the voice data in real time and provide feedback to the server, and a means for selecting an optimal rehabilitation method. This makes it possible to provide an optimal rehabilitation method in real time based on the patient's condition and maximize its effectiveness.

[1497] "Patient" refers to an individual who is a target of using the rehabilitation system.

[1498] "Voice input" refers to the words or voice data uttered by the patient.

[1499] "Terminal" refers to an electronic device that receives the patient's voice input and performs the necessary processing.

[1500] "Server" refers to a central computer system that analyzes data sent from terminals and performs necessary processing and feedback.

[1501] "Voice data" refers to a digital representation of a patient's voice input.

[1502] "Emotion engine" refers to a software or hardware component for analyzing audio data and assessing emotional state.

[1503] "Emotional state" refers to the psychological state of the patient as judged from the tone and strength of their voice.

[1504] "Cognitive status" refers to the state of a patient's cognitive function.

[1505] "Rehabilitation techniques" refer to therapies provided based on the patient's emotional and cognitive state.

[1506] "Real-time" refers to processing that occurs at a speed close to instantaneous.

[1507] This invention is a system that combines a generative AI model, a smart speaker, and an emotion engine to provide a rehabilitation method to slow the progression of dementia. This system analyzes the patient's condition from their voice and adjusts the rehabilitation content in real time to improve effectiveness.

[1508] System configuration

[1509] 1. Terminal

[1510] The device consists of a smart speaker that receives the patient's voice input. Specifically, the smart speaker picks up the patient's voice using a microphone, performs initial voice recognition, and converts it into text data. The device also has the function of sending this data to a server. For example, the device may ask the patient, "Good morning, how are you feeling today?"

[1511] 2. Server

[1512] The server receives the voice data sent from the device and converts it into text data using more advanced voice recognition technology. One possible technology is the Google Cloud Speech-to-Text API. The server then analyzes the tone and strength of the voice data to evaluate the patient's emotional and cognitive state. For example, the server evaluates the voice data, such as "I'm a little tired today," as "I feel fatigued." Based on the analysis results, the server selects the optimal rehabilitation method and sends it to the device.

[1513] 3. Emotion Engine

[1514] The emotion engine analyzes the emotional state of the voice data in real time and feeds the results back to the server. Specific analysis items include tone, stress, intonation, and other non-verbal elements. For example, if "anxiety" is detected from the intonation of a patient's voice, the result is fed back to the server.

[1515] 4. Rehabilitation methods

[1516] Rehabilitation techniques include cognitive stimulation therapy, reminiscence therapy, and reality orientation. These techniques are automatically selected by the server depending on the patient's condition. For example, if the emotional state is judged to be "calm," reminiscence therapy will be used and questions such as "Tell me about the town where you used to live" will be selected.

[1517] 5. Feedback

[1518] The device presents rehabilitation content to the patient and receives the patient's response again. The received data is sent to the server and used to determine the next rehabilitation content. For example, if the patient responds, "The town where I used to live was quiet and a very nice place," the voice data is sent to the server and analyzed.

[1519] Specific examples

[1520] Basic Scenario

[1521] In the morning, the user (a dementia patient) answers questions posed by the device. For example, to the question, "Good morning. What kind of dreams did you have today?", the user replies, "I didn't have any particular dreams." The server analyzes the voice data and evaluates the user's emotional state as "calm." The server selects the reminiscence method and sends the next question, "What is your best childhood memory?" to the device. The device presents the question to the patient, who responds. The server then analyzes the response again to determine the next rehabilitation plan.

[1522] Utilizing the Emotion Engine

[1523] The emotion engine analyzes the tone and intonation of the patient's voice. For example, if a patient replies, "I'm a little tired today," the emotion engine analyzes the tone and intonation and evaluates it as "feeling tired." This evaluation result is fed back to the server, which then adjusts the rehabilitation program based on that information.

[1524] Utilizing a variety of rehabilitation methods

[1525] If the patient's emotional state is evaluated as "anxious," the server selects reality orientation. The device asks, "What are your plans for today?" and the patient replies, "I have an appointment to meet my family today." The server analyzes this information and evaluates the patient's sense of security. This determines the next rehabilitation content.

