System

Abstract

A system is provided.SOLUTION: A system comprising: means for recording a voice; means for transmitting the recorded voice to an information processing apparatus; means for analyzing voice data and extracting a voice feature by the information processing apparatus; means for determining whether the voice feature matches registered data based on the voice feature; means for transmitting a determination result to a communication terminal; and means for displaying the determination result by the communication terminal.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] The number of victims of special frauds over the phone is increasing, with elderly people and children often being targeted in particular. The problem is that victims are easily deceived when fraudsters call pretending to be family members or acquaintances. In order to prevent such fraud, a means is needed to quickly and reliably determine whether the caller is actually a family member or acquaintance when answering a phone call. [Means for solving the problem]

[0005] The present invention provides a system including a means for recording voice, a means for transmitting the recorded voice to an information processing device, a means for the information processing device to analyze the voice data and extract voice characteristics, a means for determining whether the voice matches registered data based on the voice characteristics, a means for transmitting the determination result to a communication terminal, and a means for the communication terminal to display the determination result. With this system, a user records voice when registering a phone book, the communication terminal transmits the recorded voice data to the information processing device, and the information processing device extracts voice characteristics from the recorded voice data and registers them in a database. Furthermore, when an incoming call is received, the communication terminal records the caller's voice in real time, transmits the voice data to the information processing device, extracts voice characteristics, and compares the voice data with registered voice data to determine whether it matches. As a result, the communication terminal displays the determination result, allowing users to quickly confirm whether the caller is a registered person, thereby preventing special fraud.

[0006] A "voice recording means" is a device or software that has the functionality to record a user's spoken voice as digital data.

[0007] An "information processing device" is a computer system or server for performing analysis, feature extraction, and comparison of speech data.

[0008] "Voice data" refers to data that represents recorded user voice information in digital form.

[0009] "Audio features" are data extracted from recorded audio data that indicate specific characteristics or patterns of audio.

[0010] A "communication terminal" is a device such as a mobile phone or a landline phone that a user uses to make calls.

[0011] "Enrollment data" is information in a database that includes voice characteristics and corresponding person information.

[0012] The "determination result" is the comparison result as to whether the voice characteristics of the caller match the registered data.

[0013] A "database" is a system that systematically stores data such as voice characteristics and corresponding person information.

[0014] The "notification means" is software or hardware that has a function for notifying the user of the communication terminal of the determination result.

[0015] "When an incoming call comes in" refers to the moment when an external call comes in to the communication terminal.

[0016] "Real-time" refers to the operation of processing audio data as it is recorded and displaying the judgment results immediately. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9]1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0018] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0019] First, the terms used in the following description will be explained.

[0020] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0021] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0022] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0023] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0024] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0025] [First embodiment]

[0026] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0027] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0028] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0029] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0030] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0031] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0032] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0033] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0035] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0036] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0037] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0038] The present invention is a speech recognition system that utilizes a communication terminal such as a mobile phone or a landline phone and an information processing device that processes voice data. This system is designed as an effective means of preventing damage from special frauds.

[0039] First, we will explain the process by which a user registers the voices of family members and acquaintances. The user uses the voice recording function of the communication terminal to record the voice of each person. When recording is complete, the communication terminal sends the recorded voice data to an information processing device (server). The server analyzes the received voice data and extracts a feature vector of the voice. The name and information of the person corresponding to this voice feature vector are registered in a database.

[0040] Next, we will explain the voice recognition process when an incoming call is received. When a communication terminal receives an incoming call, it records the caller's voice in real time the moment the call starts. This recorded voice data is sent from the communication terminal to a server. The server analyzes the received voice data and extracts a feature vector of the voice. The extracted feature vector is compared with voice feature vectors already registered in a database. The result of this comparison is judged by the server and sent to the communication terminal.

[0041] Based on the results of the determination received from the server, the communication device displays on its display whether the caller is a registered person. If there is a match, the communication device displays "From Mr. / Ms. XX," indicating that the caller is a family member or acquaintance. If there is no match, the communication device displays "Unregistered," warning the user to be cautious.

[0042] As a specific example, consider the case where a user registers the voice of their son. The user uses the recording function of their communication device to record their son's voice and sends the voice data to the server. The server analyzes the voice data and registers the son's voice feature vector and name in a database. After that, when the user receives a call on their communication device and the caller is their son, the caller's voice is recorded in real time and sent to the server. The server analyzes the voice feature vector and, if it matches the son's voice feature vector in the database, the communication device displays the message "Call from your son." On the other hand, if the caller is not registered, the communication device displays the message "Unregistered caller."

[0043] In this way, the system based on the present invention allows users to receive calls with peace of mind and is effective in preventing damage from special frauds.

[0044] The processing flow will be explained below.

[0045] Step 1:

[0046] The user starts the application on the communication terminal, selects the voice registration menu of the phone book, and then enters the name of the person to be registered.

[0047] Step 2:

[0048] The user records the voice of the person to be registered into the communication terminal. When the user presses the recording completion button, the recorded voice data is saved in the communication terminal.

[0049] Step 3:

[0050] The device then transmits the stored audio data to a server, typically via the Internet.

[0051] Step 4:

[0052] The server analyzes the received voice data and extracts a voice feature vector using a voice recognition algorithm.

[0053] Step 5:

[0054] The server registers the extracted voice feature vectors together with the corresponding person's name in a database, which stores the voice feature vectors and person information.

[0055] Step 6:

[0056] When the communication terminal receives an incoming call, the user starts the call, and the terminal records the caller's voice in real time.

[0057] Step 7:

[0058] The device sends the recorded voice data of the caller to the server, where it is immediately analyzed.

[0059] Step 8:

[0060] The server analyzes the received voice data and extracts the caller's voice feature vector, which is then compared with the registered data in the database.

[0061] Step 9:

[0062] The server judges the result of matching the voice feature vector. If the caller's voice matches the registered data, the judgement result is recorded as "Registered Person." If not, it is recorded as "Unregistered Person."

[0063] Step 10:

[0064] The server transmits the result of the determination to the communication terminal, which then determines the content to display based on the received result of the determination.

[0065] Step 11:

[0066] The device displays the result on the screen. If the caller's voice matches the registered voice, the name of the registered person is displayed. If they do not match, the device displays "Unregistered."

[0067] Step 12:

[0068] The user checks the display and determines whether the caller is a family member or acquaintance. If the message "Unregistered" appears, the user becomes more vigilant.

[0069] Example 1

[0070] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0071] The purpose of this invention is to provide a communication system that protects users from crimes such as special fraud and allows them to make calls with peace of mind. In particular, there is a need to prevent fraudulent acts by recognizing the voice of the other party in real time and comparing it with pre-registered voices. This system provides a method for registering and recognizing the voices of family members and acquaintances, and comprehensively identifying suspicious callers.

[0072] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0073] In this invention, the server includes a means for analyzing voice data and extracting a voice feature vector, a means for determining whether the voice feature vector matches registered data in a database, and a means for transmitting the determination result to the communication terminal. This allows users to analyze the caller's voice in real time when a call starts and compare it with pre-registered voices. Furthermore, the determination result is transmitted to the communication terminal and displayed on the display, allowing users to immediately confirm whether the caller is a registered person. This prevents crimes such as special fraud and allows users to make calls with peace of mind.

[0074] A "communication terminal" is a device that allows a user to record voice and transmit the voice data to an information processing device.

[0075] "Recording means" refers to the function of recording audio within the communication terminal.

[0076] The "information processing device" is a device that analyzes transmitted voice data and extracts voice feature vectors.

[0077] "Audio data" refers to a recording of audio that is recorded by a communication terminal and transmitted to an information processing device.

[0078] A "voice feature vector" is a collection of data that is extracted from voice data and that quantifies individual voice features.

[0079] A "database" is a digital data structure for storing speech feature vectors and related information.

[0080] "Matching means" refers to a function that compares the extracted voice feature vector with existing data in the database and determines whether they match.

[0081] The "determination result" is information indicating the result of matching or mismatching obtained by matching the speech feature vectors.

[0082] The "display means" is a function that enables the communication terminal to visually convey the determination result to the user.

[0083] This invention is a voice recognition system using a communication terminal and an information processing device (server). The main purpose of this system is to prevent damage from special frauds, specifically by recognizing the caller's voice in real time and comparing it with pre-registered voices to detect suspicious callers.

[0084] Hardware and software used

[0085] The communication terminals used are mobile phones and landlines, which have the functions of recording voice and transmitting data via the Internet.

[0086] The server is an information processing device required to analyze voice data. Voice analysis software, such as OpenAI's Whisper or Google's Speech-to-Text, is installed on the server.

[0087] Processing flow

[0088] 1. Recording Audio

[0089] The user uses the recording function of the communication device to record the voices of family members and acquaintances, and the recorded voice data is saved in the communication device.

[0090] 2. Sending audio data

[0091] The terminal transmits the recorded voice data to an information processing device (server) via the Internet.

[0092] 3. Analysis and registration of audio data

[0093] The server analyzes the received voice data and extracts a voice feature vector. The extracted voice feature vector is then registered in a database. This saves the registrant's voice data in the database.

[0094] 4. Processing when an incoming call arrives

[0095] When the device receives an incoming call, it records the caller's voice in real time as soon as the call begins, and the recorded voice data is sent to the server.

[0096] 5. Real-time analysis and matching of voice data

[0097] The server analyzes the received voice data in real time, extracts voice feature vectors, and then compares the extracted feature vectors with existing data in the database to determine whether they match.

[0098] 6. Sending and displaying the results

[0099] The server sends the result of the judgment to the communication terminal, which then displays the result to the user. If the caller is a registered person, it will display "From Mr. / Ms. XX," and if not, it will display "Unregistered." This allows the user to easily check whether the caller is trustworthy.

[0100] Specific examples

[0101] When a user registers his son's voice

[0102] The user uses the recording function of the communication device to have their son say, "Hello, I'm your son." Once the recording is complete, the voice data is sent to the server, which performs voice analysis, extracts the voice feature vector, and registers it in a database.

[0103] When a call comes in

[0104] The device receives an incoming call and displays "Incoming call." When the call begins, the caller's voice is recorded in real time and sent to the server. The server analyzes the voice data in real time and compares it with the son's voice feature vector in the database. As a result, the message "Call from your son" is displayed.

[0105] Prompt Sentence Examples

[0106] "We will record and register your mother's voice. We will start recording, so please say, 'Hello, I'm your mother.'"

[0107] "You have an incoming call. We are verifying that the caller is a registered user."

[0108] "(For example, my son) I'll record my son's voice and say, 'Hi, I'm your son,' and we'll save it."

[0109] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0110] Step 1: Record your audio

[0111] A user uses the voice recording function of a communication terminal to record the voices of family members or acquaintances.

[0112] Input: Voice of family or friends

[0113] Specific behavior:

[0114] 1. Select "Audio Recording" from the menu on your communication device.

[0115] 2. Press the start recording button and have the subject speak.

[0116] 3. Press the Stop Recording button to save the recording.

[0117] Output: Recorded audio data

[0118] Step 2: Sending audio data

[0119] The terminal transmits the recorded voice data to an information processing device (server).

[0120] Input: Recorded audio data

[0121] Specific behavior:

[0122] 1. Select the saved audio data.

[0123] 2. Send the data to the server using your internet connection.

[0124] Output: Audio data sent to the server

[0125] Step 3: Analyze and register audio data

[0126] The server receives the audio data and analyzes it.

[0127] Input: Transmitted audio data

[0128] Specific behavior:

[0129] 1. The server analyzes the voice data using voice analysis software (e.g., OpenAI's Whisper or Google's Speech-to-Text).

[0130] 2. Extract the voice feature vector and register it in a database along with the subject's name and information.

[0131] Output: Audio feature vectors and correspondence information registered in the database

[0132] Step 4: Receiving a call

[0133] The device receives an incoming call.

[0134] Input: Incoming call signal from the caller

[0135] Specific behavior:

[0136] 1. The call will ring and the display will show "Incoming call."

[0137] 2. The user receives the call and starts the conversation.

[0138] Output: Call Start

[0139] Step 5: Record the caller's voice

[0140] The device records the caller's voice in real time.

[0141] Input: Caller's voice

[0142] Specific behavior:

[0143] 1. Start recording as soon as the call begins.

[0144] 2. Prepare to send the recording data to the server.

[0145] Output: Recorded caller voice data

[0146] Step 6: Sending audio data

[0147] The terminal transmits the recorded voice data of the caller to the server.

[0148] Input: Recorded caller voice data

[0149] Specific behavior:

[0150] 1. Send real-time audio data to a server via an internet connection.

[0151] Output: Caller's voice data sent to the server

[0152] Step 7: Analyze and match the audio data

[0153] The server analyzes the caller's voice data received and compares it with existing data in the database.

[0154] Input: Caller's voice data sent to the server

[0155] Specific behavior:

[0156] 1. The server uses speech analysis software to extract the caller's voice feature vector.

[0157] 2. Compare the extracted speech feature vector with existing speech feature vectors in the database.

[0158] 3. Determine the matching result.

