system

Abstract

There is a need for methods that can detect and prevent fraud before users become victims. [Solution] The system provided to the user includes means for monitoring the user's communications and acquiring voice data, means for converting the acquired voice data into text data, means for analyzing the text data to identify patterns indicating fraudulent activity, means for sending a warning if fraud is suspected, and means for summarizing the content of the call and notifying a third party.

Description

Technical Field

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

Background Art

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

Summary of the Invention

Problems to be Solved by the Invention

[0004] In modern society, special fraud is increasing, and particularly cases where vulnerable people such as the elderly are targeted are on the rise. Such fraud acts aim to obtain personal information via phone calls or illegally withdraw money, and the damage is enormous. Therefore, there is a need for means to detect and prevent such fraud before users are involved.

Means for Solving the Problems

[0005] This invention provides a system that monitors the content of phone calls received by a user, acquires and analyzes the voice data, and identifies specific patterns indicating fraudulent activity. This system displays warnings to the user in real time, alerting them before they become victims of fraud. Furthermore, by summarizing the call content and notifying a third party, it enables not only the user but also their family and other relevant parties to cooperate in preventing fraud. In addition, the system achieves more accurate identification by using machine learning algorithms to score the likelihood of fraud.

[0006] "Means for monitoring user communications and acquiring voice data" refers to a device or program equipped with the function of automatically monitoring calls received by a user and acquiring the content of those calls as digital data.

[0007] "Means for converting acquired audio data into text data" refers to a device or program that performs a process of converting audio signals into text information in real time using speech recognition technology.

[0008] "Means for analyzing text data to identify patterns indicating fraudulent activity" refers to algorithms that perform a process of recognizing and analyzing specific keywords and contexts that suggest fraud based on past data.

[0009] "A means of sending warnings in the event of suspected fraud" refers to a mechanism that notifies call participants or related parties of the risk when the possibility of fraud is detected.

[0010] "Means of summarizing call content and notifying a third party" refers to a system that extracts the key points of audio data, summarizes them in a short format, and electronically transmits them to designated parties.

[0011] "A means of displaying warnings to users in real time" refers to a system with an interface that provides rapid alerts by immediately displaying the judgment results on the user's terminal screen.

[0012] "A method of scoring the likelihood of fraud using machine learning algorithms" is an evaluation method that uses a learning model based on a large amount of data to analyze received information and quantify the risk of fraud. [Brief explanation of the drawing]

[0013] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2, which incorporates an emotion engine. [Figure 14] It is a sequence diagram showing the processing flow of a data processing system in Application Example 2 when a sentiment engine is combined.

Embodiments for Carrying Out the Invention

[0014] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0015] First, the terms used in the following description will be explained.

[0016] In the following embodiments, a numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0017] In the following embodiments, a numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0018] In the following embodiments, a numbered storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, and the like.

[0019] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0020] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0021] [First Embodiment]

[0022] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0023] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0024] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0025] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0026] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0027] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0028] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0029] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0030] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0031] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0032] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0033] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0034] This invention embodies the configuration and operation method of a system for protecting users from telephone fraud. This system is executed by a monitoring application installed on the user's terminal.

[0035] First, the device automatically launches the application when the user starts a call. This launch allows the device to capture the voice data during the call in real time and send it to the server. With the user's permission, the device uses streaming technology to securely relay this data to the server.

[0036] The server converts the received audio data into text using speech recognition software and analyzes the resulting text data. This analysis is performed using generative AI technology based on a pre-stored database of fraud patterns. The server evaluates the risk of fraud using a specific scoring system and determines the likelihood of fraud based on that score. If fraud is identified as likely, the server sends a warning to the terminal.

[0037] The device immediately displays this warning as a pop-up to the user, allowing them to recognize in real time that the call may be unsafe. Furthermore, the server summarizes the call and sends a message containing the key points to registered family members and other relevant parties, further supporting the user's safety.

[0038] As a concrete example, consider a scenario where an elderly user receives a call from an unknown number. If the system determines that this call may be fraudulent, the device will display a warning to the user stating "This may be a scam," and the server will simultaneously notify registered family members of this summary. This allows the user and their family to take swift action. This system leverages the advantages of machine learning and real-time monitoring technology to provide an effective means of protecting users from phone scams.

[0039] The following describes the processing flow.

[0040] Step 1:

[0041] The device automatically launches the monitoring application when the user receives a call. This prepares it to collect audio data in real time.

[0042] Step 2:

[0043] Based on user permission, the device encrypts the audio data during a call and sends it to the server in streaming format. This transmission is performed while protecting the user's privacy.

[0044] Step 3:

[0045] The server converts the received audio data into text data using speech recognition technology. This conversion is performed in real time, preparing the data for analysis.

[0046] Step 4:

[0047] The server uses a generative AI model to analyze text data and identify specific patterns and keywords that indicate fraudulent activity. The analysis is performed based on a database of past fraud cases.

[0048] Step 5:

[0049] The server scores the risk of fraud based on the analysis results. This score reflects the degree of likelihood of fraud.

[0050] Step 6:

[0051] If the score exceeds a set threshold, the server determines that it may be a scam and sends a warning to the device.

[0052] Step 7:

[0053] The device immediately displays a warning message to the user, reminding them about the security of their calls. This message is presented in an intuitive and easy-to-understand format.

[0054] Step 8:

[0055] The server summarizes the call content and sends an alert notification containing the summary to pre-configured family members and relevant parties. This helps families share the risk of fraud and take appropriate action.

[0056] (Example 1)

[0057] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0058] In recent years, with the advancement of communication technology, the methods of telephone fraud have diversified, and fraud victims, particularly the elderly, are on the rise. However, current countermeasures have difficulty detecting fraudulent activity in real time and issuing warnings to users quickly, which presents a challenge in preventing fraud.

[0059] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0060] In this invention, the server includes means for automatically launching a monitoring application when a communication device initiates a call, means for acquiring voice information in real time and transmitting it to a data processing device using streaming technology, and means for converting the voice information into text format. This enables the user to quickly analyze the content of a call and immediately issue a warning if there is a possibility of fraud.

[0061] "Communication equipment" refers to electronic devices used for making calls and data communications, specifically terminals used by users when making voice calls or other forms of communication.

[0062] A "monitoring application" refers to a software program installed on a communication device that automatically starts up when a call begins to acquire and transmit voice data.

[0063] "Voice information" refers to the content of the user's speech acquired during a call, and this information serves as input data for later analysis.

[0064] A "data processing device" refers to a computer or server that has the function of receiving, analyzing, and converting audio information.

[0065] "Streaming technology" refers to a method of transmitting audio data in real time and is used to efficiently provide audio information to data processing devices.

[0066] "Means of converting to text format" refers to the process of converting acquired audio information into text data, which is achieved using speech recognition technology.

[0067] "Generative AI technology" refers to technologies that analyze data acquired using machine learning and natural language processing to generate new information and patterns.

[0068] This system is a protective system designed to shield users from phone scams. Specifically, a monitoring application installed on the user's communication device automatically runs as soon as a call begins. This allows the communication device to acquire voice information during the call in real time and transmit it to a data processing unit (server).

[0069] The device uses widely used streaming protocols (such as WebRTC) to securely and efficiently transmit the acquired audio information to the server. The server receives this audio information and converts it into text format using speech recognition software. This conversion can utilize existing speech recognition technologies such as Google® Cloud Speech-to-Text API.

[0070] Next, the server analyzes the converted text data using generative AI technology. The analysis is performed while referring to data models related to fraudulent activity, and prompts are used to identify fraud patterns. An example of such a prompt is the instruction given to the generative AI model: "Analyze this audio data and score the likelihood that it is fraudulent." Based on the results of this analysis, the server performs a risk assessment and scores calls that may be fraudulent.

[0071] For example, if a user receives a call from an unknown number, and the server determines that it is highly likely to be a scam, a warning is sent to the communication device. This warning immediately appears as a pop-up to the user, allowing them to instantly recognize the security of the call. Furthermore, the server summarizes the call content and, depending on the user's settings, notifies third parties such as family members or acquaintances, further enhancing the user's security.

[0072] Thus, the present invention provides a system that enhances user security by incorporating real-time analysis of voice information and an immediate warning function.

[0073] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0074] Step 1:

[0075] The terminal automatically launches a monitoring application when a user initiates a call. At this time, the terminal prepares to acquire audio information and begins capturing audio data during the user's call. The input is the user's call audio, and the output is real-time audio data. The terminal then prepares to stream this audio data for subsequent processing.

[0076] Step 2:

[0077] The terminal transmits the acquired audio information to the server using streaming technology. Here, the terminal uses a real-time communication protocol such as WebRTC to compress the audio data and perform appropriate buffering, minimizing data loss during the communication process. The input is the real-time audio data acquired in step 1, and the output is the data received by the server.

[0078] Step 3:

[0079] The server receives audio information sent from the terminal and converts it into text format using speech recognition software. Specifically, the server inputs the received audio data into a recognition engine such as the Google Cloud Speech-to-Text API, generating the resulting text data. The input is the received audio data, and the output is the converted text data.

[0080] Step 4:

[0081] The server analyzes the text data converted from the audio information using a generative AI model. Here, it refers to a pre-built fraud pattern data model and inputs prompt sentences into the generative AI model to evaluate the text. The input is the converted text data and prompt sentences, and the output is the information from the analyzed fraud risk assessment.

[0082] Step 5:

[0083] The server scores the risk of fraud based on the analysis results. Specifically, it applies a scoring algorithm to the analyzed data to evaluate whether it is highly likely to be fraudulent. The input is the text data of the analysis results, and the output is a score indicating the likelihood of fraud.

[0084] Step 6:

[0085] If the server determines that a transaction is likely to be fraudulent, it sends a warning to the device. The device receives this warning and immediately displays a pop-up notification to the user stating, "This may be a scam." The input is the score of the detection result, and the output is the warning displayed to the user.

[0086] Step 7:

[0087] The server summarizes the call content using a generation AI model and notifies registered family members and related parties of the user with the summarized information. The summary extracts the key points of the call and is structured as a concise message. The input is the text data of the call, and the output is a notification message to third parties.

[0088] (Application Example 1)

[0089] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0090] In modern times, telephone scams have become more sophisticated, and there is a problem in that vulnerable users, especially the elderly, are particularly susceptible to becoming victims. This increases the risk of many people suffering financial losses and emotional distress. To solve this problem, a system is needed that can assess the risk of telephone scams in real time and immediately warn users in a visually clear and easy-to-understand manner.

[0091] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0092] In this invention, the server includes an information processing device that includes means for monitoring the user's communications and acquiring voice information, means for converting the acquired voice information into symbolic information, means for analyzing the symbolic information to identify patterns indicating fraudulent activity, and means for displaying a visual warning using augmented reality technology. This enables the user to grasp the risk of fraud in real time and take immediate action.

[0093] An "information processing device" is a computer system that performs processing such as acquiring, analyzing, and displaying audio information.

[0094] "User" refers to an individual or organization that utilizes the system based on this patent and is protected from the risk of fraud.

[0095] "Communication" refers to information transmission activities such as phone calls and the transmission of voice information performed by users.

[0096] "Voice information" refers to data composed of voices exchanged during communication.

[0097] "Symbolic information" refers to data representation obtained by converting audio information into forms such as characters and numbers.

[0098] "Fraudulent activity" refers to unethical acts aimed at obtaining profit by deceiving others.

[0099] A "pattern" is a characteristic data component used to identify fraudulent activity.

[0100] "Analysis" is data processing that extracts specific information based on acquired data.

[0101] Augmented reality technology is a technique that overlays digital information onto real-world visual information.

[0102] A "warning" is a message intended to inform the user of a potential danger and to urge them to take precautions.

[0103] To implement this invention, it is first necessary to install an information processing device in the user's terminal. This device acquires voice information in real time during a call and converts that information into symbolic information. The terminal uses voice recognition software to convert the acquired voice information into text data.

[0104] Next, the device sends this text data to the server. The server is equipped with a generative AI model that analyzes the text data to identify patterns indicating fraudulent activity. If the analysis suggests fraud, a warning is sent to the device. Furthermore, the server uses a machine learning algorithm to score the likelihood of fraud based on the analyzed data and provides the warning to the user visually using augmented reality technology.

[0105] As a concrete example, consider a scenario where a user receives an important call from an unknown number. If the server determines that the content of this call resembles a scam, it immediately informs the device, and a warning message, "This may be a scam," is displayed via the augmented reality display. Based on this information, the user can take swift action.

[0106] This system is designed to help users make calls safely and protect themselves from the risk of fraud. An example of a prompt message is, "Explain how to evaluate the fraudulent nature of a phone call in real time and display a warning message on the HMD display."

[0107] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0108] Step 1:

[0109] The terminal activates the voice acquisition module as soon as the user starts a call. This module captures the voice data during the call in real time and encodes that data in digital format. The input is the voice during the call, and the output is digital voice data.