[1526] Prompt Sentence Examples

[1527] "To provide rehabilitation techniques for dementia patients, suggest optimal questions based on their emotional and cognitive states. For example, if a patient says, 'I'm a little tired today,' analyze their speech data and generate the appropriate next question."

[1528] In this way, the system of the present invention can analyze the patient's voice data and provide optimal rehabilitation methods in real time, which can effectively stimulate the patient's cognitive function and slow the progression of dementia. The addition of an emotion engine further improves the accuracy and effectiveness of rehabilitation content.

[1529] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1530] Step 1:

[1531] The device receives voice input. The user speaks to the smart speaker. For example, the device asks, "Good morning, how are you feeling today?" and the user replies, "I'm a little tired today." This voice data is input into the device.

[1532] Input: User's voice data: "I'm a little tired today."

[1533] Output: Received audio data

[1534] Step 2:

[1535] The voice data received by the device is converted into text data using an initial voice recognition engine. For example, the voice recognition technology built into a smart speaker can be used to convert the voice into text, such as "I'm a little tired today." This text data is then sent to the server.

[1536] Input: Received audio data

[1537] Output: Text data "I'm a little tired today"

[1538] Step 3:

[1539] The server uses advanced speech recognition technology to convert the voice data sent from the device back into text data, and then uses advanced technology such as the Google Cloud Speech-to-Text API to analyze the details of the voice and generate text data.

[1540] Input: Audio data sent from the device

[1541] Output: Highly refined text data "I'm a little tired today"

[1542] Step 4:

[1543] The server performs detailed analysis of the voice data and evaluates the emotional state using an emotion engine. The server analyzes the tone, strength, and intonation of the voice to identify the emotional state. For example, it may evaluate the voice as "fatigue" based on its low tone and strength.

[1544] Input: Highly refined text data and its speech characteristics

[1545] Output: Emotional state "Feeling tired"

[1546] Step 5:

[1547] The emotion engine feeds back the analyzed emotional state to the server. The emotion engine determines "fatigue" based on the tone and intonation of the voice data and sends the evaluation result to the server.

[1548] Input: Tone and intonation analysis data of speech data

[1549] Output: Emotional state assessment result: "Feeling tired"

[1550] Step 6:

[1551] The server selects the optimal rehabilitation method based on the emotional and cognitive states. Based on the result "feeling tired," the server selects cognitive stimulation therapy as the rehabilitation method and determines its specific content. For example, it selects the question, "Do you know what day it is today?"

[1552] Input: Emotional state assessment result: "Feeling tired"

[1553] Output: The selected rehabilitation technique "Cognitive Stimulation Therapy" and the question "Do you know what day it is today?"

[1554] Step 7:

[1555] The server sends the selected rehabilitation method and its details to the device. As part of the cognitive stimulation therapy, the server sends the question "Do you know what day it is today?" to the device.

[1556] Input: Selected rehabilitation method and specific question

[1557] Output: Instruction data to the terminal

[1558] Step 8:

[1559] The device then presents the received rehabilitation information to the user via voice. For example, the device might ask the user, "Do you know what day it is today?" The user might respond, "Today is Tuesday."

[1560] Input: Rehabilitation details sent from the server

[1561] Output: Audio presentation to the user

[1562] Step 9:

[1563] The user responds to the rehabilitation content. When the user responds, "Today is Tuesday," the terminal receives the voice again. This received data is sent to the server again.

[1564] Input: User's voice reply "Today is Tuesday"

[1565] Output: Response data to the terminal

[1566] Step 10:

[1567] The server analyzes the voice data again and determines the next rehabilitation content based on the results. For example, based on the response "Today is Tuesday," the server selects the next rehabilitation method and sends the next question, "What did you do on Tuesday?" to the terminal.

[1568] Input: User response data received again

[1569] Output: Next rehabilitation technique and specific questions

[1570] This series of processes enables feedback tailored to the user's emotional and cognitive state, providing optimal rehabilitation techniques. This system is expected to slow the progression of dementia in patients.