[0159] Output: Matching result (determining whether the caller is a registered person)

[0160] Step 8: Receive and display the results

[0161] The terminal displays the judgment result received from the server.

[0162] Input: Matching result sent from the server

[0163] Specific behavior:

[0164] 1. The terminal displays the received judgment result.

[0165] 2. If the caller is a registered person, it will display "From Mr. / Ms. XX."

[0166] 3. If there is no match, the message "Unregistered" will be displayed to warn the user.

[0167] Output: Display of judgment result (registered person or unregistered person)

[0168] This series of processes allows the user to check in real time whether the caller is trustworthy.

[0169] (Application example 1)

[0170] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0171] In recent years, voice-based fraud has become more prevalent, requiring users to be vigilant against calls from unknown callers. However, conventional voice recognition systems are inadequate, particularly in their inability to display real-time results, making it difficult for users to make a quick decision. This has led to a growing demand for effective means to prevent fraud. Furthermore, there is a need for a system that can instantly display a warning on smart devices, such as smart glasses and head-mounted displays, that are linked to communication terminals.

[0172] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0173] In this invention, the server includes a means for recording voice, a means for transmitting the recorded voice to an information processing device, a means for the information processing device to analyze the voice data and extract voice characteristics, a means for determining whether the voice data matches registered data based on the voice characteristics, a means for transmitting the determination result to a communication terminal, a means for the communication terminal to display the determination result, and a means for linking with a smart device to display the determination result in real time. This allows users to immediately be alert to calls from unknown callers and prevent fraud. Furthermore, by using smart glasses or a head-mounted display, a warning is displayed directly in the user's field of vision, allowing for a quick and effective response.

[0174] "Audio recording means" means equipment or software for recording audio signals as digital data.

[0175] The "means for transmitting the recorded voice to the information processing device" is a function or protocol for transferring the recorded voice data to the information processing device through a communication line.

[0176] "Means for an information processing device to analyze voice data and extract voice features" refers to a process in which an information processing device analyzes input voice data using an algorithm and extracts features.

[0177] The "means for determining whether or not the data matches the registered data based on the voice characteristics" is a process for comparing pre-registered voice characteristic data with newly extracted voice characteristic data and determining whether or not there is a match.

[0178] The "means for transmitting the determination result to the communication terminal" is a communication means or protocol for transmitting the determination result to the communication terminal in real time.

[0179] The "means for the communication terminal to display the determination result" is a function that visually presents the determination result to the user using the display or interface of the communication terminal.

[0180] "Means for linking with a smart device and displaying the judgment results in real time" refers to a system that links with a smart device such as smart glasses or a head-mounted display and presents the judgment results to the user in real time.

[0181] "Smart glasses" are glasses-type devices that run specific applications and provide augmented reality and notification information through a display.

[0182] A "head-mounted display" is a device worn on the head and positioned within the user's field of vision to provide visual information to the user.

[0183] This invention is a voice recognition system that allows users to receive calls with peace of mind. In order to prevent damage from special fraud, the system analyzes the caller's voice in real time and compares it with pre-registered voice data to determine whether the caller is a registered person.

[0184] The system of the present invention includes the following hardware and software.

[0185] Hardware: Smart glasses (e.g., Google Glass, Vuzix Blade), smartphones

[0186] Software: Audio processing libraries (e.g., pyaudio, librosa), dedicated APIs for smart glasses (e.g., Google Glass API), local databases (e.g., SQLite)

[0187] When a user uses the system for the first time, they record the voice data they want to register in advance using a communication terminal and send it to the server. The server analyzes the received voice data and extracts a voice feature vector. This extracted voice feature vector is registered in the database. Through this process, the voices of family members and acquaintances are registered in the system.

[0188] When a call comes in, the communication terminal records the caller's voice in real time and sends the recorded voice data to the server. The server analyzes the voice data and extracts a voice feature vector. It then compares it with voice feature vectors already registered in a database and sends the result to the communication terminal. The communication terminal displays the result on smart glasses or a head-mounted display. This allows the user to confirm whether the caller is a registered person before starting the call.

[0189] As a specific example, consider the case where a user registers the voice of their son. The user uses the recording function of their communication device to record their son's voice and sends the voice data to the server. The server analyzes the voice data and registers the son's voice feature vector in a database. If the user then receives a call while cooking and the caller is their son, the communication device records the caller's voice in real time and sends it to the server. The server analyzes the voice feature vector, and if it matches the voice feature vector of the son that has already been registered, the smart glasses display will say, "Call from your son." On the other hand, if the caller is an unregistered person, the message will say, "This call is from your son. Please be careful."

[0190] This system allows users to immediately become alert to calls from unknown callers, preventing fraud before they occur. In addition, by using smart glasses or a head-mounted display, a warning can be displayed directly in the user's field of vision, enabling a quick and effective response.

[0191] An example of a prompt is as follows:

[0192] "This is a demo scene of a voice recognition security guard. It registers the voices of family members, identifies suspicious callers, and displays a warning. It compares them in real time with a recording of the son's voice. This system significantly reduces the risk of special fraud."

[0193] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0194] Step 1:

[0195] A user uses the recording function of a communication terminal to record the voices of family members or acquaintances.

[0196] Input: Audio signal (voice of user's family or acquaintance)

[0197] Processing: Use the recording function of the communication device to record the audio as digital data.

[0198] Output: Recorded audio data (digital format)

[0199] Step 2:

[0200] The communication terminal transmits the recorded voice data to the server.

[0201] Input: Recorded audio data

[0202] Processing: Upload the audio data to the server via a communication line.

[0203] Output: Audio data sent to the server

[0204] Step 3:

[0205] The server analyzes the received voice data and extracts a voice feature vector.

[0206] Input: Received audio data

[0207] Processing: The server uses an audio processing library (e.g., librosa) to analyze the audio data and generate an audio feature vector using techniques such as FFT (Fast Fourier Transform).

[0208] Output: Audio feature vector

[0209] Step 4:

[0210] The server registers the voice feature vector in a database.

[0211] Input: Audio feature vector, corresponding person name and information

[0212] Processing: The server uses a local database (e.g., SQLite) to store the voice feature vectors and person information in the database.

[0213] Output: Trained speech feature data

[0214] Step 5:

[0215] When a call is received, the communication terminal records the caller's voice in real time.

[0216] Input: Incoming voice signal (caller's voice)

[0217] Processing: Record the audio as digital data in real time using the recording function of the communication device.

[0218] Output: Recorded real-time audio data

[0219] Step 6:

[0220] The communication terminal transmits the recorded real-time voice data to the server.

[0221] Input: Recorded real-time audio data

[0222] Processing: Upload the audio data to the server via a communication line.

[0223] Output: Real-time audio data sent to the server

[0224] Step 7:

[0225] The server analyzes the real-time voice data and extracts voice feature vectors.

[0226] Input: Received real-time audio data

[0227] Processing: The server uses a speech processing library to analyze the speech data and generate a speech feature vector.

[0228] Output: Real-time audio feature vectors

[0229] Step 8:

[0230] The server matches the real-time voice feature vector with the registered voice feature vectors in the database.

[0231] Input: Real-time speech feature vector, registered speech feature vector in database

[0232] Processing: The server calculates the cosine similarity or Euclidean distance to assess the degree of match between the two vectors.

[0233] Output: Matching result (match / no match)

[0234] Step 9:

[0235] The server transmits the determination result to the communication terminal.

[0236] Input: Judgment result

[0237] Processing: The judgment result is sent to the communication terminal via the communication line.

[0238] Output: The judgment result sent to the communication terminal

[0239] Step 10:

[0240] The communication terminal displays the judgment results on smart glasses or a head-mounted display.

[0241] Input: Received verdict

[0242] Processing: Using the smart glasses' dedicated API, the judgment result is displayed on the screen.

[0243] Output: The result displayed on the screen ("This is from ____" or "This is an unregistered user")

[0244] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0245] The present invention is a system that aims to prevent more sophisticated special frauds by combining a voice recognition system with an emotion engine, and by analyzing the caller's voice when a communication terminal receives an incoming call, and also analyzing the user's emotional state.

[0246] First, we will explain the process by which a user registers the voices of family members and acquaintances. The user uses the voice recording function of the communication device to record the voice of each person. When recording is complete, the communication device sends the recorded voice data to the server. The server analyzes the received voice data and extracts a feature vector of the voice. The name and information of the person corresponding to this voice feature vector are registered in a database.

[0247] Next, we will explain the voice recognition and emotion analysis process when an incoming call is received. When a communication terminal receives an incoming call, the user starts the call. As soon as the call starts, the terminal records the caller's voice in real time. This recorded voice data is sent from the communication terminal to a server. The server analyzes the received voice data and extracts a feature vector of the voice. The extracted feature vector is compared with voice feature vectors already registered in a database, and the result of this comparison is determined by the server. The result is sent to the communication terminal and displayed on a display.

[0248] In addition, the communication terminal is equipped with an emotion engine that analyzes the user's voice during a call and recognizes the user's emotional state in real time. The recognized emotion information is immediately sent to the server, which then generates a warning message based on that information and sends it to the communication terminal.

[0249] As a specific example, consider the case where a user registers their mother's voice. The user uses the recording function of their communication device to record their mother's voice and sends the voice data to the server. The server analyzes the voice data and registers the mother's voice feature vector and name in a database. If the user's communication device then receives a call from the mother, the caller's voice is recorded in real time and sent to the server. The server analyzes the voice feature vector and, if it matches the mother's voice feature vector in the database, the communication device displays the message "It's from your mother." At the same time, the emotion engine analyzes the user's voice in real time, and if the user's emotions indicate anxiety or caution, a warning message such as "Caution required" is displayed on the communication device.

[0250] In this way, the system based on the present invention can provide more accurate measures to prevent special frauds by combining the results of analyzing the user's emotional state using an emotion engine in addition to caller recognition using voice feature vectors. It also provides an environment where users can receive calls with peace of mind, which is extremely effective in preventing fraud damage before it happens.

[0251] The processing flow will be explained below.

[0252] This invention is a system that combines a voice recognition system and an emotion engine, and when a communication terminal receives an incoming call, it analyzes the caller's voice and also analyzes the user's emotional state, thereby preventing damage from special fraud. The specific processing flow is explained below step by step.

[0253] User Voice Enrollment Process

[0254] Step 1:

[0255] The user starts the application on the communication terminal, selects the voice registration menu of the phone book, and then enters the name of the person to be registered.

[0256] Step 2:

[0257] The user records the voice of the person to be registered into the communication terminal, and when the recording is complete, the user presses the recording completion button.

[0258] Step 3:

[0259] The device sends the recorded audio data to a server via the Internet.

[0260] Step 4:

[0261] The server analyzes the received voice data and extracts a voice feature vector using a voice recognition algorithm.

[0262] Step 5:

[0263] The server registers the extracted voice feature vectors together with the names of the corresponding people in a database.

[0264] Incoming voice recognition and sentiment analysis process

[0265] Step 1:

[0266] When the communication terminal receives an incoming call, the user starts the call, and the terminal records the caller's voice in real time.

[0267] Step 2:

[0268] The device sends the recorded voice data of the caller to the server, where it is immediately analyzed.

[0269] Step 3:

[0270] The server analyzes the received voice data and extracts the caller's voice feature vector, which is then compared with the registered data in the database.

[0271] Step 4:

[0272] The server judges the result of matching the voice feature vector. If the caller's voice matches the registered data, the judgement result is recorded as "Registered Person." If not, it is recorded as "Unregistered Person."

[0273] Step 5:

[0274] The server transmits the result of the determination to the communication terminal, which then determines the content to display based on the received result of the determination.

[0275] Step 6:

[0276] The device displays the result on the screen. If the caller's voice matches the registered voice, the name of the registered person is displayed. If they do not match, the device displays "Unregistered."

[0277] Emotion recognition process by emotion engine

[0278] Step 1:

[0279] During a call, the device records the user's voice in real time and sends it to the emotion engine.

[0280] Step 2:

[0281] The emotion engine analyzes the received user voice data and extracts the user's emotion feature vector using emotion recognition algorithms.

[0282] Step 3:

[0283] The emotion engine determines the user's emotional state, such as anxiety, alertness, or calmness, based on the extracted emotion feature vector.

[0284] Step 4:

[0285] The emotion engine sends the judgment result to the server, which generates a warning message as needed based on the received emotion information.

[0286] Step 5:

[0287] The server transmits the generated warning message to the communication terminal, and the terminal displays the received warning message on its display.

[0288] This allows users to not only confirm whether the caller is a registered person, but also receive warning messages tailored to their own emotional state, making it possible to prevent damage from special frauds.

[0289] Example 2

[0290] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0291] Conventional voice recognition systems only analyze and recognize the caller's voice, so they cannot completely prevent risks such as special fraud. Also, because they do not take into account the user's emotional state, they have the problem of not being able to issue appropriate warnings even if the user is feeling anxious or cautious.

[0292] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0293] In this invention, the server includes means for analyzing voice data and extracting voice characteristics, means for determining whether the voice characteristics match registered data, and means for generating a warning message based on emotion data and transmitting it to the communication terminal, thereby enabling the user's emotional state to be analyzed in real time and a warning message to be displayed as necessary.