[0110] Step 2:

[0111] The terminal passes encoded digital audio data to speech recognition software. The speech recognition software analyzes this data and converts it into symbolic information in string format. This process converts the audio data into text data. The input is digital audio data, and the output is the converted text data.

[0112] Step 3:

[0113] The device sends the converted text data to the server using streaming technology. The server inputs the received text data into a generating AI model, which analyzes it to identify patterns of fraudulent activity. The input is text data, and the output is a score indicating the likelihood of fraud.

[0114] Step 4:

[0115] The server calculates a fraud risk score based on the analysis results obtained by the generating AI model. If fraud is suspected, it evaluates the score against specific criteria and generates an alert. The input is the analysis results, and the output is the fraud risk score and the alert.

[0116] Step 5:

[0117] If fraud is suspected, the server sends an alert to the device. The device then uses augmented reality technology to display a message on its screen to visually communicate the received alert to the user. The input is the alert from the server, and the output is the display of the warning message to the user.

[0118] Step 6:

[0119] The user reviews the fraud warning message displayed on the screen and takes appropriate action based on that information. The input is the warning message, and the output is the user's response.

[0120] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0121] This invention is a fraud prevention system that incorporates emotion recognition functionality. It not only analyzes the content of a user's phone call but also takes into account their emotions during the call, thereby enabling more accurate fraud detection.

[0122] The system is installed on the user's device and automatically activates the monitoring application when a call begins. The device collects call audio in real time and sends it securely to the server. The transmitted audio data is converted into text by the server using speech recognition technology. This text data is then analyzed by a generative AI, and at the same time, an emotion engine identifies the user's emotional state.

[0123] The emotion engine analyzes the tone and patterns of the user's voice to identify emotional states such as tension, anxiety, and confusion. This information is combined with the results of text data analysis to more reliably detect potential fraud. The server uses the emotion recognition results in the fraud risk scoring process, evaluating them comprehensively with the results of regular text analysis. If a high probability of fraud is determined, a warning is immediately issued to the device to alert the user. The content of the warning message is adjusted according to the emotions recognized by the emotion engine and presented to the user in the most appropriate way.

[0124] In addition, the server creates a call summary and notifies family members or trusted third parties, with this notification including additional information based on sentiment data to help recipients understand the situation more accurately. Sentiment data is also used for future analysis and improvement of learning models, contributing to overall system performance.

[0125] For example, if the emotion engine detects that a user is feeling anxious, the system will issue a warning earlier than usual and provide the user with advice to alleviate that anxiety. In this way, the present invention aims to dramatically improve the effectiveness of fraud prevention.

[0126] The following describes the processing flow.

[0127] Step 1:

[0128] When the user receives a call, the device automatically launches the monitoring app and prepares to acquire audio data in real time.

[0129] Step 2:

[0130] The device transmits the acquired audio data to the server using secure streaming technology. This transmission is encrypted to protect user privacy.

[0131] Step 3:

[0132] The server instantly converts the received audio data into text data using speech recognition technology. This conversion prepares the audio information into an analyzable format.

[0133] Step 4:

[0134] The server simultaneously activates an emotion engine to analyze the user's emotional state from the transmitted audio data. It determines whether the user is experiencing tension, anxiety, or anger based on factors such as tone of voice, speed, and pauses.

[0135] Step 5:

[0136] The server uses a generation AI to analyze the converted text data. This analysis identifies patterns and keywords that suggest fraud and assesses the likelihood of fraud.

[0137] Step 6:

[0138] Based on the results of the emotion engine analysis and text analysis, the server scores the fraud risk. This score is used to determine whether a warning is needed for the user.

[0139] Step 7:

[0140] The server sends a warning to the device if the scoring result exceeds a set threshold. The content of the warning message is adjusted according to the user's emotional state to prompt appropriate action.

[0141] Step 8:

[0142] The device displays a warning message to the user in real time. This display allows the user to immediately become aware of the security of their call.

[0143] Step 9:

[0144] The server summarizes the call and notifies family members or trusted third parties of the summary. This notification also includes information about the emotional state of the caller to help recipients accurately understand the situation.

[0145] (Example 2)

[0146] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0147] Traditional fraud prevention systems simply analyze the user's call content as text, issuing warnings without considering emotional shifts, resulting in low accuracy in detecting fraud. Furthermore, they struggled to provide appropriate warnings based on the actual level of risk, making them user-unfriendly.

[0148] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0149] In this invention, the server includes means for converting acoustic data into encoded data, means for analyzing the acoustic data to identify an emotional state, and means for integrating the results of the analysis of the encoded data and the results of the identification of the emotional state to evaluate the fraud risk. This makes it possible to evaluate the fraud risk with high accuracy, taking into account the user's emotional state, and to issue appropriate warnings.

[0150] "Audio data" refers to the voice information obtained from a user's phone call and is fundamental data for analyzing voice.

[0151] "Encoded data" refers to data obtained by converting acoustic data into text or other parsable formats.

[0152] "Emotional state" refers to the psychological or emotional condition derived from the user's voice, and includes tension, anxiety, and other similar states.

[0153] "Fraud risk" refers to the degree of likelihood that fraud is occurring, as assessed based on the analysis of call content and emotional state.

[0154] A "warning" refers to a cautionary message sent to a user when the risk of fraud is deemed high.

[0155] This invention is a system aimed at detecting fraudulent activity with high accuracy by monitoring and analyzing the content of user calls in real time. This system integrates the acquisition of acoustic data, analysis of emotional state, assessment of fraud risk, issuance of warnings, and notification to third parties.

[0156] The terminal automatically collects acoustic data as soon as the user starts a call and sends it to the server. This acoustic data is securely transferred to the server over the network. The server then converts the transferred acoustic data into encoded data using speech recognition technology. In this process, a speech recognition API is typically used as the specific software.

[0157] The server further analyzes the encoded data using a generation AI model to scrutinize the user's call content. Simultaneously, it uses an emotion recognition engine to evaluate the user's emotional state from the acoustic data. This emotion recognition process employs a dedicated algorithm that analyzes the tone and pitch of the voice.

[0158] The generative AI model analyzes various conversation patterns to determine the likelihood of fraudulent activity and integrates this with an analysis of the user's emotional state to comprehensively assess the fraud risk. Based on this risk assessment, if a high probability of fraud is detected, a warning is sent to the device. The warning message is displayed in real time with content optimized for the user's emotional state.

[0159] In addition, the server creates a summary of the call content and notifies a third party, along with information about the emotional state of the caller. This notification allows the recipient to understand the situation more accurately. This enables the early detection of fraudulent activity that may have occurred without the user's knowledge, and allows for appropriate countermeasures to be taken.

[0160] For example, if a user experiences anxiety during a phone call and the emotion recognition engine detects this anxiety, the system will issue a warning more quickly than usual, accompanied by a message offering advice to alleviate the anxiety. This allows the user to take immediate action.

[0161] An example of a prompt message is: "Analyze the content and emotional data of a user when they receive a fraudulent call, and provide guidance on how to offer early warnings and advice if they are feeling anxious."

[0162] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0163] Step 1:

[0164] The device automatically collects acoustic data when it detects the start of a call. The input is an audio signal, acquired in real time through the device's internal microphone. This acoustic data is transmitted to the server via network communication. The output is the acoustic data converted into a secure transmission format.

[0165] Step 2:

[0166] The server receives audio data transmitted from the terminal and converts it into encoded data using speech recognition technology. The input is audio data, which is then converted into text data through encoding. Specifically, it uses a speech recognition API to break down the audio signal into words and strings. The output is encoded data in text format.

[0167] Step 3:

[0168] The server uses a generative AI model to analyze encoded text data. The input is encoded data, and the server performs an analysis to identify patterns indicating fraudulent activity. This analysis applies various classification algorithms provided by the generative AI model. The output is the analysis results, including fraud risk information.

[0169] Step 4:

[0170] The server identifies the user's emotional state based on acoustic data. The input is acoustic data, and emotions are identified by analyzing the pitch and tone of the voice in particular. Specifically, an emotion recognition engine is used, outputting states of tension and anxiety as numerical values. The output is data indicating the emotional state.

[0171] Step 5:

[0172] The server integrates the results of the analysis of encoded data and the results of the identification of emotional states to assess fraud risk. The input consists of the analysis results and emotional data, and scoring is performed to comprehensively evaluate the integrated data from both. Specifically, a scoring algorithm is applied to quantify the likelihood of fraud. The output is the fraud risk assessment score.

[0173] Step 6:

[0174] The terminal notifies the user of warning messages received from the server. The input is the content of the warning, and the warning message is displayed in real time. Specifically, it presents the appropriate warning message visually or audibly through the user interface. The output is the warning message for the user to review.

[0175] Step 7:

[0176] The server creates a summary based on call content and sentiment data and notifies a trusted third party. The input consists of text data of the call content and sentiment information, which are used to construct the summary. Specifically, it uses a summary generation API to extract important information and adds sentiment information as supplementary data. The output is a notification message containing the summary information.

[0177] (Application Example 2)

[0178] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0179] Fraudulent activities in online transactions and communications are becoming increasingly sophisticated, raising the risk of users becoming victims of fraud. Furthermore, conventional fraud detection technologies struggle to accurately capture changes in users' emotions, making effective preventative measures against fraudulent transactions and communications necessary. In addition, there is a lack of warnings that take into account users' mental state and emotions, resulting in insufficient support for users to take appropriate measures.

[0180] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0181] In this invention, the server includes means for monitoring user communications and acquiring audio data, means for converting the acquired audio into text data, means for analyzing the text data to identify patterns indicating fraud, and means for recognizing the user's emotional state. This allows for the sending of warnings based on the user's emotional state and scoring of the likelihood of fraudulent transactions, thereby protecting the user from fraudulent activity.

[0182] "Means for monitoring user communications and acquiring voice data" refers to a function that captures telephone and other voice communications made by users in real time and records that voice information as data.

[0183] "Methods for converting acquired audio data into text data" refers to technologies that convert information obtained through audio into text information and generate it in a format that is easy to analyze.

[0184] "Means for analyzing text data to identify patterns indicating fraudulent activity" refers to a system that analyzes transcribed data to detect characteristic patterns or phrases that suggest potential fraudulent activity.

[0185] "Means for recognizing a user's emotional state" refers to technology that determines a user's emotional state at a given time by analyzing the tone of their voice and their choice of words during communication.

[0186] "Means for sending emotional state-based warnings" refers to a mechanism that generates appropriate warning messages according to the detected emotions of the user and notifies the user accordingly.

[0187] "A means of summarizing call content and notifying a third party of information including emotional data" refers to a function that concisely summarizes the communication content and adds the user's emotional information, then provides that information to a trusted third party.

[0188] "Methods for scoring the likelihood of fraud" refer to a process that uses machine learning to quantify and evaluate the risk of fraudulent activity based on acquired data.

[0189] This invention aims to integrate a fraud prevention system into an application for e-commerce sites, enabling users to proactively detect fraud risks during online transactions or customer support calls. The system utilizes smartphones and cloud servers to monitor user communications.

[0190] The device acquires voice data in real time as soon as a call begins and securely sends it to a cloud server. On the server, speech recognition technology such as Google Cloud Speech-to-Text is used to convert the voice data into text data. This text data is analyzed by generative AI models such as OpenAI® GPT-4® to detect patterns of fraudulent activity. The server also analyzes the user's voice tone to identify their emotional state. When emotional states such as tension, anxiety, or confusion are identified, the suspicion of fraudulent activity increases.

[0191] By comprehensively evaluating the analysis results and emotional state, and scoring the likelihood of fraud, if a high risk is detected, an emotion-based warning message is immediately sent to the device. For example, if a user attempts to purchase an unusually large item and an emotion of anxiety is detected, a warning such as "Please reconfirm this transaction. We also recommend contacting official support" will be sent.

[0192] Furthermore, call summaries and sentiment data will be shared with a trusted third party based on the user's consent. This sharing helps in the early detection of signs of fraud.

[0193] An example of a prompt message might be: "Generate a warning message to display if the user shows signs of anxiety or confusion when making a large transaction."

[0194] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0195] Step 1:

[0196] The terminal detects the start of a user's phone call. Real-time acquisition of audio data occurs at the start of the call. The input is the user's voice during the call; by collecting this as audio data, accurate data is provided to the system.

[0197] Step 2:

[0198] The audio data acquired by the device is securely sent to the cloud server as an audio file. The input is the audio data collected in step 1, and securely transferring this to the server enables subsequent processing.

[0199] Step 3:

[0200] The server uses speech recognition technology to convert received audio data into text data. The input here is the transmitted audio file, which is then transcribed using a speech recognition tool such as Google Cloud Speech-to-Text. The output is parseable text data.

[0201] Step 4:

[0202] The server uses a generative AI model to analyze text data and detect patterns indicating fraudulent activity. The input is the text data obtained in step 3, and a generative AI model such as GPT-4 performs pattern analysis. The output is information assessing the risk of fraudulent activity.

