System

Abstract

A system is provided.SOLUTION: A system comprising: a voice acquiring unit for analyzing a telephone conversation of an elderly person in real time; a voice analyzing unit for analyzing a voice received by a server using a generated AI model; a fraud detecting unit for comparing an analysis result with a past crime database and detecting a fraudulent act; a warning notifying unit for transmitting a warning to a user device based on a detection result of a fraudulent act; and an automatic notifying unit for notifying police and a social worker of a detection result.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] There has been a rise in telephone frauds targeting the elderly, such as "It's me, it's me" frauds, and those with dementia or impaired judgment are particularly at risk of falling victim. Current prevention measures are often manual and reactive, and have limitations in preventing fraud before it happens. To solve this problem, a system is needed that monitors elderly phone conversations in real time and quickly warns and notifies users if there is a possibility of fraud. [Means for solving the problem]

[0005] The present invention is a system including a voice capture means for analyzing elderly people's telephone conversations in real time, a data transmission means for transmitting the captured voice data to a server, a voice analysis means for analyzing the voice data received by the server using a generative AI model, a fraud detection means for comparing the analysis results with a past crime database to detect fraud, a warning notification means for sending a warning to the user's device based on the fraud detection results, and an automatic notification means for notifying the police and social workers of the detection results. The voice capture means stores multiple voice data in a buffer and transmits them in real time, and the fraud detection means uses a generative AI model to analyze specific keywords and patterns in the voice data. This can prevent elderly people from falling victim to fraud.

[0006] The "voice acquisition means" is a device or function that uses a microphone or the like to acquire the voice of the elderly person in real time while they are having a telephone conversation.

[0007] The "data transmission means" is a device or function that transmits the acquired voice data to a server using a communication means such as the Internet.

[0008] "Speech analysis means" refers to a device or function that uses an artificial intelligence model to analyze the speech data sent to the server and understand its content.

[0009] The "fraud detection means" is a device or function that compares the analysis results obtained by the voice analysis means with a database of past crimes and detects characteristics of fraudulent acts.

[0010] The "warning notification means" is a device or function that warns the elderly person's terminal of the details of fraudulent activity when such activity is detected.

[0011] An "automatic notification means" is a device or function that automatically notifies police and social workers of details of fraudulent activity when it is detected.

[0012] A "generative AI model" is an artificial intelligence model that learns from past data and is used to analyze voice data.

[0013] A "past crime database" is a database that collects data on past crime cases and uses it for cross-checking.

[0014] "Voice data" refers to data that records what a user says over the telephone. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0023] [First embodiment]

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

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

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

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

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

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

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

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

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

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

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

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

[0036] The present invention provides a system for preventing telephone fraud targeting elderly people. The system includes a user terminal, a server, and terminals of police and social workers that are linked to the server. Specific embodiments will now be described.

[0037] User device behavior

[0038] Audio Acquisition:

[0039] When a user starts a conversation using a telephone, the user's device captures the conversation voice in real time using a built-in microphone. The captured voice data is stored in a buffer and accumulated at regular intervals.

[0040] Data transmission:

[0041] The device transmits the captured voice data to a server via the Internet. This transmission is done in real time, allowing the server to quickly receive and analyze the data.

[0042] Server Operation

[0043] Audio data reception:

[0044] The server receives the voice data sent from the user's device and stores it in a specific format for analysis.

[0045] Audio Analysis:

[0046] The received voice data is analyzed using a generative AI model that has been pre-trained to identify scam-specific phrases and patterns.

[0047] Fraud Detection:

[0048] The analysis results are compared with a database of past crimes, and if there is a high possibility of fraud, a determination is made as to whether it is fraud. Fraud is determined based on specific keywords and patterns.

[0049] Warning notice:

[0050] If fraudulent activity is detected, the server first sends a warning to the user's device, which alerts the user to the content of the conversation.

[0051] Automatic notifications:

[0052] At the same time, the server sends detailed notifications to the police and social workers' devices, including basic information about the user, a summary of the conversation, and any fraud patterns detected.

[0053] Specific examples

[0054] Case 1: An example where the system intervenes before an elderly person is deceived

[0055] Consider the case where a user (elderly person) is making a phone call and the other party uses words that fit into fraudulent patterns, such as "I need money" or "I've been in an accident."

[0056] 1. The user's device acquires the conversation content in real time and sends it to the server.

[0057] 2. The server receives the audio data and the generative AI model analyzes it.

[0058] 3. After checking against a database of past crimes, it is determined that there is a high possibility of fraud.

[0059] 4. The server immediately sends a warning to the user's device, displaying the message "Possible fraud. Please be careful."

[0060] 5. At the same time, the server automatically notifies the police and social workers and sends them details.

[0061] In this way, the present invention can quickly prevent users (elderly people) from becoming victims of telephone fraud. By immediately notifying specialized agencies, even faster response is possible.

[0062] The embodiment for carrying out the present invention is a system including the main means and specific operations according to the means as described above.

[0063] The processing flow will be explained below.

[0064] Step 1: Audio capture

[0065] When a user starts a conversation using the phone, the device captures the conversation voice in real time through the built-in microphone, and the captured voice data is stored in a buffer at regular intervals.

[0066] Step 2: Send data

[0067] The device transmits the captured voice data to a server via the Internet. This transmission is done in real time, so the server can receive and analyze the data immediately.

[0068] Step 3: Receiving audio data

[0069] The server receives the voice data sent from the user's device and stores the received data in a specific format for analysis.

[0070] Step 4: Audio analysis

[0071] The server analyzes the received voice data using an artificial intelligence model, and a pre-trained generative AI model determines the content of the conversation based on this data.

[0072] Step 5: Fraud detection

[0073] The server uses the results of the voice analysis to determine whether the call is likely to be fraudulent, checking it against a database of past crimes and whether it contains specific keywords or phrases.

[0074] Step 6: Warning Notification

[0075] If there is a high possibility of fraudulent activity, the server will send a warning message to the user's device, allowing the user to pay attention to the content of the conversation.

[0076] Step 7: Automatic Notifications

[0077] The server also sends detailed notifications to police and social worker devices, including basic information about the user, a summary of the conversation, and any fraud patterns detected.

[0078] Step 8: Continuous monitoring

[0079] The server continues to monitor the conversation after the notification has been sent and will notify again if any further fraud attempts are detected. This continuous monitoring helps prevent multiple fraud attempts.

[0080] Example 1

[0081] 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."

[0082] In modern society, telephone fraud targeting the elderly is on the rise, increasing the risk of elderly people becoming victims of fraud. Existing crime prevention measures are insufficient to address this issue, and there is a particular need for a system that can detect fraudulent activity in real time and respond quickly. In addition, it is often difficult for elderly people to recognize that they have been victimized by fraud, so immediate notification to specialized agencies is necessary.

[0083] 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.

[0084] In this invention, the server includes a voice capture means for capturing conversational voices in real time when a user is having a telephone conversation, a data transmission means for transmitting the captured voice data to the server via the Internet, a means for saving the voice data received by the server in a specific format, a voice analysis means for analyzing the received voice data using a generative AI model, a fraud detection means for comparing the analysis results with a past crime database to detect fraud, a warning notification means for sending a warning to the user's device based on the fraud detection results, and an automatic notification means for notifying the police and social workers of the detection results. This enables telephone fraud against elderly people to be detected in real time and responded to promptly.

[0085] The term "voice acquisition means" refers to a device or function for acquiring conversational voice in real time when a user is having a telephone conversation.

[0086] "Data transmission means" refers to a device or function for transmitting acquired voice data to a server via the Internet.

[0087] "Received data storage means" refers to a device or function for storing voice data received by the server in a specific format suitable for analysis.

[0088] "Voice analysis means" refers to a device or function for analyzing received voice data using a generative AI model.

[0089] "Fraud detection means" refers to devices or functions that compare the analysis results with a database of past crimes to detect fraudulent activity.

[0090] "Warning notification means" refers to a device or function for sending a warning to a user's terminal based on the result of fraud detection.

[0091] "Automatic notification means" refers to devices and functions for notifying police and social workers of detection results.

[0092] A "generative AI model" is an artificial intelligence model that is pre-trained for a specific task and is used to analyze and identify fraud-specific language and patterns.

[0093] A "prompt" refers to an instruction or question that specifically indicates to the generative AI model the content and purpose of the data to be analyzed.

[0094] The present invention provides a system for preventing telephone fraud targeting the elderly. The system includes a user terminal, a server, and terminals of police and social workers linked to the server. Specific embodiments are described in detail below.

[0095] User device behavior

[0096] Audio Acquisition:

[0097] When a user starts a conversation using a telephone, the user's device captures the conversation voice in real time using a built-in microphone. The captured voice data is stored in a buffer and accumulated at regular intervals.

[0098] Data transmission:

[0099] The user's terminal transmits the acquired voice data to a server via the Internet in real time.

[0100] Server Operation

[0101] Audio data reception:

[0102] The server receives the voice data sent from the user's terminal and stores this data in a specific format (for example, WAV format).

[0103] Audio Analysis:

[0104] The server analyzes the received voice data using a generative AI model that has been trained on fraud-specific phrases and patterns and can identify signs of fraud.

[0105] Fraud Detection

[0106] The server compares the results of the generated AI model with a database of past crimes to determine whether fraud has occurred, based on specific keywords and patterns.

[0107] Alerts and automatic notifications

[0108] User warning notice:

[0109] If fraudulent activity is detected, the server will first send a warning to the user's device, which will alert the user to the content of the conversation.

[0110] Automatic notification to professional authorities:

[0111] At the same time, the server sends detailed notifications to the police and social workers' devices, including basic information about the user, a summary of the conversation, and any detected fraud patterns.

[0112] Specific examples

[0113] Let's assume that while a user (elderly person) is making a phone call, the other party uses words that fit into fraudulent patterns, such as "I need money" or "I've been in an accident."

[0114] 1. The user's device acquires the conversation content in real time and sends it to the server.

[0115] 2. The server receives the audio data and the generative AI model analyzes it.

[0116] 3. After checking against a database of past crimes, it is determined that there is a high possibility of fraud.

[0117] 4. The server immediately sends a warning to the user's device, displaying the message "Possible fraud. Please be careful."

[0118] 5. At the same time, the server automatically notifies the police and social workers and sends them details.

[0119] Prompt Sentence Examples

[0120] Determine whether the email contains words that fit the scam pattern, such as "I need money" or "I've been in an accident."

[0121] In this way, the present invention can quickly prevent elderly people from becoming victims of telephone fraud. Immediate notification to specialized agencies allows for even faster response. The main means and operations of this system are as described above.

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

[0123] Step 1: Audio capture

[0124] When a user starts a call, the device's built-in microphone is activated and captures the conversation voice in real time. The captured voice data is stored in a buffer. The data stored in the buffer is accumulated at regular intervals.

[0125] Input: User's voice

[0126] Output: Buffered audio data

[0127] Specific behavior:

[0128] A user places a call.

[0129] The device's microphone picks up the audio.

[0130] The acquired audio data is stored in a buffer.

[0131] Step 2: Send data

[0132] The user's device transmits the voice data stored in the buffer to the server via the Internet in real time at regular intervals.

[0133] Input: Buffered audio data

[0134] Output: Audio data sent to a server over the internet

[0135] Specific behavior:

[0136] The device establishes an internet connection.

[0137] The audio data stored in the buffer is sent to the server at regular intervals.

[0138] Step 3: Receiving audio data

[0139] The server receives the voice data sent from the user's terminal and saves it in a specific format suitable for analysis (for example, WAV format).

[0140] Input: Audio data sent over the internet

[0141] Output: Audio data saved in a specific format

[0142] Specific behavior:

[0143] The server starts a module for receiving audio data.

[0144] Receives audio data sent from the terminal.

[0145] Save the received data in a specific format.

[0146] Step 4: Audio analysis

[0147] The server analyzes the voice data using a generative AI model that has been trained in advance to recognize scam-specific phrases and patterns.

[0148] Input: Stored audio data

[0149] Output: Analysis result (possibility of fraud)

[0150] Specific behavior:

[0151] The server launches the generative AI model.

[0152] The voice data is analyzed based on the prompt sentence.

[0153] Example prompt: "Please rate this audio recording for signs of fraud."

[0154] Generative AI models identify potential fraud.

[0155] Step 5: Fraud detection

[0156] The server compares the analysis results of the generated AI model with a database of past crimes to determine whether there is a high possibility of fraud.

[0157] Input: Analysis results of the generative AI model

[0158] Output: Determination of likelihood of fraud

[0159] Specific behavior:

[0160] Obtain the analytical data generated by the generative AI model.

[0161] The analytical data is compared with a database of past crimes.

[0162] Score and determine the likelihood of fraud.

[0163] Step 6: Warning Notification

[0164] If fraud is detected, the server sends a warning message to the user's terminal.

[0165] Input: Possible fraud verdict

[0166] Output: The warning message sent to the user's terminal.

[0167] Specific behavior:

[0168] The server generates a warning message.

[0169] A warning message is sent to the user's device via the Internet.

[0170] The user's device will display a warning message and give an audio notification.

[0171] Step 7: Automatic Notifications

[0172] The server simultaneously sends notifications containing detailed information to the police and social worker terminals.

[0173] Input: Possible fraud verdict

[0174] Output: Notification messages sent to police and social worker devices

[0175] Specific behavior:

[0176] The server generates a notification message.

[0177] The notification message will include details such as the user's basic information, a summary of the conversation, and any fraud patterns detected.

[0178] Sending notification messages via the internet to police and social workers.

[0179] The police and social worker terminals receive and display the notification message.

[0180] (Application example 1)

[0181] 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."

[0182] Senior citizens falling victim to telephone fraud has become a social problem. These frauds are sophisticated and often result in seniors losing large amounts of money without even realizing it. A real-time, effective method to address this problem is needed.

[0183] 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.

[0184] In this invention, the server includes a voice acquisition means for analyzing the elderly person's telephone conversations in real time, a data transmission means for transmitting the acquired voice data to the server, a voice analysis means for analyzing the voice data received by the server using a generative AI model, a fraud detection means for comparing the analysis results with past crime information and detecting fraud, a warning notification means for sending a warning to the user's terminal based on the fraud detection results, an automatic notification means for notifying a local protective agency and family of the detection results, and a notification means for identifying fraud in real time, displaying a warning to the user, and sending a notification to a specialist agency and a registered guardian. This makes it possible to detect fraud and issue a warning before the elderly person falls victim to a telephone scam, and notify the specialist agency and guardian.

[0185] The "voice capture means" is a device or system that captures telephone conversations in real time and stores them as voice data.

[0186] The "data transmission means" is a device or system for transmitting the acquired voice data to the server.

[0187] A "voice analysis means" is a device or system that uses a generative AI model to analyze voice data and identify specific patterns or phrases.

[0188] The "fraud detection means" is a device or system that compares the data analyzed by the voice analysis means with past crime information to detect fraudulent acts.

[0189] The "warning notification means" is a device or system that sends a warning to a user's terminal based on the result of detecting fraudulent activity.

[0190] "Automatic notification means" means a device or system for automatically notifying local protective agencies and families of the results of a detection.

[0191] "Notification mechanism" means a device or system that identifies fraudulent activity in real time, displays a warning to the user, and sends notifications to professional organizations and registered parents.

[0192] The present invention is a system for detecting in real time the risk of elderly people falling victim to telephone fraud and responding promptly. Specifically, the system includes a voice acquisition means, a data transmission means, a voice analysis means, a fraud detection means, a warning notification means, an automatic notification means, and a notification means.

[0193] Audio acquisition means

[0194] The user's device uses a built-in microphone to capture the telephone conversation in real time. The captured voice data is temporarily stored in a buffer. This voice data is collected continuously during the call.

[0195] Data transmission method

[0196] The user's device transmits the collected voice data to a server via the Internet. This transmission is done in real time, and the voice data is immediately passed to the server.

[0197] Voice analysis methods

[0198] The server receives the voice data sent from the user's device and analyzes it using a generative AI model. The generative AI model has been trained in advance on fraud patterns and keywords, allowing it to detect signs of fraud with high accuracy.

[0199] Fraud detection measures

[0200] The analyzed voice data is then compared with past criminal records, and if there is a high possibility of fraud, fraudulent activity is detected based on relevant patterns and keywords.

[0201] Warning notification means

[0202] If fraud is detected, the server first sends a warning to the user's device, displaying a message such as "Possible fraud. Please be careful."

[0203] automatic notification means

[0204] At the same time, the server automatically processes and transmits the information to notify local protective agencies and family members of the results of the detection, enabling immediate action in the event of an emergency.

[0205] Notification means

[0206] If the server identifies fraudulent activity, it will display a warning to the user in real time and send a notification to the relevant professional organization and registered guardian. This function is expected to enable prompt action in the event of an incident.

[0207] Specific examples

[0208] For example, if an elderly person is having a phone conversation that includes the phrases "I need money" and "I had an accident," the following process takes place:

[0209] 1. Audio capture:

[0210] The user's device captures the conversation audio in real time.

[0211] 2. Data transmission:

[0212] The audio data is sent to a server via the Internet.

[0213] 3. Audio analysis:

[0214] The server analyzes the received audio data.

[0215] 4. Fraud detection:

[0216] Signs of fraud are detected.

[0217] 5. Warning notice:

[0218] The message "This may be a scam. Please be careful" will be displayed on the user's device.

[0219] 6. Automatic Notifications:

[0220] Notification will be sent to family members and local protective agencies.

[0221] Prompt Sentence Examples

[0222] Voice data: "I need money" "I had an accident"

[0223] Class Label: Fraud Pattern

[0224] As described above, the present invention provides a concrete means for reducing the risk of elderly people becoming victims of telephone fraud and for increasing safety.

[0225] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0226] Step 1:

[0227] The user's device captures the phone conversation in real time. The input is the user's voice, which is captured by a built-in microphone. The voice data is temporarily stored in a buffer.

[0228] Step 2:

[0229] The terminal transmits the acquired voice data to the server. The input is the voice data in the buffer. The data is transmitted via the Internet and reaches the server. The output is the voice data stored on the server.

[0230] Step 3:

[0231] The server analyzes the received voice data using a generative AI model. The input is the voice data stored on the server. The generative AI model has previously learned fraud patterns and analyzes the voice data to detect signs of fraud. The output is the analysis results.

[0232] Step 4:

[0233] The server compares the analysis results with past criminal information to detect fraud. The input is the analysis results from the generative AI model, which are compared with a database of past criminal activity. If the fraud patterns match, it is determined that fraud exists. The output is a detection result indicating whether or not fraud exists.

[0234] Step 5:

[0235] The server sends a warning to the user's terminal based on the fraud detection result. The input is the result that fraud has been detected. The warning message is generated and sent to the user's terminal. The output is the user's terminal with the warning message displayed.

[0236] Step 6:

[0237] The server notifies the local protection agency and the family of the detection result. The input is the fraud detection result and basic information of the user. The notification content is automatically generated and sent to the registered parent or guardian and protection agency. The output is the completed notification.

[0238] Step 7:

[0239] The server identifies fraudulent activity in real time, displays a warning to the user, and sends notifications to professional organizations and registered guardians. The input is the fraud identification and warning message. The output is the warning displayed on the user's device and the notification sent.

[0240] Through the above processing steps, the present invention provides a specific means for seniors to quickly take action before they fall victim to telephone fraud.

[0241] 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.