[0294] The "means for recording voice" refers to a function or device that allows a user to record voice using a communication terminal.

[0295] The "means for transmitting recorded voice to the information processing device" refers to a function or device for transmitting voice data recorded by a communication terminal to the information processing device via a network.

[0296] The "means for an information processing device to analyze voice data and extract voice features" refers to software or algorithms that allow the information processing device to analyze received voice data and generate voice feature vectors.

[0297] "Means for determining whether the voice characteristics match the registered data" refers to a function or software for comparing the analyzed voice characteristics with the registered data in the database and determining whether they match.

[0298] The "means for transmitting the determination result to the communication terminal" refers to a communication function or protocol for transmitting the determination result from the information processing device to the communication terminal.

[0299] The "means for the communication terminal to display the determination result" refers to a display or display function for visually presenting the determination result received by the communication terminal to the user.

[0300] The "means for the communication terminal to analyze the emotional state of the user" refers to an emotion engine or software that allows the communication terminal to analyze the user's voice during a call and recognize the user's emotional state in real time.

[0301] The "means for transmitting analyzed emotion data to the information processing device" refers to a communication function or protocol that allows the communication terminal to transmit analyzed user emotion data to the information processing device.

[0302] "Means for an information processing device to generate an alert message based on emotion data and transmit it to a communication terminal" refers to a function or software that allows an information processing device to create an alert message based on emotion data received and transmit it to a communication terminal.

[0303] This invention combines a voice recognition system with an emotion engine, and aims to analyze the caller's voice when a communication terminal receives an incoming call, and also analyze the user's emotional state. This system will enable more advanced prevention of special frauds.

[0304] Hardware and software used

[0305] 1. Communication terminal:

[0306] Voice recording function: A microphone and recording app to record the user's voice.

[0307] Display: A screen for displaying the results and warning messages.

[0308] Emotion engine: Software for analyzing user emotions.

[0309] 2. Server (information processing device):

[0310] Voice recognition software: For example, Google Voice Recognition API.

[0311] Database management system: for example, MySQL.

[0312] Sentiment analysis algorithm: A model for analyzing the user's emotional state.

[0313] Process Overview

[0314] 1. Audio Recording and Transmission:

[0315] Users use their communication terminals to record the voices of their family members and friends, and the recorded voice data is sent from the communication terminals to a server.

[0316] 2. Analysis of audio data:

[0317] The server analyzes the received voice data and extracts the voice feature vector. It uses voice recognition software to analyze the voice waveform and generate the feature vector.

[0318] 3. Register feature vectors and data:

[0319] The extracted feature vectors are registered in a database along with the names and other information of the corresponding people. This registered data is managed using a database management system.

[0320] 4. Incoming call voice recognition and emotion analysis:

[0321] When a communication device receives an incoming call, it records the caller's voice in real time and sends it to the server, which analyzes the voice data, extracts feature vectors, and compares them with the registered data in a database.

[0322] At the same time, the emotion engine installed in the communication terminal analyzes the user's voice in real time and recognizes their emotional state.

[0323] 5. Sending and viewing results:

[0324] The server sends the results of the matching and emotion analysis to the communication terminal, which then displays the caller's name and a warning message.

[0325] Specific examples

[0326] Consider the case where a user registers their mother's voice. The user uses the recording function of the communication terminal to record their mother's voice and sends the voice data to the server. The server analyzes the voice data and registers the mother's voice feature vector and name in the database.

[0327] If the user then receives a call on their communication device and the caller is their mother, the caller's voice is recorded in real time and sent to the server. The server analyzes the voice feature vector, and if it matches the mother's voice feature vector in the database, the communication device displays the message "Call from mother." At the same time, the emotion engine analyzes the user's voice in real time, and if the user is expressing anxiety or caution, a warning message such as "Caution required" is displayed on the communication device.

[0328] Prompt Sentence Examples

[0329] "Please tell me the procedure for a system that registers the mother's voice, recognizes the mother's voice in real time when a call comes in, analyzes the user's emotional state, and displays a warning message."

[0330] In this way, the present invention provides a system that combines voice recognition and emotion analysis, thereby providing an environment in which telephones can be used more safely and reducing the risk of special fraud and the like.

[0331] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0332] Step 1:

[0333] A user uses a communication terminal to record the voice of a family member or friend. The user presses the record button on the communication terminal, and then presses the save button after recording is complete. The input is the user's voice, and the voice data is saved in the communication terminal as output.

[0334] Step 2:

[0335] The communication terminal sends the recorded voice data to the server. Once the recording is complete, the voice data is sent to the server via the network. The input is the voice data stored in the communication terminal, and the voice data that is transferred to the server as output is generated. During this process, the progress of the communication is displayed on the terminal.

[0336] Step 3:

[0337] The server analyzes the voice data and extracts voice features. The server analyzes the received voice data using voice recognition software (e.g., Google Voice Recognition API) and generates a voice feature vector. The input is the voice data transferred to the server, and the output is the extracted voice feature vector.

[0338] Step 4:

[0339] The server registers the extracted voice feature vector in a database. The server saves the name and information of the person corresponding to the voice feature vector in a database (e.g., MySQL). The input is the voice feature vector and the corresponding information, and the output is newly registered data in the database.

[0340] Step 5:

[0341] The communication terminal records the caller's voice in real time. When a call comes in, the communication terminal records the caller's voice as soon as the call starts. The input is the caller's voice when the call comes in, and the real-time recorded voice data is generated as the output.

[0342] Step 6:

[0343] The communication terminal sends the recorded voice data to the server. The voice data recorded in real time is sent to the server using asynchronous communication technology (e.g., WebSocket). The input is the voice data recorded in real time, and the voice data transferred to the server is generated as the output.

[0344] Step 7:

[0345] The server analyzes the incoming voice data and extracts a voice feature vector. The server analyzes the received voice data and generates a voice feature vector. The input is the voice data transferred to the server, and the output is the extracted voice feature vector.

[0346] Step 8:

[0347] The server compares the voice feature vector with the registered data in the database and determines whether they match. The server compares the extracted voice feature vector with the registered data in the database and determines whether they match. The input is the voice feature vector and the registered data in the database, and the output is the result of the determination of whether they match.

[0348] Step 9:

[0349] The server transmits the determination result to the communication terminal. The server transmits the determination result to the communication terminal. The input is the determination result, and the determination result is generated and transferred to the communication terminal as output.

[0350] Step 10:

[0351] The communication terminal displays the judgment result and begins emotion analysis. The judgment result is displayed on the communication terminal's display, and at the same time the emotion engine begins analyzing the user's voice in real time. The input is the judgment result from the server and the user's voice during the call, and the output is the result displayed on the display and emotion analysis data.

[0352] Step 11:

[0353] The communication terminal analyzes the user's emotional state and sends the results to the server. The emotion engine then sends the analyzed emotional data to the server. The input is the user's voice and the analysis results by the emotion engine, and the output is the emotional data sent to the server.

[0354] Step 12:

[0355] The server generates a warning message based on the emotion data and sends it to the communication terminal. The server generates a warning message based on the received emotion data and sends it to the communication terminal. The input is emotion data, and the warning message that is transferred to the communication terminal is generated as the output.

[0356] Step 13:

[0357] The communication terminal displays the warning message. The communication terminal displays the warning message received from the server on the display. The input is the warning message from the server, and the warning message displayed on the display is generated as the output.

[0358] (Application example 2)

[0359] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0360] Currently, fraud methods are becoming more sophisticated, and special frauds over the phone are on the rise. Many of these methods target the elderly, and the resulting losses are particularly significant, posing a problem. Conventional voice recognition systems primarily focus on identifying callers, without taking into account the user's emotional state. As a result, even if the caller is registered, fraudulent calls may not be detected. Therefore, there is a need for a system that can prevent fraud and risky calls by analyzing the user's emotional state.

[0361] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0362] In this invention, the server includes means for analyzing the emotional state of the user from the voice during the call, means for generating a warning message based on the emotion analysis result and transmitting it to the communication terminal, means for recording the voice, means for transmitting the recorded voice to the information processing device, means for the information processing device to analyze the voice data and extract voice characteristics, means for determining whether the voice data matches registered data based on the voice characteristics, means for transmitting the determination result to the communication terminal, and means for the communication terminal to display the determination result. This enables more advanced prevention of special frauds by taking into account both the caller's recognition and the user's emotional state.

[0363] "Audio recording means" means a device or software for capturing and recording audio data during a call.

[0364] The "means for transmitting the recorded voice to the information processing device" is a function or device that transfers the recorded voice data to a server or processing device via communication or a network.

[0365] The "means for an information processing device to analyze voice data and extract voice features" refers to an algorithm or device for identifying and extracting specific voice patterns or features from received voice data.

[0366] "Means for determining whether the voice characteristics match registered data" refers to a function or device for comparing the extracted voice characteristics with pre-registered voice data and determining whether there is a match.

[0367] The "means for transmitting the determination result to the communication terminal" is a function or device that transmits the voice feature matching result to the terminal via communication or a network.

[0368] The "means for the communication terminal to display the determination result" refers to a display device or software for visually notifying the user of the received determination result.

[0369] The "means for analyzing the user's emotional state from the voice during a call" refers to an algorithm or device for determining the user's emotions in real time from the voice data acquired during a call.

[0370] The "means for generating a warning message based on the emotion analysis result and transmitting it to the communication terminal" refers to a function or device for creating a warning message based on the recognized emotional state and transmitting it to the communication terminal.

[0371] The present invention provides a security system that prevents special frauds by analyzing the caller's voice when a communication terminal receives an incoming call and also analyzing the user's emotional state. Specific embodiments for carrying out the present invention will be described below.

[0372] System Configuration

[0373] The system includes the following major hardware and software components:

[0374] Communication device: Smartphone with voice recording and display functions

[0375] Server: Voice data analyzer, database, emotion engine

[0376] Software: Speech recognition library (e.g., speech_recognition), emotion recognition library (e.g., emotion_recognition), voice feature recognition library (e.g., voice_recognition)

[0377] Program processing and natural language explanation

[0378] 1. Voice registration:

[0379] The user uses the voice recording function of the communication device to record the voices of family members or acquaintances. After the recording is complete, the communication device sends the recorded voice data to the server.

[0380] The server analyzes the received voice data and extracts voice feature vectors. The extracted voice feature vectors and the corresponding person's name and information are registered in a database.

[0381] 2. Incoming call voice recognition and sentiment analysis:

[0382] When a communication terminal receives an incoming call, the user starts the call. The terminal records the caller's voice in real time as soon as the call starts. This recorded voice data is sent from the communication terminal to a server.

[0383] The server analyzes the received voice data and extracts the voice feature vector. The extracted feature vector is compared with the voice feature vectors already registered in the database, and the result is judged and sent to the communication terminal. The judgment result is displayed on the communication terminal.

[0384] During a call, the emotion engine installed in the communication terminal analyzes the user's voice and recognizes the user's emotional state in real time. The recognized emotional information is immediately sent to the server, which then generates a warning message based on that information and sends it to the communication terminal.

[0385] Real-world examples

[0386] For example, consider the case where a user has registered their mother's voice in advance. The user records the mother's voice using a communication terminal and sends the voice data to the server. The server analyzes the voice data and registers the mother's voice feature vector and name in a database.

[0387] Next, when the user's communication device receives a call and recognizes that the caller is the mother, the caller's voice is recorded in real time and sent to the server. The server analyzes the voice feature vector, and if it matches the voice feature vector of the mother in the database, the communication device displays the message "Call from mother." At the same time, the emotion engine analyzes the user's voice in real time, and if the user's emotion indicates anxiety or caution, a warning message such as "Caution required" is displayed.

[0388] Prompt Sentence Examples

[0389] Also, when using a generative AI model to generate an appropriate warning message, enter the following prompt:

[0390] The user's emotional state is anxious:

[0391] "We need to be extra careful as the user has expressed anxiety. Do you want to continue communicating?"

[0392] Called party is unknown:

[0393] "This caller is not registered and may be a scam. Would you like to continue the call?"

[0394] The present invention makes it possible to provide an advanced security system that analyzes both the caller's identity and the user's emotional state in real time, preventing fraud and dangerous calls before they occur.

[0395] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0396] Step 1:

[0397] Voice registration

[0398] Input: Audio data obtained using the user's recording function

[0399] How it works: A user uses the voice recording function of their communication device to record the voice of a family member or friend. After pressing the record button and finishing recording, the communication device automatically sends the voice data to the server.

[0400] Data processing: The server analyzes the received voice data and extracts voice feature vectors.

[0401] Output: The extracted speech feature vectors are registered in a database.

[0402] Step 2:

[0403] Voice recognition for incoming calls

[0404] Input: Caller's voice data recorded when receiving a call

[0405] Operation: When a communication terminal receives an incoming call, the user starts the call. At the same time, the terminal records the caller's voice in real time and sends this voice data to the server.