[1571] (Application example 2)

[1572] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1573] Rehabilitation for dementia patients requires selecting the optimal method to suit each individual patient's condition and providing effective treatment. However, with conventional methods, it has been difficult to accurately grasp the patient's emotional and cognitive state in real time and provide the optimal rehabilitation method based on that. In particular, more advanced technology is required to realize rehabilitation in a virtual environment. The purpose of this project is to solve these issues.

[1574] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1575] In this invention, the server includes means for receiving voice input from the patient, means for transmitting the received voice data to the server, means for analyzing the voice data to grasp the emotional and cognitive states of the patient, means for selecting an optimal rehabilitation method based on the grasped states, means for transmitting the selected rehabilitation method to the terminal, means for the terminal to present rehabilitation content to the patient and obtain feedback, and means for presenting rehabilitation content in a virtual environment. This makes it possible to provide the rehabilitation method optimal for the patient's condition in real time, thereby effectively slowing the progression of dementia.

[1576] Key Word Definitions

[1577] "Means for receiving patient voice input" refers to a device for collecting voice data from a dementia patient, often a microphone or smart speaker.

[1578] "Means for transmitting received voice data to a server" refers to a function for transferring collected voice data to a central computer (server) via a network.

[1579] "Means for analyzing voice data to understand the emotional and cognitive state of a patient" refers to algorithms or analytical software for analyzing received voice data and determining the emotional and cognitive state of a patient.

[1580] "Means for selecting the most appropriate rehabilitation method based on the identified condition" refers to the logic and process for automatically selecting the rehabilitation method that is most appropriate for the patient's current condition based on the results of the analysis.

[1581] "Means for transmitting the selected rehabilitation method to the terminal" refers to a communication means for transmitting information about the rehabilitation method selected by the server to the patient's terminal.

[1582] "Means for the device to present rehabilitation content to the patient and obtain feedback" refers to an interface that presents selected rehabilitation content through audio or a virtual environment and collects the patient's reactions and answers again.

[1583] "Means for presenting rehabilitation content in a virtual environment" refers to software or hardware that uses virtual reality or augmented reality technology to provide patients with visual or auditory rehabilitation content.

[1584] MODE FOR CARRYING OUT THE INVENTION

[1585] To illustrate an embodiment of the present invention, the following system configuration and operation will be described.

[1586] System Configuration

[1587] This system consists of the following main hardware and software:

[1588] Hardware

[1589] Smart glasses or head-mounted displays (HMD)

[1590] microphone

[1591] Cloud Server

[1592] software

[1593] Speech recognition library (e.g., speech_recognition library)

[1594] Sentiment analysis engine (e.g., custom-developed emotion_engine module)

[1595] Rehabilitation method selection logic (e.g., custom-developed rehab_methods module)

[1596] Virtual environment interface (e.g., a custom-developed virtual_environment module)

[1597] A natural language explanation of how to proceed

[1598] 1. Receiving and sending audio data

[1599] The user's voice input is received by a microphone attached to the smart glasses or head-mounted display, and the received voice data is transmitted to a cloud server in real time.

[1600] 2. Analysis of audio data

[1601] The server converts the received voice data into text data using a speech recognition library. It then uses an emotion analysis engine to analyze the tone and strength of the text data to understand the user's emotional and cognitive state. Based on the results of this analysis, the server selects the most appropriate rehabilitation method for the user.

[1602] 3. Selection of rehabilitation method

[1603] Based on the analyzed data, the server uses a rehabilitation method selection logic to determine the appropriate rehabilitation content. The rehabilitation method is automatically selected from cognitive stimulation therapy, reminiscence therapy, and reality orientation.

[1604] 4. Presentation of rehabilitation content

[1605] The selected rehabilitation content is sent from the cloud server to smart glasses or a head-mounted display, and the user can receive the rehabilitation content visually or audibly through virtual reality or augmented reality using a virtual environment interface.

[1606] 5. Get feedback

[1607] If the user provides vocal feedback on the rehabilitation content, the audio is again picked up by the microphone and sent to the cloud server, and this process further adjusts the next rehabilitation content.

[1608] Specific examples

[1609] For example, if a user is asked, "What kind of dream did you have today?" and the user answers, "I didn't have any particular dreams," this voice data is sent to the server, where it is analyzed and the emotional state is evaluated as "calm" based on the tone and intensity of the voice. Based on this evaluation, the server selects a reminiscence method and presents the next question, "What is your best childhood memory?" A nostalgic scene is displayed within the virtual environment.