[0242] The present invention provides a system for preventing telephone fraud targeting the elderly, which further incorporates an emotion engine that recognizes the user's emotions. The system includes a user terminal, a server, and terminals of police and social workers that are linked to the server. Specific embodiments will now be described.

[0243] User device behavior

[0244] Audio Acquisition:

[0245] When a user starts a conversation using a telephone, the user's device captures the conversation voice in real time using a built-in microphone. The captured voice data is stored in a buffer and accumulated at regular intervals.

[0246] Data transmission:

[0247] The device transmits the captured voice data to a server via the Internet. This transmission is done in real time, allowing the server to quickly receive and analyze the data.

[0248] Server Operation

[0249] Audio data reception:

[0250] The server receives the voice data sent from the user's device and stores it in a specific format for analysis.

[0251] Audio Analysis:

[0252] The server analyzes the received voice data using a generative AI model and emotion engine. The generative AI model is pre-trained and can identify scam-specific language and patterns. The emotion engine recognizes emotions from the user's tone of voice and speaking rate, and reflects them in the analysis results.

[0253] Fraud Detection:

[0254] The server compares the voice analysis results with a criminal database to determine the likelihood of fraud. Fraud detection is based on specific keywords and patterns, as well as the user's emotional state. For example, if the user shows signs of anxiety or nervousness, it is determined that fraud is likely.

[0255] Warning notice:

[0256] If fraudulent activity is detected, the server will first send a warning message to the user's device, which will include appropriate wording based on the user's emotional state, encouraging the user to be more careful about the content of the conversation.

[0257] Automatic notifications:

[0258] At the same time, the server sends detailed notifications to the police and social workers' devices, including basic information about the user, a summary of the conversation, detected fraud patterns, and information about the user's emotional state.

[0259] Specific examples

[0260] Case 1: System behavior when elderly people feel uneasy about fraudsters

[0261] Consider the case where a user (elderly person) is making a phone call and the other party uses words that fit into fraudulent patterns, such as "I need money" or "I've been in an accident," and the user feels anxious or tense during the conversation.

[0262] 1. The user's device captures the conversation content and the user's tone of voice in real time and sends them to the server.

[0263] 2. The server receives the voice data, and the generative AI model and emotion engine analyze this data.

[0264] 3. After checking against a database of past crimes, it is determined that there is a high possibility of fraud and that the user is feeling uneasy.

[0265] 4. The server immediately sends a warning message to the user's device, stating, "This may be a scam. Please remain calm."

[0266] 5. At the same time, the server automatically notifies the police and social workers and sends them details.

[0267] In this way, the present invention can quickly prevent users (elderly people) from falling victim to telephone fraud, and by using an emotion engine, it can provide more accurate judgments and warnings. Immediate notification to specialized institutions enables a quick response.

[0268] The embodiment for carrying out the present invention is a system including the main means and specific operations according to the means as described above.

[0269] The processing flow will be explained below.

[0270] Step 1: Audio capture

[0271] When a user starts a conversation using the phone, the device captures the conversation voice in real time through the built-in microphone. The captured voice data is stored in a buffer and accumulated at regular intervals.

[0272] Step 2: Send data

[0273] The device transmits the captured audio data to a server via the Internet. This transmission is done in real time, so the server can receive and analyze the data immediately.

[0274] Step 3: Receiving audio data

[0275] The server receives the voice data sent from the user's device and stores it in a specific format for analysis.

[0276] Step 4: Audio analysis

[0277] The server analyzes the received voice data using a generative AI model and an emotion engine. The generative AI model is pre-trained and can identify scam-specific language and patterns. The emotion engine recognizes emotions from the user's tone of voice and speaking rate, and reflects them in the analysis results.

[0278] Step 5: Fraud detection

[0279] The server compares the voice analysis results with a criminal database to determine whether the person is likely to be committing fraud. The server determines whether the person is committing fraud based on certain keywords and patterns, as well as the user's emotional state. For example, if the user shows signs of anxiety or nervousness, the server determines that the person is likely to be committing fraud.

[0280] Step 6: Warning Notification

[0281] If fraudulent activity is detected, the server will send a warning message to the user's device, which will include appropriate wording based on the user's emotional state, encouraging the user to be more careful about the content of the conversation.

[0282] Step 7: Automatic Notifications

[0283] The server sends detailed notifications to police and social workers, including basic information about the user, a summary of the conversation, any fraud patterns detected, and information about the user's emotional state.

[0284] Step 8: Continuous monitoring

[0285] The server continues to monitor the conversation after the notification has been sent and will notify again if any further fraud attempts are detected. This continuous monitoring helps prevent multiple fraud attempts from occurring.

[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] Previously, countermeasures against telephone fraud targeting the elderly were limited in their ability to prevent fraud before it occurred. Furthermore, there was no way to respond quickly when a fraudulent activity was in progress, making elderly people vulnerable to victimization. Furthermore, detecting fraud requires advanced analysis that takes into account emotional changes, but conventional technology was unable to adequately address this issue.

[0289] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0290] In this invention, the server includes a voice acquisition means for acquiring the elderly person's telephone conversation in real time, a data transmission means for transmitting the acquired voice data to the server via the Internet, a voice analysis means for analyzing the voice data received by the server using a pre-trained generative AI model and an emotion engine, a fraud detection means for comparing the analysis results with past crime data and determining whether fraud has occurred, a warning notification means for sending a warning to the user's device based on the fraud determination result, and an automatic notification means for notifying the police and social workers of the determination result. This makes it possible to detect fraud in real time while taking into account the elderly person's emotional state, and to quickly issue a warning and take action.

[0291] "Voice capture means" is a collective term for the hardware and software used to capture a user's telephone conversation in real time.

[0292] The "data transmission means" is a means for transmitting the acquired voice data to a server via the Internet.

[0293] A "server" is a device or system that includes a central processing unit and associated peripherals for receiving and analyzing audio data.

[0294] A "generative AI model" is an artificial intelligence model that is pre-trained for a specific purpose and is used to identify specific keywords and patterns of fraudulent activity.

[0295] The "emotion engine" is a software model that analyzes the tone and rate of a user's voice to recognize their emotional state.

[0296] "Speech analysis means" refers to means for analyzing received speech data using a generative AI model and an emotion engine.

[0297] "Fraud detection means" refers to a means for comparing the analysis results with a database of past crimes to determine whether or not fraud has occurred.

[0298] The "warning notification means" is a means for sending a warning to the user's terminal based on the result of the fraudulent activity determination.

[0299] "Automatic notification means" refers to a means for notifying police and social workers of fraud determination results.

[0300] MODE FOR CARRYING OUT THE INVENTION

[0301] The present invention provides a system for preventing telephone fraud targeting elderly people, and we will explain how to implement it in detail. This system includes a user's terminal, a server, and terminals of police and social workers that are linked to the server.

[0302] Hardware and software used

[0303] The following hardware and software are used to implement this system:

[0304] User device: Smartphone with built-in microphone and internet connection

[0305] Server: A server in a data center with a high-performance processor and large storage capacity

[0306] Generative AI models: AI models pre-trained to identify fraud-specific language and patterns

[0307] Emotion Engine: Software for recognizing emotions from the user's tone of voice and speech rate

[0308] User device behavior

[0309] When a user starts a conversation using a telephone, the user's device uses a built-in microphone to capture the conversation audio in real time. The captured audio data is stored in a buffer, and data is accumulated at regular intervals. The device then transmits the buffered audio data to a server via the Internet. The transmission is performed in real time and is sent continuously without any divisions.

[0310] Server Operation

[0311] The server receives the voice data sent from the user's device and stores it in a specific format. The server then analyzes the received voice data using a generative AI model and an emotion engine. The generative AI model is pre-trained to identify keywords and patterns specific to fraud. The emotion engine recognizes emotions from the user's tone and speed of voice and reflects them in the analysis results.

[0312] Based on the results of the above analysis, the server compares the voice analysis results with a criminal database to determine the possibility of fraud, which is determined based on specific keywords and patterns, as well as the user's emotional state.

[0313] Alerts and automatic notifications

[0314] When fraudulent activity is detected, the server first sends a warning message to the user's device. The warning contains appropriate wording based on the user's emotional state, encouraging the user to be more careful about the content of the conversation. At the same time, the server sends a notification with detailed information to the police and social worker's devices. The notification content includes the user's basic information, a summary of the conversation, the detected fraud pattern, and information about the user's emotional state.

[0315] Specific examples

[0316] Case 1: System behavior when elderly people feel uneasy about fraudsters

[0317] 1. Consider a scenario where a user (elderly person) is making a phone call and the other party uses words that fit into fraudulent patterns, such as "I need money" or "I've been in an accident," and the user feels anxious or tense during the conversation.

[0318] 2. The user's device captures the conversation content and the user's tone of voice in real time and sends them to the server.

[0319] 3. The server receives the voice data and the generative AI model and emotion engine analyze this data.

[0320] 4. After checking against a database of past crimes, it is determined that there is a high possibility of fraud and that the user is feeling uneasy.

[0321] 5. The server immediately sends a warning message to the user's device, stating, "This may be a scam. Please remain calm."

[0322] 6. At the same time, the server automatically notifies the police and social workers and sends them details.

[0323] Prompt Sentence Examples

[0324] "A 72-year-old elderly person, Mr. A, received a phone call saying, 'Your grandson has been in an accident.' Hearing this, Mr. A became very anxious. How would you analyze this situation and detect possible fraud?"

[0325] Using this prompt, the generative AI model and emotion engine can analyze fraud patterns and user emotions and take appropriate action.

[0326] The above is an embodiment of the present invention. The present invention is a system that detects fraudulent activity while taking into account the emotional state of seniors in real time, and provides prompt warning and response.

[0327] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0328] Step 1:

[0329] Audio Acquisition

[0330] When a user starts a conversation using the phone, the device uses the built-in microphone to capture the conversation audio in real time. The user's voice input is stored as data in a buffer. This voice data is updated at regular intervals (e.g., every second). If the user is talking on the phone with a friend, the device simultaneously captures the friend's voice and the user's voice and records them in a buffer.

[0331] Step 2:

[0332] Data transmission

[0333] The device transmits the voice data stored in the buffer to the server via the Internet. The user's device converts the voice data stored in the buffer into packets and sends them to the server. This transmission is done in real time, and the data is processed so that it is sent continuously without being divided. The user's device transmits a packet of voice data via the Internet every second.

[0334] Step 3:

[0335] Audio data reception

[0336] The server receives voice data sent from the user's device. It stores voice data packets received from the Internet in a specific format in storage. The server restores the data immediately after receiving it and stores it in storage as an audio file for analysis. One second after the user starts the call, the server receives the first voice data packet and stores it in a specified directory.

[0337] Step 4:

[0338] Audio analysis

[0339] The server analyzes the received voice data using a generative AI model and an emotion engine. The input data is the received audio file, which is fed to the generative AI model for analysis. The generative AI model is pre-trained to identify keywords and patterns specific to fraud. The emotion engine recognizes emotions from the tone and rate of the user's voice and combines the analysis results with the output of the generative AI model. For example, the server analyzes whether the user uttered a phrase such as "I need money" and also determines whether the tone of voice is unstable.

[0340] Step 5:

[0341] Fraud Detection

[0342] The server compares the voice analysis results with a database of past crimes to determine the likelihood of fraud. The input data is the analysis results from the generative AI model and emotion engine, and is compared with the criminal database to determine fraud. The server matches the analyzed data with pre-registered criminal data and evaluates the likelihood of fraud based on specific keywords and patterns, as well as the user's emotional state. For example, if the phrase "I had an accident" is detected, it is determined that there is a high possibility of fraud.

[0343] Step 6:

[0344] Warning notice

[0345] If fraud is detected, the server first sends a warning message to the user's device. The input data is the fraud judgment result, and the warning message is created based on that. The warning message contains appropriate wording based on the user's emotional state and is sent to the user's device. For example, if the system detects fraud, a message saying "There is a possibility of fraud. Please remain calm" is displayed on the user's smartphone.

[0346] Step 7:

[0347] Automatic notifications

[0348] At the same time, the server sends a notification containing detailed information to the police and social worker terminals. The input data is the fraud detection result and the user's basic information, which constitutes the notification content. The notification content includes the user's basic information, a summary of the conversation, the detected fraud pattern, and information about the user's emotional state. For example, if fraud is confirmed, the server immediately connects to the police system and sends detailed information. The social worker also receives the information.

[0349] In this way, the system prevents telephone fraud against seniors in real time and enables prompt and appropriate responses.

[0350] (Application example 2)

[0351] 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."

[0352] Telephone scams targeting seniors often involve sophisticated tactics by scammers who exploit seniors' anxieties and nervousness to defraud them of their money. Current technology struggles to detect fraud in real time, and there is a lack of mechanisms to warn seniors before a fraud occurs. Furthermore, there are no fraud detection and warning systems that take into account the emotional state of seniors. Therefore, there is a need for a system that can quickly and effectively protect seniors from fraud.

[0353] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a voice acquisition means for receiving voice data, a data transmission means for transmitting the acquired voice data in real time, a voice analysis means for analyzing the voice data using a generative AI model, a fraud detection means for detecting fraud by comparing the analysis results with a past crime database, an emotion recognition means for analyzing the user's emotional state, a warning notification means for sending a warning based on the fraud detection result, and a notification means for alerting the user based on the possibility of fraud. This enables real-time fraud detection and warning notification that takes into account the emotional state of the elderly.

[0354] "Elderly" refers to individuals who are older and generally more likely to be targeted by fraud.

[0355] "Telephone conversation" refers to a voice conversation conducted over the telephone.

[0356] "Audio capture means" refers to devices or software for recording telephone conversations in real time.

[0357] "Data transmission means" refers to a device or software for transmitting acquired voice data to a server.

[0358] A "generative AI model" is a pre-trained artificial intelligence model that is used to analyze voice data to detect fraudulent activity.

[0359] "Voice analysis means" refers to a device or software for analyzing voice data using a generative AI model.

[0360] "Fraud detection methods" refer to devices or software that compare voice analysis results with a database of past crimes to detect fraudulent activity.

[0361] "Warning notification means" refers to a device or software for sending a warning message to a user's terminal based on the results of fraudulent activity detection.

[0362] "Automatic notification means" refers to devices or software that notify police and social workers based on the detection of fraudulent activity.

[0363] "Emotion recognition means" refers to devices or software that analyze the user's tone of voice, speaking rate, etc. to recognize their emotional state.

[0364] "Notification means" means a device or software that sends a message to a user to warn them of possible fraud.

[0365] Overall system overview

[0366] This invention is a system for preventing telephone fraud targeting the elderly. The system includes a user terminal, a server, and police and social worker terminals. The system has the following main functions: a voice acquisition means, a data transmission means, a voice analysis means, a fraud detection means, an emotion recognition means, a warning notification means, an automatic notification means, and a notification means.

[0367] User terminal operation

[0368] Audio acquisition method:

[0369] When a user initiates a call, the user's device captures the conversation in real time using a built-in microphone. The voice data is stored in a buffer and accumulated at regular intervals.

[0370] Data transmission method:

[0371] The device transmits the captured audio data to a server via the Internet. This transmission is done in real time, allowing the server to quickly receive and analyze the data.

[0372] Server Operation

[0373] Audio data reception:

[0374] The server receives the voice data sent from the user's device and stores it in a specific format.

[0375] Audio analysis methods:

[0376] The server analyzes the received voice data using a generative AI model that has been pre-trained to identify scam-specific phrases and patterns.

[0377] Emotion recognition means:

[0378] The server recognizes the user's emotions from the tone of their voice, the rate at which they speak, etc. This improves the accuracy of the analysis if the user is feeling anxious or nervous.

[0379] Fraud detection measures:

[0380] The results of voice analysis and emotion recognition are combined and compared with a database of past crimes to determine the possibility of fraud.

[0381] Warning notification means:

[0382] If fraud is detected, the server first sends a warning message to the user's device. The message content is adjusted based on the user's emotional state, for example, "There is a possibility of fraud. Please remain calm."

[0383] Automatic notification method:

[0384] At the same time, the server sends a notification to police and social workers with detailed information, including basic information about the user, a summary of the conversation, any fraud patterns detected, and information about the user's emotional state.

[0385] Means of notification:

[0386] To ease the user's anxiety and tension, appropriate steps and contact information may also be provided.

[0387] Specific examples

[0388] Audio capture and data transmission:

[0389] When an elderly person (user) starts a call, the device's built-in microphone picks up the voice, and this data is sent to the server in real time.

[0390] Speech analysis and emotion recognition:

[0391] The server analyzes the received voice data and identifies fraud patterns such as "I need money" or "I've had an accident." At the same time, the emotion engine recognizes emotions such as anxiety and tension from the user's voice.

[0392] Fraud detection and warning notifications:

[0393] If it is determined that there is a high possibility of fraud, a warning message such as "This may be fraud. Please remain calm" will be displayed on the user's device.

[0394] Automatic notifications:

[0395] Similarly, police and social workers will be automatically notified of details of fraudulent activity, facilitating a swift response.

[0396] An example of a prompt sentence could be set up for the generative AI model as follows: "If the user shows signs of anxiety or tension during the phone call, determine that there is a high possibility of fraud. Analyze fraudulent patterns of words and the user's emotional state and display a warning."

[0397] In this way, the system also takes into account the user's emotional state and can quickly and effectively protect seniors from fraud.

[0398] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0399] Step 1: Audio capture

[0400] When a user starts a call, the user's device uses the built-in microphone to capture the conversation in real time. The audio data is stored in a local buffer, and data is accumulated at regular intervals. The input is the audio during the call, and the output is the audio data stored in the buffer.

[0401] Step 2: Send data

[0402] The user's terminal transmits the voice data stored in the local buffer to the server in real time. The data is uploaded to the server via the Internet using a data transmission means, so the input is the voice data in the buffer and the output is the voice data transmitted to the server.

[0403] Step 3: Receiving audio data

[0404] The server receives the voice data sent from the user terminal. The received voice data is stored in a specific format for analysis. The input is the voice data sent from the terminal, and the output is the voice data stored in the data storage in the server.

[0405] Step 4: Audio analysis

[0406] The server analyzes the stored voice data using a generative AI model. First, it converts the voice data into text, then analyzes the text to identify scam-specific phrases and patterns. The input is the stored voice data, and the output is the resulting text data.

[0407] Step 5: Emotion Recognition

[0408] During the voice analysis, the server uses emotion recognition means to analyze the user's tone of voice and speaking rate to determine the user's emotional state. The input is the voice data and analyzed text data, and the output is an evaluation of the user's emotional state.

[0409] Step 6: Fraud detection

[0410] The server compares the results of voice analysis and emotion recognition with a database of past crimes to assess the likelihood of fraud. If certain keywords or patterns match, it determines that there is a high possibility of fraud. The input is the analysis results and emotion evaluation results, and the output is an assessment of the likelihood of fraud.

[0411] Step 7: Warning Notification

[0412] If fraud is detected, the server sends a warning message to the user's device. The message content is adjusted based on the analysis results and the user's emotional state. For example, it might say, "There is a possibility of fraud. Please remain calm." The input is the fraud detection result, and the output is the warning message displayed on the user's device.

[0413] Step 8: Automatic Notifications

[0414] At the same time, the server also notifies the police and social workers. The notification content includes the user's basic information, a summary of the conversation, detected fraud patterns, and information about the user's emotional state. The input is the fraud detection results and related information, and the output is detailed information sent to the police and social workers.