[0406] Data processing: The server analyzes the received voice data and extracts voice feature vectors. The extracted feature vectors are compared with previously registered voice feature vectors.

[0407] Output: The matching result is sent to the communication terminal and displayed on the screen.

[0408] Step 3:

[0409] Emotional state analysis

[0410] Input: Voice data of the user during the call

[0411] How it works: The emotion engine in the communication device analyzes the user's voice in real time during a call.

[0412] Data Processing: The emotion engine identifies the user's emotional state (e.g., anxiety, alertness, etc.).

[0413] Output: The analysis results are sent to the server.

[0414] Step 4:

[0415] Warning message generation and notification

[0416] Input: Emotion analysis results and voice feature matching results

[0417] Operation: The server generates a warning message as needed based on the emotion analysis results and voice feature matching results.

[0418] Data processing: The generated message is sent to the communication terminal.

[0419] Output: A warning message is displayed on the communication terminal.

[0420] Specific operation example

[0421] For example, when a user whose mother's voice has already been registered receives a call from her mother, the process is as follows.

[0422] In step 1, the mother's voice is pre-registered.

[0423] In step 2, the moment a call is received, the caller's voice is recorded in real time and sent to the server. The server then performs a voice analysis and compares it with pre-registered voice characteristics. If the caller is confirmed to be the mother, the message "This is from your mother" is displayed on the communication device.

[0424] In step 3, the user's voice during the call is analyzed in real time, and if the user shows any anxiety or alertness, their emotional state is sent to the server.

[0425] If the emotional state is recognized as anxious in step 4, the server generates a warning message such as "Caution required" which is displayed on the communication terminal.

[0426] Examples of prompts include:

[0427] The user's emotional state is anxious:

[0428] "We need to be extra careful as the user has expressed anxiety. Do you want to continue communicating?"

[0429] Called party is unknown:

[0430] "This caller is not registered and may be a scam. Would you like to continue the call?"

[0431] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0432] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0433] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0434] [Second embodiment]

[0435] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0436] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0437] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0438] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0439] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0440] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0441] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0442] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0443] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0444] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0445] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0446] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0447] The present invention is a speech recognition system that utilizes a communication terminal such as a mobile phone or a landline phone and an information processing device that processes voice data. This system is designed as an effective means of preventing damage from special frauds.

[0448] First, we will explain the process by which a user registers the voices of family members and acquaintances. The user uses the voice recording function of the communication terminal to record the voice of each person. When recording is complete, the communication terminal sends the recorded voice data to an information processing device (server). The server analyzes the received voice data and extracts a feature vector of the voice. The name and information of the person corresponding to this voice feature vector are registered in a database.

[0449] Next, we will explain the voice recognition process when an incoming call is received. When a communication terminal receives an incoming call, it records the caller's voice in real time the moment the call starts. This recorded voice data is sent from the communication terminal to a server. The server analyzes the received voice data and extracts a feature vector of the voice. The extracted feature vector is compared with voice feature vectors already registered in a database. The result of this comparison is judged by the server and sent to the communication terminal.

[0450] Based on the results of the determination received from the server, the communication device displays on its display whether the caller is a registered person. If there is a match, the communication device displays "From Mr. / Ms. XX," indicating that the caller is a family member or acquaintance. If there is no match, the communication device displays "Unregistered," warning the user to be cautious.

[0451] As a specific example, consider the case where a user registers the voice of their son. The user uses the recording function of their communication device to record their son's voice and sends the voice data to the server. The server analyzes the voice data and registers the son's voice feature vector and name in a database. After that, when the user receives a call on their communication device and the caller is their son, the caller's voice is recorded in real time and sent to the server. The server analyzes the voice feature vector and, if it matches the son's voice feature vector in the database, the communication device displays the message "Call from your son." On the other hand, if the caller is not registered, the communication device displays the message "Unregistered caller."

[0452] In this way, the system based on the present invention allows users to receive calls with peace of mind and is effective in preventing damage from special frauds.

[0453] The processing flow will be explained below.

[0454] Step 1:

[0455] The user starts the application on the communication terminal, selects the voice registration menu of the phone book, and then enters the name of the person to be registered.

[0456] Step 2:

[0457] The user records the voice of the person to be registered into the communication terminal. When the user presses the recording completion button, the recorded voice data is saved in the communication terminal.

[0458] Step 3:

[0459] The device then transmits the stored audio data to a server, typically via the Internet.

[0460] Step 4:

[0461] The server analyzes the received voice data and extracts a voice feature vector using a voice recognition algorithm.

[0462] Step 5:

[0463] The server registers the extracted voice feature vectors together with the corresponding person's name in a database, which stores the voice feature vectors and person information.

[0464] Step 6:

[0465] When the communication terminal receives an incoming call, the user starts the call, and the terminal records the caller's voice in real time.

[0466] Step 7:

[0467] The device sends the recorded voice data of the caller to the server, where it is immediately analyzed.

[0468] Step 8:

[0469] The server analyzes the received voice data and extracts the caller's voice feature vector, which is then compared with the registered data in the database.

[0470] Step 9:

[0471] The server judges the result of matching the voice feature vector. If the caller's voice matches the registered data, the judgement result is recorded as "Registered Person." If not, it is recorded as "Unregistered Person."

[0472] Step 10:

[0473] The server transmits the result of the determination to the communication terminal, which then determines the content to display based on the received result of the determination.

[0474] Step 11:

[0475] The device displays the result on the screen. If the caller's voice matches the registered voice, the name of the registered person is displayed. If they do not match, the device displays "Unregistered."

[0476] Step 12:

[0477] The user checks the display and determines whether the caller is a family member or acquaintance. If the message "Unregistered" appears, the user becomes more vigilant.

[0478] Example 1

[0479] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0480] The purpose of this invention is to provide a communication system that protects users from crimes such as special fraud and allows them to make calls with peace of mind. In particular, there is a need to prevent fraudulent acts by recognizing the voice of the other party in real time and comparing it with pre-registered voices. This system provides a method for registering and recognizing the voices of family members and acquaintances, and comprehensively identifying suspicious callers.

[0481] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0482] In this invention, the server includes a means for analyzing voice data and extracting a voice feature vector, a means for determining whether the voice feature vector matches registered data in a database, and a means for transmitting the determination result to the communication terminal. This allows users to analyze the caller's voice in real time when a call starts and compare it with pre-registered voices. Furthermore, the determination result is transmitted to the communication terminal and displayed on the display, allowing users to immediately confirm whether the caller is a registered person. This prevents crimes such as special fraud and allows users to make calls with peace of mind.

[0483] A "communication terminal" is a device that allows a user to record voice and transmit the voice data to an information processing device.

[0484] "Recording means" refers to the function of recording audio within the communication terminal.

[0485] The "information processing device" is a device that analyzes transmitted voice data and extracts voice feature vectors.

[0486] "Audio data" refers to a recording of audio that is recorded by a communication terminal and transmitted to an information processing device.

[0487] A "voice feature vector" is a collection of data that is extracted from voice data and that quantifies individual voice features.

[0488] A "database" is a digital data structure for storing speech feature vectors and related information.

[0489] "Matching means" refers to a function that compares the extracted voice feature vector with existing data in the database and determines whether they match.

[0490] The "determination result" is information indicating the result of matching or mismatching obtained by matching the speech feature vectors.

[0491] The "display means" is a function that enables the communication terminal to visually convey the determination result to the user.

[0492] This invention is a voice recognition system using a communication terminal and an information processing device (server). The main purpose of this system is to prevent damage from special frauds, specifically by recognizing the caller's voice in real time and comparing it with pre-registered voices to detect suspicious callers.

[0493] Hardware and software used

[0494] The communication terminals used are mobile phones and landlines, which have the functions of recording voice and transmitting data via the Internet.

[0495] The server is an information processing device required to analyze voice data. Voice analysis software, such as OpenAI's Whisper or Google's Speech-to-Text, is installed on the server.

[0496] Processing flow

[0497] 1. Recording Audio

[0498] The user uses the recording function of the communication device to record the voices of family members and acquaintances, and the recorded voice data is saved in the communication device.

[0499] 2. Sending audio data

[0500] The terminal transmits the recorded voice data to an information processing device (server) via the Internet.

[0501] 3. Analysis and registration of audio data

[0502] The server analyzes the received voice data and extracts a voice feature vector. The extracted voice feature vector is then registered in a database. This saves the registrant's voice data in the database.

[0503] 4. Processing when an incoming call arrives

[0504] When the device receives an incoming call, it records the caller's voice in real time as soon as the call begins, and the recorded voice data is sent to the server.

[0505] 5. Real-time analysis and matching of voice data

[0506] The server analyzes the received voice data in real time, extracts voice feature vectors, and then compares the extracted feature vectors with existing data in the database to determine whether they match.

[0507] 6. Sending and displaying the results

[0508] The server sends the result of the judgment to the communication terminal, which then displays the result to the user. If the caller is a registered person, it will display "From Mr. / Ms. XX," and if not, it will display "Unregistered." This allows the user to easily check whether the caller is trustworthy.

[0509] Specific examples

[0510] When a user registers his son's voice

[0511] The user uses the recording function of the communication device to have their son say, "Hello, I'm your son." Once the recording is complete, the voice data is sent to the server, which performs voice analysis, extracts the voice feature vector, and registers it in a database.

[0512] When a call comes in

[0513] The device receives an incoming call and displays "Incoming call." When the call begins, the caller's voice is recorded in real time and sent to the server. The server analyzes the voice data in real time and compares it with the son's voice feature vector in the database. As a result, the message "Call from your son" is displayed.

[0514] Prompt Sentence Examples

[0515] "We will record and register your mother's voice. We will start recording, so please say, 'Hello, I'm your mother.'"

[0516] "You have an incoming call. We are verifying that the caller is a registered user."

[0517] "(For example, my son) I'll record my son's voice and say, 'Hi, I'm your son,' and we'll save it."

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

[0519] Step 1: Record your audio

[0520] A user uses the voice recording function of a communication terminal to record the voices of family members or acquaintances.

[0521] Input: Voice of family or friends

[0522] Specific behavior:

[0523] 1. Select "Audio Recording" from the menu on your communication device.

[0524] 2. Press the start recording button and have the subject speak.

[0525] 3. Press the Stop Recording button to save the recording.

[0526] Output: Recorded audio data

[0527] Step 2: Sending audio data

[0528] The terminal transmits the recorded voice data to an information processing device (server).

[0529] Input: Recorded audio data

[0530] Specific behavior:

[0531] 1. Select the saved audio data.

[0532] 2. Send the data to the server using your internet connection.

[0533] Output: Audio data sent to the server

[0534] Step 3: Analyze and register audio data

[0535] The server receives the audio data and analyzes it.

[0536] Input: Transmitted audio data

[0537] Specific behavior:

[0538] 1. The server analyzes the voice data using voice analysis software (e.g., OpenAI's Whisper or Google's Speech-to-Text).

[0539] 2. Extract the voice feature vector and register it in a database along with the subject's name and information.

[0540] Output: Audio feature vectors and correspondence information registered in the database

[0541] Step 4: Receiving a call

[0542] The device receives an incoming call.

[0543] Input: Incoming call signal from the caller

[0544] Specific behavior:

[0545] 1. The call will ring and the display will show "Incoming call."

[0546] 2. The user receives the call and starts the conversation.

[0547] Output: Call Start

[0548] Step 5: Record the caller's voice

[0549] The device records the caller's voice in real time.

[0550] Input: Caller's voice

[0551] Specific behavior:

[0552] 1. Start recording as soon as the call begins.

[0553] 2. Prepare to send the recording data to the server.

[0554] Output: Recorded caller voice data

[0555] Step 6: Sending audio data

[0556] The terminal transmits the recorded voice data of the caller to the server.

[0557] Input: Recorded caller voice data

[0558] Specific behavior:

[0559] 1. Send real-time audio data to a server via an internet connection.

[0560] Output: Caller's voice data sent to the server

[0561] Step 7: Analyze and match the audio data

[0562] The server analyzes the caller's voice data received and compares it with existing data in the database.

[0563] Input: Caller's voice data sent to the server

[0564] Specific behavior:

[0565] 1. The server uses speech analysis software to extract the caller's voice feature vector.

[0566] 2. Compare the extracted speech feature vector with existing speech feature vectors in the database.

[0567] 3. Determine the matching result.

[0568] Output: Matching result (determining whether the caller is a registered person)

[0569] Step 8: Receive and display the results

[0570] The terminal displays the judgment result received from the server.

[0571] Input: Matching result sent from the server

[0572] Specific behavior:

[0573] 1. The terminal displays the received judgment result.

[0574] 2. If the caller is a registered person, it will display "From Mr. / Ms. XX."

[0575] 3. If there is no match, the message "Unregistered" will be displayed to warn the user.

[0576] Output: Display of judgment result (registered person or unregistered person)

[0577] This series of processes allows the user to check in real time whether the caller is trustworthy.