[1610] Prompt Sentence Examples

[1611] Emotional state: Calm

[1612] Q: Tell me about the town where you used to live.

[1613] In this way, the system can analyze the user's voice data and provide optimal rehabilitation techniques in real time, which can effectively stimulate the user's cognitive functions and slow the progression of dementia.

[1614] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1615] Program processing flow

[1616] Step 1:

[1617] The device receives the user's voice input. The voice data is captured by a microphone and sent to a cloud server in real time. The input is the user's voice, and the output is sent to the server as voice data.

[1618] Step 2:

[1619] The server receives the voice data and converts it into text data using a speech recognition library. The input is voice data and the output is text data. The speech recognition library analyzes the sound wave data and converts it into text based on a language model.

[1620] Step 3:

[1621] The server inputs the converted text data into an emotion analysis engine, which analyzes the emotional and cognitive states. The input is text data, and the output is an evaluation of the emotional and cognitive states. The emotion analysis engine takes into account non-verbal elements such as tone, strength, and intonation during analysis.

[1622] Step 4:

[1623] Based on the analysis results, the server uses the rehabilitation method selection logic to select the optimal rehabilitation method. The input is the evaluation results of the emotional and cognitive states, and the output is the selected rehabilitation method. The rehabilitation method selection logic compares the application conditions of each rehabilitation method with the current evaluation results and selects the most appropriate method.

[1624] Step 5:

[1625] The server sends the selected rehabilitation technique to the terminal. The input is the selected rehabilitation technique, and the output is the rehabilitation technique information sent to the terminal. The selected information is delivered to the terminal using a communication protocol.

[1626] Step 6:

[1627] The device presents rehabilitation content to the user through a virtual environment interface. The input is rehabilitation method information, and the output is a virtual rehabilitation environment that the user perceives visually or audibly. Visual content is displayed and played on a head-mounted display, and audio content is displayed and played on speakers.

[1628] Step 7:

[1629] The user responds to the presented rehabilitation content and takes action, and their reactions are collected again as voice data by the device. The input is the user's voice response, and the output is again sent to the server as voice data.

[1630] Step 8:

[1631] The server analyzes the user's feedback and adjusts the next rehabilitation content. The input is the user's voice data and the rehabilitation method results, and the output is the adjusted rehabilitation method. The server reuses the emotion analysis engine and rehabilitation method selection logic to reselect the optimal rehabilitation method that reflects the feedback.

[1632] Through these steps, the system can adapt flexibly to the user's condition and provide effective rehabilitation.

[1633] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1634] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1635] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1636] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1637] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1638] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1639] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1640] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1641] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1642] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1643] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1644] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1645] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1646] 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.

[1647] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1648] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1649] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1650] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1651] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1652] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1653] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1654] The following is further disclosed regarding the above embodiment.

[1655] (Claim 1)

[1656] means for receiving patient voice input;

[1657] means for transmitting the received voice data to a server;

[1658] a means for the server to analyze the voice data and grasp the emotional and cognitive states of the patient;

[1659] A means for selecting the most appropriate rehabilitation method based on the grasped condition;

[1660] A means for transmitting the selected rehabilitation technique to the terminal;

[1661] A means for the device to present rehabilitation content to the patient and obtain feedback;

[1662] A system including:

[1663] (Claim 2)

[1664] 10. The system of claim 1, further comprising means in the server for analyzing the tone and intensity of the patient's voice data to identify emotions.

[1665] (Claim 3)

[1666] 2. The system according to claim 1, further comprising a server means for automatically selecting one of cognitive stimulation therapy, reminiscence therapy, and reality orientation as a rehabilitation technique.

[1667] "Example 1"

[1668] (Claim 1)

[1669] means for receiving patient voice input;

[1670] means for transmitting the received voice data to an information processing device;

[1671] a means for an information processing device to analyze the voice data and grasp the emotional and cognitive states of the patient;

[1672] A means for selecting an optimal retraining method based on the identified state;

[1673] means for transmitting the selected retraining technique to the terminal;

[1674] a means for the device to present the retraining content to the patient and obtain feedback;

[1675] A system including:

[1676] (Claim 2)

[1677] 10. The system of claim 1, further comprising means for analyzing the tone and intensity of the patient's voice data to identify emotions.