[0415] At each step, different hardware and software in the system work together to effectively prevent elderly people from falling victim to fraud.

[0416] 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.

[0417] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0418] 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.

[0419] [Second embodiment]

[0420] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0421] 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.

[0422] 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).

[0423] 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.

[0424] 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.

[0425] 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).

[0426] 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.

[0427] 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.

[0428] 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.

[0429] 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.

[0430] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0431] 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."

[0432] The present invention provides a system for preventing telephone fraud targeting elderly people. The system includes a user terminal, a server, and terminals of police and social workers that are linked to the server. Specific embodiments will now be described.

[0433] User device behavior

[0434] Audio Acquisition:

[0435] When a user starts a conversation using a telephone, the user's device captures the conversation voice in real time using a built-in microphone. The captured voice data is stored in a buffer and accumulated at regular intervals.

[0436] Data transmission:

[0437] The device transmits the captured voice data to a server via the Internet. This transmission is done in real time, allowing the server to quickly receive and analyze the data.

[0438] Server Operation

[0439] Audio data reception:

[0440] The server receives the voice data sent from the user's device and stores it in a specific format for analysis.

[0441] Audio Analysis:

[0442] The received voice data is analyzed using a generative AI model that has been pre-trained to identify scam-specific phrases and patterns.

[0443] Fraud Detection:

[0444] The analysis results are compared with a database of past crimes, and if there is a high possibility of fraud, a determination is made as to whether it is fraud. Fraud is determined based on specific keywords and patterns.

[0445] Warning notice:

[0446] If fraudulent activity is detected, the server first sends a warning to the user's device, which alerts the user to the content of the conversation.

[0447] Automatic notifications:

[0448] At the same time, the server sends detailed notifications to the police and social workers' devices, including basic information about the user, a summary of the conversation, and any fraud patterns detected.

[0449] Specific examples

[0450] Case 1: An example where the system intervenes before an elderly person is deceived

[0451] Consider the case where a user (elderly person) is making a phone call and the other party uses words that fit into fraudulent patterns, such as "I need money" or "I've been in an accident."

[0452] 1. The user's device acquires the conversation content in real time and sends it to the server.

[0453] 2. The server receives the audio data and the generative AI model analyzes it.

[0454] 3. After checking against a database of past crimes, it is determined that there is a high possibility of fraud.

[0455] 4. The server immediately sends a warning to the user's device, displaying the message "Possible fraud. Please be careful."

[0456] 5. At the same time, the server automatically notifies the police and social workers and sends them details.

[0457] In this way, the present invention can quickly prevent users (elderly people) from becoming victims of telephone fraud. By immediately notifying specialized agencies, even faster response is possible.

[0458] The embodiment for carrying out the present invention is a system including the main means and specific operations according to the means as described above.

[0459] The processing flow will be explained below.

[0460] Step 1: Audio capture

[0461] When a user starts a conversation using the phone, the device captures the conversation voice in real time through the built-in microphone, and the captured voice data is stored in a buffer at regular intervals.

[0462] Step 2: Send data

[0463] The device transmits the captured voice data to a server via the Internet. This transmission is done in real time, so the server can receive and analyze the data immediately.

[0464] Step 3: Receiving audio data

[0465] The server receives the voice data sent from the user's device and stores the received data in a specific format for analysis.

[0466] Step 4: Audio analysis

[0467] The server analyzes the received voice data using an artificial intelligence model, and a pre-trained generative AI model determines the content of the conversation based on this data.

[0468] Step 5: Fraud detection

[0469] The server uses the results of the voice analysis to determine whether the call is likely to be fraudulent, checking it against a database of past crimes and whether it contains specific keywords or phrases.

[0470] Step 6: Warning Notification

[0471] If there is a high possibility of fraudulent activity, the server will send a warning message to the user's device, allowing the user to pay attention to the content of the conversation.

[0472] Step 7: Automatic Notifications

[0473] The server also sends detailed notifications to police and social worker devices, including basic information about the user, a summary of the conversation, and any fraud patterns detected.

[0474] Step 8: Continuous monitoring

[0475] The server continues to monitor the conversation after the notification has been sent and will notify again if any further fraud attempts are detected. This continuous monitoring helps prevent multiple fraud attempts.

[0476] Example 1

[0477] 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."

[0478] In modern society, telephone fraud targeting the elderly is on the rise, increasing the risk of elderly people becoming victims of fraud. Existing crime prevention measures are insufficient to address this issue, and there is a particular need for a system that can detect fraudulent activity in real time and respond quickly. In addition, it is often difficult for elderly people to recognize that they have been victimized by fraud, so immediate notification to specialized agencies is necessary.

[0479] 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.

[0480] In this invention, the server includes a voice capture means for capturing conversational voices in real time when a user is having a telephone conversation, a data transmission means for transmitting the captured voice data to the server via the Internet, a means for saving the voice data received by the server in a specific format, a voice analysis means for analyzing the received voice data using a generative AI model, a fraud detection means for comparing the analysis results with a past crime database to detect fraud, a warning notification means for sending a warning to the user's device based on the fraud detection results, and an automatic notification means for notifying the police and social workers of the detection results. This enables telephone fraud against elderly people to be detected in real time and responded to promptly.

[0481] The term "voice acquisition means" refers to a device or function for acquiring conversational voice in real time when a user is having a telephone conversation.

[0482] "Data transmission means" refers to a device or function for transmitting acquired voice data to a server via the Internet.

[0483] "Received data storage means" refers to a device or function for storing voice data received by the server in a specific format suitable for analysis.

[0484] "Voice analysis means" refers to a device or function for analyzing received voice data using a generative AI model.

[0485] "Fraud detection means" refers to devices or functions that compare the analysis results with a database of past crimes to detect fraudulent activity.

[0486] "Warning notification means" refers to a device or function for sending a warning to a user's terminal based on the result of fraud detection.

[0487] "Automatic notification means" refers to devices and functions for notifying police and social workers of detection results.

[0488] A "generative AI model" is an artificial intelligence model that is pre-trained for a specific task and is used to analyze and identify fraud-specific language and patterns.

[0489] A "prompt" refers to an instruction or question that specifically indicates to the generative AI model the content and purpose of the data to be analyzed.

[0490] The present invention provides a system for preventing telephone fraud targeting the elderly. The system includes a user terminal, a server, and terminals of police and social workers linked to the server. Specific embodiments are described in detail below.

[0491] User device behavior

[0492] Audio Acquisition:

[0493] When a user starts a conversation using a telephone, the user's device captures the conversation voice in real time using a built-in microphone. The captured voice data is stored in a buffer and accumulated at regular intervals.

[0494] Data transmission:

[0495] The user's terminal transmits the acquired voice data to a server via the Internet in real time.

[0496] Server Operation

[0497] Audio data reception:

[0498] The server receives the voice data sent from the user's terminal and stores this data in a specific format (for example, WAV format).

[0499] Audio Analysis:

[0500] The server analyzes the received voice data using a generative AI model that has been trained on fraud-specific phrases and patterns and can identify signs of fraud.

[0501] Fraud Detection

[0502] The server compares the results of the generated AI model with a database of past crimes to determine whether fraud has occurred, based on specific keywords and patterns.

[0503] Alerts and automatic notifications

[0504] User warning notice:

[0505] If fraudulent activity is detected, the server will first send a warning to the user's device, which will alert the user to the content of the conversation.

[0506] Automatic notification to professional authorities:

[0507] At the same time, the server sends detailed notifications to the police and social workers' devices, including basic information about the user, a summary of the conversation, and any detected fraud patterns.

[0508] Specific examples

[0509] Let's assume that while a user (elderly person) is making a phone call, the other party uses words that fit into fraudulent patterns, such as "I need money" or "I've been in an accident."

[0510] 1. The user's device acquires the conversation content in real time and sends it to the server.

[0511] 2. The server receives the audio data and the generative AI model analyzes it.

[0512] 3. After checking against a database of past crimes, it is determined that there is a high possibility of fraud.

[0513] 4. The server immediately sends a warning to the user's device, displaying the message "Possible fraud. Please be careful."

[0514] 5. At the same time, the server automatically notifies the police and social workers and sends them details.

[0515] Prompt Sentence Examples

[0516] Determine whether the email contains words that fit the scam pattern, such as "I need money" or "I've been in an accident."

[0517] In this way, the present invention can quickly prevent elderly people from becoming victims of telephone fraud. Immediate notification to specialized agencies allows for even faster response. The main means and operations of this system are as described above.

[0518] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0519] Step 1: Audio capture

[0520] When a user starts a call, the device's built-in microphone is activated and captures the conversation voice in real time. The captured voice data is stored in a buffer. The data stored in the buffer is accumulated at regular intervals.

[0521] Input: User's voice

[0522] Output: Buffered audio data

[0523] Specific behavior:

[0524] A user places a call.

[0525] The device's microphone picks up the audio.

[0526] The acquired audio data is stored in a buffer.

[0527] Step 2: Send data

[0528] The user's device transmits the voice data stored in the buffer to the server via the Internet in real time at regular intervals.

[0529] Input: Buffered audio data

[0530] Output: Audio data sent to a server over the internet

[0531] Specific behavior:

[0532] The device establishes an internet connection.

[0533] The audio data stored in the buffer is sent to the server at regular intervals.

[0534] Step 3: Receiving audio data

[0535] The server receives the voice data sent from the user's terminal and saves it in a specific format suitable for analysis (for example, WAV format).

[0536] Input: Audio data sent over the internet

[0537] Output: Audio data saved in a specific format

[0538] Specific behavior:

[0539] The server starts a module for receiving audio data.

[0540] Receives audio data sent from the terminal.

[0541] Save the received data in a specific format.

[0542] Step 4: Audio analysis

[0543] The server analyzes the voice data using a generative AI model that has been trained in advance to recognize scam-specific phrases and patterns.

[0544] Input: Stored audio data

[0545] Output: Analysis result (possibility of fraud)

[0546] Specific behavior:

[0547] The server launches the generative AI model.

[0548] The voice data is analyzed based on the prompt sentence.

[0549] Example prompt: "Please rate this audio recording for signs of fraud."

[0550] Generative AI models identify potential fraud.

[0551] Step 5: Fraud detection

[0552] The server compares the analysis results of the generated AI model with a database of past crimes to determine whether there is a high possibility of fraud.

[0553] Input: Analysis results of the generative AI model

[0554] Output: Determination of likelihood of fraud

[0555] Specific behavior:

[0556] Obtain the analytical data generated by the generative AI model.

[0557] The analytical data is compared with a database of past crimes.

[0558] Score and determine the likelihood of fraud.

[0559] Step 6: Warning Notification

[0560] If fraud is detected, the server sends a warning message to the user's terminal.

[0561] Input: Possible fraud verdict

[0562] Output: The warning message sent to the user's terminal.

[0563] Specific behavior:

[0564] The server generates a warning message.

[0565] A warning message is sent to the user's device via the Internet.

[0566] The user's device will display a warning message and give an audio notification.

[0567] Step 7: Automatic Notifications

[0568] The server simultaneously sends notifications containing detailed information to the police and social worker terminals.

[0569] Input: Possible fraud verdict

[0570] Output: Notification messages sent to police and social worker devices

[0571] Specific behavior:

[0572] The server generates a notification message.

[0573] The notification message will include details such as the user's basic information, a summary of the conversation, and any fraud patterns detected.

[0574] Sending notification messages via the internet to police and social workers.

[0575] The police and social worker terminals receive and display the notification message.

[0576] (Application example 1)

[0577] 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."

[0578] Senior citizens falling victim to telephone fraud has become a social problem. These frauds are sophisticated and often result in seniors losing large amounts of money without even realizing it. A real-time, effective method to address this problem is needed.

[0579] 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.

[0580] In this invention, the server includes a voice acquisition means for analyzing the elderly person's telephone conversations in real time, a data transmission means for transmitting the acquired voice data to the server, a voice analysis means for analyzing the voice data received by the server using a generative AI model, a fraud detection means for comparing the analysis results with past crime information and detecting fraud, a warning notification means for sending a warning to the user's terminal based on the fraud detection results, an automatic notification means for notifying a local protective agency and family of the detection results, and a notification means for identifying fraud in real time, displaying a warning to the user, and sending a notification to a specialist agency and a registered guardian. This makes it possible to detect fraud and issue a warning before the elderly person falls victim to a telephone scam, and notify the specialist agency and guardian.

[0581] The "voice capture means" is a device or system that captures telephone conversations in real time and stores them as voice data.

[0582] The "data transmission means" is a device or system for transmitting the acquired voice data to the server.

[0583] A "voice analysis means" is a device or system that uses a generative AI model to analyze voice data and identify specific patterns or phrases.

[0584] The "fraud detection means" is a device or system that compares the data analyzed by the voice analysis means with past crime information to detect fraudulent acts.

[0585] The "warning notification means" is a device or system that sends a warning to a user's terminal based on the result of detecting fraudulent activity.

[0586] "Automatic notification means" means a device or system for automatically notifying local protective agencies and families of the results of a detection.

[0587] "Notification mechanism" means a device or system that identifies fraudulent activity in real time, displays a warning to the user, and sends notifications to professional organizations and registered parents.

[0588] The present invention is a system for detecting in real time the risk of elderly people falling victim to telephone fraud and responding promptly. Specifically, the system includes a voice acquisition means, a data transmission means, a voice analysis means, a fraud detection means, a warning notification means, an automatic notification means, and a notification means.

[0589] Audio acquisition means

[0590] The user's device uses a built-in microphone to capture the telephone conversation in real time. The captured voice data is temporarily stored in a buffer. This voice data is collected continuously during the call.

[0591] Data transmission method

[0592] The user's device transmits the collected voice data to a server via the Internet. This transmission is done in real time, and the voice data is immediately passed to the server.

[0593] Voice analysis methods

[0594] The server receives the voice data sent from the user's device and analyzes it using a generative AI model. The generative AI model has been trained in advance on fraud patterns and keywords, allowing it to detect signs of fraud with high accuracy.

[0595] Fraud detection measures

[0596] The analyzed voice data is then compared with past criminal records, and if there is a high possibility of fraud, fraudulent activity is detected based on relevant patterns and keywords.

[0597] Warning notification means

[0598] If fraud is detected, the server first sends a warning to the user's device, displaying a message such as "Possible fraud. Please be careful."

[0599] automatic notification means

[0600] At the same time, the server automatically processes and transmits the information to notify local protective agencies and family members of the results of the detection, enabling immediate action in the event of an emergency.

[0601] Notification means

[0602] If the server identifies fraudulent activity, it will display a warning to the user in real time and send a notification to the relevant professional organization and registered guardian. This function is expected to enable prompt action in the event of an incident.

[0603] Specific examples

[0604] For example, if an elderly person is having a phone conversation that includes the phrases "I need money" and "I had an accident," the following process takes place:

[0605] 1. Audio capture:

[0606] The user's device captures the conversation audio in real time.

[0607] 2. Data transmission:

[0608] The audio data is sent to a server via the Internet.

[0609] 3. Audio analysis:

[0610] The server analyzes the received audio data.

[0611] 4. Fraud detection:

[0612] Signs of fraud are detected.

[0613] 5. Warning notice:

[0614] The message "This may be a scam. Please be careful" will be displayed on the user's device.

[0615] 6. Automatic Notifications:

[0616] Notification will be sent to family members and local protective agencies.

[0617] Prompt Sentence Examples

[0618] Voice data: "I need money" "I had an accident"

[0619] Class Label: Fraud Pattern

[0620] As described above, the present invention provides a concrete means for reducing the risk of elderly people becoming victims of telephone fraud and for increasing safety.

[0621] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0622] Step 1:

[0623] The user's device captures the phone conversation in real time. The input is the user's voice, which is captured by a built-in microphone. The voice data is temporarily stored in a buffer.

[0624] Step 2:

[0625] The terminal transmits the acquired voice data to the server. The input is the voice data in the buffer. The data is transmitted via the Internet and reaches the server. The output is the voice data stored on the server.

[0626] Step 3:

[0627] The server analyzes the received voice data using a generative AI model. The input is the voice data stored on the server. The generative AI model has previously learned fraud patterns and analyzes the voice data to detect signs of fraud. The output is the analysis results.

[0628] Step 4:

[0629] The server compares the analysis results with past criminal information to detect fraud. The input is the analysis results from the generative AI model, which are compared with a database of past criminal activity. If the fraud patterns match, it is determined that fraud exists. The output is a detection result indicating whether or not fraud exists.

[0630] Step 5:

[0631] The server sends a warning to the user's terminal based on the fraud detection result. The input is the result that fraud has been detected. The warning message is generated and sent to the user's terminal. The output is the user's terminal with the warning message displayed.

[0632] Step 6:

[0633] The server notifies the local protection agency and the family of the detection result. The input is the fraud detection result and basic information of the user. The notification content is automatically generated and sent to the registered parent or guardian and protection agency. The output is the completed notification.

[0634] Step 7:

[0635] The server identifies fraudulent activity in real time, displays a warning to the user, and sends notifications to professional organizations and registered guardians. The input is the fraud identification and warning message. The output is the warning displayed on the user's device and the notification sent.

[0636] Through the above processing steps, the present invention provides a specific means for seniors to quickly take action before they fall victim to telephone fraud.

[0637] 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.

[0638] The present invention provides a system for preventing telephone fraud targeting the elderly, which further incorporates an emotion engine that recognizes the user's emotions. The system includes a user terminal, a server, and terminals of police and social workers that are linked to the server. Specific embodiments will now be described.

[0639] User device behavior

[0640] Audio Acquisition:

[0641] When a user starts a conversation using a telephone, the user's device captures the conversation voice in real time using a built-in microphone. The captured voice data is stored in a buffer and accumulated at regular intervals.

[0642] Data transmission:

[0643] The device transmits the captured voice data to a server via the Internet. This transmission is done in real time, allowing the server to quickly receive and analyze the data.

[0644] Server Operation

[0645] Audio data reception:

[0646] The server receives the voice data sent from the user's device and stores it in a specific format for analysis.

[0647] Audio Analysis:

[0648] The server analyzes the received voice data using a generative AI model and emotion engine. The generative AI model is pre-trained and can identify scam-specific language and patterns. The emotion engine recognizes emotions from the user's tone of voice and speaking rate, and reflects them in the analysis results.

[0649] Fraud Detection:

[0650] The server compares the voice analysis results with a criminal database to determine the likelihood of fraud. Fraud detection is based on specific keywords and patterns, as well as the user's emotional state. For example, if the user shows signs of anxiety or nervousness, it is determined that fraud is likely.

[0651] Warning notice:

[0652] If fraudulent activity is detected, the server will first send a warning message to the user's device, which will include appropriate wording based on the user's emotional state, encouraging the user to be more careful about the content of the conversation.

[0653] Automatic notifications:

[0654] At the same time, the server sends detailed notifications to the police and social workers' devices, including basic information about the user, a summary of the conversation, detected fraud patterns, and information about the user's emotional state.

[0655] Specific examples

[0656] Case 1: System behavior when elderly people feel uneasy about fraudsters

[0657] Consider the case where a user (elderly person) is making a phone call and the other party uses words that fit into fraudulent patterns, such as "I need money" or "I've been in an accident," and the user feels anxious or tense during the conversation.

[0658] 1. The user's device captures the conversation content and the user's tone of voice in real time and sends them to the server.

[0659] 2. The server receives the voice data, and the generative AI model and emotion engine analyze this data.