[0203] Step 5:

[0204] Simultaneously, the server analyzes the tone of the voice data and recognizes the user's emotional state. The input is the voice data itself, and emotion recognition is performed using a specific algorithm. The output is the user's emotional state (e.g., tension, anxiety).

[0205] Step 6:

[0206] The server integrates the results of fraud risk assessment and sentiment recognition to calculate a risk score. A machine learning algorithm is used for this, with the input being the results of steps 4 and 5. The output is the overall risk score.

[0207] Step 7:

[0208] A warning message based on the user's emotional state is immediately sent to the user's device. The input is the risk score calculated in step 6, and an appropriate warning is generated. The output is the warning message displayed to the user.

[0209] Step 8:

[0210] The server summarizes the call content and notifies a trusted third party, including emotional data. The input consists of audio text data and emotional state information, which are used to create the notification. The output is detailed notification information sent to the third party.

[0211] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0212] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0213] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0214] [Second Embodiment]

[0215] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0216] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0217] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0218] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0219] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0220] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0221] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0222] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0223] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0224] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0225] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0226] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0227] This invention embodies the configuration and operation method of a system for protecting users from telephone fraud. This system is executed by a monitoring application installed on the user's terminal.

[0228] First, the device automatically launches the application when the user starts a call. This launch allows the device to capture the voice data during the call in real time and send it to the server. With the user's permission, the device uses streaming technology to securely relay this data to the server.

[0229] The server converts the received audio data into text using speech recognition software and analyzes the resulting text data. This analysis is performed using generative AI technology based on a pre-stored database of fraud patterns. The server evaluates the risk of fraud using a specific scoring system and determines the likelihood of fraud based on that score. If fraud is identified as likely, the server sends a warning to the terminal.

[0230] The device immediately displays this warning as a pop-up to the user, allowing them to recognize in real time that the call may be unsafe. Furthermore, the server summarizes the call and sends a message containing the key points to registered family members and other relevant parties, further supporting the user's safety.

[0231] As a concrete example, consider a scenario where an elderly user receives a call from an unknown number. If the system determines that this call may be fraudulent, the device will display a warning to the user stating "This may be a scam," and the server will simultaneously notify registered family members of this summary. This allows the user and their family to take swift action. This system leverages the advantages of machine learning and real-time monitoring technology to provide an effective means of protecting users from phone scams.

[0232] The following describes the processing flow.

[0233] Step 1:

[0234] The device automatically launches the monitoring application when the user receives a call. This prepares it to collect audio data in real time.

[0235] Step 2:

[0236] Based on user permission, the device encrypts the audio data during a call and sends it to the server in streaming format. This transmission is performed while protecting the user's privacy.

[0237] Step 3:

[0238] The server converts the received audio data into text data using speech recognition technology. This conversion is performed in real time, preparing the data for analysis.

[0239] Step 4:

[0240] The server uses a generative AI model to analyze text data and identify specific patterns and keywords that indicate fraudulent activity. The analysis is performed based on a database of past fraud cases.

[0241] Step 5:

[0242] The server scores the risk of fraud based on the analysis results. This score reflects the degree of likelihood of fraud.

[0243] Step 6:

[0244] If the score exceeds a set threshold, the server determines that it may be a scam and sends a warning to the device.

[0245] Step 7:

[0246] The device immediately displays a warning message to the user, reminding them about the security of their calls. This message is presented in an intuitive and easy-to-understand format.

[0247] Step 8:

[0248] The server summarizes the call content and sends an alert notification containing the summary to pre-configured family members and relevant parties. This helps families share the risk of fraud and take appropriate action.

[0249] (Example 1)

[0250] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0251] In recent years, with the advancement of communication technology, the methods of telephone fraud have diversified, and fraud victims, particularly the elderly, are on the rise. However, current countermeasures have difficulty detecting fraudulent activity in real time and issuing warnings to users quickly, which presents a challenge in preventing fraud.

[0252] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0253] In this invention, the server includes means for automatically launching a monitoring application when a communication device initiates a call, means for acquiring voice information in real time and transmitting it to a data processing device using streaming technology, and means for converting the voice information into text format. This enables the user to quickly analyze the content of a call and immediately issue a warning if there is a possibility of fraud.

[0254] "Communication equipment" refers to electronic devices used for making calls and data communications, specifically terminals used by users when making voice calls or other forms of communication.

[0255] A "monitoring application" refers to a software program installed on a communication device that automatically starts up when a call begins to acquire and transmit voice data.

[0256] "Voice information" refers to the content of the user's speech acquired during a call, and this information serves as input data for later analysis.

[0257] A "data processing device" refers to a computer or server that has the function of receiving, analyzing, and converting audio information.

[0258] "Streaming technology" refers to a method of transmitting audio data in real time and is used to efficiently provide audio information to data processing devices.

[0259] "Means of converting to text format" refers to the process of converting acquired audio information into text data, which is achieved using speech recognition technology.

[0260] "Generative AI technology" refers to technologies that analyze data acquired using machine learning and natural language processing to generate new information and patterns.

[0261] This system is a protective system designed to shield users from phone scams. Specifically, a monitoring application installed on the user's communication device automatically runs as soon as a call begins. This allows the communication device to acquire voice information during the call in real time and transmit it to a data processing unit (server).

[0262] The device uses widely used streaming protocols (such as WebRTC) to securely and efficiently transmit the acquired audio information to the server. The server receives this audio information and converts it into text format using speech recognition software. Existing speech recognition technologies, such as the Google Cloud Speech-to-Text API, can be used for this conversion.

[0263] Next, the server analyzes the converted text data using generative AI technology. The analysis is performed while referring to data models related to fraudulent activity, and prompts are used to identify fraud patterns. An example of such a prompt is the instruction given to the generative AI model: "Analyze this audio data and score the likelihood that it is fraudulent." Based on the results of this analysis, the server performs a risk assessment and scores calls that may be fraudulent.

[0264] For example, if a user receives a call from an unknown number, and the server determines that it is highly likely to be a scam, a warning is sent to the communication device. This warning immediately appears as a pop-up to the user, allowing them to instantly recognize the security of the call. Furthermore, the server summarizes the call content and, depending on the user's settings, notifies third parties such as family members or acquaintances, further enhancing the user's security.

[0265] Thus, the present invention provides a system that enhances user security by incorporating real-time analysis of voice information and an immediate warning function.

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

[0267] Step 1:

[0268] The terminal automatically launches a monitoring application when a user initiates a call. At this time, the terminal prepares to acquire audio information and begins capturing audio data during the user's call. The input is the user's call audio, and the output is real-time audio data. The terminal then prepares to stream this audio data for subsequent processing.

[0269] Step 2:

[0270] The terminal transmits the acquired audio information to the server using streaming technology. Here, the terminal uses a real-time communication protocol such as WebRTC to compress the audio data and perform appropriate buffering, minimizing data loss during the communication process. The input is the real-time audio data acquired in step 1, and the output is the data received by the server.

[0271] Step 3:

[0272] The server receives audio information sent from the terminal and converts it into text format using speech recognition software. Specifically, the server inputs the received audio data into a recognition engine such as the Google Cloud Speech-to-Text API, generating the resulting text data. The input is the received audio data, and the output is the converted text data.

[0273] Step 4:

[0274] The server analyzes the text data converted from the audio information using a generative AI model. Here, it refers to a pre-built fraud pattern data model and inputs prompt sentences into the generative AI model to evaluate the text. The input is the converted text data and prompt sentences, and the output is the information from the analyzed fraud risk assessment.

[0275] Step 5:

[0276] The server scores the risk of fraud based on the analysis results. Specifically, it applies a scoring algorithm to the analyzed data to evaluate whether it is highly likely to be fraudulent. The input is the text data of the analysis results, and the output is a score indicating the likelihood of fraud.

[0277] Step 6:

[0278] If the server determines that a transaction is likely to be fraudulent, it sends a warning to the device. The device receives this warning and immediately displays a pop-up notification to the user stating, "This may be a scam." The input is the score of the detection result, and the output is the warning displayed to the user.

[0279] Step 7:

[0280] The server summarizes the call content using a generative AI model and notifies the family members and related parties registered by the user of the summary information. The summary extracts the important points of the call and is composed as a concise message. The input is the text data of the call, and the output is the notification message to a third party.

[0281] (Application Example 1)

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

[0283] In modern times, the methods of telephone fraud have become sophisticated, and there is a problem that users who are unfamiliar with information, especially the elderly, are likely to suffer from such fraud. As a result, there is an increasing risk that many people will suffer economic losses and mental harm. To solve this problem, a system that can determine the risk of telephone fraud in real time and immediately warn the user in a visual and easy-to-understand manner is needed.

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

[0285] In this invention, the server includes a means by which an information processing device monitors the user's communication and acquires voice information, a means for converting the acquired voice information into symbolic information, a means for analyzing the symbolic information to identify a pattern indicating fraud, and a means for visually displaying a warning using augmented reality technology. As a result, the user can grasp the risk of fraud in real time and can immediately take countermeasures.

[0286] The "information processing device" is a computer system for executing processes such as acquisition, analysis, and display of voice information.

[0287] [[ID=2,7]] The "user" is an individual or organization that uses the system based on this patent and is protected from the risk of fraud.

[0288] "Communication" refers to information transmission activities such as phone calls and the transmission of voice information performed by users.

[0289] "Voice information" refers to data composed of voices exchanged during communication.

[0290] "Symbolic information" refers to data representation obtained by converting audio information into forms such as characters and numbers.

[0291] "Fraudulent activity" refers to unethical acts aimed at obtaining profit by deceiving others.

[0292] A "pattern" is a characteristic data component used to identify fraudulent activity.

[0293] "Analysis" is data processing that extracts specific information based on acquired data.

[0294] Augmented reality technology is a technique that overlays digital information onto real-world visual information.

[0295] A "warning" is a message intended to inform the user of a potential danger and to urge them to take precautions.

[0296] To implement this invention, it is first necessary to install an information processing device in the user's terminal. This device acquires voice information in real time during a call and converts that information into symbolic information. The terminal uses voice recognition software to convert the acquired voice information into text data.

[0297] Next, the device sends this text data to the server. The server is equipped with a generative AI model that analyzes the text data to identify patterns indicating fraudulent activity. If the analysis suggests fraud, a warning is sent to the device. Furthermore, the server uses a machine learning algorithm to score the likelihood of fraud based on the analyzed data and provides the warning to the user visually using augmented reality technology.

[0298] As a concrete example, consider a scenario where a user receives an important call from an unknown number. If the server determines that the content of this call resembles a scam, it immediately informs the device, and a warning message, "This may be a scam," is displayed via the augmented reality display. Based on this information, the user can take swift action.

[0299] This system is designed to help users make calls safely and protect themselves from the risk of fraud. An example of a prompt message is, "Explain how to evaluate the fraudulent nature of a phone call in real time and display a warning message on the HMD display."

[0300] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0301] Step 1:

[0302] The terminal activates the voice acquisition module as soon as the user starts a call. This module captures the voice data during the call in real time and encodes that data in digital format. The input is the voice during the call, and the output is digital voice data.

[0303] Step 2:

[0304] The terminal passes encoded digital audio data to speech recognition software. The speech recognition software analyzes this data and converts it into symbolic information in string format. This process converts the audio data into text data. The input is digital audio data, and the output is the converted text data.

[0305] Step 3:

[0306] The device sends the converted text data to the server using streaming technology. The server inputs the received text data into a generating AI model, which analyzes it to identify patterns of fraudulent activity. The input is text data, and the output is a score indicating the likelihood of fraud.

[0307] Step 4:

[0308] Based on the analysis results obtained by the generative AI model, the server calculates a fraud risk score. Thus, if there is suspicion of fraud, the score is evaluated according to specific criteria, and an alert is generated. The input is the analysis result, and the output is the fraud risk score and the alert.

[0309] Step 5:

[0310] If it is determined that there is suspicion of fraud, the server sends an alert to the terminal. The terminal uses augmented reality technology to display a message on the display to visually convey the received alert to the user. The input is the alert from the server, and the output is the display of a warning message to the user.

[0311] Step 6:

[0312] The user checks the fraud warning message displayed on the display and takes appropriate countermeasures based on that information. The input is the warning message, and the output is the response by the user's actions.

[0313] Furthermore, an emotion engine for estimating the user's emotion may be combined. That is, the specific processing unit 290 may estimate the user's emotion using the emotion recognition model 59 and perform specific processing using the user's emotion.

[0314] The present invention is a fraud prevention system incorporating an emotion recognition function, which not only analyzes the user's call content but also considers the emotion at that time to perform more accurate fraud discrimination.

[0315] The system is installed on the user's device and automatically activates the monitoring application when a call begins. The device collects call audio in real time and sends it securely to the server. The transmitted audio data is converted into text by the server using speech recognition technology. This text data is then analyzed by a generative AI, and at the same time, an emotion engine identifies the user's emotional state.