[0578] (Application example 1)

[0579] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0580] In recent years, voice-based fraud has become more prevalent, requiring users to be vigilant against calls from unknown callers. However, conventional voice recognition systems are inadequate, particularly in their inability to display real-time results, making it difficult for users to make a quick decision. This has led to a growing demand for effective means to prevent fraud. Furthermore, there is a need for a system that can instantly display a warning on smart devices, such as smart glasses and head-mounted displays, that are linked to communication terminals.

[0581] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0582] In this invention, the server includes a means for recording voice, a means for transmitting the recorded voice to an information processing device, a means for the information processing device to analyze the voice data and extract voice characteristics, a means for determining whether the voice data matches registered data based on the voice characteristics, a means for transmitting the determination result to a communication terminal, a means for the communication terminal to display the determination result, and a means for linking with a smart device to display the determination result in real time. This allows users to immediately be alert to calls from unknown callers and prevent fraud. Furthermore, by using smart glasses or a head-mounted display, a warning is displayed directly in the user's field of vision, allowing for a quick and effective response.

[0583] "Audio recording means" means equipment or software for recording audio signals as digital data.

[0584] The "means for transmitting the recorded voice to the information processing device" is a function or protocol for transferring the recorded voice data to the information processing device through a communication line.

[0585] "Means for an information processing device to analyze voice data and extract voice features" refers to a process in which an information processing device analyzes input voice data using an algorithm and extracts features.

[0586] The "means for determining whether or not the data matches the registered data based on the voice characteristics" is a process for comparing pre-registered voice characteristic data with newly extracted voice characteristic data and determining whether or not there is a match.

[0587] The "means for transmitting the determination result to the communication terminal" is a communication means or protocol for transmitting the determination result to the communication terminal in real time.

[0588] The "means for the communication terminal to display the determination result" is a function that visually presents the determination result to the user using the display or interface of the communication terminal.

[0589] "Means for linking with a smart device and displaying the judgment results in real time" refers to a system that links with a smart device such as smart glasses or a head-mounted display and presents the judgment results to the user in real time.

[0590] "Smart glasses" are glasses-type devices that run specific applications and provide augmented reality and notification information through a display.

[0591] A "head-mounted display" is a device worn on the head and positioned within the user's field of vision to provide visual information to the user.

[0592] This invention is a voice recognition system that allows users to receive calls with peace of mind. In order to prevent damage from special fraud, the system analyzes the caller's voice in real time and compares it with pre-registered voice data to determine whether the caller is a registered person.

[0593] The system of the present invention includes the following hardware and software.

[0594] Hardware: Smart glasses (e.g., Google Glass, Vuzix Blade), smartphones

[0595] Software: Audio processing libraries (e.g., pyaudio, librosa), dedicated APIs for smart glasses (e.g., Google Glass API), local databases (e.g., SQLite)

[0596] When a user uses the system for the first time, they record the voice data they want to register in advance using a communication terminal and send it to the server. The server analyzes the received voice data and extracts a voice feature vector. This extracted voice feature vector is registered in the database. Through this process, the voices of family members and acquaintances are registered in the system.

[0597] When a call comes in, the communication terminal records the caller's voice in real time and sends the recorded voice data to the server. The server analyzes the voice data and extracts a voice feature vector. It then compares it with voice feature vectors already registered in a database and sends the result to the communication terminal. The communication terminal displays the result on smart glasses or a head-mounted display. This allows the user to confirm whether the caller is a registered person before starting the call.

[0598] As a specific example, consider the case where a user registers the voice of their son. The user uses the recording function of their communication device to record their son's voice and sends the voice data to the server. The server analyzes the voice data and registers the son's voice feature vector in a database. If the user then receives a call while cooking and the caller is their son, the communication device records the caller's voice in real time and sends it to the server. The server analyzes the voice feature vector, and if it matches the voice feature vector of the son that has already been registered, the smart glasses display will say, "Call from your son." On the other hand, if the caller is an unregistered person, the message will say, "This call is from your son. Please be careful."

[0599] This system allows users to immediately become alert to calls from unknown callers, preventing fraud before they occur. In addition, by using smart glasses or a head-mounted display, a warning can be displayed directly in the user's field of vision, enabling a quick and effective response.

[0600] An example of a prompt is as follows:

[0601] "This is a demo scene of a voice recognition security guard. It registers the voices of family members, identifies suspicious callers, and displays a warning. It compares them in real time with a recording of the son's voice. This system significantly reduces the risk of special fraud."

[0602] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0603] Step 1:

[0604] A user uses the recording function of a communication terminal to record the voices of family members or acquaintances.

[0605] Input: Audio signal (voice of user's family or acquaintance)

[0606] Processing: Use the recording function of the communication device to record the audio as digital data.

[0607] Output: Recorded audio data (digital format)

[0608] Step 2:

[0609] The communication terminal transmits the recorded voice data to the server.

[0610] Input: Recorded audio data

[0611] Processing: Upload the audio data to the server via a communication line.

[0612] Output: Audio data sent to the server

[0613] Step 3:

[0614] The server analyzes the received voice data and extracts a voice feature vector.

[0615] Input: Received audio data

[0616] Processing: The server uses an audio processing library (e.g., librosa) to analyze the audio data and generate an audio feature vector using techniques such as FFT (Fast Fourier Transform).

[0617] Output: Audio feature vector

[0618] Step 4:

[0619] The server registers the voice feature vector in a database.

[0620] Input: Audio feature vector, corresponding person name and information

[0621] Processing: The server uses a local database (e.g., SQLite) to store the voice feature vectors and person information in the database.

[0622] Output: Trained speech feature data

[0623] Step 5:

[0624] When a call is received, the communication terminal records the caller's voice in real time.

[0625] Input: Incoming voice signal (caller's voice)

[0626] Processing: Record the audio as digital data in real time using the recording function of the communication device.

[0627] Output: Recorded real-time audio data

[0628] Step 6:

[0629] The communication terminal transmits the recorded real-time voice data to the server.

[0630] Input: Recorded real-time audio data

[0631] Processing: Upload the audio data to the server via a communication line.

[0632] Output: Real-time audio data sent to the server

[0633] Step 7:

[0634] The server analyzes the real-time voice data and extracts voice feature vectors.

[0635] Input: Received real-time audio data

[0636] Processing: The server uses a speech processing library to analyze the speech data and generate a speech feature vector.

[0637] Output: Real-time audio feature vectors

[0638] Step 8:

[0639] The server matches the real-time voice feature vector with the registered voice feature vectors in the database.

[0640] Input: Real-time speech feature vector, registered speech feature vector in database

[0641] Processing: The server calculates the cosine similarity or Euclidean distance to assess the degree of match between the two vectors.

[0642] Output: Matching result (match / no match)

[0643] Step 9:

[0644] The server transmits the determination result to the communication terminal.

[0645] Input: Judgment result

[0646] Processing: The judgment result is sent to the communication terminal via the communication line.

[0647] Output: The judgment result sent to the communication terminal

[0648] Step 10:

[0649] The communication terminal displays the judgment results on smart glasses or a head-mounted display.

[0650] Input: Received verdict

[0651] Processing: Using the smart glasses' dedicated API, the judgment result is displayed on the screen.

[0652] Output: The result displayed on the screen ("This is from ____" or "This is an unregistered user")

[0653] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0654] The present invention is a system that aims to prevent more sophisticated special frauds by combining a voice recognition system with an emotion engine, and by analyzing the caller's voice when a communication terminal receives an incoming call, and also analyzing the user's emotional state.

[0655] First, we will explain the process by which a user registers the voices of family members and acquaintances. The user uses the voice recording function of the communication device to record the voice of each person. When recording is complete, the communication device sends the recorded voice data to the server. The server analyzes the received voice data and extracts a feature vector of the voice. The name and information of the person corresponding to this voice feature vector are registered in a database.

[0656] Next, we will explain the voice recognition and emotion analysis process when an incoming call is received. When a communication terminal receives an incoming call, the user starts the call. As soon as the call starts, the terminal records the caller's voice in real time. This recorded voice data is sent from the communication terminal to a server. The server analyzes the received voice data and extracts a feature vector of the voice. The extracted feature vector is compared with voice feature vectors already registered in a database, and the result of this comparison is determined by the server. The result is sent to the communication terminal and displayed on a display.

[0657] In addition, the communication terminal is equipped with an emotion engine that analyzes the user's voice during a call and recognizes the user's emotional state in real time. The recognized emotion information is immediately sent to the server, which then generates a warning message based on that information and sends it to the communication terminal.

[0658] As a specific example, consider the case where a user registers their mother's voice. The user uses the recording function of their communication device to record their mother's voice and sends the voice data to the server. The server analyzes the voice data and registers the mother's voice feature vector and name in a database. If the user's communication device then receives a call from the mother, the caller's voice is recorded in real time and sent to the server. The server analyzes the voice feature vector and, if it matches the mother's voice feature vector in the database, the communication device displays the message "It's from your mother." At the same time, the emotion engine analyzes the user's voice in real time, and if the user's emotions indicate anxiety or caution, a warning message such as "Caution required" is displayed on the communication device.

[0659] In this way, the system based on the present invention can provide more accurate measures to prevent special frauds by combining the results of analyzing the user's emotional state using an emotion engine in addition to caller recognition using voice feature vectors. It also provides an environment where users can receive calls with peace of mind, which is extremely effective in preventing fraud damage before it happens.

[0660] The processing flow will be explained below.

[0661] This invention is a system that combines a voice recognition system and an emotion engine, and when a communication terminal receives an incoming call, it analyzes the caller's voice and also analyzes the user's emotional state, thereby preventing damage from special fraud. The specific processing flow is explained below step by step.

[0662] User Voice Enrollment Process

[0663] Step 1:

[0664] The user starts the application on the communication terminal, selects the voice registration menu of the phone book, and then enters the name of the person to be registered.

[0665] Step 2:

[0666] The user records the voice of the person to be registered into the communication terminal, and when the recording is complete, the user presses the recording completion button.

[0667] Step 3:

[0668] The device sends the recorded audio data to a server via the Internet.

[0669] Step 4:

[0670] The server analyzes the received voice data and extracts a voice feature vector using a voice recognition algorithm.

[0671] Step 5:

[0672] The server registers the extracted voice feature vectors together with the names of the corresponding people in a database.

[0673] Incoming voice recognition and sentiment analysis process

[0674] Step 1:

[0675] When the communication terminal receives an incoming call, the user starts the call, and the terminal records the caller's voice in real time.

[0676] Step 2:

[0677] The device sends the recorded voice data of the caller to the server, where it is immediately analyzed.

[0678] Step 3:

[0679] The server analyzes the received voice data and extracts the caller's voice feature vector, which is then compared with the registered data in the database.

[0680] Step 4:

[0681] The server judges the result of matching the voice feature vector. If the caller's voice matches the registered data, the judgement result is recorded as "Registered Person." If not, it is recorded as "Unregistered Person."

[0682] Step 5:

[0683] The server transmits the result of the determination to the communication terminal, which then determines the content to display based on the received result of the determination.

[0684] Step 6:

[0685] The device displays the result on the screen. If the caller's voice matches the registered voice, the name of the registered person is displayed. If they do not match, the device displays "Unregistered."

[0686] Emotion recognition process by emotion engine

[0687] Step 1:

[0688] During a call, the device records the user's voice in real time and sends it to the emotion engine.

[0689] Step 2:

[0690] The emotion engine analyzes the received user voice data and extracts the user's emotion feature vector using emotion recognition algorithms.

[0691] Step 3:

[0692] The emotion engine determines the user's emotional state, such as anxiety, alertness, or calmness, based on the extracted emotion feature vector.

[0693] Step 4:

[0694] The emotion engine sends the judgment result to the server, which generates a warning message as needed based on the received emotion information.

[0695] Step 5:

[0696] The server transmits the generated warning message to the communication terminal, and the terminal displays the received warning message on its display.

[0697] This allows users to not only confirm whether the caller is a registered person, but also receive warning messages tailored to their own emotional state, making it possible to prevent damage from special frauds.

[0698] Example 2

[0699] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0700] Conventional voice recognition systems only analyze and recognize the caller's voice, so they cannot completely prevent risks such as special fraud. Also, because they do not take into account the user's emotional state, they have the problem of not being able to issue appropriate warnings even if the user is feeling anxious or cautious.

[0701] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0702] In this invention, the server includes means for analyzing voice data and extracting voice characteristics, means for determining whether the voice characteristics match registered data, and means for generating a warning message based on emotion data and transmitting it to the communication terminal, thereby enabling the user's emotional state to be analyzed in real time and a warning message to be displayed as necessary.

[0703] The "means for recording voice" refers to a function or device that allows a user to record voice using a communication terminal.

[0704] The "means for transmitting recorded voice to the information processing device" refers to a function or device for transmitting voice data recorded by a communication terminal to the information processing device via a network.

[0705] The "means for an information processing device to analyze voice data and extract voice features" refers to software or algorithms that allow the information processing device to analyze received voice data and generate voice feature vectors.

[0706] "Means for determining whether the voice characteristics match the registered data" refers to a function or software for comparing the analyzed voice characteristics with the registered data in the database and determining whether they match.

[0707] The "means for transmitting the determination result to the communication terminal" refers to a communication function or protocol for transmitting the determination result from the information processing device to the communication terminal.