[0660] 3. After checking against a database of past crimes, it is determined that there is a high possibility of fraud and that the user is feeling uneasy.

[0661] 4. The server immediately sends a warning message to the user's device, stating, "This may be a scam. Please remain calm."

[0662] 5. At the same time, the server automatically notifies the police and social workers and sends them details.

[0663] In this way, the present invention can quickly prevent users (elderly people) from falling victim to telephone fraud, and by using an emotion engine, it can provide more accurate judgments and warnings. Immediate notification to specialized institutions enables a quick response.

[0664] The embodiment for carrying out the present invention is a system including the main means and specific operations according to the means as described above.

[0665] The processing flow will be explained below.

[0666] Step 1: Audio capture

[0667] When a user starts a conversation using the phone, the device captures the conversation voice in real time through the built-in microphone. The captured voice data is stored in a buffer and accumulated at regular intervals.

[0668] Step 2: Send data

[0669] The device transmits the captured audio data to a server via the Internet. This transmission is done in real time, so the server can receive and analyze the data immediately.

[0670] Step 3: Receiving audio data

[0671] The server receives the voice data sent from the user's device and stores it in a specific format for analysis.

[0672] Step 4: Audio analysis

[0673] The server analyzes the received voice data using a generative AI model and an emotion engine. The generative AI model is pre-trained and can identify scam-specific language and patterns. The emotion engine recognizes emotions from the user's tone of voice and speaking rate, and reflects them in the analysis results.

[0674] Step 5: Fraud detection

[0675] The server compares the voice analysis results with a criminal database to determine whether the person is likely to be committing fraud. The server determines whether the person is committing fraud based on certain keywords and patterns, as well as the user's emotional state. For example, if the user shows signs of anxiety or nervousness, the server determines that the person is likely to be committing fraud.

[0676] Step 6: Warning Notification

[0677] If fraudulent activity is detected, the server will send a warning message to the user's device, which will include appropriate wording based on the user's emotional state, encouraging the user to be more careful about the content of the conversation.

[0678] Step 7: Automatic Notifications

[0679] The server sends detailed notifications to police and social workers, including basic information about the user, a summary of the conversation, any fraud patterns detected, and information about the user's emotional state.

[0680] Step 8: Continuous monitoring

[0681] The server continues to monitor the conversation after the notification has been sent and will notify again if any further fraud attempts are detected. This continuous monitoring helps prevent multiple fraud attempts from occurring.

[0682] Example 2

[0683] 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."

[0684] Previously, countermeasures against telephone fraud targeting the elderly were limited in their ability to prevent fraud before it occurred. Furthermore, there was no way to respond quickly when a fraudulent activity was in progress, making elderly people vulnerable to victimization. Furthermore, detecting fraud requires advanced analysis that takes into account emotional changes, but conventional technology was unable to adequately address this issue.

[0685] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0686] In this invention, the server includes a voice acquisition means for acquiring the elderly person's telephone conversation in real time, a data transmission means for transmitting the acquired voice data to the server via the Internet, a voice analysis means for analyzing the voice data received by the server using a pre-trained generative AI model and an emotion engine, a fraud detection means for comparing the analysis results with past crime data and determining whether fraud has occurred, a warning notification means for sending a warning to the user's device based on the fraud determination result, and an automatic notification means for notifying the police and social workers of the determination result. This makes it possible to detect fraud in real time while taking into account the elderly person's emotional state, and to quickly issue a warning and take action.

[0687] "Voice capture means" is a collective term for the hardware and software used to capture a user's telephone conversation in real time.

[0688] The "data transmission means" is a means for transmitting the acquired voice data to a server via the Internet.

[0689] A "server" is a device or system that includes a central processing unit and associated peripherals for receiving and analyzing audio data.

[0690] A "generative AI model" is an artificial intelligence model that is pre-trained for a specific purpose and is used to identify specific keywords and patterns of fraudulent activity.

[0691] The "emotion engine" is a software model that analyzes the tone and rate of a user's voice to recognize their emotional state.

[0692] "Speech analysis means" refers to means for analyzing received speech data using a generative AI model and an emotion engine.

[0693] "Fraud detection means" refers to a means for comparing the analysis results with a database of past crimes to determine whether or not fraud has occurred.

[0694] The "warning notification means" is a means for sending a warning to the user's terminal based on the result of the fraudulent activity determination.

[0695] "Automatic notification means" refers to a means for notifying police and social workers of fraud determination results.

[0696] MODE FOR CARRYING OUT THE INVENTION

[0697] The present invention provides a system for preventing telephone fraud targeting elderly people, and we will explain how to implement it in detail. This system includes a user's terminal, a server, and terminals of police and social workers that are linked to the server.

[0698] Hardware and software used

[0699] The following hardware and software are used to implement this system:

[0700] User device: Smartphone with built-in microphone and internet connection

[0701] Server: A server in a data center with a high-performance processor and large storage capacity

[0702] Generative AI models: AI models pre-trained to identify fraud-specific language and patterns

[0703] Emotion Engine: Software for recognizing emotions from the user's tone of voice and speech rate

[0704] User device behavior

[0705] When a user starts a conversation using a telephone, the user's device uses a built-in microphone to capture the conversation audio in real time. The captured audio data is stored in a buffer, and data is accumulated at regular intervals. The device then transmits the buffered audio data to a server via the Internet. The transmission is performed in real time and is sent continuously without any divisions.

[0706] Server Operation

[0707] The server receives the voice data sent from the user's device and stores it in a specific format. The server then analyzes the received voice data using a generative AI model and an emotion engine. The generative AI model is pre-trained to identify keywords and patterns specific to fraud. The emotion engine recognizes emotions from the user's tone and speed of voice and reflects them in the analysis results.

[0708] Based on the results of the above analysis, the server compares the voice analysis results with a criminal database to determine the possibility of fraud, which is determined based on specific keywords and patterns, as well as the user's emotional state.

[0709] Alerts and automatic notifications

[0710] When fraudulent activity is detected, the server first sends a warning message to the user's device. The warning contains appropriate wording based on the user's emotional state, encouraging the user to be more careful about the content of the conversation. At the same time, the server sends a notification with detailed information to the police and social worker's devices. The notification content includes the user's basic information, a summary of the conversation, the detected fraud pattern, and information about the user's emotional state.

[0711] Specific examples

[0712] Case 1: System behavior when elderly people feel uneasy about fraudsters

[0713] 1. Consider a scenario where a user (elderly person) is making a phone call and the other party uses words that fit into fraudulent patterns, such as "I need money" or "I've been in an accident," and the user feels anxious or tense during the conversation.

[0714] 2. The user's device captures the conversation content and the user's tone of voice in real time and sends them to the server.

[0715] 3. The server receives the voice data and the generative AI model and emotion engine analyze this data.

[0716] 4. After checking against a database of past crimes, it is determined that there is a high possibility of fraud and that the user is feeling uneasy.

[0717] 5. The server immediately sends a warning message to the user's device, stating, "This may be a scam. Please remain calm."

[0718] 6. At the same time, the server automatically notifies the police and social workers and sends them details.

[0719] Prompt Sentence Examples

[0720] "A 72-year-old elderly person, Mr. A, received a phone call saying, 'Your grandson has been in an accident.' Hearing this, Mr. A became very anxious. How would you analyze this situation and detect possible fraud?"

[0721] Using this prompt, the generative AI model and emotion engine can analyze fraud patterns and user emotions and take appropriate action.

[0722] The above is an embodiment of the present invention. The present invention is a system that detects fraudulent activity while taking into account the emotional state of seniors in real time, and provides prompt warning and response.

[0723] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0724] Step 1:

[0725] Audio Acquisition

[0726] When a user starts a conversation using the phone, the device uses the built-in microphone to capture the conversation audio in real time. The user's voice input is stored as data in a buffer. This voice data is updated at regular intervals (e.g., every second). If the user is talking on the phone with a friend, the device simultaneously captures the friend's voice and the user's voice and records them in a buffer.

[0727] Step 2:

[0728] Data transmission

[0729] The device transmits the voice data stored in the buffer to the server via the Internet. The user's device converts the voice data stored in the buffer into packets and sends them to the server. This transmission is done in real time, and the data is processed so that it is sent continuously without being divided. The user's device transmits a packet of voice data via the Internet every second.

[0730] Step 3:

[0731] Audio data reception

[0732] The server receives voice data sent from the user's device. It stores voice data packets received from the Internet in a specific format in storage. The server restores the data immediately after receiving it and stores it in storage as an audio file for analysis. One second after the user starts the call, the server receives the first voice data packet and stores it in a specified directory.

[0733] Step 4:

[0734] Audio analysis

[0735] The server analyzes the received voice data using a generative AI model and an emotion engine. The input data is the received audio file, which is fed to the generative AI model for analysis. The generative AI model is pre-trained to identify keywords and patterns specific to fraud. The emotion engine recognizes emotions from the tone and rate of the user's voice and combines the analysis results with the output of the generative AI model. For example, the server analyzes whether the user uttered a phrase such as "I need money" and also determines whether the tone of voice is unstable.

[0736] Step 5:

[0737] Fraud Detection

[0738] The server compares the voice analysis results with a database of past crimes to determine the likelihood of fraud. The input data is the analysis results from the generative AI model and emotion engine, and is compared with the criminal database to determine fraud. The server matches the analyzed data with pre-registered criminal data and evaluates the likelihood of fraud based on specific keywords and patterns, as well as the user's emotional state. For example, if the phrase "I had an accident" is detected, it is determined that there is a high possibility of fraud.

[0739] Step 6:

[0740] Warning notice

[0741] If fraud is detected, the server first sends a warning message to the user's device. The input data is the fraud judgment result, and the warning message is created based on that. The warning message contains appropriate wording based on the user's emotional state and is sent to the user's device. For example, if the system detects fraud, a message saying "There is a possibility of fraud. Please remain calm" is displayed on the user's smartphone.

[0742] Step 7:

[0743] Automatic notifications

[0744] At the same time, the server sends a notification containing detailed information to the police and social worker terminals. The input data is the fraud detection result and the user's basic information, which constitutes the notification content. The notification content includes the user's basic information, a summary of the conversation, the detected fraud pattern, and information about the user's emotional state. For example, if fraud is confirmed, the server immediately connects to the police system and sends detailed information. The social worker also receives the information.

[0745] In this way, the system prevents telephone fraud against seniors in real time and enables prompt and appropriate responses.

[0746] (Application example 2)

[0747] 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."

[0748] Telephone scams targeting seniors often involve sophisticated tactics by scammers who exploit seniors' anxieties and nervousness to defraud them of their money. Current technology struggles to detect fraud in real time, and there is a lack of mechanisms to warn seniors before a fraud occurs. Furthermore, there are no fraud detection and warning systems that take into account the emotional state of seniors. Therefore, there is a need for a system that can quickly and effectively protect seniors from fraud.

[0749] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a voice acquisition means for receiving voice data, a data transmission means for transmitting the acquired voice data in real time, a voice analysis means for analyzing the voice data using a generative AI model, a fraud detection means for detecting fraud by comparing the analysis results with a past crime database, an emotion recognition means for analyzing the user's emotional state, a warning notification means for sending a warning based on the fraud detection result, and a notification means for alerting the user based on the possibility of fraud. This enables real-time fraud detection and warning notification that takes into account the emotional state of the elderly.

[0750] "Elderly" refers to individuals who are older and generally more likely to be targeted by fraud.

[0751] "Telephone conversation" refers to a voice conversation conducted over the telephone.

[0752] "Audio capture means" refers to devices or software for recording telephone conversations in real time.

[0753] "Data transmission means" refers to a device or software for transmitting acquired voice data to a server.

[0754] A "generative AI model" is a pre-trained artificial intelligence model that is used to analyze voice data to detect fraudulent activity.

[0755] "Voice analysis means" refers to a device or software for analyzing voice data using a generative AI model.

[0756] "Fraud detection methods" refer to devices or software that compare voice analysis results with a database of past crimes to detect fraudulent activity.

[0757] "Warning notification means" refers to a device or software for sending a warning message to a user's terminal based on the results of fraudulent activity detection.

[0758] "Automatic notification means" refers to devices or software that notify police and social workers based on the detection of fraudulent activity.

[0759] "Emotion recognition means" refers to devices or software that analyze the user's tone of voice, speaking rate, etc. to recognize their emotional state.

[0760] "Notification means" means a device or software that sends a message to a user to warn them of possible fraud.

[0761] Overall system overview

[0762] This invention is a system for preventing telephone fraud targeting the elderly. The system includes a user terminal, a server, and police and social worker terminals. The system has the following main functions: a voice acquisition means, a data transmission means, a voice analysis means, a fraud detection means, an emotion recognition means, a warning notification means, an automatic notification means, and a notification means.

[0763] User terminal operation

[0764] Audio acquisition method:

[0765] When a user initiates a call, the user's device captures the conversation in real time using a built-in microphone. The voice data is stored in a buffer and accumulated at regular intervals.

[0766] Data transmission method:

[0767] The device transmits the captured audio data to a server via the Internet. This transmission is done in real time, allowing the server to quickly receive and analyze the data.

[0768] Server Operation

[0769] Audio data reception:

[0770] The server receives the voice data sent from the user's device and stores it in a specific format.

[0771] Audio analysis methods:

[0772] The server analyzes the received voice data using a generative AI model that has been pre-trained to identify scam-specific phrases and patterns.

[0773] Emotion recognition means:

[0774] The server recognizes the user's emotions from the tone of their voice, the rate at which they speak, etc. This improves the accuracy of the analysis if the user is feeling anxious or nervous.

[0775] Fraud detection measures:

[0776] The results of voice analysis and emotion recognition are combined and compared with a database of past crimes to determine the possibility of fraud.

[0777] Warning notification means:

[0778] If fraud is detected, the server first sends a warning message to the user's device. The message content is adjusted based on the user's emotional state, for example, "There is a possibility of fraud. Please remain calm."

[0779] Automatic notification method:

[0780] At the same time, the server sends a notification to police and social workers with detailed information, including basic information about the user, a summary of the conversation, any fraud patterns detected, and information about the user's emotional state.

[0781] Means of notification:

[0782] To ease the user's anxiety and tension, appropriate steps and contact information may also be provided.

[0783] Specific examples

[0784] Audio capture and data transmission:

[0785] When an elderly person (user) starts a call, the device's built-in microphone picks up the voice, and this data is sent to the server in real time.

[0786] Speech analysis and emotion recognition:

[0787] The server analyzes the received voice data and identifies fraud patterns such as "I need money" or "I've had an accident." At the same time, the emotion engine recognizes emotions such as anxiety and tension from the user's voice.

[0788] Fraud detection and warning notifications:

[0789] If it is determined that there is a high possibility of fraud, a warning message such as "This may be fraud. Please remain calm" will be displayed on the user's device.

[0790] Automatic notifications:

[0791] Similarly, police and social workers will be automatically notified of details of fraudulent activity, facilitating a swift response.

[0792] An example of a prompt sentence could be set up for the generative AI model as follows: "If the user shows signs of anxiety or tension during the phone call, determine that there is a high possibility of fraud. Analyze fraudulent patterns of words and the user's emotional state and display a warning."

[0793] In this way, the system also takes into account the user's emotional state and can quickly and effectively protect seniors from fraud.

[0794] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0795] Step 1: Audio capture

[0796] When a user starts a call, the user's device uses the built-in microphone to capture the conversation in real time. The audio data is stored in a local buffer, and data is accumulated at regular intervals. The input is the audio during the call, and the output is the audio data stored in the buffer.

[0797] Step 2: Send data

[0798] The user's terminal transmits the voice data stored in the local buffer to the server in real time. The data is uploaded to the server via the Internet using a data transmission means, so the input is the voice data in the buffer and the output is the voice data transmitted to the server.

[0799] Step 3: Receiving audio data

[0800] The server receives the voice data sent from the user terminal. The received voice data is stored in a specific format for analysis. The input is the voice data sent from the terminal, and the output is the voice data stored in the data storage in the server.

[0801] Step 4: Audio analysis

[0802] The server analyzes the stored voice data using a generative AI model. First, it converts the voice data into text, then analyzes the text to identify scam-specific phrases and patterns. The input is the stored voice data, and the output is the resulting text data.

[0803] Step 5: Emotion Recognition

[0804] During the voice analysis, the server uses emotion recognition means to analyze the user's tone of voice and speaking rate to determine the user's emotional state. The input is the voice data and analyzed text data, and the output is an evaluation of the user's emotional state.

[0805] Step 6: Fraud detection

[0806] The server compares the results of voice analysis and emotion recognition with a database of past crimes to assess the likelihood of fraud. If certain keywords or patterns match, it determines that there is a high possibility of fraud. The input is the analysis results and emotion evaluation results, and the output is an assessment of the likelihood of fraud.

[0807] Step 7: Warning Notification

[0808] If fraud is detected, the server sends a warning message to the user's device. The message content is adjusted based on the analysis results and the user's emotional state. For example, it might say, "There is a possibility of fraud. Please remain calm." The input is the fraud detection result, and the output is the warning message displayed on the user's device.

[0809] Step 8: Automatic Notifications

[0810] At the same time, the server also notifies the police and social workers. The notification content includes the user's basic information, a summary of the conversation, detected fraud patterns, and information about the user's emotional state. The input is the fraud detection results and related information, and the output is detailed information sent to the police and social workers.

[0811] At each step, different hardware and software in the system work together to effectively prevent elderly people from falling victim to fraud.

[0812] 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.

[0813] 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.

[0814] 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.

[0815] [Third embodiment]

[0816] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0817] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0818] 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).

[0819] 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.

[0820] 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.

[0821] 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).

[0822] 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.

[0823] 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.

[0824] 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.

[0825] 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.

[0826] 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.

[0827] 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."

[0828] The present invention provides a system for preventing telephone fraud targeting elderly people. The system includes a user terminal, a server, and terminals of police and social workers that are linked to the server. Specific embodiments will now be described.

[0829] User device behavior

[0830] Audio Acquisition:

[0831] When a user starts a conversation using a telephone, the user's device captures the conversation voice in real time using a built-in microphone. The captured voice data is stored in a buffer and accumulated at regular intervals.

[0832] Data transmission:

[0833] The device transmits the captured voice data to a server via the Internet. This transmission is done in real time, allowing the server to quickly receive and analyze the data.

[0834] Server Operation

[0835] Audio data reception:

[0836] The server receives the voice data sent from the user's device and stores it in a specific format for analysis.

[0837] Audio Analysis:

[0838] The received voice data is analyzed using a generative AI model that has been pre-trained to identify scam-specific phrases and patterns.

[0839] Fraud Detection:

[0840] The analysis results are compared with a database of past crimes, and if there is a high possibility of fraud, a determination is made as to whether it is fraud. Fraud is determined based on specific keywords and patterns.

[0841] Warning notice:

[0842] If fraudulent activity is detected, the server first sends a warning to the user's device, which alerts the user to the content of the conversation.

[0843] Automatic notifications:

[0844] At the same time, the server sends detailed notifications to the police and social workers' devices, including basic information about the user, a summary of the conversation, and any fraud patterns detected.

[0845] Specific examples

[0846] Case 1: An example where the system intervenes before an elderly person is deceived

[0847] Consider the case where a user (elderly person) is making a phone call and the other party uses words that fit into fraudulent patterns, such as "I need money" or "I've been in an accident."