[0316] The emotion engine analyzes the tone and patterns of the user's voice to identify emotional states such as tension, anxiety, and confusion. This information is combined with the results of text data analysis to more reliably detect potential fraud. The server uses the emotion recognition results in the fraud risk scoring process, evaluating them comprehensively with the results of regular text analysis. If a high probability of fraud is determined, a warning is immediately issued to the device to alert the user. The content of the warning message is adjusted according to the emotions recognized by the emotion engine and presented to the user in the most appropriate way.

[0317] In addition, the server creates a call summary and notifies family members or trusted third parties, with this notification including additional information based on sentiment data to help recipients understand the situation more accurately. Sentiment data is also used for future analysis and improvement of learning models, contributing to overall system performance.

[0318] For example, if the emotion engine detects that a user is feeling anxious, the system will issue a warning earlier than usual and provide the user with advice to alleviate that anxiety. In this way, the present invention aims to dramatically improve the effectiveness of fraud prevention.

[0319] The following describes the processing flow.

[0320] Step 1:

[0321] When the user receives a call, the device automatically launches the monitoring app and prepares to acquire audio data in real time.

[0322] Step 2:

[0323] The device transmits the acquired audio data to the server using secure streaming technology. This transmission is encrypted to protect user privacy.

[0324] Step 3:

[0325] The server instantly converts the received audio data into text data using speech recognition technology. This conversion prepares the audio information into an analyzable format.

[0326] Step 4:

[0327] The server simultaneously activates an emotion engine to analyze the user's emotional state from the transmitted audio data. It determines whether the user is experiencing tension, anxiety, or anger based on factors such as tone of voice, speed, and pauses.

[0328] Step 5:

[0329] The server uses a generation AI to analyze the converted text data. This analysis identifies patterns and keywords that suggest fraud and assesses the likelihood of fraud.

[0330] Step 6:

[0331] Based on the results of the emotion engine analysis and text analysis, the server scores the fraud risk. This score is used to determine whether a warning is needed for the user.

[0332] Step 7:

[0333] The server sends a warning to the device if the scoring result exceeds a set threshold. The content of the warning message is adjusted according to the user's emotional state to prompt appropriate action.

[0334] Step 8:

[0335] The device displays a warning message to the user in real time. This display allows the user to immediately become aware of the security of their call.

[0336] Step 9:

[0337] The server summarizes the call and notifies family members or trusted third parties of the summary. This notification also includes information about the emotional state of the caller to help recipients accurately understand the situation.

[0338] (Example 2)

[0339] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0340] Traditional fraud prevention systems simply analyze the user's call content as text, issuing warnings without considering emotional shifts, resulting in low accuracy in detecting fraud. Furthermore, they struggled to provide appropriate warnings based on the actual level of risk, making them user-unfriendly.

[0341] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0342] In this invention, the server includes means for converting acoustic data into encoded data, means for analyzing the acoustic data to identify an emotional state, and means for integrating the results of the analysis of the encoded data and the results of the identification of the emotional state to evaluate the fraud risk. This makes it possible to evaluate the fraud risk with high accuracy, taking into account the user's emotional state, and to issue appropriate warnings.

[0343] "Audio data" refers to the voice information obtained from a user's phone call and is fundamental data for analyzing voice.

[0344] "Encoded data" refers to data obtained by converting acoustic data into text or other parsable formats.

[0345] "Emotional state" refers to the psychological or emotional condition derived from the user's voice, and includes tension, anxiety, and other similar states.

[0346] "Fraud risk" refers to the degree of likelihood that fraud is occurring, as assessed based on the analysis of call content and emotional state.

[0347] A "warning" refers to a cautionary message sent to a user when the risk of fraud is deemed high.

[0348] This invention is a system aimed at detecting fraudulent activity with high accuracy by monitoring and analyzing the content of user calls in real time. This system integrates the acquisition of acoustic data, analysis of emotional state, assessment of fraud risk, issuance of warnings, and notification to third parties.

[0349] The terminal automatically collects acoustic data as soon as the user starts a call and sends it to the server. This acoustic data is securely transferred to the server over the network. The server then converts the transferred acoustic data into encoded data using speech recognition technology. In this process, a speech recognition API is typically used as the specific software.

[0350] The server further analyzes the encoded data using a generation AI model to scrutinize the user's call content. Simultaneously, it uses an emotion recognition engine to evaluate the user's emotional state from the acoustic data. This emotion recognition process employs a dedicated algorithm that analyzes the tone and pitch of the voice.

[0351] The generative AI model analyzes various conversation patterns to determine the likelihood of fraudulent activity and integrates this with an analysis of the user's emotional state to comprehensively assess the fraud risk. Based on this risk assessment, if a high probability of fraud is detected, a warning is sent to the device. The warning message is displayed in real time with content optimized for the user's emotional state.

[0352] In addition, the server creates a summary of the call content and notifies a third party, along with information about the emotional state of the caller. This notification allows the recipient to understand the situation more accurately. This enables the early detection of fraudulent activity that may have occurred without the user's knowledge, and allows for appropriate countermeasures to be taken.

[0353] For example, if a user experiences anxiety during a phone call and the emotion recognition engine detects this anxiety, the system will issue a warning more quickly than usual, accompanied by a message offering advice to alleviate the anxiety. This allows the user to take immediate action.

[0354] An example of a prompt message would be: "Analyze the content and emotional data of a user when they receive a fraudulent call, and provide a way to offer early warnings and advice if they are feeling anxious."

[0355] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0356] Step 1:

[0357] The device automatically collects acoustic data when it detects the start of a call. The input is an audio signal, acquired in real time through the device's internal microphone. This acoustic data is transmitted to the server via network communication. The output is the acoustic data converted into a secure transmission format.

[0358] Step 2:

[0359] The server receives audio data transmitted from the terminal and converts it into encoded data using speech recognition technology. The input is audio data, which is then converted into text data through encoding. Specifically, it uses a speech recognition API to break down the audio signal into words and strings. The output is encoded data in text format.

[0360] Step 3:

[0361] The server uses a generative AI model to analyze encoded text data. The input is encoded data, and the server performs an analysis to identify patterns indicating fraudulent activity. This analysis applies various classification algorithms provided by the generative AI model. The output is the analysis results, including fraud risk information.

[0362] Step 4:

[0363] The server identifies the user's emotional state based on acoustic data. The input is acoustic data, and emotions are identified by analyzing the pitch and tone of the voice in particular. Specifically, an emotion recognition engine is used, outputting states of tension and anxiety as numerical values. The output is data indicating the emotional state.

[0364] Step 5:

[0365] The server integrates the results of the analysis of encoded data and the results of the identification of emotional states to assess fraud risk. The input consists of the analysis results and emotional data, and scoring is performed to comprehensively evaluate the integrated data from both. Specifically, a scoring algorithm is applied to quantify the likelihood of fraud. The output is the fraud risk assessment score.

[0366] Step 6:

[0367] The terminal notifies the user of warning messages received from the server. The input is the content of the warning, and the warning message is displayed in real time. Specifically, it presents the appropriate warning message visually or audibly through the user interface. The output is the warning message for the user to review.

[0368] Step 7:

[0369] The server creates a summary based on call content and sentiment data and notifies a trusted third party. The input consists of text data of the call content and sentiment information, which are used to construct the summary. Specifically, it uses a summary generation API to extract important information and adds sentiment information as supplementary data. The output is a notification message containing the summary information.

[0370] (Application Example 2)

[0371] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the smart glasses 214 as the "terminal".

[0372] Fraudulent activities in online transactions and communications are becoming increasingly sophisticated, raising the risk of users becoming victims of fraud. Furthermore, conventional fraud detection technologies struggle to accurately capture changes in users' emotions, making effective preventative measures against fraudulent transactions and communications necessary. In addition, there is a lack of warnings that take into account users' mental state and emotions, resulting in insufficient support for users to take appropriate measures.

[0373] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0374] In this invention, the server includes means for monitoring user communications and acquiring audio data, means for converting the acquired audio into text data, means for analyzing the text data to identify patterns indicating fraud, and means for recognizing the user's emotional state. This allows for the sending of warnings based on the user's emotional state and scoring of the likelihood of fraudulent transactions, thereby protecting the user from fraudulent activity.

[0375] "Means for monitoring user communications and acquiring voice data" refers to a function that captures telephone and other voice communications made by users in real time and records that voice information as data.

[0376] "Methods for converting acquired audio data into text data" refers to technologies that convert information obtained through audio into text information and generate it in a format that is easy to analyze.

[0377] "Means for analyzing text data to identify patterns indicating fraudulent activity" refers to a system that analyzes transcribed data to detect characteristic patterns or phrases that suggest potential fraudulent activity.

[0378] "Means for recognizing a user's emotional state" refers to technology that determines a user's emotional state at a given time by analyzing the tone of their voice and their choice of words during communication.

[0379] "Means for sending emotional state-based warnings" refers to a mechanism that generates appropriate warning messages according to the detected emotions of the user and notifies the user accordingly.

[0380] "A means of summarizing call content and notifying a third party of information including emotional data" refers to a function that concisely summarizes the communication content and adds the user's emotional information, then provides that information to a trusted third party.

[0381] "Methods for scoring the likelihood of fraud" refer to a process that uses machine learning to quantify and evaluate the risk of fraudulent activity based on acquired data.

[0382] This invention aims to integrate a fraud prevention system into an application for e-commerce sites, enabling users to proactively detect fraud risks during online transactions or customer support calls. The system utilizes smartphones and cloud servers to monitor user communications.

[0383] The device acquires voice data in real time as soon as a call begins and securely sends it to a cloud server. On the server, speech recognition technology such as Google Cloud Speech-to-Text is used to convert the voice data into text data. This text data is analyzed by generative AI models such as OpenAI GPT-4 to detect patterns of fraudulent activity. The server also analyzes the user's voice tone to identify their emotional state. When emotional states such as tension, anxiety, or confusion are identified, the suspicion of fraudulent activity increases.

[0384] By comprehensively evaluating the analysis results and emotional state, and scoring the likelihood of fraud, if a high risk is detected, an emotion-based warning message is immediately sent to the device. For example, if a user attempts to purchase an unusually large item and an emotion of anxiety is detected, a warning such as "Please reconfirm this transaction. We also recommend contacting official support" will be sent.

[0385] Furthermore, call summaries and sentiment data will be shared with a trusted third party based on the user's consent. This sharing helps in the early detection of signs of fraud.

[0386] An example of a prompt message might be: "Generate a warning message to display if the user shows signs of anxiety or confusion when making a large transaction."

[0387] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0388] Step 1:

[0389] The terminal detects the start of a user's phone call. Real-time acquisition of audio data occurs at the start of the call. The input is the user's voice during the call; by collecting this as audio data, accurate data is provided to the system.

[0390] Step 2:

[0391] The audio data acquired by the device is securely sent to the cloud server as an audio file. The input is the audio data collected in step 1, and securely transferring this to the server enables subsequent processing.

[0392] Step 3:

[0393] The server uses speech recognition technology to convert received audio data into text data. The input here is the transmitted audio file, which is then transcribed using a speech recognition tool such as Google Cloud Speech-to-Text. The output is parseable text data.

[0394] Step 4:

[0395] The server uses a generative AI model to analyze text data and detect patterns indicating fraudulent activity. The input is the text data obtained in step 3, and a generative AI model such as GPT-4 performs pattern analysis. The output is information assessing the risk of fraudulent activity.

[0396] Step 5:

[0397] Simultaneously, the server analyzes the tone of the voice data and recognizes the user's emotional state. The input is the voice data itself, and emotion recognition is performed using a specific algorithm. The output is the user's emotional state (e.g., tension, anxiety).

[0398] Step 6:

[0399] The server integrates the results of fraud risk assessment and sentiment recognition to calculate a risk score. A machine learning algorithm is used for this, with the input being the results of steps 4 and 5. The output is the overall risk score.

[0400] Step 7:

[0401] A warning message based on the user's emotional state is immediately sent to the user's device. The input is the risk score calculated in step 6, and an appropriate warning is generated. The output is the warning message displayed to the user.

[0402] Step 8:

[0403] The server summarizes the call content and notifies a trusted third party, including emotional data. The input consists of audio text data and emotional state information, which are used to create the notification. The output is detailed notification information sent to the third party.

[0404] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0405] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0406] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0407] [Third Embodiment]

[0408] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0409] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0410] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0411] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0412] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0413] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0414] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0415] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0416] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0417] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0418] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0419] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0420] This invention embodies the configuration and operation method of a system for protecting users from telephone fraud. This system is executed by a monitoring application installed on the user's terminal.

[0421] First, the device automatically launches the application when the user starts a call. This launch allows the device to capture the voice data during the call in real time and send it to the server. With the user's permission, the device uses streaming technology to securely relay this data to the server.

[0422] The server converts the received audio data into text using speech recognition software and analyzes the resulting text data. This analysis is performed using generative AI technology based on a pre-stored database of fraud patterns. The server evaluates the risk of fraud using a specific scoring system and determines the likelihood of fraud based on that score. If fraud is identified as likely, the server sends a warning to the terminal.