[0708] The "means for the communication terminal to display the determination result" refers to a display or display function for visually presenting the determination result received by the communication terminal to the user.

[0709] The "means for the communication terminal to analyze the emotional state of the user" refers to an emotion engine or software that allows the communication terminal to analyze the user's voice during a call and recognize the user's emotional state in real time.

[0710] The "means for transmitting analyzed emotion data to the information processing device" refers to a communication function or protocol that allows the communication terminal to transmit analyzed user emotion data to the information processing device.

[0711] "Means for an information processing device to generate an alert message based on emotion data and transmit it to a communication terminal" refers to a function or software that allows an information processing device to create an alert message based on emotion data received and transmit it to a communication terminal.

[0712] This invention combines a voice recognition system with an emotion engine, and aims to analyze the caller's voice when a communication terminal receives an incoming call, and also analyze the user's emotional state. This system will enable more advanced prevention of special frauds.

[0713] Hardware and software used

[0714] 1. Communication terminal:

[0715] Voice recording function: A microphone and recording app to record the user's voice.

[0716] Display: A screen for displaying the results and warning messages.

[0717] Emotion engine: Software for analyzing user emotions.

[0718] 2. Server (information processing device):

[0719] Voice recognition software: For example, Google Voice Recognition API.

[0720] Database management system: for example, MySQL.

[0721] Sentiment analysis algorithm: A model for analyzing the user's emotional state.

[0722] Process Overview

[0723] 1. Audio Recording and Transmission:

[0724] Users use their communication terminals to record the voices of their family members and friends, and the recorded voice data is sent from the communication terminals to a server.

[0725] 2. Analysis of audio data:

[0726] The server analyzes the received voice data and extracts the voice feature vector. It uses voice recognition software to analyze the voice waveform and generate the feature vector.

[0727] 3. Register feature vectors and data:

[0728] The extracted feature vectors are registered in a database along with the names and other information of the corresponding people. This registered data is managed using a database management system.

[0729] 4. Incoming call voice recognition and emotion analysis:

[0730] When a communication device receives an incoming call, it records the caller's voice in real time and sends it to the server, which analyzes the voice data, extracts feature vectors, and compares them with the registered data in a database.

[0731] At the same time, the emotion engine installed in the communication terminal analyzes the user's voice in real time and recognizes their emotional state.

[0732] 5. Sending and viewing results:

[0733] The server sends the results of the matching and emotion analysis to the communication terminal, which then displays the caller's name and a warning message.

[0734] Specific examples

[0735] Consider the case where a user registers their mother's voice. The user uses the recording function of the communication terminal to record their mother's voice and sends the voice data to the server. The server analyzes the voice data and registers the mother's voice feature vector and name in the database.

[0736] If the user then receives a call on their communication device and the caller is their mother, the caller's voice is recorded in real time and sent to the server. The server analyzes the voice feature vector, and if it matches the mother's voice feature vector in the database, the communication device displays the message "Call from mother." At the same time, the emotion engine analyzes the user's voice in real time, and if the user is expressing anxiety or caution, a warning message such as "Caution required" is displayed on the communication device.

[0737] Prompt Sentence Examples

[0738] "Please tell me the procedure for a system that registers the mother's voice, recognizes the mother's voice in real time when a call comes in, analyzes the user's emotional state, and displays a warning message."

[0739] In this way, the present invention provides a system that combines voice recognition and emotion analysis, thereby providing an environment in which telephones can be used more safely and reducing the risk of special fraud and the like.

[0740] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0741] Step 1:

[0742] A user uses a communication terminal to record the voice of a family member or friend. The user presses the record button on the communication terminal, and then presses the save button after recording is complete. The input is the user's voice, and the voice data is saved in the communication terminal as output.

[0743] Step 2:

[0744] The communication terminal sends the recorded voice data to the server. Once the recording is complete, the voice data is sent to the server via the network. The input is the voice data stored in the communication terminal, and the voice data that is transferred to the server as output is generated. During this process, the progress of the communication is displayed on the terminal.

[0745] Step 3:

[0746] The server analyzes the voice data and extracts voice features. The server analyzes the received voice data using voice recognition software (e.g., Google Voice Recognition API) and generates a voice feature vector. The input is the voice data transferred to the server, and the output is the extracted voice feature vector.

[0747] Step 4:

[0748] The server registers the extracted voice feature vector in a database. The server saves the name and information of the person corresponding to the voice feature vector in a database (e.g., MySQL). The input is the voice feature vector and the corresponding information, and the output is newly registered data in the database.

[0749] Step 5:

[0750] The communication terminal records the caller's voice in real time. When a call comes in, the communication terminal records the caller's voice as soon as the call starts. The input is the caller's voice when the call comes in, and the real-time recorded voice data is generated as the output.

[0751] Step 6:

[0752] The communication terminal sends the recorded voice data to the server. The voice data recorded in real time is sent to the server using asynchronous communication technology (e.g., WebSocket). The input is the voice data recorded in real time, and the voice data transferred to the server is generated as the output.

[0753] Step 7:

[0754] The server analyzes the incoming voice data and extracts a voice feature vector. The server analyzes the received voice data and generates a voice feature vector. The input is the voice data transferred to the server, and the output is the extracted voice feature vector.

[0755] Step 8:

[0756] The server compares the voice feature vector with the registered data in the database and determines whether they match. The server compares the extracted voice feature vector with the registered data in the database and determines whether they match. The input is the voice feature vector and the registered data in the database, and the output is the result of the determination of whether they match.

[0757] Step 9:

[0758] The server transmits the determination result to the communication terminal. The server transmits the determination result to the communication terminal. The input is the determination result, and the determination result is generated and transferred to the communication terminal as output.

[0759] Step 10:

[0760] The communication terminal displays the judgment result and begins emotion analysis. The judgment result is displayed on the communication terminal's display, and at the same time the emotion engine begins analyzing the user's voice in real time. The input is the judgment result from the server and the user's voice during the call, and the output is the result displayed on the display and emotion analysis data.

[0761] Step 11:

[0762] The communication terminal analyzes the user's emotional state and sends the results to the server. The emotion engine then sends the analyzed emotional data to the server. The input is the user's voice and the analysis results by the emotion engine, and the output is the emotional data sent to the server.

[0763] Step 12:

[0764] The server generates a warning message based on the emotion data and sends it to the communication terminal. The server generates a warning message based on the received emotion data and sends it to the communication terminal. The input is emotion data, and the warning message that is transferred to the communication terminal is generated as the output.

[0765] Step 13:

[0766] The communication terminal displays the warning message. The communication terminal displays the warning message received from the server on the display. The input is the warning message from the server, and the warning message displayed on the display is generated as the output.

[0767] (Application example 2)

[0768] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0769] Currently, fraud methods are becoming more sophisticated, and special frauds over the phone are on the rise. Many of these methods target the elderly, and the resulting losses are particularly significant, posing a problem. Conventional voice recognition systems primarily focus on identifying callers, without taking into account the user's emotional state. As a result, even if the caller is registered, fraudulent calls may not be detected. Therefore, there is a need for a system that can prevent fraud and risky calls by analyzing the user's emotional state.

[0770] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0771] In this invention, the server includes means for analyzing the emotional state of the user from the voice during the call, means for generating a warning message based on the emotion analysis result and transmitting it to the communication terminal, means for recording the voice, means for transmitting the recorded voice to the information processing device, means for the information processing device to analyze the voice data and extract voice characteristics, means for determining whether the voice data matches registered data based on the voice characteristics, means for transmitting the determination result to the communication terminal, and means for the communication terminal to display the determination result. This enables more advanced prevention of special frauds by taking into account both the caller's recognition and the user's emotional state.

[0772] "Audio recording means" means a device or software for capturing and recording audio data during a call.

[0773] The "means for transmitting the recorded voice to the information processing device" is a function or device that transfers the recorded voice data to a server or processing device via communication or a network.

[0774] The "means for an information processing device to analyze voice data and extract voice features" refers to an algorithm or device for identifying and extracting specific voice patterns or features from received voice data.

[0775] "Means for determining whether the voice characteristics match registered data" refers to a function or device for comparing the extracted voice characteristics with pre-registered voice data and determining whether there is a match.

[0776] The "means for transmitting the determination result to the communication terminal" is a function or device that transmits the voice feature matching result to the terminal via communication or a network.

[0777] The "means for the communication terminal to display the determination result" refers to a display device or software for visually notifying the user of the received determination result.

[0778] The "means for analyzing the user's emotional state from the voice during a call" refers to an algorithm or device for determining the user's emotions in real time from the voice data acquired during a call.

[0779] The "means for generating a warning message based on the emotion analysis result and transmitting it to the communication terminal" refers to a function or device for creating a warning message based on the recognized emotional state and transmitting it to the communication terminal.

[0780] The present invention provides a security system that prevents special frauds by analyzing the caller's voice when a communication terminal receives an incoming call and also analyzing the user's emotional state. Specific embodiments for carrying out the present invention will be described below.

[0781] System Configuration

[0782] The system includes the following major hardware and software components:

[0783] Communication device: Smartphone with voice recording and display functions

[0784] Server: Voice data analyzer, database, emotion engine

[0785] Software: Speech recognition library (e.g., speech_recognition), emotion recognition library (e.g., emotion_recognition), voice feature recognition library (e.g., voice_recognition)

[0786] Program processing and natural language explanation

[0787] 1. Voice registration:

[0788] The user uses the voice recording function of the communication device to record the voices of family members or acquaintances. After the recording is complete, the communication device sends the recorded voice data to the server.

[0789] The server analyzes the received voice data and extracts voice feature vectors. The extracted voice feature vectors and the corresponding person's name and information are registered in a database.

[0790] 2. Incoming call voice recognition and sentiment analysis:

[0791] When a communication terminal receives an incoming call, the user starts the call. The terminal records the caller's voice in real time as soon as the call starts. This recorded voice data is sent from the communication terminal to a server.

[0792] The server analyzes the received voice data and extracts the voice feature vector. The extracted feature vector is compared with the voice feature vectors already registered in the database, and the result is judged and sent to the communication terminal. The judgment result is displayed on the communication terminal.

[0793] During a call, the emotion engine installed in the communication terminal analyzes the user's voice and recognizes the user's emotional state in real time. The recognized emotional information is immediately sent to the server, which then generates a warning message based on that information and sends it to the communication terminal.

[0794] Real-world examples

[0795] For example, consider the case where a user has registered their mother's voice in advance. The user records the mother's voice using a communication terminal and sends the voice data to the server. The server analyzes the voice data and registers the mother's voice feature vector and name in a database.

[0796] Next, when the user's communication device receives a call and recognizes that the caller is the mother, the caller's voice is recorded in real time and sent to the server. The server analyzes the voice feature vector, and if it matches the voice feature vector of the mother in the database, the communication device displays the message "Call from mother." At the same time, the emotion engine analyzes the user's voice in real time, and if the user's emotion indicates anxiety or caution, a warning message such as "Caution required" is displayed.

[0797] Prompt Sentence Examples

[0798] Also, when using a generative AI model to generate an appropriate warning message, enter the following prompt:

[0799] The user's emotional state is anxious:

[0800] "We need to be extra careful as the user has expressed anxiety. Do you want to continue communicating?"

[0801] Called party is unknown:

[0802] "This caller is not registered and may be a scam. Would you like to continue the call?"

[0803] The present invention makes it possible to provide an advanced security system that analyzes both the caller's identity and the user's emotional state in real time, preventing fraud and dangerous calls before they occur.

[0804] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0805] Step 1:

[0806] Voice registration

[0807] Input: Audio data obtained using the user's recording function

[0808] How it works: A user uses the voice recording function of their communication device to record the voice of a family member or friend. After pressing the record button and finishing recording, the communication device automatically sends the voice data to the server.

[0809] Data processing: The server analyzes the received voice data and extracts voice feature vectors.

[0810] Output: The extracted speech feature vectors are registered in a database.

[0811] Step 2:

[0812] Voice recognition for incoming calls

[0813] Input: Caller's voice data recorded when receiving a call

[0814] Operation: When a communication terminal receives an incoming call, the user starts the call. At the same time, the terminal records the caller's voice in real time and sends this voice data to the server.

[0815] Data processing: The server analyzes the received voice data and extracts voice feature vectors. The extracted feature vectors are compared with previously registered voice feature vectors.

[0816] Output: The matching result is sent to the communication terminal and displayed on the screen.

[0817] Step 3:

[0818] Emotional state analysis

[0819] Input: Voice data of the user during the call

[0820] How it works: The emotion engine in the communication device analyzes the user's voice in real time during a call.

[0821] Data Processing: The emotion engine identifies the user's emotional state (e.g., anxiety, alertness, etc.).

[0822] Output: The analysis results are sent to the server.

[0823] Step 4:

[0824] Warning message generation and notification

[0825] Input: Emotion analysis results and voice feature matching results

[0826] Operation: The server generates a warning message as needed based on the emotion analysis results and voice feature matching results.

[0827] Data processing: The generated message is sent to the communication terminal.