[0848] 1. The user's device acquires the conversation content in real time and sends it to the server.

[0849] 2. The server receives the audio data and the generative AI model analyzes it.

[0850] 3. After checking against a database of past crimes, it is determined that there is a high possibility of fraud.

[0851] 4. The server immediately sends a warning to the user's device, displaying the message "Possible fraud. Please be careful."

[0852] 5. At the same time, the server automatically notifies the police and social workers and sends them details.

[0853] In this way, the present invention can quickly prevent users (elderly people) from becoming victims of telephone fraud. By immediately notifying specialized agencies, even faster response is possible.

[0854] The embodiment for carrying out the present invention is a system including the main means and specific operations according to the means as described above.

[0855] The processing flow will be explained below.

[0856] Step 1: Audio capture

[0857] When a user starts a conversation using the phone, the device captures the conversation voice in real time through the built-in microphone, and the captured voice data is stored in a buffer at regular intervals.

[0858] Step 2: Send data

[0859] The device transmits the captured voice data to a server via the Internet. This transmission is done in real time, so the server can receive and analyze the data immediately.

[0860] Step 3: Receiving audio data

[0861] The server receives the voice data sent from the user's device and stores the received data in a specific format for analysis.

[0862] Step 4: Audio analysis

[0863] The server analyzes the received voice data using an artificial intelligence model, and a pre-trained generative AI model determines the content of the conversation based on this data.

[0864] Step 5: Fraud detection

[0865] The server uses the results of the voice analysis to determine whether the call is likely to be fraudulent, checking it against a database of past crimes and whether it contains specific keywords or phrases.

[0866] Step 6: Warning Notification

[0867] If there is a high possibility of fraudulent activity, the server will send a warning message to the user's device, allowing the user to pay attention to the content of the conversation.

[0868] Step 7: Automatic Notifications

[0869] The server also sends detailed notifications to police and social worker devices, including basic information about the user, a summary of the conversation, and any fraud patterns detected.

[0870] Step 8: Continuous monitoring

[0871] The server continues to monitor the conversation after the notification has been sent and will notify again if any further fraud attempts are detected. This continuous monitoring helps prevent multiple fraud attempts.

[0872] Example 1

[0873] 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."

[0874] In modern society, telephone fraud targeting the elderly is on the rise, increasing the risk of elderly people becoming victims of fraud. Existing crime prevention measures are insufficient to address this issue, and there is a particular need for a system that can detect fraudulent activity in real time and respond quickly. In addition, it is often difficult for elderly people to recognize that they have been victimized by fraud, so immediate notification to specialized agencies is necessary.

[0875] 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.

[0876] In this invention, the server includes a voice capture means for capturing conversational voices in real time when a user is having a telephone conversation, a data transmission means for transmitting the captured voice data to the server via the Internet, a means for saving the voice data received by the server in a specific format, a voice analysis means for analyzing the received voice data using a generative AI model, a fraud detection means for comparing the analysis results with a past crime database to detect fraud, a warning notification means for sending a warning to the user's device based on the fraud detection results, and an automatic notification means for notifying the police and social workers of the detection results. This enables telephone fraud against elderly people to be detected in real time and responded to promptly.

[0877] The term "voice acquisition means" refers to a device or function for acquiring conversational voice in real time when a user is having a telephone conversation.

[0878] "Data transmission means" refers to a device or function for transmitting acquired voice data to a server via the Internet.

[0879] "Received data storage means" refers to a device or function for storing voice data received by the server in a specific format suitable for analysis.

[0880] "Voice analysis means" refers to a device or function for analyzing received voice data using a generative AI model.

[0881] "Fraud detection means" refers to devices or functions that compare the analysis results with a database of past crimes to detect fraudulent activity.

[0882] "Warning notification means" refers to a device or function for sending a warning to a user's terminal based on the result of fraud detection.

[0883] "Automatic notification means" refers to devices and functions for notifying police and social workers of detection results.

[0884] A "generative AI model" is an artificial intelligence model that is pre-trained for a specific task and is used to analyze and identify fraud-specific language and patterns.

[0885] A "prompt" refers to an instruction or question that specifically indicates to the generative AI model the content and purpose of the data to be analyzed.

[0886] The present invention provides a system for preventing telephone fraud targeting the elderly. The system includes a user terminal, a server, and terminals of police and social workers linked to the server. Specific embodiments are described in detail below.

[0887] User device behavior

[0888] Audio Acquisition:

[0889] When a user starts a conversation using a telephone, the user's device captures the conversation voice in real time using a built-in microphone. The captured voice data is stored in a buffer and accumulated at regular intervals.

[0890] Data transmission:

[0891] The user's terminal transmits the acquired voice data to a server via the Internet in real time.

[0892] Server Operation

[0893] Audio data reception:

[0894] The server receives the voice data sent from the user's terminal and stores this data in a specific format (for example, WAV format).

[0895] Audio Analysis:

[0896] The server analyzes the received voice data using a generative AI model that has been trained on fraud-specific phrases and patterns and can identify signs of fraud.

[0897] Fraud Detection

[0898] The server compares the results of the generated AI model with a database of past crimes to determine whether fraud has occurred, based on specific keywords and patterns.

[0899] Alerts and automatic notifications

[0900] User warning notice:

[0901] If fraudulent activity is detected, the server will first send a warning to the user's device, which will alert the user to the content of the conversation.

[0902] Automatic notification to professional authorities:

[0903] At the same time, the server sends detailed notifications to the police and social workers' devices, including basic information about the user, a summary of the conversation, and any detected fraud patterns.

[0904] Specific examples

[0905] Let's assume that while a user (elderly person) is making a phone call, the other party uses words that fit into fraudulent patterns, such as "I need money" or "I've been in an accident."

[0906] 1. The user's device acquires the conversation content in real time and sends it to the server.

[0907] 2. The server receives the audio data and the generative AI model analyzes it.

[0908] 3. After checking against a database of past crimes, it is determined that there is a high possibility of fraud.

[0909] 4. The server immediately sends a warning to the user's device, displaying the message "Possible fraud. Please be careful."

[0910] 5. At the same time, the server automatically notifies the police and social workers and sends them details.

[0911] Prompt Sentence Examples

[0912] Determine whether the email contains words that fit the scam pattern, such as "I need money" or "I've been in an accident."

[0913] In this way, the present invention can quickly prevent elderly people from becoming victims of telephone fraud. Immediate notification to specialized agencies allows for even faster response. The main means and operations of this system are as described above.

[0914] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0915] Step 1: Audio capture

[0916] When a user starts a call, the device's built-in microphone is activated and captures the conversation voice in real time. The captured voice data is stored in a buffer. The data stored in the buffer is accumulated at regular intervals.

[0917] Input: User's voice

[0918] Output: Buffered audio data

[0919] Specific behavior:

[0920] A user places a call.

[0921] The device's microphone picks up the audio.

[0922] The acquired audio data is stored in a buffer.

[0923] Step 2: Send data

[0924] The user's device transmits the voice data stored in the buffer to the server via the Internet in real time at regular intervals.

[0925] Input: Buffered audio data

[0926] Output: Audio data sent to a server over the internet

[0927] Specific behavior:

[0928] The device establishes an internet connection.

[0929] The audio data stored in the buffer is sent to the server at regular intervals.

[0930] Step 3: Receiving audio data

[0931] The server receives the voice data sent from the user's terminal and saves it in a specific format suitable for analysis (for example, WAV format).

[0932] Input: Audio data sent over the internet

[0933] Output: Audio data saved in a specific format

[0934] Specific behavior:

[0935] The server starts a module for receiving audio data.

[0936] Receives audio data sent from the terminal.

[0937] Save the received data in a specific format.

[0938] Step 4: Audio analysis

[0939] The server analyzes the voice data using a generative AI model that has been trained in advance to recognize scam-specific phrases and patterns.

[0940] Input: Stored audio data

[0941] Output: Analysis result (possibility of fraud)

[0942] Specific behavior:

[0943] The server launches the generative AI model.

[0944] The voice data is analyzed based on the prompt sentence.

[0945] Example prompt: "Please rate this audio recording for signs of fraud."

[0946] Generative AI models identify potential fraud.

[0947] Step 5: Fraud detection

[0948] The server compares the analysis results of the generated AI model with a database of past crimes to determine whether there is a high possibility of fraud.

[0949] Input: Analysis results of the generative AI model

[0950] Output: Determination of likelihood of fraud

[0951] Specific behavior:

[0952] Obtain the analytical data generated by the generative AI model.

[0953] The analytical data is compared with a database of past crimes.

[0954] Score and determine the likelihood of fraud.

[0955] Step 6: Warning Notification

[0956] If fraud is detected, the server sends a warning message to the user's terminal.

[0957] Input: Possible fraud verdict

[0958] Output: The warning message sent to the user's terminal.

[0959] Specific behavior:

[0960] The server generates a warning message.

[0961] A warning message is sent to the user's device via the Internet.

[0962] The user's device will display a warning message and give an audio notification.

[0963] Step 7: Automatic Notifications

[0964] The server simultaneously sends notifications containing detailed information to the police and social worker terminals.

[0965] Input: Possible fraud verdict

[0966] Output: Notification messages sent to police and social worker devices

[0967] Specific behavior:

[0968] The server generates a notification message.

[0969] The notification message will include details such as the user's basic information, a summary of the conversation, and any fraud patterns detected.

[0970] Sending notification messages via the internet to police and social workers.

[0971] The police and social worker terminals receive and display the notification message.

[0972] (Application example 1)

[0973] 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."

[0974] Senior citizens falling victim to telephone fraud has become a social problem. These frauds are sophisticated and often result in seniors losing large amounts of money without even realizing it. A real-time, effective method to address this problem is needed.

[0975] 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.

[0976] In this invention, the server includes a voice acquisition means for analyzing the elderly person's telephone conversations in real time, a data transmission means for transmitting the acquired voice data to the server, a voice analysis means for analyzing the voice data received by the server using a generative AI model, a fraud detection means for comparing the analysis results with past crime information and detecting fraud, a warning notification means for sending a warning to the user's terminal based on the fraud detection results, an automatic notification means for notifying a local protective agency and family of the detection results, and a notification means for identifying fraud in real time, displaying a warning to the user, and sending a notification to a specialist agency and a registered guardian. This makes it possible to detect fraud and issue a warning before the elderly person falls victim to a telephone scam, and notify the specialist agency and guardian.

[0977] The "voice capture means" is a device or system that captures telephone conversations in real time and stores them as voice data.

[0978] The "data transmission means" is a device or system for transmitting the acquired voice data to the server.

[0979] A "voice analysis means" is a device or system that uses a generative AI model to analyze voice data and identify specific patterns or phrases.

[0980] The "fraud detection means" is a device or system that compares the data analyzed by the voice analysis means with past crime information to detect fraudulent acts.

[0981] The "warning notification means" is a device or system that sends a warning to a user's terminal based on the result of detecting fraudulent activity.

[0982] "Automatic notification means" means a device or system for automatically notifying local protective agencies and families of the results of a detection.

[0983] "Notification mechanism" means a device or system that identifies fraudulent activity in real time, displays a warning to the user, and sends notifications to professional organizations and registered parents.

[0984] The present invention is a system for detecting in real time the risk of elderly people falling victim to telephone fraud and responding promptly. Specifically, the system includes a voice acquisition means, a data transmission means, a voice analysis means, a fraud detection means, a warning notification means, an automatic notification means, and a notification means.

[0985] Audio acquisition means

[0986] The user's device uses a built-in microphone to capture the telephone conversation in real time. The captured voice data is temporarily stored in a buffer. This voice data is collected continuously during the call.

[0987] Data transmission method

[0988] The user's device transmits the collected voice data to a server via the Internet. This transmission is done in real time, and the voice data is immediately passed to the server.

[0989] Voice analysis methods

[0990] The server receives the voice data sent from the user's device and analyzes it using a generative AI model. The generative AI model has been trained in advance on fraud patterns and keywords, allowing it to detect signs of fraud with high accuracy.

[0991] Fraud detection measures

[0992] The analyzed voice data is then compared with past criminal records, and if there is a high possibility of fraud, fraudulent activity is detected based on relevant patterns and keywords.

[0993] Warning notification means

[0994] If fraud is detected, the server first sends a warning to the user's device, displaying a message such as "Possible fraud. Please be careful."

[0995] automatic notification means

[0996] At the same time, the server automatically processes and transmits the information to notify local protective agencies and family members of the results of the detection, enabling immediate action in the event of an emergency.

[0997] Notification means

[0998] If the server identifies fraudulent activity, it will display a warning to the user in real time and send a notification to the relevant professional organization and registered guardian. This function is expected to enable prompt action in the event of an incident.

[0999] Specific examples

[1000] For example, if an elderly person is having a phone conversation that includes the phrases "I need money" and "I had an accident," the following process takes place:

[1001] 1. Audio capture:

[1002] The user's device captures the conversation audio in real time.

[1003] 2. Data transmission:

[1004] The audio data is sent to a server via the Internet.

[1005] 3. Audio analysis:

[1006] The server analyzes the received audio data.

[1007] 4. Fraud detection:

[1008] Signs of fraud are detected.

[1009] 5. Warning notice:

[1010] The message "This may be a scam. Please be careful" will be displayed on the user's device.

[1011] 6. Automatic Notifications:

[1012] Notification will be sent to family members and local protective agencies.

[1013] Prompt Sentence Examples

[1014] Voice data: "I need money" "I had an accident"

[1015] Class Label: Fraud Pattern

[1016] As described above, the present invention provides a concrete means for reducing the risk of elderly people becoming victims of telephone fraud and for increasing safety.

[1017] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1018] Step 1:

[1019] The user's device captures the phone conversation in real time. The input is the user's voice, which is captured by a built-in microphone. The voice data is temporarily stored in a buffer.

[1020] Step 2:

[1021] The terminal transmits the acquired voice data to the server. The input is the voice data in the buffer. The data is transmitted via the Internet and reaches the server. The output is the voice data stored on the server.

[1022] Step 3:

[1023] The server analyzes the received voice data using a generative AI model. The input is the voice data stored on the server. The generative AI model has previously learned fraud patterns and analyzes the voice data to detect signs of fraud. The output is the analysis results.

[1024] Step 4:

[1025] The server compares the analysis results with past criminal information to detect fraud. The input is the analysis results from the generative AI model, which are compared with a database of past criminal activity. If the fraud patterns match, it is determined that fraud exists. The output is a detection result indicating whether or not fraud exists.

[1026] Step 5:

[1027] The server sends a warning to the user's terminal based on the fraud detection result. The input is the result that fraud has been detected. The warning message is generated and sent to the user's terminal. The output is the user's terminal with the warning message displayed.

[1028] Step 6:

[1029] The server notifies the local protection agency and the family of the detection result. The input is the fraud detection result and basic information of the user. The notification content is automatically generated and sent to the registered parent or guardian and protection agency. The output is the completed notification.

[1030] Step 7:

[1031] The server identifies fraudulent activity in real time, displays a warning to the user, and sends notifications to professional organizations and registered guardians. The input is the fraud identification and warning message. The output is the warning displayed on the user's device and the notification sent.

[1032] Through the above processing steps, the present invention provides a specific means for seniors to quickly take action before they fall victim to telephone fraud.

[1033] 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.

[1034] The present invention provides a system for preventing telephone fraud targeting the elderly, which further incorporates an emotion engine that recognizes the user's emotions. The system includes a user terminal, a server, and terminals of police and social workers that are linked to the server. Specific embodiments will now be described.

[1035] User device behavior

[1036] Audio Acquisition:

[1037] When a user starts a conversation using a telephone, the user's device captures the conversation voice in real time using a built-in microphone. The captured voice data is stored in a buffer and accumulated at regular intervals.

[1038] Data transmission:

[1039] The device transmits the captured voice data to a server via the Internet. This transmission is done in real time, allowing the server to quickly receive and analyze the data.

[1040] Server Operation

[1041] Audio data reception:

[1042] The server receives the voice data sent from the user's device and stores it in a specific format for analysis.

[1043] Audio Analysis:

[1044] The server analyzes the received voice data using a generative AI model and emotion engine. The generative AI model is pre-trained and can identify scam-specific language and patterns. The emotion engine recognizes emotions from the user's tone of voice and speaking rate, and reflects them in the analysis results.

[1045] Fraud Detection:

[1046] The server compares the voice analysis results with a criminal database to determine the likelihood of fraud. Fraud detection is based on specific keywords and patterns, as well as the user's emotional state. For example, if the user shows signs of anxiety or nervousness, it is determined that fraud is likely.

[1047] Warning notice:

[1048] If fraudulent activity is detected, the server will first send a warning message to the user's device, which will include appropriate wording based on the user's emotional state, encouraging the user to be more careful about the content of the conversation.

[1049] Automatic notifications:

[1050] At the same time, the server sends detailed notifications to the police and social workers' devices, including basic information about the user, a summary of the conversation, detected fraud patterns, and information about the user's emotional state.

[1051] Specific examples

[1052] Case 1: System behavior when elderly people feel uneasy about fraudsters

[1053] Consider the case where a user (elderly person) is making a phone call and the other party uses words that fit into fraudulent patterns, such as "I need money" or "I've been in an accident," and the user feels anxious or tense during the conversation.

[1054] 1. The user's device captures the conversation content and the user's tone of voice in real time and sends them to the server.

[1055] 2. The server receives the voice data, and the generative AI model and emotion engine analyze this data.

[1056] 3. After checking against a database of past crimes, it is determined that there is a high possibility of fraud and that the user is feeling uneasy.

[1057] 4. The server immediately sends a warning message to the user's device, stating, "This may be a scam. Please remain calm."

[1058] 5. At the same time, the server automatically notifies the police and social workers and sends them details.

[1059] In this way, the present invention can quickly prevent users (elderly people) from falling victim to telephone fraud, and by using an emotion engine, it can provide more accurate judgments and warnings. Immediate notification to specialized institutions enables a quick response.

[1060] The embodiment for carrying out the present invention is a system including the main means and specific operations according to the means as described above.

[1061] The processing flow will be explained below.

[1062] Step 1: Audio capture

[1063] When a user starts a conversation using the phone, the device captures the conversation voice in real time through the built-in microphone. The captured voice data is stored in a buffer and accumulated at regular intervals.

[1064] Step 2: Send data

[1065] The device transmits the captured audio data to a server via the Internet. This transmission is done in real time, so the server can receive and analyze the data immediately.

[1066] Step 3: Receiving audio data

[1067] The server receives the voice data sent from the user's device and stores it in a specific format for analysis.

[1068] Step 4: Audio analysis

[1069] The server analyzes the received voice data using a generative AI model and an emotion engine. The generative AI model is pre-trained and can identify scam-specific language and patterns. The emotion engine recognizes emotions from the user's tone of voice and speaking rate, and reflects them in the analysis results.

[1070] Step 5: Fraud detection

[1071] The server compares the voice analysis results with a criminal database to determine whether the person is likely to be committing fraud. The server determines whether the person is committing fraud based on certain keywords and patterns, as well as the user's emotional state. For example, if the user shows signs of anxiety or nervousness, the server determines that the person is likely to be committing fraud.

[1072] Step 6: Warning Notification

[1073] If fraudulent activity is detected, the server will send a warning message to the user's device, which will include appropriate wording based on the user's emotional state, encouraging the user to be more careful about the content of the conversation.