[0423] The device immediately displays this warning as a pop-up to the user, allowing them to recognize in real time that the call may be unsafe. Furthermore, the server summarizes the call and sends a message containing the key points to registered family members and other relevant parties, further supporting the user's safety.

[0424] As a concrete example, consider a scenario where an elderly user receives a call from an unknown number. If the system determines that this call may be fraudulent, the device will display a warning to the user stating "This may be a scam," and the server will simultaneously notify registered family members of this summary. This allows the user and their family to take swift action. This system leverages the advantages of machine learning and real-time monitoring technology to provide an effective means of protecting users from phone scams.

[0425] The following describes the processing flow.

[0426] Step 1:

[0427] The device automatically launches the monitoring application when the user receives a call. This prepares it to collect audio data in real time.

[0428] Step 2:

[0429] Based on user permission, the device encrypts the audio data during a call and sends it to the server in streaming format. This transmission is performed while protecting the user's privacy.

[0430] Step 3:

[0431] The server converts the received audio data into text data using speech recognition technology. This conversion is performed in real time, preparing the data for analysis.

[0432] Step 4:

[0433] The server uses a generative AI model to analyze text data and identify specific patterns and keywords that indicate fraudulent activity. The analysis is performed based on a database of past fraud cases.

[0434] Step 5:

[0435] The server scores the risk of fraud based on the analysis results. This score reflects the degree of likelihood of fraud.

[0436] Step 6:

[0437] If the score exceeds a set threshold, the server determines that it may be a scam and sends a warning to the device.

[0438] Step 7:

[0439] The device immediately displays a warning message to the user, reminding them about the security of their calls. This message is presented in an intuitive and easy-to-understand format.

[0440] Step 8:

[0441] The server summarizes the call content and sends an alert notification containing the summary to pre-configured family members and relevant parties. This helps families share the risk of fraud and take appropriate action.

[0442] (Example 1)

[0443] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0444] In recent years, with the advancement of communication technology, the methods of telephone fraud have diversified, and fraud victims, particularly the elderly, are on the rise. However, current countermeasures have difficulty detecting fraudulent activity in real time and issuing warnings to users quickly, which presents a challenge in preventing fraud.

[0445] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0446] In this invention, the server includes means for automatically launching a monitoring application when a communication device initiates a call, means for acquiring voice information in real time and transmitting it to a data processing device using streaming technology, and means for converting the voice information into text format. This enables the user to quickly analyze the content of a call and immediately issue a warning if there is a possibility of fraud.

[0447] "Communication equipment" refers to electronic devices used for making calls and data communications, specifically terminals used by users when making voice calls or other forms of communication.

[0448] A "monitoring application" refers to a software program installed on a communication device that automatically starts up when a call begins to acquire and transmit voice data.

[0449] "Voice information" refers to the content of the user's speech acquired during a call, and this information serves as input data for later analysis.

[0450] A "data processing device" refers to a computer or server that has the function of receiving, analyzing, and converting audio information.

[0451] "Streaming technology" refers to a method of transmitting audio data in real time and is used to efficiently provide audio information to data processing devices.

[0452] "Means of converting to text format" refers to the process of converting acquired audio information into text data, which is achieved using speech recognition technology.

[0453] "Generative AI technology" refers to technologies that analyze data acquired using machine learning and natural language processing to generate new information and patterns.

[0454] This system is a protective system designed to shield users from phone scams. Specifically, a monitoring application installed on the user's communication device automatically runs as soon as a call begins. This allows the communication device to acquire voice information during the call in real time and transmit it to a data processing unit (server).

[0455] The device uses widely used streaming protocols (such as WebRTC) to securely and efficiently transmit the acquired audio information to the server. The server receives this audio information and converts it into text format using speech recognition software. Existing speech recognition technologies, such as the Google Cloud Speech-to-Text API, can be used for this conversion.

[0456] Next, the server analyzes the converted text data using generative AI technology. The analysis is performed while referring to data models related to fraudulent activity, and prompts are used to identify fraud patterns. An example of such a prompt is the instruction given to the generative AI model: "Analyze this audio data and score the likelihood that it is fraudulent." Based on the results of this analysis, the server performs a risk assessment and scores calls that may be fraudulent.

[0457] For example, if a user receives a call from an unknown number, and the server determines that it is highly likely to be a scam, a warning is sent to the communication device. This warning immediately appears as a pop-up to the user, allowing them to instantly recognize the security of the call. Furthermore, the server summarizes the call content and, depending on the user's settings, notifies third parties such as family members or acquaintances, further enhancing the user's security.

[0458] Thus, the present invention provides a system that enhances user security by incorporating real-time analysis of voice information and an immediate warning function.

[0459] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0460] Step 1:

[0461] The terminal automatically launches a monitoring application when a user initiates a call. At this time, the terminal prepares to acquire audio information and begins capturing audio data during the user's call. The input is the user's call audio, and the output is real-time audio data. The terminal then prepares to stream this audio data for subsequent processing.

[0462] Step 2:

[0463] The terminal transmits the acquired audio information to the server using streaming technology. Here, the terminal uses a real-time communication protocol such as WebRTC to compress the audio data and perform appropriate buffering, minimizing data loss during the communication process. The input is the real-time audio data acquired in step 1, and the output is the data received by the server.

[0464] Step 3:

[0465] The server receives audio information sent from the terminal and converts it into text format using speech recognition software. Specifically, the server inputs the received audio data into a recognition engine such as the Google Cloud Speech-to-Text API, generating the resulting text data. The input is the received audio data, and the output is the converted text data.

[0466] Step 4:

[0467] The server analyzes the text data converted from the audio information using a generative AI model. Here, it refers to a pre-built fraud pattern data model and inputs prompt sentences into the generative AI model to evaluate the text. The input is the converted text data and prompt sentences, and the output is the information from the analyzed fraud risk assessment.

[0468] Step 5:

[0469] The server scores the risk of fraud based on the analysis results. Specifically, it applies a scoring algorithm to the analyzed data to evaluate whether it is highly likely to be fraudulent. The input is the text data of the analysis results, and the output is a score indicating the likelihood of fraud.

[0470] Step 6:

[0471] If the server determines that a transaction is likely to be fraudulent, it sends a warning to the device. The device receives this warning and immediately displays a pop-up notification to the user stating, "This may be a scam." The input is the score of the detection result, and the output is the warning displayed to the user.

[0472] Step 7:

[0473] The server summarizes the call content using a generation AI model and notifies registered family members and related parties of the user with the summarized information. The summary extracts the key points of the call and is structured as a concise message. The input is the text data of the call, and the output is a notification message to third parties.

[0474] (Application Example 1)

[0475] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0476] In modern times, telephone scams have become more sophisticated, and there is a problem in that vulnerable users, especially the elderly, are particularly susceptible to becoming victims. This increases the risk of many people suffering financial losses and emotional distress. To solve this problem, a system is needed that can assess the risk of telephone scams in real time and immediately warn users in a visually clear and easy-to-understand manner.

[0477] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0478] In this invention, the server includes an information processing device that includes means for monitoring the user's communications and acquiring voice information, means for converting the acquired voice information into symbolic information, means for analyzing the symbolic information to identify patterns indicating fraudulent activity, and means for displaying a visual warning using augmented reality technology. This enables the user to grasp the risk of fraud in real time and take immediate action.

[0479] An "information processing device" is a computer system that performs processing such as acquiring, analyzing, and displaying audio information.

[0480] "User" refers to an individual or organization that utilizes the system based on this patent and is protected from the risk of fraud.

[0481] "Communication" refers to information transmission activities such as phone calls and the transmission of voice information performed by users.

[0482] "Voice information" refers to data composed of voices exchanged during communication.

[0483] "Symbolic information" refers to data representation obtained by converting audio information into forms such as characters and numbers.

[0484] "Fraudulent activity" refers to unethical acts aimed at obtaining profit by deceiving others.

[0485] A "pattern" is a characteristic data component used to identify fraudulent activity.

[0486] "Analysis" is data processing that extracts specific information based on acquired data.

[0487] Augmented reality technology is a technique that overlays digital information onto real-world visual information.

[0488] A "warning" is a message intended to inform the user of a potential danger and to urge them to take precautions.

[0489] To implement this invention, it is first necessary to install an information processing device in the user's terminal. This device acquires voice information in real time during a call and converts that information into symbolic information. The terminal uses voice recognition software to convert the acquired voice information into text data.

[0490] Next, the device sends this text data to the server. The server is equipped with a generative AI model that analyzes the text data to identify patterns indicating fraudulent activity. If the analysis suggests fraud, a warning is sent to the device. Furthermore, the server uses a machine learning algorithm to score the likelihood of fraud based on the analyzed data and provides the warning to the user visually using augmented reality technology.

[0491] As a concrete example, consider a scenario where a user receives an important call from an unknown number. If the server determines that the content of this call resembles a scam, it immediately informs the device, and a warning message, "This may be a scam," is displayed via the augmented reality display. Based on this information, the user can take swift action.

[0492] This system is designed to help users make calls safely and protect themselves from the risk of fraud. An example of a prompt message is, "Explain how to evaluate the fraudulent nature of a phone call in real time and display a warning message on the HMD display."

[0493] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0494] Step 1:

[0495] The terminal activates the voice acquisition module as soon as the user starts a call. This module captures the voice data during the call in real time and encodes that data in digital format. The input is the voice during the call, and the output is digital voice data.

[0496] Step 2:

[0497] The terminal passes encoded digital audio data to speech recognition software. The speech recognition software analyzes this data and converts it into symbolic information in string format. This process converts the audio data into text data. The input is digital audio data, and the output is the converted text data.

[0498] Step 3:

[0499] The device sends the converted text data to the server using streaming technology. The server inputs the received text data into a generating AI model, which analyzes it to identify patterns of fraudulent activity. The input is text data, and the output is a score indicating the likelihood of fraud.

[0500] Step 4:

[0501] The server calculates a fraud risk score based on the analysis results obtained by the generating AI model. If fraud is suspected, it evaluates the score against specific criteria and generates an alert. The input is the analysis results, and the output is the fraud risk score and the alert.

[0502] Step 5:

[0503] If fraud is suspected, the server sends an alert to the device. The device then uses augmented reality technology to display a message on its screen to visually communicate the received alert to the user. The input is the alert from the server, and the output is the display of the warning message to the user.

[0504] Step 6:

[0505] The user reviews the fraud warning message displayed on the screen and takes appropriate action based on that information. The input is the warning message, and the output is the user's response.

[0506] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0507] This invention is a fraud prevention system that incorporates emotion recognition functionality. It not only analyzes the content of a user's phone call but also takes into account their emotions during the call, thereby enabling more accurate fraud detection.

[0508] The system is installed on the user's device and automatically activates the monitoring application when a call begins. The device collects call audio in real time and sends it securely to the server. The transmitted audio data is converted into text by the server using speech recognition technology. This text data is then analyzed by a generative AI, and at the same time, an emotion engine identifies the user's emotional state.

[0509] The emotion engine analyzes the tone and patterns of the user's voice to identify emotional states such as tension, anxiety, and confusion. This information is combined with the results of text data analysis to more reliably detect potential fraud. The server uses the emotion recognition results in the fraud risk scoring process, evaluating them comprehensively with the results of regular text analysis. If a high probability of fraud is determined, a warning is immediately issued to the device to alert the user. The content of the warning message is adjusted according to the emotions recognized by the emotion engine and presented to the user in the most appropriate way.

[0510] In addition, the server creates a call summary and notifies family members or trusted third parties, with this notification including additional information based on sentiment data to help recipients understand the situation more accurately. Sentiment data is also used for future analysis and improvement of learning models, contributing to overall system performance.

[0511] For example, if the emotion engine detects that a user is feeling anxious, the system will issue a warning earlier than usual and provide the user with advice to alleviate that anxiety. In this way, the present invention aims to dramatically improve the effectiveness of fraud prevention.

[0512] The following describes the processing flow.

[0513] Step 1:

[0514] When the user receives a call, the device automatically launches the monitoring app and prepares to acquire audio data in real time.

[0515] Step 2:

[0516] The device transmits the acquired audio data to the server using secure streaming technology. This transmission is encrypted to protect user privacy.

[0517] Step 3:

[0518] The server instantly converts the received audio data into text data using speech recognition technology. This conversion prepares the audio information into an analyzable format.

[0519] Step 4:

[0520] The server simultaneously activates an emotion engine to analyze the user's emotional state from the transmitted audio data. It determines whether the user is experiencing tension, anxiety, or anger based on factors such as tone of voice, speed, and pauses.

[0521] Step 5:

[0522] The server uses a generation AI to analyze the converted text data. This analysis identifies patterns and keywords that suggest fraud and assesses the likelihood of fraud.

[0523] Step 6:

[0524] Based on the results of the emotion engine analysis and text analysis, the server scores the fraud risk. This score is used to determine whether a warning is needed for the user.

[0525] Step 7:

[0526] The server sends a warning to the device if the scoring result exceeds a set threshold. The content of the warning message is adjusted according to the user's emotional state to prompt appropriate action.