[0828] Output: A warning message is displayed on the communication terminal.

[0829] Specific operation example

[0830] For example, when a user whose mother's voice has already been registered receives a call from her mother, the process is as follows.

[0831] In step 1, the mother's voice is pre-registered.

[0832] In step 2, the moment a call is received, the caller's voice is recorded in real time and sent to the server. The server then performs a voice analysis and compares it with pre-registered voice characteristics. If the caller is confirmed to be the mother, the message "This is from your mother" is displayed on the communication device.

[0833] In step 3, the user's voice during the call is analyzed in real time, and if the user shows any anxiety or alertness, their emotional state is sent to the server.

[0834] If the emotional state is recognized as anxious in step 4, the server generates a warning message such as "Caution required" which is displayed on the communication terminal.

[0835] Examples of prompts include:

[0836] The user's emotional state is anxious:

[0837] "We need to be extra careful as the user has expressed anxiety. Do you want to continue communicating?"

[0838] Called party is unknown:

[0839] "This caller is not registered and may be a scam. Would you like to continue the call?"

[0840] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0841] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0842] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0843] [Third embodiment]

[0844] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0845] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0846] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0847] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0848] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0849] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0850] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0851] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0852] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0853] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0854] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0855] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0856] The present invention is a speech recognition system that utilizes a communication terminal such as a mobile phone or a landline phone and an information processing device that processes voice data. This system is designed as an effective means of preventing damage from special frauds.

[0857] First, we will explain the process by which a user registers the voices of family members and acquaintances. The user uses the voice recording function of the communication terminal to record the voice of each person. When recording is complete, the communication terminal sends the recorded voice data to an information processing device (server). The server analyzes the received voice data and extracts a feature vector of the voice. The name and information of the person corresponding to this voice feature vector are registered in a database.

[0858] Next, we will explain the voice recognition process when an incoming call is received. When a communication terminal receives an incoming call, it records the caller's voice in real time the moment the call starts. This recorded voice data is sent from the communication terminal to a server. The server analyzes the received voice data and extracts a feature vector of the voice. The extracted feature vector is compared with voice feature vectors already registered in a database. The result of this comparison is judged by the server and sent to the communication terminal.

[0859] Based on the results of the determination received from the server, the communication device displays on its display whether the caller is a registered person. If there is a match, the communication device displays "From Mr. / Ms. XX," indicating that the caller is a family member or acquaintance. If there is no match, the communication device displays "Unregistered," warning the user to be cautious.

[0860] As a specific example, consider the case where a user registers the voice of their son. The user uses the recording function of their communication device to record their son's voice and sends the voice data to the server. The server analyzes the voice data and registers the son's voice feature vector and name in a database. After that, when the user receives a call on their communication device and the caller is their son, the caller's voice is recorded in real time and sent to the server. The server analyzes the voice feature vector and, if it matches the son's voice feature vector in the database, the communication device displays the message "Call from your son." On the other hand, if the caller is not registered, the communication device displays the message "Unregistered caller."

[0861] In this way, the system based on the present invention allows users to receive calls with peace of mind and is effective in preventing damage from special frauds.

[0862] The processing flow will be explained below.

[0863] Step 1:

[0864] The user starts the application on the communication terminal, selects the voice registration menu of the phone book, and then enters the name of the person to be registered.

[0865] Step 2:

[0866] The user records the voice of the person to be registered into the communication terminal. When the user presses the recording completion button, the recorded voice data is saved in the communication terminal.

[0867] Step 3:

[0868] The device then transmits the stored audio data to a server, typically via the Internet.

[0869] Step 4:

[0870] The server analyzes the received voice data and extracts a voice feature vector using a voice recognition algorithm.

[0871] Step 5:

[0872] The server registers the extracted voice feature vectors together with the corresponding person's name in a database, which stores the voice feature vectors and person information.

[0873] Step 6:

[0874] When the communication terminal receives an incoming call, the user starts the call, and the terminal records the caller's voice in real time.

[0875] Step 7:

[0876] The device sends the recorded voice data of the caller to the server, where it is immediately analyzed.

[0877] Step 8:

[0878] The server analyzes the received voice data and extracts the caller's voice feature vector, which is then compared with the registered data in the database.

[0879] Step 9:

[0880] The server judges the result of matching the voice feature vector. If the caller's voice matches the registered data, the judgement result is recorded as "Registered Person." If not, it is recorded as "Unregistered Person."

[0881] Step 10:

[0882] The server transmits the result of the determination to the communication terminal, which then determines the content to display based on the received result of the determination.

[0883] Step 11:

[0884] The device displays the result on the screen. If the caller's voice matches the registered voice, the name of the registered person is displayed. If they do not match, the device displays "Unregistered."

[0885] Step 12:

[0886] The user checks the display and determines whether the caller is a family member or acquaintance. If the message "Unregistered" appears, the user becomes more vigilant.

[0887] Example 1

[0888] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0889] The purpose of this invention is to provide a communication system that protects users from crimes such as special fraud and allows them to make calls with peace of mind. In particular, there is a need to prevent fraudulent acts by recognizing the voice of the other party in real time and comparing it with pre-registered voices. This system provides a method for registering and recognizing the voices of family members and acquaintances, and comprehensively identifying suspicious callers.

[0890] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0891] In this invention, the server includes a means for analyzing voice data and extracting a voice feature vector, a means for determining whether the voice feature vector matches registered data in a database, and a means for transmitting the determination result to the communication terminal. This allows users to analyze the caller's voice in real time when a call starts and compare it with pre-registered voices. Furthermore, the determination result is transmitted to the communication terminal and displayed on the display, allowing users to immediately confirm whether the caller is a registered person. This prevents crimes such as special fraud and allows users to make calls with peace of mind.

[0892] A "communication terminal" is a device that allows a user to record voice and transmit the voice data to an information processing device.

[0893] "Recording means" refers to the function of recording audio within the communication terminal.

[0894] The "information processing device" is a device that analyzes transmitted voice data and extracts voice feature vectors.

[0895] "Audio data" refers to a recording of audio that is recorded by a communication terminal and transmitted to an information processing device.

[0896] A "voice feature vector" is a collection of data that is extracted from voice data and that quantifies individual voice features.

[0897] A "database" is a digital data structure for storing speech feature vectors and related information.

[0898] "Matching means" refers to a function that compares the extracted voice feature vector with existing data in the database and determines whether they match.

[0899] The "determination result" is information indicating the result of matching or mismatching obtained by matching the speech feature vectors.

[0900] The "display means" is a function that enables the communication terminal to visually convey the determination result to the user.

[0901] This invention is a voice recognition system using a communication terminal and an information processing device (server). The main purpose of this system is to prevent damage from special frauds, specifically by recognizing the caller's voice in real time and comparing it with pre-registered voices to detect suspicious callers.

[0902] Hardware and software used

[0903] The communication terminals used are mobile phones and landlines, which have the functions of recording voice and transmitting data via the Internet.

[0904] The server is an information processing device required to analyze voice data. Voice analysis software, such as OpenAI's Whisper or Google's Speech-to-Text, is installed on the server.

[0905] Processing flow

[0906] 1. Recording Audio

[0907] The user uses the recording function of the communication device to record the voices of family members and acquaintances, and the recorded voice data is saved in the communication device.

[0908] 2. Sending audio data

[0909] The terminal transmits the recorded voice data to an information processing device (server) via the Internet.

[0910] 3. Analysis and registration of audio data

[0911] The server analyzes the received voice data and extracts a voice feature vector. The extracted voice feature vector is then registered in a database. This saves the registrant's voice data in the database.

[0912] 4. Processing when an incoming call arrives

[0913] When the device receives an incoming call, it records the caller's voice in real time as soon as the call begins, and the recorded voice data is sent to the server.

[0914] 5. Real-time analysis and matching of voice data

[0915] The server analyzes the received voice data in real time, extracts voice feature vectors, and then compares the extracted feature vectors with existing data in the database to determine whether they match.

[0916] 6. Sending and displaying the results

[0917] The server sends the result of the judgment to the communication terminal, which then displays the result to the user. If the caller is a registered person, it will display "From Mr. / Ms. XX," and if not, it will display "Unregistered." This allows the user to easily check whether the caller is trustworthy.

[0918] Specific examples

[0919] When a user registers his son's voice

[0920] The user uses the recording function of the communication device to have their son say, "Hello, I'm your son." Once the recording is complete, the voice data is sent to the server, which performs voice analysis, extracts the voice feature vector, and registers it in a database.

[0921] When a call comes in

[0922] The device receives an incoming call and displays "Incoming call." When the call begins, the caller's voice is recorded in real time and sent to the server. The server analyzes the voice data in real time and compares it with the son's voice feature vector in the database. As a result, the message "Call from your son" is displayed.

[0923] Prompt Sentence Examples

[0924] "We will record and register your mother's voice. We will start recording, so please say, 'Hello, I'm your mother.'"

[0925] "You have an incoming call. We are verifying that the caller is a registered user."

[0926] "(For example, my son) I'll record my son's voice and say, 'Hi, I'm your son,' and we'll save it."

[0927] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0928] Step 1: Record your audio

[0929] A user uses the voice recording function of a communication terminal to record the voices of family members or acquaintances.

[0930] Input: Voice of family or friends

[0931] Specific behavior:

[0932] 1. Select "Audio Recording" from the menu on your communication device.

[0933] 2. Press the start recording button and have the subject speak.

[0934] 3. Press the Stop Recording button to save the recording.

[0935] Output: Recorded audio data

[0936] Step 2: Sending audio data

[0937] The terminal transmits the recorded voice data to an information processing device (server).

[0938] Input: Recorded audio data

[0939] Specific behavior:

[0940] 1. Select the saved audio data.

[0941] 2. Send the data to the server using your internet connection.

[0942] Output: Audio data sent to the server

[0943] Step 3: Analyze and register audio data

[0944] The server receives the audio data and analyzes it.

[0945] Input: Transmitted audio data

[0946] Specific behavior:

[0947] 1. The server analyzes the voice data using voice analysis software (e.g., OpenAI's Whisper or Google's Speech-to-Text).

[0948] 2. Extract the voice feature vector and register it in a database along with the subject's name and information.

[0949] Output: Audio feature vectors and correspondence information registered in the database

[0950] Step 4: Receiving a call

[0951] The device receives an incoming call.

[0952] Input: Incoming call signal from the caller

[0953] Specific behavior:

[0954] 1. The call will ring and the display will show "Incoming call."

[0955] 2. The user receives the call and starts the conversation.

[0956] Output: Call Start

[0957] Step 5: Record the caller's voice

[0958] The device records the caller's voice in real time.

[0959] Input: Caller's voice

[0960] Specific behavior:

[0961] 1. Start recording as soon as the call begins.

[0962] 2. Prepare to send the recording data to the server.

[0963] Output: Recorded caller voice data

[0964] Step 6: Sending audio data

[0965] The terminal transmits the recorded voice data of the caller to the server.

[0966] Input: Recorded caller voice data

[0967] Specific behavior:

[0968] 1. Send real-time audio data to a server via an internet connection.

[0969] Output: Caller's voice data sent to the server

[0970] Step 7: Analyze and match the audio data

[0971] The server analyzes the caller's voice data received and compares it with existing data in the database.

[0972] Input: Caller's voice data sent to the server

[0973] Specific behavior:

[0974] 1. The server uses speech analysis software to extract the caller's voice feature vector.

[0975] 2. Compare the extracted speech feature vector with existing speech feature vectors in the database.

[0976] 3. Determine the matching result.

[0977] Output: Matching result (determining whether the caller is a registered person)

[0978] Step 8: Receive and display the results

[0979] The terminal displays the judgment result received from the server.

[0980] Input: Matching result sent from the server

[0981] Specific behavior:

[0982] 1. The terminal displays the received judgment result.

[0983] 2. If the caller is a registered person, it will display "From Mr. / Ms. XX."

[0984] 3. If there is no match, the message "Unregistered" will be displayed to warn the user.

[0985] Output: Display of judgment result (registered person or unregistered person)

[0986] This series of processes allows the user to check in real time whether the caller is trustworthy.

[0987] (Application example 1)

[0988] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0989] In recent years, voice-based fraud has become more prevalent, requiring users to be vigilant against calls from unknown callers. However, conventional voice recognition systems are inadequate, particularly in their inability to display real-time results, making it difficult for users to make a quick decision. This has led to a growing demand for effective means to prevent fraud. Furthermore, there is a need for a system that can instantly display a warning on smart devices, such as smart glasses and head-mounted displays, that are linked to communication terminals.

[0990] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0991] In this invention, the server includes a means for recording voice, a means for transmitting the recorded voice to an information processing device, a means for the information processing device to analyze the voice data and extract voice characteristics, a means for determining whether the voice data matches registered data based on the voice characteristics, a means for transmitting the determination result to a communication terminal, a means for the communication terminal to display the determination result, and a means for linking with a smart device to display the determination result in real time. This allows users to immediately be alert to calls from unknown callers and prevent fraud. Furthermore, by using smart glasses or a head-mounted display, a warning is displayed directly in the user's field of vision, allowing for a quick and effective response.