[1074] Step 7: Automatic Notifications

[1075] The server sends detailed notifications to police and social workers, including basic information about the user, a summary of the conversation, any fraud patterns detected, and information about the user's emotional state.

[1076] Step 8: Continuous monitoring

[1077] The server continues to monitor the conversation after the notification has been sent and will notify again if any further fraud attempts are detected. This continuous monitoring helps prevent multiple fraud attempts from occurring.

[1078] Example 2

[1079] 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."

[1080] Previously, countermeasures against telephone fraud targeting the elderly were limited in their ability to prevent fraud before it occurred. Furthermore, there was no way to respond quickly when a fraudulent activity was in progress, making elderly people vulnerable to victimization. Furthermore, detecting fraud requires advanced analysis that takes into account emotional changes, but conventional technology was unable to adequately address this issue.

[1081] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1082] In this invention, the server includes a voice acquisition means for acquiring the elderly person's telephone conversation in real time, a data transmission means for transmitting the acquired voice data to the server via the Internet, a voice analysis means for analyzing the voice data received by the server using a pre-trained generative AI model and an emotion engine, a fraud detection means for comparing the analysis results with past crime data and determining whether fraud has occurred, a warning notification means for sending a warning to the user's device based on the fraud determination result, and an automatic notification means for notifying the police and social workers of the determination result. This makes it possible to detect fraud in real time while taking into account the elderly person's emotional state, and to quickly issue a warning and take action.

[1083] "Voice capture means" is a collective term for the hardware and software used to capture a user's telephone conversation in real time.

[1084] The "data transmission means" is a means for transmitting the acquired voice data to a server via the Internet.

[1085] A "server" is a device or system that includes a central processing unit and associated peripherals for receiving and analyzing audio data.

[1086] A "generative AI model" is an artificial intelligence model that is pre-trained for a specific purpose and is used to identify specific keywords and patterns of fraudulent activity.

[1087] The "emotion engine" is a software model that analyzes the tone and rate of a user's voice to recognize their emotional state.

[1088] "Speech analysis means" refers to means for analyzing received speech data using a generative AI model and an emotion engine.

[1089] "Fraud detection means" refers to a means for comparing the analysis results with a database of past crimes to determine whether or not fraud has occurred.

[1090] The "warning notification means" is a means for sending a warning to the user's terminal based on the result of the fraudulent activity determination.

[1091] "Automatic notification means" refers to a means for notifying police and social workers of fraud determination results.

[1092] MODE FOR CARRYING OUT THE INVENTION

[1093] The present invention provides a system for preventing telephone fraud targeting elderly people, and we will explain how to implement it in detail. This system includes a user's terminal, a server, and terminals of police and social workers that are linked to the server.

[1094] Hardware and software used

[1095] The following hardware and software are used to implement this system:

[1096] User device: Smartphone with built-in microphone and internet connection

[1097] Server: A server in a data center with a high-performance processor and large storage capacity

[1098] Generative AI models: AI models pre-trained to identify fraud-specific language and patterns

[1099] Emotion Engine: Software for recognizing emotions from the user's tone of voice and speech rate

[1100] User device behavior

[1101] When a user starts a conversation using a telephone, the user's device uses a built-in microphone to capture the conversation audio in real time. The captured audio data is stored in a buffer, and data is accumulated at regular intervals. The device then transmits the buffered audio data to a server via the Internet. The transmission is performed in real time and is sent continuously without any divisions.

[1102] Server Operation

[1103] The server receives the voice data sent from the user's device and stores it in a specific format. The server then analyzes the received voice data using a generative AI model and an emotion engine. The generative AI model is pre-trained to identify keywords and patterns specific to fraud. The emotion engine recognizes emotions from the user's tone and speed of voice and reflects them in the analysis results.

[1104] Based on the results of the above analysis, the server compares the voice analysis results with a criminal database to determine the possibility of fraud, which is determined based on specific keywords and patterns, as well as the user's emotional state.

[1105] Alerts and automatic notifications

[1106] When fraudulent activity is detected, the server first sends a warning message to the user's device. The warning contains appropriate wording based on the user's emotional state, encouraging the user to be more careful about the content of the conversation. At the same time, the server sends a notification with detailed information to the police and social worker's devices. The notification content includes the user's basic information, a summary of the conversation, the detected fraud pattern, and information about the user's emotional state.

[1107] Specific examples

[1108] Case 1: System behavior when elderly people feel uneasy about fraudsters

[1109] 1. Consider a scenario where a user (elderly person) is making a phone call and the other party uses words that fit into fraudulent patterns, such as "I need money" or "I've been in an accident," and the user feels anxious or tense during the conversation.

[1110] 2. The user's device captures the conversation content and the user's tone of voice in real time and sends them to the server.

[1111] 3. The server receives the voice data and the generative AI model and emotion engine analyze this data.

[1112] 4. After checking against a database of past crimes, it is determined that there is a high possibility of fraud and that the user is feeling uneasy.

[1113] 5. The server immediately sends a warning message to the user's device, stating, "This may be a scam. Please remain calm."

[1114] 6. At the same time, the server automatically notifies the police and social workers and sends them details.

[1115] Prompt Sentence Examples

[1116] "A 72-year-old elderly person, Mr. A, received a phone call saying, 'Your grandson has been in an accident.' Hearing this, Mr. A became very anxious. How would you analyze this situation and detect possible fraud?"

[1117] Using this prompt, the generative AI model and emotion engine can analyze fraud patterns and user emotions and take appropriate action.

[1118] The above is an embodiment of the present invention. The present invention is a system that detects fraudulent activity while taking into account the emotional state of seniors in real time, and provides prompt warning and response.

[1119] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1120] Step 1:

[1121] Audio Acquisition

[1122] When a user starts a conversation using the phone, the device uses the built-in microphone to capture the conversation audio in real time. The user's voice input is stored as data in a buffer. This voice data is updated at regular intervals (e.g., every second). If the user is talking on the phone with a friend, the device simultaneously captures the friend's voice and the user's voice and records them in a buffer.

[1123] Step 2:

[1124] Data transmission

[1125] The device transmits the voice data stored in the buffer to the server via the Internet. The user's device converts the voice data stored in the buffer into packets and sends them to the server. This transmission is done in real time, and the data is processed so that it is sent continuously without being divided. The user's device transmits a packet of voice data via the Internet every second.

[1126] Step 3:

[1127] Audio data reception

[1128] The server receives voice data sent from the user's device. It stores voice data packets received from the Internet in a specific format in storage. The server restores the data immediately after receiving it and stores it in storage as an audio file for analysis. One second after the user starts the call, the server receives the first voice data packet and stores it in a specified directory.

[1129] Step 4:

[1130] Audio analysis

[1131] The server analyzes the received voice data using a generative AI model and an emotion engine. The input data is the received audio file, which is fed to the generative AI model for analysis. The generative AI model is pre-trained to identify keywords and patterns specific to fraud. The emotion engine recognizes emotions from the tone and rate of the user's voice and combines the analysis results with the output of the generative AI model. For example, the server analyzes whether the user uttered a phrase such as "I need money" and also determines whether the tone of voice is unstable.

[1132] Step 5:

[1133] Fraud Detection

[1134] The server compares the voice analysis results with a database of past crimes to determine the likelihood of fraud. The input data is the analysis results from the generative AI model and emotion engine, and is compared with the criminal database to determine fraud. The server matches the analyzed data with pre-registered criminal data and evaluates the likelihood of fraud based on specific keywords and patterns, as well as the user's emotional state. For example, if the phrase "I had an accident" is detected, it is determined that there is a high possibility of fraud.

[1135] Step 6:

[1136] Warning notice

[1137] If fraud is detected, the server first sends a warning message to the user's device. The input data is the fraud judgment result, and the warning message is created based on that. The warning message contains appropriate wording based on the user's emotional state and is sent to the user's device. For example, if the system detects fraud, a message saying "There is a possibility of fraud. Please remain calm" is displayed on the user's smartphone.

[1138] Step 7:

[1139] Automatic notifications

[1140] At the same time, the server sends a notification containing detailed information to the police and social worker terminals. The input data is the fraud detection result and the user's basic information, which constitutes the notification content. The notification content includes the user's basic information, a summary of the conversation, the detected fraud pattern, and information about the user's emotional state. For example, if fraud is confirmed, the server immediately connects to the police system and sends detailed information. The social worker also receives the information.

[1141] In this way, the system prevents telephone fraud against seniors in real time and enables prompt and appropriate responses.

[1142] (Application example 2)

[1143] 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."

[1144] Telephone scams targeting seniors often involve sophisticated tactics by scammers who exploit seniors' anxieties and nervousness to defraud them of their money. Current technology struggles to detect fraud in real time, and there is a lack of mechanisms to warn seniors before a fraud occurs. Furthermore, there are no fraud detection and warning systems that take into account the emotional state of seniors. Therefore, there is a need for a system that can quickly and effectively protect seniors from fraud.

[1145] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a voice acquisition means for receiving voice data, a data transmission means for transmitting the acquired voice data in real time, a voice analysis means for analyzing the voice data using a generative AI model, a fraud detection means for detecting fraud by comparing the analysis results with a past crime database, an emotion recognition means for analyzing the user's emotional state, a warning notification means for sending a warning based on the fraud detection result, and a notification means for alerting the user based on the possibility of fraud. This enables real-time fraud detection and warning notification that takes into account the emotional state of the elderly.

[1146] "Elderly" refers to individuals who are older and generally more likely to be targeted by fraud.

[1147] "Telephone conversation" refers to a voice conversation conducted over the telephone.

[1148] "Audio capture means" refers to devices or software for recording telephone conversations in real time.

[1149] "Data transmission means" refers to a device or software for transmitting acquired voice data to a server.

[1150] A "generative AI model" is a pre-trained artificial intelligence model that is used to analyze voice data to detect fraudulent activity.

[1151] "Voice analysis means" refers to a device or software for analyzing voice data using a generative AI model.

[1152] "Fraud detection methods" refer to devices or software that compare voice analysis results with a database of past crimes to detect fraudulent activity.

[1153] "Warning notification means" refers to a device or software for sending a warning message to a user's terminal based on the results of fraudulent activity detection.

[1154] "Automatic notification means" refers to devices or software that notify police and social workers based on the detection of fraudulent activity.

[1155] "Emotion recognition means" refers to devices or software that analyze the user's tone of voice, speaking rate, etc. to recognize their emotional state.

[1156] "Notification means" means a device or software that sends a message to a user to warn them of possible fraud.

[1157] Overall system overview

[1158] This invention is a system for preventing telephone fraud targeting the elderly. The system includes a user terminal, a server, and police and social worker terminals. The system has the following main functions: a voice acquisition means, a data transmission means, a voice analysis means, a fraud detection means, an emotion recognition means, a warning notification means, an automatic notification means, and a notification means.

[1159] User terminal operation

[1160] Audio acquisition method:

[1161] When a user initiates a call, the user's device captures the conversation in real time using a built-in microphone. The voice data is stored in a buffer and accumulated at regular intervals.

[1162] Data transmission method:

[1163] The device transmits the captured audio data to a server via the Internet. This transmission is done in real time, allowing the server to quickly receive and analyze the data.

[1164] Server Operation

[1165] Audio data reception:

[1166] The server receives the voice data sent from the user's device and stores it in a specific format.

[1167] Audio analysis methods:

[1168] The server analyzes the received voice data using a generative AI model that has been pre-trained to identify scam-specific phrases and patterns.

[1169] Emotion recognition means:

[1170] The server recognizes the user's emotions from the tone of their voice, the rate at which they speak, etc. This improves the accuracy of the analysis if the user is feeling anxious or nervous.

[1171] Fraud detection measures:

[1172] The results of voice analysis and emotion recognition are combined and compared with a database of past crimes to determine the possibility of fraud.

[1173] Warning notification means:

[1174] If fraud is detected, the server first sends a warning message to the user's device. The message content is adjusted based on the user's emotional state, for example, "There is a possibility of fraud. Please remain calm."

[1175] Automatic notification method:

[1176] At the same time, the server sends a notification to police and social workers with detailed information, including basic information about the user, a summary of the conversation, any fraud patterns detected, and information about the user's emotional state.

[1177] Means of notification:

[1178] To ease the user's anxiety and tension, appropriate steps and contact information may also be provided.

[1179] Specific examples

[1180] Audio capture and data transmission:

[1181] When an elderly person (user) starts a call, the device's built-in microphone picks up the voice, and this data is sent to the server in real time.

[1182] Speech analysis and emotion recognition:

[1183] The server analyzes the received voice data and identifies fraud patterns such as "I need money" or "I've had an accident." At the same time, the emotion engine recognizes emotions such as anxiety and tension from the user's voice.

[1184] Fraud detection and warning notifications:

[1185] If it is determined that there is a high possibility of fraud, a warning message such as "This may be fraud. Please remain calm" will be displayed on the user's device.

[1186] Automatic notifications:

[1187] Similarly, police and social workers will be automatically notified of details of fraudulent activity, facilitating a swift response.

[1188] An example of a prompt sentence could be set up for the generative AI model as follows: "If the user shows signs of anxiety or tension during the phone call, determine that there is a high possibility of fraud. Analyze fraudulent patterns of words and the user's emotional state and display a warning."

[1189] In this way, the system also takes into account the user's emotional state and can quickly and effectively protect seniors from fraud.

[1190] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1191] Step 1: Audio capture

[1192] When a user starts a call, the user's device uses the built-in microphone to capture the conversation in real time. The audio data is stored in a local buffer, and data is accumulated at regular intervals. The input is the audio during the call, and the output is the audio data stored in the buffer.

[1193] Step 2: Send data

[1194] The user's terminal transmits the voice data stored in the local buffer to the server in real time. The data is uploaded to the server via the Internet using a data transmission means, so the input is the voice data in the buffer and the output is the voice data transmitted to the server.

[1195] Step 3: Receiving audio data

[1196] The server receives the voice data sent from the user terminal. The received voice data is stored in a specific format for analysis. The input is the voice data sent from the terminal, and the output is the voice data stored in the data storage in the server.

[1197] Step 4: Audio analysis

[1198] The server analyzes the stored voice data using a generative AI model. First, it converts the voice data into text, then analyzes the text to identify scam-specific phrases and patterns. The input is the stored voice data, and the output is the resulting text data.

[1199] Step 5: Emotion Recognition

[1200] During the voice analysis, the server uses emotion recognition means to analyze the user's tone of voice and speaking rate to determine the user's emotional state. The input is the voice data and analyzed text data, and the output is an evaluation of the user's emotional state.

[1201] Step 6: Fraud detection

[1202] The server compares the results of voice analysis and emotion recognition with a database of past crimes to assess the likelihood of fraud. If certain keywords or patterns match, it determines that there is a high possibility of fraud. The input is the analysis results and emotion evaluation results, and the output is an assessment of the likelihood of fraud.

[1203] Step 7: Warning Notification

[1204] If fraud is detected, the server sends a warning message to the user's device. The message content is adjusted based on the analysis results and the user's emotional state. For example, it might say, "There is a possibility of fraud. Please remain calm." The input is the fraud detection result, and the output is the warning message displayed on the user's device.

[1205] Step 8: Automatic Notifications

[1206] At the same time, the server also notifies the police and social workers. The notification content includes the user's basic information, a summary of the conversation, detected fraud patterns, and information about the user's emotional state. The input is the fraud detection results and related information, and the output is detailed information sent to the police and social workers.

[1207] At each step, different hardware and software in the system work together to effectively prevent elderly people from falling victim to fraud.

[1208] 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.

[1209] 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.

[1210] 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.

[1211] [Fourth embodiment]

[1212] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1213] 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.

[1214] 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).

[1215] 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.

[1216] 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.

[1217] 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).

[1218] 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.

[1219] 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.

[1220] 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.

[1221] 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.

[1222] 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.

[1223] 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.

[1224] 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."

[1225] The present invention provides a system for preventing telephone fraud targeting elderly people. The system includes a user terminal, a server, and terminals of police and social workers that are linked to the server. Specific embodiments will now be described.

[1226] User device behavior

[1227] Audio Acquisition:

[1228] When a user starts a conversation using a telephone, the user's device captures the conversation voice in real time using a built-in microphone. The captured voice data is stored in a buffer and accumulated at regular intervals.

[1229] Data transmission:

[1230] The device transmits the captured voice data to a server via the Internet. This transmission is done in real time, allowing the server to quickly receive and analyze the data.

[1231] Server Operation

[1232] Audio data reception:

[1233] The server receives the voice data sent from the user's device and stores it in a specific format for analysis.

[1234] Audio Analysis:

[1235] The received voice data is analyzed using a generative AI model that has been pre-trained to identify scam-specific phrases and patterns.

[1236] Fraud Detection:

[1237] The analysis results are compared with a database of past crimes, and if there is a high possibility of fraud, a determination is made as to whether it is fraud. Fraud is determined based on specific keywords and patterns.

[1238] Warning notice:

[1239] If fraudulent activity is detected, the server first sends a warning to the user's device, which alerts the user to the content of the conversation.

[1240] Automatic notifications:

[1241] At the same time, the server sends detailed notifications to the police and social workers' devices, including basic information about the user, a summary of the conversation, and any fraud patterns detected.

[1242] Specific examples

[1243] Case 1: An example where the system intervenes before an elderly person is deceived

[1244] Consider the case where a user (elderly person) is making a phone call and the other party uses words that fit into fraudulent patterns, such as "I need money" or "I've been in an accident."

[1245] 1. The user's device acquires the conversation content in real time and sends it to the server.

[1246] 2. The server receives the audio data and the generative AI model analyzes it.

[1247] 3. After checking against a database of past crimes, it is determined that there is a high possibility of fraud.

[1248] 4. The server immediately sends a warning to the user's device, displaying the message "Possible fraud. Please be careful."

[1249] 5. At the same time, the server automatically notifies the police and social workers and sends them details.

[1250] In this way, the present invention can quickly prevent users (elderly people) from becoming victims of telephone fraud. By immediately notifying specialized agencies, even faster response is possible.

[1251] The embodiment for carrying out the present invention is a system including the main means and specific operations according to the means as described above.

[1252] The processing flow will be explained below.

[1253] Step 1: Audio capture

[1254] When a user starts a conversation using the phone, the device captures the conversation voice in real time through the built-in microphone, and the captured voice data is stored in a buffer at regular intervals.

[1255] Step 2: Send data

[1256] The device transmits the captured voice data to a server via the Internet. This transmission is done in real time, so the server can receive and analyze the data immediately.

[1257] Step 3: Receiving audio data

[1258] The server receives the voice data sent from the user's device and stores the received data in a specific format for analysis.

[1259] Step 4: Audio analysis

[1260] The server analyzes the received voice data using an artificial intelligence model, and a pre-trained generative AI model determines the content of the conversation based on this data.

[1261] Step 5: Fraud detection

[1262] The server uses the results of the voice analysis to determine whether the call is likely to be fraudulent, checking it against a database of past crimes and whether it contains specific keywords or phrases.

[1263] Step 6: Warning Notification

[1264] If there is a high possibility of fraudulent activity, the server will send a warning message to the user's device, allowing the user to pay attention to the content of the conversation.

[1265] Step 7: Automatic Notifications

[1266] The server also sends detailed notifications to police and social worker devices, including basic information about the user, a summary of the conversation, and any fraud patterns detected.

[1267] Step 8: Continuous monitoring

[1268] The server continues to monitor the conversation after the notification has been sent and will notify again if any further fraud attempts are detected. This continuous monitoring helps prevent multiple fraud attempts.