[0527] Step 8:

[0528] The device displays a warning message to the user in real time. This display allows the user to immediately become aware of the security of their call.

[0529] Step 9:

[0530] The server summarizes the call and notifies family members or trusted third parties of the summary. This notification also includes information about the emotional state of the caller to help recipients accurately understand the situation.

[0531] (Example 2)

[0532] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0533] Traditional fraud prevention systems simply analyze the user's call content as text, issuing warnings without considering emotional shifts, resulting in low accuracy in detecting fraud. Furthermore, they struggled to provide appropriate warnings based on the actual level of risk, making them user-unfriendly.

[0534] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0535] In this invention, the server includes means for converting acoustic data into encoded data, means for analyzing the acoustic data to identify an emotional state, and means for integrating the results of the analysis of the encoded data and the results of the identification of the emotional state to evaluate the fraud risk. This makes it possible to evaluate the fraud risk with high accuracy, taking into account the user's emotional state, and to issue appropriate warnings.

[0536] "Audio data" refers to the voice information obtained from a user's phone call and is fundamental data for analyzing voice.

[0537] "Encoded data" refers to data obtained by converting acoustic data into text or other parsable formats.

[0538] "Emotional state" refers to the psychological or emotional condition derived from the user's voice, and includes tension, anxiety, and other similar states.

[0539] "Fraud risk" refers to the degree of likelihood that fraud is occurring, as assessed based on the analysis of call content and emotional state.

[0540] A "warning" refers to a cautionary message sent to a user when the risk of fraud is deemed high.

[0541] This invention is a system aimed at detecting fraudulent activity with high accuracy by monitoring and analyzing the content of user calls in real time. This system integrates the acquisition of acoustic data, analysis of emotional state, assessment of fraud risk, issuance of warnings, and notification to third parties.

[0542] The terminal automatically collects acoustic data as soon as the user starts a call and sends it to the server. This acoustic data is securely transferred to the server over the network. The server then converts the transferred acoustic data into encoded data using speech recognition technology. In this process, a speech recognition API is typically used as the specific software.

[0543] The server further analyzes the encoded data using a generation AI model to scrutinize the user's call content. Simultaneously, it uses an emotion recognition engine to evaluate the user's emotional state from the acoustic data. This emotion recognition process employs a dedicated algorithm that analyzes the tone and pitch of the voice.

[0544] The generative AI model analyzes various conversation patterns to determine the likelihood of fraudulent activity and integrates this with an analysis of the user's emotional state to comprehensively assess the fraud risk. Based on this risk assessment, if a high probability of fraud is detected, a warning is sent to the device. The warning message is displayed in real time with content optimized for the user's emotional state.

[0545] In addition, the server creates a summary of the call content and notifies a third party, along with information about the emotional state of the caller. This notification allows the recipient to understand the situation more accurately. This enables the early detection of fraudulent activity that may have occurred without the user's knowledge, and allows for appropriate countermeasures to be taken.

[0546] For example, if a user experiences anxiety during a phone call and the emotion recognition engine detects this anxiety, the system will issue a warning more quickly than usual, accompanied by a message offering advice to alleviate the anxiety. This allows the user to take immediate action.

[0547] An example of a prompt message would be: "Analyze the content and emotional data of a user when they receive a fraudulent call, and provide a way to offer early warnings and advice if they are feeling anxious."

[0548] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0549] Step 1:

[0550] The device automatically collects acoustic data when it detects the start of a call. The input is an audio signal, acquired in real time through the device's internal microphone. This acoustic data is transmitted to the server via network communication. The output is the acoustic data converted into a secure transmission format.

[0551] Step 2:

[0552] The server receives audio data transmitted from the terminal and converts it into encoded data using speech recognition technology. The input is audio data, which is then converted into text data through encoding. Specifically, it uses a speech recognition API to break down the audio signal into words and strings. The output is encoded data in text format.

[0553] Step 3:

[0554] The server uses a generative AI model to analyze encoded text data. The input is encoded data, and the server performs an analysis to identify patterns indicating fraudulent activity. This analysis applies various classification algorithms provided by the generative AI model. The output is the analysis results, including fraud risk information.

[0555] Step 4:

[0556] The server identifies the user's emotional state based on acoustic data. The input is acoustic data, and emotions are identified by analyzing the pitch and tone of the voice in particular. Specifically, an emotion recognition engine is used, outputting states of tension and anxiety as numerical values. The output is data indicating the emotional state.

[0557] Step 5:

[0558] The server integrates the results of the analysis of encoded data and the results of the identification of emotional states to assess fraud risk. The input consists of the analysis results and emotional data, and scoring is performed to comprehensively evaluate the integrated data from both. Specifically, a scoring algorithm is applied to quantify the likelihood of fraud. The output is the fraud risk assessment score.

[0559] Step 6:

[0560] The terminal notifies the user of warning messages received from the server. The input is the content of the warning, and the warning message is displayed in real time. Specifically, it presents the appropriate warning message visually or audibly through the user interface. The output is the warning message for the user to review.

[0561] Step 7:

[0562] The server creates a summary based on call content and sentiment data and notifies a trusted third party. The input consists of text data of the call content and sentiment information, which are used to construct the summary. Specifically, it uses a summary generation API to extract important information and adds sentiment information as supplementary data. The output is a notification message containing the summary information.

[0563] (Application Example 2)

[0564] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0565] Fraudulent activities in online transactions and communications are becoming increasingly sophisticated, raising the risk of users becoming victims of fraud. Furthermore, conventional fraud detection technologies struggle to accurately capture changes in users' emotions, making effective preventative measures against fraudulent transactions and communications necessary. In addition, there is a lack of warnings that take into account users' mental state and emotions, resulting in insufficient support for users to take appropriate measures.

[0566] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0567] In this invention, the server includes means for monitoring user communications and acquiring audio data, means for converting the acquired audio into text data, means for analyzing the text data to identify patterns indicating fraud, and means for recognizing the user's emotional state. This allows for the sending of warnings based on the user's emotional state and scoring of the likelihood of fraudulent transactions, thereby protecting the user from fraudulent activity.

[0568] "Means for monitoring user communications and acquiring voice data" refers to a function that captures telephone and other voice communications made by users in real time and records that voice information as data.

[0569] "Methods for converting acquired audio data into text data" refers to technologies that convert information obtained through audio into text information and generate it in a format that is easy to analyze.

[0570] "Means for analyzing text data to identify patterns indicating fraudulent activity" refers to a system that analyzes transcribed data to detect characteristic patterns or phrases that suggest potential fraudulent activity.

[0571] "Means for recognizing a user's emotional state" refers to technology that determines a user's emotional state at a given time by analyzing the tone of their voice and their choice of words during communication.

[0572] "Means for sending emotional state-based warnings" refers to a mechanism that generates appropriate warning messages according to the detected emotions of the user and notifies the user accordingly.

[0573] "A means of summarizing call content and notifying a third party of information including emotional data" refers to a function that concisely summarizes the communication content and adds the user's emotional information, then provides that information to a trusted third party.

[0574] "Methods for scoring the likelihood of fraud" refer to a process that uses machine learning to quantify and evaluate the risk of fraudulent activity based on acquired data.

[0575] This invention aims to integrate a fraud prevention system into an application for e-commerce sites, enabling users to proactively detect fraud risks during online transactions or customer support calls. The system utilizes smartphones and cloud servers to monitor user communications.

[0576] The device acquires voice data in real time as soon as a call begins and securely sends it to a cloud server. On the server, speech recognition technology such as Google Cloud Speech-to-Text is used to convert the voice data into text data. This text data is analyzed by generative AI models such as OpenAI GPT-4 to detect patterns of fraudulent activity. The server also analyzes the user's voice tone to identify their emotional state. When emotional states such as tension, anxiety, or confusion are identified, the suspicion of fraudulent activity increases.

[0577] By comprehensively evaluating the analysis results and emotional state, and scoring the likelihood of fraud, if a high risk is detected, an emotion-based warning message is immediately sent to the device. For example, if a user attempts to purchase an unusually large item and an emotion of anxiety is detected, a warning such as "Please reconfirm this transaction. We also recommend contacting official support" will be sent.

[0578] Furthermore, call summaries and sentiment data will be shared with a trusted third party based on the user's consent. This sharing helps in the early detection of signs of fraud.

[0579] An example of a prompt message might be: "Generate a warning message to display if the user shows signs of anxiety or confusion when making a large transaction."

[0580] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0581] Step 1:

[0582] The terminal detects the start of a user's phone call. Real-time acquisition of audio data occurs at the start of the call. The input is the user's voice during the call; by collecting this as audio data, accurate data is provided to the system.

[0583] Step 2:

[0584] The audio data acquired by the device is securely sent to the cloud server as an audio file. The input is the audio data collected in step 1, and securely transferring this to the server enables subsequent processing.

[0585] Step 3:

[0586] The server uses speech recognition technology to convert received audio data into text data. The input here is the transmitted audio file, which is then transcribed using a speech recognition tool such as Google Cloud Speech-to-Text. The output is parseable text data.

[0587] Step 4:

[0588] The server uses a generative AI model to analyze text data and detect patterns indicating fraudulent activity. The input is the text data obtained in step 3, and a generative AI model such as GPT-4 performs pattern analysis. The output is information assessing the risk of fraudulent activity.

[0589] Step 5:

[0590] Simultaneously, the server analyzes the tone of the voice data and recognizes the user's emotional state. The input is the voice data itself, and emotion recognition is performed using a specific algorithm. The output is the user's emotional state (e.g., tension, anxiety).

[0591] Step 6:

[0592] The server integrates the results of fraud risk assessment and sentiment recognition to calculate a risk score. A machine learning algorithm is used for this, with the input being the results of steps 4 and 5. The output is the overall risk score.

[0593] Step 7:

[0594] A warning message based on the user's emotional state is immediately sent to the user's device. The input is the risk score calculated in step 6, and an appropriate warning is generated. The output is the warning message displayed to the user.

[0595] Step 8:

[0596] The server summarizes the call content and notifies a trusted third party, including emotional data. The input consists of audio text data and emotional state information, which are used to create the notification. The output is detailed notification information sent to the third party.

[0597] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0598] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0599] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0600] [Fourth Embodiment]

[0601] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0602] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0603] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0604] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0605] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0606] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0607] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0608] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0609] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0610] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0611] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0612] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0613] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0614] This invention embodies the configuration and operation method of a system for protecting users from telephone fraud. This system is executed by a monitoring application installed on the user's terminal.

[0615] First, the device automatically launches the application when the user starts a call. This launch allows the device to capture the voice data during the call in real time and send it to the server. With the user's permission, the device uses streaming technology to securely relay this data to the server.

[0616] The server converts the received audio data into text using speech recognition software and analyzes the resulting text data. This analysis is performed using generative AI technology based on a pre-stored database of fraud patterns. The server evaluates the risk of fraud using a specific scoring system and determines the likelihood of fraud based on that score. If fraud is identified as likely, the server sends a warning to the terminal.

[0617] The device immediately displays this warning as a pop-up to the user, allowing them to recognize in real time that the call may be unsafe. Furthermore, the server summarizes the call and sends a message containing the key points to registered family members and other relevant parties, further supporting the user's safety.

[0618] As a concrete example, consider a scenario where an elderly user receives a call from an unknown number. If the system determines that this call may be fraudulent, the device will display a warning to the user stating "This may be a scam," and the server will simultaneously notify registered family members of this summary. This allows the user and their family to take swift action. This system leverages the advantages of machine learning and real-time monitoring technology to provide an effective means of protecting users from phone scams.

[0619] The following describes the processing flow.

[0620] Step 1:

[0621] The device automatically launches the monitoring application when the user receives a call. This prepares it to collect audio data in real time.

[0622] Step 2:

[0623] Based on user permission, the device encrypts the audio data during a call and sends it to the server in streaming format. This transmission is performed while protecting the user's privacy.

[0624] Step 3:

[0625] The server converts the received audio data into text data using speech recognition technology. This conversion is performed in real time, preparing the data for analysis.

[0626] Step 4:

[0627] The server uses a generative AI model to analyze text data and identify specific patterns and keywords that indicate fraudulent activity. The analysis is performed based on a database of past fraud cases.

[0628] Step 5:

[0629] The server scores the risk of fraud based on the analysis results. This score reflects the degree of likelihood of fraud.

[0630] Step 6:

[0631] If the score exceeds a set threshold, the server determines that it may be a scam and sends a warning to the device.

[0632] Step 7:

[0633] The device immediately displays a warning message to the user, reminding them about the security of their calls. This message is presented in an intuitive and easy-to-understand format.

[0634] Step 8:

[0635] The server summarizes the call content and sends an alert notification containing the summary to pre-configured family members and relevant parties. This helps families share the risk of fraud and take appropriate action.

[0636] (Example 1)

[0637] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0638] In recent years, with the advancement of communication technology, the methods of telephone fraud have diversified, and fraud victims, particularly the elderly, are on the rise. However, current countermeasures have difficulty detecting fraudulent activity in real time and issuing warnings to users quickly, which presents a challenge in preventing fraud.