[1678] (Claim 3)

[1679] 10. The system of claim 1, further comprising means for an information processing device for automatically selecting one of cognitive stimulation therapy, reminiscence therapy, and reality-focused therapy as the retraining technique.

[1680] "Application Example 1"

[1681] (Claim 1)

[1682] means for receiving patient voice input;

[1683] means for transmitting the received voice data to a server;

[1684] a means for the server to analyze the voice data and grasp the emotional and cognitive states of the patient;

[1685] A means for selecting the most appropriate rehabilitation method based on the grasped condition;

[1686] A means for transmitting the selected rehabilitation technique to the terminal;

[1687] A means for the device to present rehabilitation content to the patient and obtain feedback;

[1688] means for converting the patient's voice data into text data and assessing the patient's emotional state using an emotion analysis model;

[1689] A means for generating specific prompt sentences and providing rehabilitation techniques based on the evaluation results;

[1690] A system including:

[1691] (Claim 2)

[1692] 10. The system of claim 1, further comprising server means for converting received voice data into text data and analyzing tone and intensity to assess emotional state.

[1693] (Claim 3)

[1694] 2. The system according to claim 1, further comprising means for automatically selecting one of cognitive stimulation therapy, reminiscence therapy, and reality orientation as a rehabilitation technique, and providing specific prompt sentences based on the evaluation results.

[1695] "Example 2: Combining Emotion Engines"

[1696] (Claim 1)

[1697] means for receiving patient voice input;

[1698] means for transmitting the received voice data to a server;

[1699] a means for the server to analyze the voice data and grasp the emotional and cognitive states of the patient;

[1700] The emotion engine analyzes the voice data in real time and provides feedback to the server.

[1701] A means for selecting the most appropriate rehabilitation method based on the grasped condition;

[1702] A means for transmitting the selected rehabilitation technique to the terminal;

[1703] A means for presenting the selected rehabilitation content to the patient;

[1704] means for receiving and transmitting patient feedback to the server;

[1705] A system including:

[1706] (Claim 2)

[1707] 10. The system of claim 1, further comprising a server means for analyzing the tone and intensity of the received voice data to identify emotions.

[1708] (Claim 3)

[1709] 2. The system according to claim 1, further comprising a server means for automatically selecting one of cognitive stimulation therapy, reminiscence therapy, and reality orientation as a rehabilitation technique.

[1710] "Application example 2 when combining emotion engines"

[1711] New Claims

[1712] (Claim 1)

[1713] means for receiving patient voice input;

[1714] means for transmitting the received voice data to a server;

[1715] a means for the server to analyze the voice data and grasp the emotional and cognitive states of the patient;

[1716] A means for selecting the most appropriate rehabilitation method based on the grasped condition;

[1717] A means for transmitting the selected rehabilitation technique to the terminal;

[1718] A means for the device to present rehabilitation content to the patient and obtain feedback;

[1719] a means for presenting rehabilitation content in a virtual environment;

[1720] A system including:

[1721] (Claim 2)

[1722] 10. The system of claim 1, further comprising means in the server for analyzing the tone and intensity of the patient's voice data to identify emotions.

[1723] (Claim 3)

[1724] 2. The system according to claim 1, further comprising a server means for automatically selecting one of cognitive stimulation therapy, reminiscence therapy, and reality orientation as a rehabilitation technique. [Explanation of symbols]

[1725] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for receiving patient voice input; means for transmitting the received voice data to a server; a means for the server to analyze the voice data and grasp the emotional and cognitive states of the patient; A means for selecting the most appropriate rehabilitation method based on the grasped condition; A means for transmitting the selected rehabilitation technique to the terminal; A means for the device to present rehabilitation content to the patient and obtain feedback; A system including:

2. 10. The system of claim 1, further comprising means in the server for analyzing the tone and intensity of the patient's voice data to identify emotions.

3. 2. The system according to claim 1, further comprising a server means for automatically selecting one of cognitive stimulation therapy, reminiscence therapy, and reality orientation as a rehabilitation technique.

Images