[1269] Example 1

[1270] 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."

[1271] In modern society, telephone fraud targeting the elderly is on the rise, increasing the risk of elderly people becoming victims of fraud. Existing crime prevention measures are insufficient to address this issue, and there is a particular need for a system that can detect fraudulent activity in real time and respond quickly. In addition, it is often difficult for elderly people to recognize that they have been victimized by fraud, so immediate notification to specialized agencies is necessary.

[1272] 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.

[1273] In this invention, the server includes a voice capture means for capturing conversational voices in real time when a user is having a telephone conversation, a data transmission means for transmitting the captured voice data to the server via the Internet, a means for saving the voice data received by the server in a specific format, a voice analysis means for analyzing the received voice data using a generative AI model, a fraud detection means for comparing the analysis results with a past crime database to detect fraud, a warning notification means for sending a warning to the user's device based on the fraud detection results, and an automatic notification means for notifying the police and social workers of the detection results. This enables telephone fraud against elderly people to be detected in real time and responded to promptly.

[1274] The term "voice acquisition means" refers to a device or function for acquiring conversational voice in real time when a user is having a telephone conversation.

[1275] "Data transmission means" refers to a device or function for transmitting acquired voice data to a server via the Internet.

[1276] "Received data storage means" refers to a device or function for storing voice data received by the server in a specific format suitable for analysis.

[1277] "Voice analysis means" refers to a device or function for analyzing received voice data using a generative AI model.

[1278] "Fraud detection means" refers to devices or functions that compare the analysis results with a database of past crimes to detect fraudulent activity.

[1279] "Warning notification means" refers to a device or function for sending a warning to a user's terminal based on the result of fraud detection.

[1280] "Automatic notification means" refers to devices and functions for notifying police and social workers of detection results.

[1281] A "generative AI model" is an artificial intelligence model that is pre-trained for a specific task and is used to analyze and identify fraud-specific language and patterns.

[1282] A "prompt" refers to an instruction or question that specifically indicates to the generative AI model the content and purpose of the data to be analyzed.

[1283] The present invention provides a system for preventing telephone fraud targeting the elderly. The system includes a user terminal, a server, and terminals of police and social workers linked to the server. Specific embodiments are described in detail below.

[1284] User device behavior

[1285] Audio Acquisition:

[1286] When a user starts a conversation using a telephone, the user's device captures the conversation voice in real time using a built-in microphone. The captured voice data is stored in a buffer and accumulated at regular intervals.

[1287] Data transmission:

[1288] The user's terminal transmits the acquired voice data to a server via the Internet in real time.

[1289] Server Operation

[1290] Audio data reception:

[1291] The server receives the voice data sent from the user's terminal and stores this data in a specific format (for example, WAV format).

[1292] Audio Analysis:

[1293] The server analyzes the received voice data using a generative AI model that has been trained on fraud-specific phrases and patterns and can identify signs of fraud.

[1294] Fraud Detection

[1295] The server compares the results of the generated AI model with a database of past crimes to determine whether fraud has occurred, based on specific keywords and patterns.

[1296] Alerts and automatic notifications

[1297] User warning notice:

[1298] If fraudulent activity is detected, the server will first send a warning to the user's device, which will alert the user to the content of the conversation.

[1299] Automatic notification to professional authorities:

[1300] At the same time, the server sends detailed notifications to the police and social workers' devices, including basic information about the user, a summary of the conversation, and any detected fraud patterns.

[1301] Specific examples

[1302] Let's assume that while a user (elderly person) is making a phone call, the other party uses words that fit into fraudulent patterns, such as "I need money" or "I've been in an accident."

[1303] 1. The user's device acquires the conversation content in real time and sends it to the server.

[1304] 2. The server receives the audio data and the generative AI model analyzes it.

[1305] 3. After checking against a database of past crimes, it is determined that there is a high possibility of fraud.

[1306] 4. The server immediately sends a warning to the user's device, displaying the message "Possible fraud. Please be careful."

[1307] 5. At the same time, the server automatically notifies the police and social workers and sends them details.

[1308] Prompt Sentence Examples

[1309] Determine whether the email contains words that fit the scam pattern, such as "I need money" or "I've been in an accident."

[1310] In this way, the present invention can quickly prevent elderly people from becoming victims of telephone fraud. Immediate notification to specialized agencies allows for even faster response. The main means and operations of this system are as described above.

[1311] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1312] Step 1: Audio capture

[1313] When a user starts a call, the device's built-in microphone is activated and captures the conversation voice in real time. The captured voice data is stored in a buffer. The data stored in the buffer is accumulated at regular intervals.

[1314] Input: User's voice

[1315] Output: Buffered audio data

[1316] Specific behavior:

[1317] A user places a call.

[1318] The device's microphone picks up the audio.

[1319] The acquired audio data is stored in a buffer.

[1320] Step 2: Send data

[1321] The user's device transmits the voice data stored in the buffer to the server via the Internet in real time at regular intervals.

[1322] Input: Buffered audio data

[1323] Output: Audio data sent to a server over the internet

[1324] Specific behavior:

[1325] The device establishes an internet connection.

[1326] The audio data stored in the buffer is sent to the server at regular intervals.

[1327] Step 3: Receiving audio data

[1328] The server receives the voice data sent from the user's terminal and saves it in a specific format suitable for analysis (for example, WAV format).

[1329] Input: Audio data sent over the internet

[1330] Output: Audio data saved in a specific format

[1331] Specific behavior:

[1332] The server starts a module for receiving audio data.

[1333] Receives audio data sent from the terminal.

[1334] Save the received data in a specific format.

[1335] Step 4: Audio analysis

[1336] The server analyzes the voice data using a generative AI model that has been trained in advance to recognize scam-specific phrases and patterns.

[1337] Input: Stored audio data

[1338] Output: Analysis result (possibility of fraud)

[1339] Specific behavior:

[1340] The server launches the generative AI model.

[1341] The voice data is analyzed based on the prompt sentence.

[1342] Example prompt: "Please rate this audio recording for signs of fraud."

[1343] Generative AI models identify potential fraud.

[1344] Step 5: Fraud detection

[1345] The server compares the analysis results of the generated AI model with a database of past crimes to determine whether there is a high possibility of fraud.

[1346] Input: Analysis results of the generative AI model

[1347] Output: Determination of likelihood of fraud

[1348] Specific behavior:

[1349] Obtain the analytical data generated by the generative AI model.

[1350] The analytical data is compared with a database of past crimes.

[1351] Score and determine the likelihood of fraud.

[1352] Step 6: Warning Notification

[1353] If fraud is detected, the server sends a warning message to the user's terminal.

[1354] Input: Possible fraud verdict

[1355] Output: The warning message sent to the user's terminal.

[1356] Specific behavior:

[1357] The server generates a warning message.

[1358] A warning message is sent to the user's device via the Internet.

[1359] The user's device will display a warning message and give an audio notification.

[1360] Step 7: Automatic Notifications

[1361] The server simultaneously sends notifications containing detailed information to the police and social worker terminals.

[1362] Input: Possible fraud verdict

[1363] Output: Notification messages sent to police and social worker devices

[1364] Specific behavior:

[1365] The server generates a notification message.

[1366] The notification message will include details such as the user's basic information, a summary of the conversation, and any fraud patterns detected.

[1367] Sending notification messages via the internet to police and social workers.

[1368] The police and social worker terminals receive and display the notification message.

[1369] (Application example 1)

[1370] 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."

[1371] Senior citizens falling victim to telephone fraud has become a social problem. These frauds are sophisticated and often result in seniors losing large amounts of money without even realizing it. A real-time, effective method to address this problem is needed.

[1372] 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.

[1373] In this invention, the server includes a voice acquisition means for analyzing the elderly person's telephone conversations in real time, a data transmission means for transmitting the acquired voice data to the server, a voice analysis means for analyzing the voice data received by the server using a generative AI model, a fraud detection means for comparing the analysis results with past crime information and detecting fraud, a warning notification means for sending a warning to the user's terminal based on the fraud detection results, an automatic notification means for notifying a local protective agency and family of the detection results, and a notification means for identifying fraud in real time, displaying a warning to the user, and sending a notification to a specialist agency and a registered guardian. This makes it possible to detect fraud and issue a warning before the elderly person falls victim to a telephone scam, and notify the specialist agency and guardian.

[1374] The "voice capture means" is a device or system that captures telephone conversations in real time and stores them as voice data.

[1375] The "data transmission means" is a device or system for transmitting the acquired voice data to the server.

[1376] A "voice analysis means" is a device or system that uses a generative AI model to analyze voice data and identify specific patterns or phrases.

[1377] The "fraud detection means" is a device or system that compares the data analyzed by the voice analysis means with past crime information to detect fraudulent acts.

[1378] The "warning notification means" is a device or system that sends a warning to a user's terminal based on the result of detecting fraudulent activity.

[1379] "Automatic notification means" means a device or system for automatically notifying local protective agencies and families of the results of a detection.

[1380] "Notification mechanism" means a device or system that identifies fraudulent activity in real time, displays a warning to the user, and sends notifications to professional organizations and registered parents.

[1381] The present invention is a system for detecting in real time the risk of elderly people falling victim to telephone fraud and responding promptly. Specifically, the system includes a voice acquisition means, a data transmission means, a voice analysis means, a fraud detection means, a warning notification means, an automatic notification means, and a notification means.

[1382] Audio acquisition means

[1383] The user's device uses a built-in microphone to capture the telephone conversation in real time. The captured voice data is temporarily stored in a buffer. This voice data is collected continuously during the call.

[1384] Data transmission method

[1385] The user's device transmits the collected voice data to a server via the Internet. This transmission is done in real time, and the voice data is immediately passed to the server.

[1386] Voice analysis methods

[1387] The server receives the voice data sent from the user's device and analyzes it using a generative AI model. The generative AI model has been trained in advance on fraud patterns and keywords, allowing it to detect signs of fraud with high accuracy.

[1388] Fraud detection measures

[1389] The analyzed voice data is then compared with past criminal records, and if there is a high possibility of fraud, fraudulent activity is detected based on relevant patterns and keywords.

[1390] Warning notification means

[1391] If fraud is detected, the server first sends a warning to the user's device, displaying a message such as "Possible fraud. Please be careful."

[1392] automatic notification means

[1393] At the same time, the server automatically processes and transmits the information to notify local protective agencies and family members of the results of the detection, enabling immediate action in the event of an emergency.

[1394] Notification means

[1395] If the server identifies fraudulent activity, it will display a warning to the user in real time and send a notification to the relevant professional organization and registered guardian. This function is expected to enable prompt action in the event of an incident.

[1396] Specific examples

[1397] For example, if an elderly person is having a phone conversation that includes the phrases "I need money" and "I had an accident," the following process takes place:

[1398] 1. Audio capture:

[1399] The user's device captures the conversation audio in real time.

[1400] 2. Data transmission:

[1401] The audio data is sent to a server via the Internet.

[1402] 3. Audio analysis:

[1403] The server analyzes the received audio data.

[1404] 4. Fraud detection:

[1405] Signs of fraud are detected.

[1406] 5. Warning notice:

[1407] The message "This may be a scam. Please be careful" will be displayed on the user's device.

[1408] 6. Automatic Notifications:

[1409] Notification will be sent to family members and local protective agencies.

[1410] Prompt Sentence Examples

[1411] Voice data: "I need money" "I had an accident"

[1412] Class Label: Fraud Pattern

[1413] As described above, the present invention provides a concrete means for reducing the risk of elderly people becoming victims of telephone fraud and for increasing safety.

[1414] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1415] Step 1:

[1416] The user's device captures the phone conversation in real time. The input is the user's voice, which is captured by a built-in microphone. The voice data is temporarily stored in a buffer.

[1417] Step 2:

[1418] The terminal transmits the acquired voice data to the server. The input is the voice data in the buffer. The data is transmitted via the Internet and reaches the server. The output is the voice data stored on the server.

[1419] Step 3:

[1420] The server analyzes the received voice data using a generative AI model. The input is the voice data stored on the server. The generative AI model has previously learned fraud patterns and analyzes the voice data to detect signs of fraud. The output is the analysis results.

[1421] Step 4:

[1422] The server compares the analysis results with past criminal information to detect fraud. The input is the analysis results from the generative AI model, which are compared with a database of past criminal activity. If the fraud patterns match, it is determined that fraud exists. The output is a detection result indicating whether or not fraud exists.

[1423] Step 5:

[1424] The server sends a warning to the user's terminal based on the fraud detection result. The input is the result that fraud has been detected. The warning message is generated and sent to the user's terminal. The output is the user's terminal with the warning message displayed.

[1425] Step 6:

[1426] The server notifies the local protection agency and the family of the detection result. The input is the fraud detection result and basic information of the user. The notification content is automatically generated and sent to the registered parent or guardian and protection agency. The output is the completed notification.

[1427] Step 7:

[1428] The server identifies fraudulent activity in real time, displays a warning to the user, and sends notifications to professional organizations and registered guardians. The input is the fraud identification and warning message. The output is the warning displayed on the user's device and the notification sent.

[1429] Through the above processing steps, the present invention provides a specific means for seniors to quickly take action before they fall victim to telephone fraud.

[1430] 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.

[1431] The present invention provides a system for preventing telephone fraud targeting the elderly, which further incorporates an emotion engine that recognizes the user's emotions. The system includes a user terminal, a server, and terminals of police and social workers that are linked to the server. Specific embodiments will now be described.

[1432] User device behavior

[1433] Audio Acquisition:

[1434] When a user starts a conversation using a telephone, the user's device captures the conversation voice in real time using a built-in microphone. The captured voice data is stored in a buffer and accumulated at regular intervals.

[1435] Data transmission:

[1436] The device transmits the captured voice data to a server via the Internet. This transmission is done in real time, allowing the server to quickly receive and analyze the data.

[1437] Server Operation

[1438] Audio data reception:

[1439] The server receives the voice data sent from the user's device and stores it in a specific format for analysis.

[1440] Audio Analysis:

[1441] The server analyzes the received voice data using a generative AI model and emotion engine. The generative AI model is pre-trained and can identify scam-specific language and patterns. The emotion engine recognizes emotions from the user's tone of voice and speaking rate, and reflects them in the analysis results.

[1442] Fraud Detection:

[1443] The server compares the voice analysis results with a criminal database to determine the likelihood of fraud. Fraud detection is based on specific keywords and patterns, as well as the user's emotional state. For example, if the user shows signs of anxiety or nervousness, it is determined that fraud is likely.

[1444] Warning notice:

[1445] If fraudulent activity is detected, the server will first send a warning message to the user's device, which will include appropriate wording based on the user's emotional state, encouraging the user to be more careful about the content of the conversation.

[1446] Automatic notifications:

[1447] At the same time, the server sends detailed notifications to the police and social workers' devices, including basic information about the user, a summary of the conversation, detected fraud patterns, and information about the user's emotional state.

[1448] Specific examples

[1449] Case 1: System behavior when elderly people feel uneasy about fraudsters

[1450] Consider the case where a user (elderly person) is making a phone call and the other party uses words that fit into fraudulent patterns, such as "I need money" or "I've been in an accident," and the user feels anxious or tense during the conversation.

[1451] 1. The user's device captures the conversation content and the user's tone of voice in real time and sends them to the server.

[1452] 2. The server receives the voice data, and the generative AI model and emotion engine analyze this data.

[1453] 3. After checking against a database of past crimes, it is determined that there is a high possibility of fraud and that the user is feeling uneasy.

[1454] 4. The server immediately sends a warning message to the user's device, stating, "This may be a scam. Please remain calm."

[1455] 5. At the same time, the server automatically notifies the police and social workers and sends them details.

[1456] In this way, the present invention can quickly prevent users (elderly people) from falling victim to telephone fraud, and by using an emotion engine, it can provide more accurate judgments and warnings. Immediate notification to specialized institutions enables a quick response.

[1457] The embodiment for carrying out the present invention is a system including the main means and specific operations according to the means as described above.

[1458] The processing flow will be explained below.

[1459] Step 1: Audio capture

[1460] When a user starts a conversation using the phone, the device captures the conversation voice in real time through the built-in microphone. The captured voice data is stored in a buffer and accumulated at regular intervals.

[1461] Step 2: Send data

[1462] The device transmits the captured audio data to a server via the Internet. This transmission is done in real time, so the server can receive and analyze the data immediately.

[1463] Step 3: Receiving audio data

[1464] The server receives the voice data sent from the user's device and stores it in a specific format for analysis.

[1465] Step 4: Audio analysis

[1466] The server analyzes the received voice data using a generative AI model and an emotion engine. The generative AI model is pre-trained and can identify scam-specific language and patterns. The emotion engine recognizes emotions from the user's tone of voice and speaking rate, and reflects them in the analysis results.

[1467] Step 5: Fraud detection

[1468] The server compares the voice analysis results with a criminal database to determine whether the person is likely to be committing fraud. The server determines whether the person is committing fraud based on certain keywords and patterns, as well as the user's emotional state. For example, if the user shows signs of anxiety or nervousness, the server determines that the person is likely to be committing fraud.

[1469] Step 6: Warning Notification

[1470] If fraudulent activity is detected, the server will send a warning message to the user's device, which will include appropriate wording based on the user's emotional state, encouraging the user to be more careful about the content of the conversation.

[1471] Step 7: Automatic Notifications

[1472] The server sends detailed notifications to police and social workers, including basic information about the user, a summary of the conversation, any fraud patterns detected, and information about the user's emotional state.

[1473] Step 8: Continuous monitoring

[1474] The server continues to monitor the conversation after the notification has been sent and will notify again if any further fraud attempts are detected. This continuous monitoring helps prevent multiple fraud attempts from occurring.

[1475] Example 2

[1476] 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."

[1477] Previously, countermeasures against telephone fraud targeting the elderly were limited in their ability to prevent fraud before it occurred. Furthermore, there was no way to respond quickly when a fraudulent activity was in progress, making elderly people vulnerable to victimization. Furthermore, detecting fraud requires advanced analysis that takes into account emotional changes, but conventional technology was unable to adequately address this issue.

[1478] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1479] In this invention, the server includes a voice acquisition means for acquiring the elderly person's telephone conversation in real time, a data transmission means for transmitting the acquired voice data to the server via the Internet, a voice analysis means for analyzing the voice data received by the server using a pre-trained generative AI model and an emotion engine, a fraud detection means for comparing the analysis results with past crime data and determining whether fraud has occurred, a warning notification means for sending a warning to the user's device based on the fraud determination result, and an automatic notification means for notifying the police and social workers of the determination result. This makes it possible to detect fraud in real time while taking into account the elderly person's emotional state, and to quickly issue a warning and take action.

[1480] "Voice capture means" is a collective term for the hardware and software used to capture a user's telephone conversation in real time.

[1481] The "data transmission means" is a means for transmitting the acquired voice data to a server via the Internet.

[1482] A "server" is a device or system that includes a central processing unit and associated peripherals for receiving and analyzing audio data.

[1483] A "generative AI model" is an artificial intelligence model that is pre-trained for a specific purpose and is used to identify specific keywords and patterns of fraudulent activity.

[1484] The "emotion engine" is a software model that analyzes the tone and rate of a user's voice to recognize their emotional state.

[1485] "Speech analysis means" refers to means for analyzing received speech data using a generative AI model and an emotion engine.

[1486] "Fraud detection means" refers to a means for comparing the analysis results with a database of past crimes to determine whether or not fraud has occurred.