[0639] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0640] In this invention, the server includes means for automatically launching a monitoring application when a communication device initiates a call, means for acquiring voice information in real time and transmitting it to a data processing device using streaming technology, and means for converting the voice information into text format. This enables the user to quickly analyze the content of a call and immediately issue a warning if there is a possibility of fraud.

[0641] "Communication equipment" refers to electronic devices used for making calls and data communications, specifically terminals used by users when making voice calls or other forms of communication.

[0642] A "monitoring application" refers to a software program installed on a communication device that automatically starts up when a call begins to acquire and transmit voice data.

[0643] "Voice information" refers to the content of the user's speech acquired during a call, and this information serves as input data for later analysis.

[0644] A "data processing device" refers to a computer or server that has the function of receiving, analyzing, and converting audio information.

[0645] "Streaming technology" refers to a method of transmitting audio data in real time and is used to efficiently provide audio information to data processing devices.

[0646] "Means of converting to text format" refers to the process of converting acquired audio information into text data, which is achieved using speech recognition technology.

[0647] "Generative AI technology" refers to technologies that analyze data acquired using machine learning and natural language processing to generate new information and patterns.

[0648] This system is a protective system designed to shield users from phone scams. Specifically, a monitoring application installed on the user's communication device automatically runs as soon as a call begins. This allows the communication device to acquire voice information during the call in real time and transmit it to a data processing unit (server).

[0649] The device uses widely used streaming protocols (such as WebRTC) to securely and efficiently transmit the acquired audio information to the server. The server receives this audio information and converts it into text format using speech recognition software. Existing speech recognition technologies, such as the Google Cloud Speech-to-Text API, can be used for this conversion.

[0650] Next, the server analyzes the converted text data using generative AI technology. The analysis is performed while referring to data models related to fraudulent activity, and prompts are used to identify fraud patterns. An example of such a prompt is the instruction given to the generative AI model: "Analyze this audio data and score the likelihood that it is fraudulent." Based on the results of this analysis, the server performs a risk assessment and scores calls that may be fraudulent.

[0651] For example, if a user receives a call from an unknown number, and the server determines that it is highly likely to be a scam, a warning is sent to the communication device. This warning immediately appears as a pop-up to the user, allowing them to instantly recognize the security of the call. Furthermore, the server summarizes the call content and, depending on the user's settings, notifies third parties such as family members or acquaintances, further enhancing the user's security.

[0652] Thus, the present invention provides a system that enhances user security by incorporating real-time analysis of voice information and an immediate warning function.

[0653] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0654] Step 1:

[0655] The terminal automatically launches a monitoring application when a user initiates a call. At this time, the terminal prepares to acquire audio information and begins capturing audio data during the user's call. The input is the user's call audio, and the output is real-time audio data. The terminal then prepares to stream this audio data for subsequent processing.

[0656] Step 2:

[0657] The terminal transmits the acquired audio information to the server using streaming technology. Here, the terminal uses a real-time communication protocol such as WebRTC to compress the audio data and perform appropriate buffering, minimizing data loss during the communication process. The input is the real-time audio data acquired in step 1, and the output is the data received by the server.

[0658] Step 3:

[0659] The server receives audio information sent from the terminal and converts it into text format using speech recognition software. Specifically, the server inputs the received audio data into a recognition engine such as the Google Cloud Speech-to-Text API, generating the resulting text data. The input is the received audio data, and the output is the converted text data.

[0660] Step 4:

[0661] The server analyzes the text data converted from the audio information using a generative AI model. Here, it refers to a pre-built fraud pattern data model and inputs prompt sentences into the generative AI model to evaluate the text. The input is the converted text data and prompt sentences, and the output is the information from the analyzed fraud risk assessment.

[0662] Step 5:

[0663] The server scores the risk of fraud based on the analysis results. Specifically, it applies a scoring algorithm to the analyzed data to evaluate whether it is highly likely to be fraudulent. The input is the text data of the analysis results, and the output is a score indicating the likelihood of fraud.

[0664] Step 6:

[0665] If the server determines that a transaction is likely to be fraudulent, it sends a warning to the device. The device receives this warning and immediately displays a pop-up notification to the user stating, "This may be a scam." The input is the score of the detection result, and the output is the warning displayed to the user.

[0666] Step 7:

[0667] The server summarizes the call content using a generation AI model and notifies registered family members and related parties of the user with the summarized information. The summary extracts the key points of the call and is structured as a concise message. The input is the text data of the call, and the output is a notification message to third parties.

[0668] (Application Example 1)

[0669] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0670] In modern times, telephone scams have become more sophisticated, and there is a problem in that vulnerable users, especially the elderly, are particularly susceptible to becoming victims. This increases the risk of many people suffering financial losses and emotional distress. To solve this problem, a system is needed that can assess the risk of telephone scams in real time and immediately warn users in a visually clear and easy-to-understand manner.

[0671] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0672] In this invention, the server includes an information processing device that includes means for monitoring the user's communications and acquiring voice information, means for converting the acquired voice information into symbolic information, means for analyzing the symbolic information to identify patterns indicating fraudulent activity, and means for displaying a visual warning using augmented reality technology. This enables the user to grasp the risk of fraud in real time and take immediate action.

[0673] An "information processing device" is a computer system that performs processing such as acquiring, analyzing, and displaying audio information.

[0674] "User" refers to an individual or organization that utilizes the system based on this patent and is protected from the risk of fraud.

[0675] "Communication" refers to information transmission activities such as phone calls and the transmission of voice information performed by users.

[0676] "Voice information" refers to data composed of voices exchanged during communication.

[0677] "Symbolic information" refers to data representation obtained by converting audio information into forms such as characters and numbers.

[0678] "Fraudulent activity" refers to unethical acts aimed at obtaining profit by deceiving others.

[0679] A "pattern" is a characteristic data component used to identify fraudulent activity.

[0680] "Analysis" is data processing that extracts specific information based on acquired data.

[0681] Augmented reality technology is a technique that overlays digital information onto real-world visual information.

[0682] A "warning" is a message intended to inform the user of a potential danger and to urge them to take precautions.

[0683] To implement this invention, it is first necessary to install an information processing device in the user's terminal. This device acquires voice information in real time during a call and converts that information into symbolic information. The terminal uses voice recognition software to convert the acquired voice information into text data.

[0684] Next, the device sends this text data to the server. The server is equipped with a generative AI model that analyzes the text data to identify patterns indicating fraudulent activity. If the analysis suggests fraud, a warning is sent to the device. Furthermore, the server uses a machine learning algorithm to score the likelihood of fraud based on the analyzed data and provides the warning to the user visually using augmented reality technology.

[0685] As a concrete example, consider a scenario where a user receives an important call from an unknown number. If the server determines that the content of this call resembles a scam, it immediately informs the device, and a warning message, "This may be a scam," is displayed via the augmented reality display. Based on this information, the user can take swift action.

[0686] This system is designed to help users make calls safely and protect themselves from the risk of fraud. An example of a prompt message is, "Explain how to evaluate the fraudulent nature of a phone call in real time and display a warning message on the HMD display."

[0687] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0688] Step 1:

[0689] The terminal activates the voice acquisition module as soon as the user starts a call. This module captures the voice data during the call in real time and encodes that data in digital format. The input is the voice during the call, and the output is digital voice data.

[0690] Step 2:

[0691] The terminal passes encoded digital audio data to speech recognition software. The speech recognition software analyzes this data and converts it into symbolic information in string format. This process converts the audio data into text data. The input is digital audio data, and the output is the converted text data.

[0692] Step 3:

[0693] The device sends the converted text data to the server using streaming technology. The server inputs the received text data into a generating AI model, which analyzes it to identify patterns of fraudulent activity. The input is text data, and the output is a score indicating the likelihood of fraud.

[0694] Step 4:

[0695] The server calculates a fraud risk score based on the analysis results obtained by the generating AI model. If fraud is suspected, it evaluates the score against specific criteria and generates an alert. The input is the analysis results, and the output is the fraud risk score and the alert.

[0696] Step 5:

[0697] If fraud is suspected, the server sends an alert to the device. The device then uses augmented reality technology to display a message on its screen to visually communicate the received alert to the user. The input is the alert from the server, and the output is the display of the warning message to the user.

[0698] Step 6:

[0699] The user reviews the fraud warning message displayed on the screen and takes appropriate action based on that information. The input is the warning message, and the output is the user's response.

[0700] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0701] This invention is a fraud prevention system that incorporates emotion recognition functionality. It not only analyzes the content of a user's phone call but also takes into account their emotions during the call, thereby enabling more accurate fraud detection.

[0702] The system is installed on the user's device and automatically activates the monitoring application when a call begins. The device collects call audio in real time and sends it securely to the server. The transmitted audio data is converted into text by the server using speech recognition technology. This text data is then analyzed by a generative AI, and at the same time, an emotion engine identifies the user's emotional state.

[0703] The emotion engine analyzes the tone and patterns of the user's voice to identify emotional states such as tension, anxiety, and confusion. This information is combined with the results of text data analysis to more reliably detect potential fraud. The server uses the emotion recognition results in the fraud risk scoring process, evaluating them comprehensively with the results of regular text analysis. If a high probability of fraud is determined, a warning is immediately issued to the device to alert the user. The content of the warning message is adjusted according to the emotions recognized by the emotion engine and presented to the user in the most appropriate way.

[0704] In addition, the server creates a call summary and notifies family members or trusted third parties, with this notification including additional information based on sentiment data to help recipients understand the situation more accurately. Sentiment data is also used for future analysis and improvement of learning models, contributing to overall system performance.

[0705] For example, if the emotion engine detects that a user is feeling anxious, the system will issue a warning earlier than usual and provide the user with advice to alleviate that anxiety. In this way, the present invention aims to dramatically improve the effectiveness of fraud prevention.

[0706] The following describes the processing flow.

[0707] Step 1:

[0708] When the user receives a call, the device automatically launches the monitoring app and prepares to acquire audio data in real time.

[0709] Step 2:

[0710] The device transmits the acquired audio data to the server using secure streaming technology. This transmission is encrypted to protect user privacy.

[0711] Step 3:

[0712] The server instantly converts the received audio data into text data using speech recognition technology. This conversion prepares the audio information into an analyzable format.

[0713] Step 4:

[0714] The server simultaneously activates an emotion engine to analyze the user's emotional state from the transmitted audio data. It determines whether the user is experiencing tension, anxiety, or anger based on factors such as tone of voice, speed, and pauses.

[0715] Step 5:

[0716] The server uses a generation AI to analyze the converted text data. This analysis identifies patterns and keywords that suggest fraud and assesses the likelihood of fraud.

[0717] Step 6:

[0718] Based on the results of the emotion engine analysis and text analysis, the server scores the fraud risk. This score is used to determine whether a warning is needed for the user.

[0719] Step 7:

[0720] The server sends a warning to the device if the scoring result exceeds a set threshold. The content of the warning message is adjusted according to the user's emotional state to prompt appropriate action.

[0721] Step 8:

[0722] The device displays a warning message to the user in real time. This display allows the user to immediately become aware of the security of their call.

[0723] Step 9:

[0724] The server summarizes the call and notifies family members or trusted third parties of the summary. This notification also includes information about the emotional state of the caller to help recipients accurately understand the situation.

[0725] (Example 2)

[0726] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0727] Traditional fraud prevention systems simply analyze the user's call content as text, issuing warnings without considering emotional shifts, resulting in low accuracy in detecting fraud. Furthermore, they struggled to provide appropriate warnings based on the actual level of risk, making them user-unfriendly.

[0728] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0729] In this invention, the server includes means for converting acoustic data into encoded data, means for analyzing the acoustic data to identify an emotional state, and means for integrating the results of the analysis of the encoded data and the results of the identification of the emotional state to evaluate the fraud risk. This makes it possible to evaluate the fraud risk with high accuracy, taking into account the user's emotional state, and to issue appropriate warnings.

[0730] "Audio data" refers to the voice information obtained from a user's phone call and is fundamental data for analyzing voice.

[0731] "Encoded data" refers to data obtained by converting acoustic data into text or other parsable formats.

[0732] "Emotional state" refers to the psychological or emotional condition derived from the user's voice, and includes tension, anxiety, and other similar states.

[0733] "Fraud risk" refers to the degree of likelihood that fraud is occurring, as assessed based on the analysis of call content and emotional state.

[0734] A "warning" refers to a cautionary message sent to a user when the risk of fraud is deemed high.

[0735] This invention is a system aimed at detecting fraudulent activity with high accuracy by monitoring and analyzing the content of user calls in real time. This system integrates the acquisition of acoustic data, analysis of emotional state, assessment of fraud risk, issuance of warnings, and notification to third parties.

[0736] The terminal automatically collects acoustic data as soon as the user starts a call and sends it to the server. This acoustic data is securely transferred to the server over the network. The server then converts the transferred acoustic data into encoded data using speech recognition technology. In this process, a speech recognition API is typically used as the specific software.

[0737] The server further analyzes the encoded data using a generation AI model to scrutinize the user's call content. Simultaneously, it uses an emotion recognition engine to evaluate the user's emotional state from the acoustic data. This emotion recognition process employs a dedicated algorithm that analyzes the tone and pitch of the voice.

[0738] The generative AI model analyzes various conversation patterns to determine the likelihood of fraudulent activity and integrates this with an analysis of the user's emotional state to comprehensively assess the fraud risk. Based on this risk assessment, if a high probability of fraud is detected, a warning is sent to the device. The warning message is displayed in real time with content optimized for the user's emotional state.

[0739] In addition, the server creates a summary of the call content and notifies a third party, along with information about the emotional state of the caller. This notification allows the recipient to understand the situation more accurately. This enables the early detection of fraudulent activity that may have occurred without the user's knowledge, and allows for appropriate countermeasures to be taken.

[0740] For example, if a user experiences anxiety during a phone call and the emotion recognition engine detects this anxiety, the system will issue a warning more quickly than usual, accompanied by a message offering advice to alleviate the anxiety. This allows the user to take immediate action.

[0741] An example of a prompt message would be: "Analyze the content and emotional data of a user when they receive a fraudulent call, and provide a way to offer early warnings and advice if they are feeling anxious."

[0742] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0743] Step 1:

[0744] The device automatically collects acoustic data when it detects the start of a call. The input is an audio signal, acquired in real time through the device's internal microphone. This acoustic data is transmitted to the server via network communication. The output is the acoustic data converted into a secure transmission format.

[0745] Step 2:

[0746] The server receives audio data transmitted from the terminal and converts it into encoded data using speech recognition technology. The input is audio data, which is then converted into text data through encoding. Specifically, it uses a speech recognition API to break down the audio signal into words and strings. The output is encoded data in text format.

[0747] Step 3:

[0748] The server uses a generative AI model to analyze encoded text data. The input is encoded data, and the server performs an analysis to identify patterns indicating fraudulent activity. This analysis applies various classification algorithms provided by the generative AI model. The output is the analysis results, including fraud risk information.

[0749] Step 4:

[0750] The server identifies the user's emotional state based on acoustic data. The input is acoustic data, and emotions are identified by analyzing the pitch and tone of the voice in particular. Specifically, an emotion recognition engine is used, outputting states of tension and anxiety as numerical values. The output is data indicating the emotional state.

[0751] Step 5:

[0752] The server integrates the results of the analysis of encoded data and the results of the identification of emotional states to assess fraud risk. The input consists of the analysis results and emotional data, and scoring is performed to comprehensively evaluate the integrated data from both. Specifically, a scoring algorithm is applied to quantify the likelihood of fraud. The output is the fraud risk assessment score.

[0753] Step 6:

[0754] The terminal notifies the user of warning messages received from the server. The input is the content of the warning, and the warning message is displayed in real time. Specifically, it presents the appropriate warning message visually or audibly through the user interface. The output is the warning message for the user to review.

[0755] Step 7:

[0756] The server creates a summary based on call content and sentiment data and notifies a trusted third party. The input consists of text data of the call content and sentiment information, which are used to construct the summary. Specifically, it uses a summary generation API to extract important information and adds sentiment information as supplementary data. The output is a notification message containing the summary information.

[0757] (Application Example 2)

[0758] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0759] Fraudulent activities in online transactions and communications are becoming increasingly sophisticated, raising the risk of users becoming victims of fraud. Furthermore, conventional fraud detection technologies struggle to accurately capture changes in users' emotions, making effective preventative measures against fraudulent transactions and communications necessary. In addition, there is a lack of warnings that take into account users' mental state and emotions, resulting in insufficient support for users to take appropriate measures.

[0760] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0761] In this invention, the server includes means for monitoring user communications and acquiring audio data, means for converting the acquired audio into text data, means for analyzing the text data to identify patterns indicating fraud, and means for recognizing the user's emotional state. This allows for the sending of warnings based on the user's emotional state and scoring of the likelihood of fraudulent transactions, thereby protecting the user from fraudulent activity.

[0762] "Means for monitoring user communications and acquiring voice data" refers to a function that captures telephone and other voice communications made by users in real time and records that voice information as data.

[0763] "Methods for converting acquired audio data into text data" refers to technologies that convert information obtained through audio into text information and generate it in a format that is easy to analyze.

[0764] "Means for analyzing text data to identify patterns indicating fraudulent activity" refers to a system that analyzes transcribed data to detect characteristic patterns or phrases that suggest potential fraudulent activity.

[0765] "Means for recognizing a user's emotional state" refers to technology that determines a user's emotional state at a given time by analyzing the tone of their voice and their choice of words during communication.

[0766] "Means for sending emotional state-based warnings" refers to a mechanism that generates appropriate warning messages according to the detected emotions of the user and notifies the user accordingly.

[0767] "A means of summarizing call content and notifying a third party of information including emotional data" refers to a function that concisely summarizes the communication content and adds the user's emotional information, then provides that information to a trusted third party.

[0768] "Methods for scoring the likelihood of fraud" refer to a process that uses machine learning to quantify and evaluate the risk of fraudulent activity based on acquired data.

[0769] This invention aims to integrate a fraud prevention system into an application for e-commerce sites, enabling users to proactively detect fraud risks during online transactions or customer support calls. The system utilizes smartphones and cloud servers to monitor user communications.

[0770] The device acquires voice data in real time as soon as a call begins and securely sends it to a cloud server. On the server, speech recognition technology such as Google Cloud Speech-to-Text is used to convert the voice data into text data. This text data is analyzed by generative AI models such as OpenAI GPT-4 to detect patterns of fraudulent activity. The server also analyzes the user's voice tone to identify their emotional state. When emotional states such as tension, anxiety, or confusion are identified, the suspicion of fraudulent activity increases.

[0771] By comprehensively evaluating the analysis results and emotional state, and scoring the likelihood of fraud, if a high risk is detected, an emotion-based warning message is immediately sent to the device. For example, if a user attempts to purchase an unusually large item and an emotion of anxiety is detected, a warning such as "Please reconfirm this transaction. We also recommend contacting official support" will be sent.

[0772] Furthermore, call summaries and sentiment data will be shared with a trusted third party based on the user's consent. This sharing helps in the early detection of signs of fraud.

[0773] An example of a prompt message might be: "Generate a warning message to display if the user shows signs of anxiety or confusion when making a large transaction."

[0774] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0775] Step 1:

[0776] The terminal detects the start of a user's phone call. Real-time acquisition of audio data occurs at the start of the call. The input is the user's voice during the call; by collecting this as audio data, accurate data is provided to the system.

[0777] Step 2:

[0778] The audio data acquired by the device is securely sent to the cloud server as an audio file. The input is the audio data collected in step 1, and securely transferring this to the server enables subsequent processing.

[0779] Step 3:

[0780] The server uses speech recognition technology to convert received audio data into text data. The input here is the transmitted audio file, which is then transcribed using a speech recognition tool such as Google Cloud Speech-to-Text. The output is parseable text data.

[0781] Step 4:

[0782] The server uses a generative AI model to analyze text data and detect patterns indicating fraudulent activity. The input is the text data obtained in step 3, and a generative AI model such as GPT-4 performs pattern analysis. The output is information assessing the risk of fraudulent activity.

[0783] Step 5:

[0784] Simultaneously, the server analyzes the tone of the voice data and recognizes the user's emotional state. The input is the voice data itself, and emotion recognition is performed using a specific algorithm. The output is the user's emotional state (e.g., tension, anxiety).

[0785] Step 6:

[0786] The server integrates the results of fraud risk assessment and sentiment recognition to calculate a risk score. A machine learning algorithm is used for this, with the input being the results of steps 4 and 5. The output is the overall risk score.

[0787] Step 7:

[0788] A warning message based on the user's emotional state is immediately sent to the user's device. The input is the risk score calculated in step 6, and an appropriate warning is generated. The output is the warning message displayed to the user.

[0789] Step 8:

[0790] The server summarizes the call content and notifies a trusted third party, including emotional data. The input consists of audio text data and emotional state information, which are used to create the notification. The output is detailed notification information sent to the third party.

[0791] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0792] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0793] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0794] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0795] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0796] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0797] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0798] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0799] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0800] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0801] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0802] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0803] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0804] 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.

[0805] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0806] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0807] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0808] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0809] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0810] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0811] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted as being incorporated by reference.

[0812] The following is further disclosed regarding the embodiments described above.

[0813] (Claim 1)

[0814] A means of monitoring user communications and acquiring voice data,

[0815] A means of converting acquired audio data into text data,

[0816] A means of analyzing text data to identify patterns indicating fraudulent activity,

[0817] A means of sending a warning in case of suspected fraud,

[0818] A means of summarizing the content of a call and notifying a third party,

[0819] A system that includes this.

[0820] (Claim 2)

[0821] The system according to claim 1, comprising means for displaying a warning to the user in real time.

[0822] (Claim 3)

[0823] The system according to claim 1, comprising means for scoring the likelihood of fraud using a machine learning algorithm based on analyzed data.

[0824] "Example 1"

[0825] (Claim 1)

[0826] A means to automatically launch a monitoring application when a communication device initiates a call,

[0827] A means for acquiring audio information in real time and transmitting it to a data processing device using streaming technology,

[0828] A means of converting audio information into text format,

[0829] A method for analyzing text data using a data model related to fraudulent activity,

[0830] A method for performing a risk assessment based on the analyzed information and scoring the likelihood of fraud,

[0831] A means of sending a warning to a communication device in case of fraud concerns,

[0832] A means of summarizing the content of the call and notifying the relevant parties,

[0833] A system that includes this.

[0834] (Claim 2)

[0835] The system according to claim 1, further comprising means for immediately displaying a warning to the user.

[0836] (Claim 3)

[0837] The system according to claim 1, comprising means for identifying fraud patterns using generative AI technology in the analysis process.

[0838] "Application Example 1"

[0839] (Claim 1)

[0840] The information processing device includes means for monitoring the user's communications and acquiring voice information,

[0841] A means of converting acquired audio information into symbolic information,

[0842] A means for analyzing symbolic information to identify patterns indicating fraudulent activity,

[0843] A means of sending a warning in case of suspected fraud,

[0844] A means of summarizing the content of communications and notifying a third party,

[0845] A means of displaying a visual warning using augmented reality technology,

[0846] A system that includes this.

[0847] (Claim 2)

[0848] The system according to claim 1, wherein the information processing device includes means for visually displaying a warning to the user in real time using augmented reality technology.

[0849] (Claim 3)

[0850] The system according to claim 1, comprising means for scoring the likelihood of fraud using a machine learning algorithm based on the analyzed information, and displaying the results using augmented reality technology.

[0851] "Example 2 of combining an emotion engine"

[0852] (Claim 1)

[0853] A means of monitoring user communications and acquiring acoustic data,

[0854] A means of converting acquired acoustic data into encoded data,

[0855] A means of analyzing acoustic data to identify emotional states,

[0856] A means for analyzing encoded data to identify patterns indicating fraudulent activity,

[0857] A method for evaluating fraud risk by integrating the results of encoding data analysis and the results of emotional state identification,

[0858] A means of sending a warning when it is determined that there is a high possibility of fraud,

[0859] A means of summarizing the content of a call and notifying a third party,

[0860] A system that includes this.

[0861] (Claim 2)

[0862] The system according to claim 1, comprising means for displaying a warning to the user in real time, and adjusting the content of the warning according to the recognized emotional state.

[0863] (Claim 3)

[0864] The system according to claim 1, comprising a means for scoring the likelihood of fraud using a learning algorithm based on analyzed data, and also reflecting the results of identifying emotional states in the scoring.

[0865] "Application example 2 of combining emotional engines"

[0866] (Claim 1)

[0867] A means of monitoring user communications and acquiring voice data,

[0868] A means of converting acquired audio data into text data,

[0869] A means of analyzing text data to identify patterns indicating fraudulent activity,

[0870] A means of recognizing the user's emotional state,

[0871] A means of sending emotionally charged warnings in cases of suspected fraud,

[0872] A means of summarizing the content of a call and notifying a third party of the information, including emotional data,

[0873] A system that includes this.

[0874] (Claim 2)

[0875] The system according to claim 1, further comprising means for displaying a warning to the user in real time according to their emotional state.

[0876] (Claim 3)

[0877] The system according to claim 1, comprising means for scoring the likelihood of fraud using a machine learning algorithm based on analyzed data and emotion recognition results. [Explanation of Symbols]

[0878] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of monitoring user communications and acquiring voice data, A means of converting acquired audio data into text data, A means of analyzing text data to identify patterns indicating fraudulent activity, A means of sending a warning in case of suspected fraud, A means of summarizing the content of a call and notifying a third party, A system that includes this.

2. The system according to claim 1, comprising means for displaying a warning to the user in real time.

3. The system according to claim 1, comprising means for scoring the likelihood of fraud using a machine learning algorithm based on analyzed data.

Images