[1487] The "warning notification means" is a means for sending a warning to the user's terminal based on the result of the fraudulent activity determination.

[1488] "Automatic notification means" refers to a means for notifying police and social workers of fraud determination results.

[1489] MODE FOR CARRYING OUT THE INVENTION

[1490] The present invention provides a system for preventing telephone fraud targeting elderly people, and we will explain how to implement it in detail. This system includes a user's terminal, a server, and terminals of police and social workers that are linked to the server.

[1491] Hardware and software used

[1492] The following hardware and software are used to implement this system:

[1493] User device: Smartphone with built-in microphone and internet connection

[1494] Server: A server in a data center with a high-performance processor and large storage capacity

[1495] Generative AI models: AI models pre-trained to identify fraud-specific language and patterns

[1496] Emotion Engine: Software for recognizing emotions from the user's tone of voice and speech rate

[1497] User device behavior

[1498] When a user starts a conversation using a telephone, the user's device uses a built-in microphone to capture the conversation audio in real time. The captured audio data is stored in a buffer, and data is accumulated at regular intervals. The device then transmits the buffered audio data to a server via the Internet. The transmission is performed in real time and is sent continuously without any divisions.

[1499] Server Operation

[1500] The server receives the voice data sent from the user's device and stores it in a specific format. The server then analyzes the received voice data using a generative AI model and an emotion engine. The generative AI model is pre-trained to identify keywords and patterns specific to fraud. The emotion engine recognizes emotions from the user's tone and speed of voice and reflects them in the analysis results.

[1501] Based on the results of the above analysis, the server compares the voice analysis results with a criminal database to determine the possibility of fraud, which is determined based on specific keywords and patterns, as well as the user's emotional state.

[1502] Alerts and automatic notifications

[1503] When fraudulent activity is detected, the server first sends a warning message to the user's device. The warning contains appropriate wording based on the user's emotional state, encouraging the user to be more careful about the content of the conversation. At the same time, the server sends a notification with detailed information to the police and social worker's devices. The notification content includes the user's basic information, a summary of the conversation, the detected fraud pattern, and information about the user's emotional state.

[1504] Specific examples

[1505] Case 1: System behavior when elderly people feel uneasy about fraudsters

[1506] 1. Consider a scenario where a user (elderly person) is making a phone call and the other party uses words that fit into fraudulent patterns, such as "I need money" or "I've been in an accident," and the user feels anxious or tense during the conversation.

[1507] 2. The user's device captures the conversation content and the user's tone of voice in real time and sends them to the server.

[1508] 3. The server receives the voice data and the generative AI model and emotion engine analyze this data.

[1509] 4. After checking against a database of past crimes, it is determined that there is a high possibility of fraud and that the user is feeling uneasy.

[1510] 5. The server immediately sends a warning message to the user's device, stating, "This may be a scam. Please remain calm."

[1511] 6. At the same time, the server automatically notifies the police and social workers and sends them details.

[1512] Prompt Sentence Examples

[1513] "A 72-year-old elderly person, Mr. A, received a phone call saying, 'Your grandson has been in an accident.' Hearing this, Mr. A became very anxious. How would you analyze this situation and detect possible fraud?"

[1514] Using this prompt, the generative AI model and emotion engine can analyze fraud patterns and user emotions and take appropriate action.

[1515] The above is an embodiment of the present invention. The present invention is a system that detects fraudulent activity while taking into account the emotional state of seniors in real time, and provides prompt warning and response.

[1516] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1517] Step 1:

[1518] Audio Acquisition

[1519] When a user starts a conversation using the phone, the device uses the built-in microphone to capture the conversation audio in real time. The user's voice input is stored as data in a buffer. This voice data is updated at regular intervals (e.g., every second). If the user is talking on the phone with a friend, the device simultaneously captures the friend's voice and the user's voice and records them in a buffer.

[1520] Step 2:

[1521] Data transmission

[1522] The device transmits the voice data stored in the buffer to the server via the Internet. The user's device converts the voice data stored in the buffer into packets and sends them to the server. This transmission is done in real time, and the data is processed so that it is sent continuously without being divided. The user's device transmits a packet of voice data via the Internet every second.

[1523] Step 3:

[1524] Audio data reception

[1525] The server receives voice data sent from the user's device. It stores voice data packets received from the Internet in a specific format in storage. The server restores the data immediately after receiving it and stores it in storage as an audio file for analysis. One second after the user starts the call, the server receives the first voice data packet and stores it in a specified directory.

[1526] Step 4:

[1527] Audio analysis

[1528] The server analyzes the received voice data using a generative AI model and an emotion engine. The input data is the received audio file, which is fed to the generative AI model for analysis. The generative AI model is pre-trained to identify keywords and patterns specific to fraud. The emotion engine recognizes emotions from the tone and rate of the user's voice and combines the analysis results with the output of the generative AI model. For example, the server analyzes whether the user uttered a phrase such as "I need money" and also determines whether the tone of voice is unstable.

[1529] Step 5:

[1530] Fraud Detection

[1531] The server compares the voice analysis results with a database of past crimes to determine the likelihood of fraud. The input data is the analysis results from the generative AI model and emotion engine, and is compared with the criminal database to determine fraud. The server matches the analyzed data with pre-registered criminal data and evaluates the likelihood of fraud based on specific keywords and patterns, as well as the user's emotional state. For example, if the phrase "I had an accident" is detected, it is determined that there is a high possibility of fraud.

[1532] Step 6:

[1533] Warning notice

[1534] If fraud is detected, the server first sends a warning message to the user's device. The input data is the fraud judgment result, and the warning message is created based on that. The warning message contains appropriate wording based on the user's emotional state and is sent to the user's device. For example, if the system detects fraud, a message saying "There is a possibility of fraud. Please remain calm" is displayed on the user's smartphone.

[1535] Step 7:

[1536] Automatic notifications

[1537] At the same time, the server sends a notification containing detailed information to the police and social worker terminals. The input data is the fraud detection result and the user's basic information, which constitutes the notification content. The notification content includes the user's basic information, a summary of the conversation, the detected fraud pattern, and information about the user's emotional state. For example, if fraud is confirmed, the server immediately connects to the police system and sends detailed information. The social worker also receives the information.

[1538] In this way, the system prevents telephone fraud against seniors in real time and enables prompt and appropriate responses.

[1539] (Application example 2)

[1540] 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."

[1541] Telephone scams targeting seniors often involve sophisticated tactics by scammers who exploit seniors' anxieties and nervousness to defraud them of their money. Current technology struggles to detect fraud in real time, and there is a lack of mechanisms to warn seniors before a fraud occurs. Furthermore, there are no fraud detection and warning systems that take into account the emotional state of seniors. Therefore, there is a need for a system that can quickly and effectively protect seniors from fraud.

[1542] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a voice acquisition means for receiving voice data, a data transmission means for transmitting the acquired voice data in real time, a voice analysis means for analyzing the voice data using a generative AI model, a fraud detection means for detecting fraud by comparing the analysis results with a past crime database, an emotion recognition means for analyzing the user's emotional state, a warning notification means for sending a warning based on the fraud detection result, and a notification means for alerting the user based on the possibility of fraud. This enables real-time fraud detection and warning notification that takes into account the emotional state of the elderly.

[1543] "Elderly" refers to individuals who are older and generally more likely to be targeted by fraud.

[1544] "Telephone conversation" refers to a voice conversation conducted over the telephone.

[1545] "Audio capture means" refers to devices or software for recording telephone conversations in real time.

[1546] "Data transmission means" refers to a device or software for transmitting acquired voice data to a server.

[1547] A "generative AI model" is a pre-trained artificial intelligence model that is used to analyze voice data to detect fraudulent activity.

[1548] "Voice analysis means" refers to a device or software for analyzing voice data using a generative AI model.

[1549] "Fraud detection methods" refer to devices or software that compare voice analysis results with a database of past crimes to detect fraudulent activity.

[1550] "Warning notification means" refers to a device or software for sending a warning message to a user's terminal based on the results of fraudulent activity detection.

[1551] "Automatic notification means" refers to devices or software that notify police and social workers based on the detection of fraudulent activity.

[1552] "Emotion recognition means" refers to devices or software that analyze the user's tone of voice, speaking rate, etc. to recognize their emotional state.

[1553] "Notification means" means a device or software that sends a message to a user to warn them of possible fraud.

[1554] Overall system overview

[1555] This invention is a system for preventing telephone fraud targeting the elderly. The system includes a user terminal, a server, and police and social worker terminals. The system has the following main functions: a voice acquisition means, a data transmission means, a voice analysis means, a fraud detection means, an emotion recognition means, a warning notification means, an automatic notification means, and a notification means.

[1556] User terminal operation

[1557] Audio acquisition method:

[1558] When a user initiates a call, the user's device captures the conversation in real time using a built-in microphone. The voice data is stored in a buffer and accumulated at regular intervals.

[1559] Data transmission method:

[1560] The device transmits the captured audio data to a server via the Internet. This transmission is done in real time, allowing the server to quickly receive and analyze the data.

[1561] Server Operation

[1562] Audio data reception:

[1563] The server receives the voice data sent from the user's device and stores it in a specific format.

[1564] Audio analysis methods:

[1565] The server analyzes the received voice data using a generative AI model that has been pre-trained to identify scam-specific phrases and patterns.

[1566] Emotion recognition means:

[1567] The server recognizes the user's emotions from the tone of their voice, the rate at which they speak, etc. This improves the accuracy of the analysis if the user is feeling anxious or nervous.

[1568] Fraud detection measures:

[1569] The results of voice analysis and emotion recognition are combined and compared with a database of past crimes to determine the possibility of fraud.

[1570] Warning notification means:

[1571] If fraud is detected, the server first sends a warning message to the user's device. The message content is adjusted based on the user's emotional state, for example, "There is a possibility of fraud. Please remain calm."

[1572] Automatic notification method:

[1573] At the same time, the server sends a notification to police and social workers with detailed information, including basic information about the user, a summary of the conversation, any fraud patterns detected, and information about the user's emotional state.

[1574] Means of notification:

[1575] To ease the user's anxiety and tension, appropriate steps and contact information may also be provided.

[1576] Specific examples

[1577] Audio capture and data transmission:

[1578] When an elderly person (user) starts a call, the device's built-in microphone picks up the voice, and this data is sent to the server in real time.

[1579] Speech analysis and emotion recognition:

[1580] The server analyzes the received voice data and identifies fraud patterns such as "I need money" or "I've had an accident." At the same time, the emotion engine recognizes emotions such as anxiety and tension from the user's voice.

[1581] Fraud detection and warning notifications:

[1582] If it is determined that there is a high possibility of fraud, a warning message such as "This may be fraud. Please remain calm" will be displayed on the user's device.

[1583] Automatic notifications:

[1584] Similarly, police and social workers will be automatically notified of details of fraudulent activity, facilitating a swift response.

[1585] An example of a prompt sentence could be set up for the generative AI model as follows: "If the user shows signs of anxiety or tension during the phone call, determine that there is a high possibility of fraud. Analyze fraudulent patterns of words and the user's emotional state and display a warning."

[1586] In this way, the system also takes into account the user's emotional state and can quickly and effectively protect seniors from fraud.

[1587] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1588] Step 1: Audio capture

[1589] When a user starts a call, the user's device uses the built-in microphone to capture the conversation in real time. The audio data is stored in a local buffer, and data is accumulated at regular intervals. The input is the audio during the call, and the output is the audio data stored in the buffer.

[1590] Step 2: Send data

[1591] The user's terminal transmits the voice data stored in the local buffer to the server in real time. The data is uploaded to the server via the Internet using a data transmission means, so the input is the voice data in the buffer and the output is the voice data transmitted to the server.

[1592] Step 3: Receiving audio data

[1593] The server receives the voice data sent from the user terminal. The received voice data is stored in a specific format for analysis. The input is the voice data sent from the terminal, and the output is the voice data stored in the data storage in the server.

[1594] Step 4: Audio analysis

[1595] The server analyzes the stored voice data using a generative AI model. First, it converts the voice data into text, then analyzes the text to identify scam-specific phrases and patterns. The input is the stored voice data, and the output is the resulting text data.

[1596] Step 5: Emotion Recognition

[1597] During the voice analysis, the server uses emotion recognition means to analyze the user's tone of voice and speaking rate to determine the user's emotional state. The input is the voice data and analyzed text data, and the output is an evaluation of the user's emotional state.

[1598] Step 6: Fraud detection

[1599] The server compares the results of voice analysis and emotion recognition with a database of past crimes to assess the likelihood of fraud. If certain keywords or patterns match, it determines that there is a high possibility of fraud. The input is the analysis results and emotion evaluation results, and the output is an assessment of the likelihood of fraud.

[1600] Step 7: Warning Notification

[1601] If fraud is detected, the server sends a warning message to the user's device. The message content is adjusted based on the analysis results and the user's emotional state. For example, it might say, "There is a possibility of fraud. Please remain calm." The input is the fraud detection result, and the output is the warning message displayed on the user's device.

[1602] Step 8: Automatic Notifications

[1603] At the same time, the server also notifies the police and social workers. The notification content includes the user's basic information, a summary of the conversation, detected fraud patterns, and information about the user's emotional state. The input is the fraud detection results and related information, and the output is detailed information sent to the police and social workers.

[1604] At each step, different hardware and software in the system work together to effectively prevent elderly people from falling victim to fraud.

[1605] 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.

[1606] 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.

[1607] 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.

[1608] 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.

[1609] FIG. 9 illustrates an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1610] 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.

[1611] 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).

[1612] 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.

[1613] 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."

[1614] 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.

[1615] 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).

[1616] 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.

[1617] 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.

[1618] 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.

[1619] 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.

[1620] 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.

[1621] 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.

[1622] 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.

[1623] 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.

[1624] 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.

[1625] 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.

[1626] The following is further disclosed regarding the above embodiment.

[1627] (Claim 1)

[1628] a voice acquisition means for analyzing elderly people's telephone conversations in real time;

[1629] a data transmission means for transmitting the acquired voice data to a server;

[1630] a voice analysis means for analyzing the voice data received by the server using a generative AI model;

[1631] a fraud detection means for comparing the analysis results with a historical crime database to detect fraudulent activity;

[1632] a warning notification means for sending a warning to a user's terminal based on the result of the fraud detection;

[1633] Automated notification measures to inform police and social workers of detection results;

[1634] A system including:

[1635] (Claim 2)

[1636] The voice acquisition means stores a plurality of voice data in a buffer and transmits them in real time.

[1637] 10. The system of claim 1.

[1638] (Claim 3)

[1639] The fraud detection method uses a generative AI model to analyze voice data for specific keywords and patterns.

[1640] 10. The system of claim 1.

[1641] "Example 1"

[1642] (Claim 1)

[1643] a voice acquisition means for acquiring a conversation voice in real time when a user is having a telephone conversation;

[1644] a data transmission means for transmitting the acquired voice data to a server via the Internet;

[1645] a means for storing the received audio data in a specific format by the server;

[1646] a voice analysis means for analyzing the received voice data using a generative AI model;

[1647] a fraud detection means for comparing the analysis results with a historical crime database to detect fraudulent activity;

[1648] a warning notification means for sending a warning to a user's terminal based on the result of the fraud detection;

[1649] Automated notification measures to inform police and social workers of detection results;

[1650] A system including:

[1651] (Claim 2)

[1652] 2. The system according to claim 1, wherein the voice acquisition means stores voice data in a buffer, accumulates the data at regular intervals, and transmits the data in real time.

[1653] (Claim 3)

[1654] 10. The system of claim 1, wherein the fraud detection means uses a generative AI model to analyze the voice data for specific keywords and patterns based on prompts.

[1655] "Application Example 1"

[1656] (Claim 1)

[1657] a voice acquisition means for analyzing elderly people's telephone conversations in real time;

[1658] a data transmission means for transmitting the acquired voice data to a server;

[1659] a voice analysis means for analyzing the voice data received by the server using a generative AI model;

[1660] a fraud detection means for comparing the analysis results with past criminal information to detect fraudulent activity;

[1661] a warning notification means for sending a warning to a user's terminal based on the result of the fraud detection;

[1662] an automated notification mechanism for notifying local protective agencies and families of the results of the detection;

[1663] a notification mechanism for identifying fraudulent activity in real time, alerting the user and sending notifications to professional organizations and registered parents;

[1664] A system including:

[1665] (Claim 2)

[1666] 2. The system according to claim 1, wherein the voice acquisition means acquires voice during a call in real time and transmits the voice to the server.

[1667] (Claim 3)

[1668] 10. The system of claim 1, wherein the fraud detection means uses a generative AI model to analyze specific phrases and patterns in the voice data to assess the likelihood of fraud.

[1669] "Example 2: Combining Emotion Engines"

[1670] (Claim 1)

[1671] a voice acquisition means for acquiring a telephone conversation of an elderly person in real time;

[1672] a data transmission means for transmitting the acquired voice data to a server via the Internet;

[1673] a speech analysis means for analyzing the speech data received by the server using a pre-trained generative AI model and an emotion engine;

[1674] a fraud detection means for comparing the analysis results with past crime data to determine fraudulent activity;

[1675] a warning notification means for sending a warning to a user's terminal based on the result of the fraudulent activity determination;

[1676] An automated notification method for notifying police and social workers of the results of the assessment;

[1677] A system including:

[1678] (Claim 2)

[1679] 2. The system according to claim 1, wherein the voice acquisition means stores the conversation voice in a buffer and transmits it to the server in real time.

[1680] (Claim 3)

[1681] 10. The system of claim 1, wherein the fraud detection means uses a generative AI model to identify specific keywords and patterns in the voice data and an emotion engine to analyze the user's emotional state.

[1682] "Application example 2 when combining emotion engines"

[1683] Claims

[1684] (Claim 1)

[1685] a voice acquisition means for analyzing elderly people's telephone conversations in real time;

[1686] a data transmission means for transmitting the acquired voice data to a server;

[1687] a voice analysis means for analyzing the voice data received by the server using a generative AI model;

[1688] a fraud detection means for comparing the analysis results with a historical crime database to detect fraudulent activity;

[1689] a warning notification means for sending a warning to a user's terminal based on the result of the fraud detection;

[1690] an automated means of notifying police and social workers of the findings;

[1691] emotion recognition means for analyzing the emotional state of a user;

[1692] a notification means for alerting a user based on the possibility of fraud;

[1693] A system including:

[1694] (Claim 2)

[1695] The voice acquisition means stores a plurality of voice data in a buffer and transmits them in real time.

[1696] 10. The system of claim 1.

[1697] (Claim 3)

[1698] The fraud detection method uses a generative AI model to analyze voice data for specific keywords and patterns.

[1699] 10. The system of claim 1. [Explanation of symbols]

[1700] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a voice acquisition means for analyzing elderly people's telephone conversations in real time; a data transmission means for transmitting the acquired voice data to a server; a voice analysis means for analyzing the voice data received by the server using a generative AI model; a fraud detection means for comparing the analysis results with a historical crime database to detect fraudulent activity; a warning notification means for sending a warning to a user's terminal based on the result of the fraud detection; Automated notification measures to inform police and social workers of detection results; A system including:

2. The voice acquisition means stores a plurality of voice data in a buffer and transmits them in real time. The system of claim 1 .

3. The fraud detection method uses a generative AI model to analyze voice data for specific keywords and patterns. The system of claim 1 .

Images