[0992] "Audio recording means" means equipment or software for recording audio signals as digital data.

[0993] The "means for transmitting the recorded voice to the information processing device" is a function or protocol for transferring the recorded voice data to the information processing device through a communication line.

[0994] "Means for an information processing device to analyze voice data and extract voice features" refers to a process in which an information processing device analyzes input voice data using an algorithm and extracts features.

[0995] The "means for determining whether or not the data matches the registered data based on the voice characteristics" is a process for comparing pre-registered voice characteristic data with newly extracted voice characteristic data and determining whether or not there is a match.

[0996] The "means for transmitting the determination result to the communication terminal" is a communication means or protocol for transmitting the determination result to the communication terminal in real time.

[0997] The "means for the communication terminal to display the determination result" is a function that visually presents the determination result to the user using the display or interface of the communication terminal.

[0998] "Means for linking with a smart device and displaying the judgment results in real time" refers to a system that links with a smart device such as smart glasses or a head-mounted display and presents the judgment results to the user in real time.

[0999] "Smart glasses" are glasses-type devices that run specific applications and provide augmented reality and notification information through a display.

[1000] A "head-mounted display" is a device worn on the head and positioned within the user's field of vision to provide visual information to the user.

[1001] This invention is a voice recognition system that allows users to receive calls with peace of mind. In order to prevent damage from special fraud, the system analyzes the caller's voice in real time and compares it with pre-registered voice data to determine whether the caller is a registered person.

[1002] The system of the present invention includes the following hardware and software.

[1003] Hardware: Smart glasses (e.g., Google Glass, Vuzix Blade), smartphones

[1004] Software: Audio processing libraries (e.g., pyaudio, librosa), dedicated APIs for smart glasses (e.g., Google Glass API), local databases (e.g., SQLite)

[1005] When a user uses the system for the first time, they record the voice data they want to register in advance using a communication terminal and send it to the server. The server analyzes the received voice data and extracts a voice feature vector. This extracted voice feature vector is registered in the database. Through this process, the voices of family members and acquaintances are registered in the system.

[1006] When a call comes in, the communication terminal records the caller's voice in real time and sends the recorded voice data to the server. The server analyzes the voice data and extracts a voice feature vector. It then compares it with voice feature vectors already registered in a database and sends the result to the communication terminal. The communication terminal displays the result on smart glasses or a head-mounted display. This allows the user to confirm whether the caller is a registered person before starting the call.

[1007] As a specific example, consider the case where a user registers the voice of their son. The user uses the recording function of their communication device to record their son's voice and sends the voice data to the server. The server analyzes the voice data and registers the son's voice feature vector in a database. If the user then receives a call while cooking and the caller is their son, the communication device records the caller's voice in real time and sends it to the server. The server analyzes the voice feature vector, and if it matches the voice feature vector of the son that has already been registered, the smart glasses display will say, "Call from your son." On the other hand, if the caller is an unregistered person, the message will say, "This call is from your son. Please be careful."

[1008] This system allows users to immediately become alert to calls from unknown callers, preventing fraud before they occur. In addition, by using smart glasses or a head-mounted display, a warning can be displayed directly in the user's field of vision, enabling a quick and effective response.

[1009] An example of a prompt is as follows:

[1010] "This is a demo scene of a voice recognition security guard. It registers the voices of family members, identifies suspicious callers, and displays a warning. It compares them in real time with a recording of the son's voice. This system significantly reduces the risk of special fraud."

[1011] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1012] Step 1:

[1013] A user uses the recording function of a communication terminal to record the voices of family members or acquaintances.

[1014] Input: Audio signal (voice of user's family or acquaintance)

[1015] Processing: Use the recording function of the communication device to record the audio as digital data.

[1016] Output: Recorded audio data (digital format)

[1017] Step 2:

[1018] The communication terminal transmits the recorded voice data to the server.

[1019] Input: Recorded audio data

[1020] Processing: Upload the audio data to the server via a communication line.

[1021] Output: Audio data sent to the server

[1022] Step 3:

[1023] The server analyzes the received voice data and extracts a voice feature vector.

[1024] Input: Received audio data

[1025] Processing: The server uses an audio processing library (e.g., librosa) to analyze the audio data and generate an audio feature vector using techniques such as FFT (Fast Fourier Transform).

[1026] Output: Audio feature vector

[1027] Step 4:

[1028] The server registers the voice feature vector in a database.

[1029] Input: Audio feature vector, corresponding person name and information

[1030] Processing: The server uses a local database (e.g., SQLite) to store the voice feature vectors and person information in the database.

[1031] Output: Trained speech feature data

[1032] Step 5:

[1033] When a call is received, the communication terminal records the caller's voice in real time.

[1034] Input: Incoming voice signal (caller's voice)

[1035] Processing: Record the audio as digital data in real time using the recording function of the communication device.

[1036] Output: Recorded real-time audio data

[1037] Step 6:

[1038] The communication terminal transmits the recorded real-time voice data to the server.

[1039] Input: Recorded real-time audio data

[1040] Processing: Upload the audio data to the server via a communication line.

[1041] Output: Real-time audio data sent to the server

[1042] Step 7:

[1043] The server analyzes the real-time voice data and extracts voice feature vectors.

[1044] Input: Received real-time audio data

[1045] Processing: The server uses a speech processing library to analyze the speech data and generate a speech feature vector.

[1046] Output: Real-time audio feature vectors

[1047] Step 8:

[1048] The server matches the real-time voice feature vector with the registered voice feature vectors in the database.

[1049] Input: Real-time speech feature vector, registered speech feature vector in database

[1050] Processing: The server calculates the cosine similarity or Euclidean distance to assess the degree of match between the two vectors.

[1051] Output: Matching result (match / no match)

[1052] Step 9:

[1053] The server transmits the determination result to the communication terminal.

[1054] Input: Judgment result

[1055] Processing: The judgment result is sent to the communication terminal via the communication line.

[1056] Output: The judgment result sent to the communication terminal

[1057] Step 10:

[1058] The communication terminal displays the judgment results on smart glasses or a head-mounted display.

[1059] Input: Received verdict

[1060] Processing: Using the smart glasses' dedicated API, the judgment result is displayed on the screen.

[1061] Output: The result displayed on the screen ("This is from ____" or "This is an unregistered user")

[1062] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1063] The present invention is a system that aims to prevent more sophisticated special frauds by combining a voice recognition system with an emotion engine, and by analyzing the caller's voice when a communication terminal receives an incoming call, and also analyzing the user's emotional state.

[1064] First, we will explain the process by which a user registers the voices of family members and acquaintances. The user uses the voice recording function of the communication device to record the voice of each person. When recording is complete, the communication device sends the recorded voice data to the server. The server analyzes the received voice data and extracts a feature vector of the voice. The name and information of the person corresponding to this voice feature vector are registered in a database.

[1065] Next, we will explain the voice recognition and emotion analysis process when an incoming call is received. When a communication terminal receives an incoming call, the user starts the call. As soon as the call starts, the terminal records the caller's voice in real time. This recorded voice data is sent from the communication terminal to a server. The server analyzes the received voice data and extracts a feature vector of the voice. The extracted feature vector is compared with voice feature vectors already registered in a database, and the result of this comparison is determined by the server. The result is sent to the communication terminal and displayed on a display.

[1066] In addition, the communication terminal is equipped with an emotion engine that analyzes the user's voice during a call and recognizes the user's emotional state in real time. The recognized emotion information is immediately sent to the server, which then generates a warning message based on that information and sends it to the communication terminal.

[1067] As a specific example, consider the case where a user registers their mother's voice. The user uses the recording function of their communication device to record their mother's voice and sends the voice data to the server. The server analyzes the voice data and registers the mother's voice feature vector and name in a database. If the user's communication device then receives a call from the mother, the caller's voice is recorded in real time and sent to the server. The server analyzes the voice feature vector and, if it matches the mother's voice feature vector in the database, the communication device displays the message "It's from your mother." At the same time, the emotion engine analyzes the user's voice in real time, and if the user's emotions indicate anxiety or caution, a warning message such as "Caution required" is displayed on the communication device.

[1068] In this way, the system based on the present invention can provide more accurate measures to prevent special frauds by combining the results of analyzing the user's emotional state using an emotion engine in addition to caller recognition using voice feature vectors. It also provides an environment where users can receive calls with peace of mind, which is extremely effective in preventing fraud damage before it happens.

[1069] The processing flow will be explained below.

[1070] This invention is a system that combines a voice recognition system and an emotion engine, and when a communication terminal receives an incoming call, it analyzes the caller's voice and also analyzes the user's emotional state, thereby preventing damage from special fraud. The specific processing flow is explained below step by step.

[1071] User Voice Enrollment Process

[1072] Step 1:

[1073] The user starts the application on the communication terminal, selects the voice registration menu of the phone book, and then enters the name of the person to be registered.

[1074] Step 2:

[1075] The user records the voice of the person to be registered into the communication terminal, and when the recording is complete, the user presses the recording completion button.

[1076] Step 3:

[1077] The device sends the recorded audio data to a server via the Internet.

[1078] Step 4:

[1079] The server analyzes the received voice data and extracts a voice feature vector using a voice recognition algorithm.

[1080] Step 5:

[1081] The server registers the extracted voice feature vectors together with the names of the corresponding people in a database.

[1082] Incoming voice recognition and sentiment analysis process

[1083] Step 1:

[1084] When the communication terminal receives an incoming call, the user starts the call, and the terminal records the caller's voice in real time.

[1085] Step 2:

[1086] The device sends the recorded voice data of the caller to the server, where it is immediately analyzed.

[1087] Step 3:

[1088] The server analyzes the received voice data and extracts the caller's voice feature vector, which is then compared with the registered data in the database.

[1089] Step 4:

[1090] The server judges the result of matching the voice feature vector. If the caller's voice matches the registered data, the judgement result is recorded as "Registered Person." If not, it is recorded as "Unregistered Person."

[1091] Step 5:

[1092] The server transmits the result of the determination to the communication terminal, which then determines the content to display based on the received result of the determination.

[1093] Step 6:

[1094] The device displays the result on the screen. If the caller's voice matches the registered voice, the name of the registered person is displayed. If they do not match, the device displays "Unregistered."

[1095] Emotion recognition process by emotion engine

[1096] Step 1:

[1097] During a call, the device records the user's voice in real time and sends it to the emotion engine.

[1098] Step 2:

[1099] The emotion engine analyzes the received user voice data and extracts the user's emotion feature vector using emotion recognition algorithms.

[1100] Step 3:

[1101] The emotion engine determines the user's emotional state, such as anxiety, alertness, or calmness, based on the extracted emotion feature vector.

[1102] Step 4:

[1103] The emotion engine sends the judgment result to the server, which generates a warning message as needed based on the received emotion information.

[1104] Step 5:

[1105] The server transmits the generated warning message to the communication terminal, and the terminal displays the received warning message on its display.

[1106] This allows users to not only confirm whether the caller is a registered person, but also receive warning messages tailored to their own emotional state, making it possible to prevent damage from special frauds.

[1107] Example 2

[1108] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1109] Conventional voice recognition systems only analyze and recognize the caller's voice, so they cannot completely prevent risks such as special fraud. Also, because they do not take into account the user's emotional state, they have the problem of not being able to issue appropriate warnings even if the user is feeling anxious or cautious.

[1110] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1111] In this invention, the server includes means for analyzing voice data and extracting voice characteristics, means for determining whether the voice characteristics match registered data, and means for generating a warning message based on emotion data and transmitting it to the communication terminal, thereby enabling the user's emotional state to be analyzed in real time and a warning message to be displayed as necessary.

[1112] The "means for recording voice" refers to a function or device that allows a user to record voice using a communication terminal.

[1113] The "means for transmitting recorded voice to the information processing device" refers to a function or device for transmitting voice data recorded by a communication terminal to the information processing device via a network.

[1114] The "means for an information processing device to analyze voice data and extract voice features" refers to software or algorithms that allow the information processing device to analyze received voice data and generate voice feature vectors.

[1115] "Means for determining whether the voice characteristics match the registered data" refers to a function or software for comparing the analyzed voice characteristics with the registered data in the database and determining whether they match.

[1116] The "means for transmitting the determination result to the communication terminal" refers to a communication function or protocol for transmitting the determination result from the information processing device to the communication terminal.

[1117] The "means for the communication terminal to display the determination result" refers to a display or display function for visually presenting the determination result received by the communication terminal to the user.

[1118] The "means for the communication terminal to analyze the emotional state of the user" refers to an emotion engine or software that allows the communication terminal to analyze the user's voice during a call and recognize the user's emotional state in real time.

[1119] The "means for transmitting analyzed emotion data to the information processing device" refers to a communication function or protocol that allows the communication terminal to transmit analyzed user emotion data to the information processing device.

[1120] "Means for an information processing device to generate an alert message based on emotion data and transmit it to a communication terminal" refers to a function or software that allows an information processing device to create an alert message based on emotion data received and transmit it to a communication terminal.

[1121] This invention combines a voice recognition system with an emotion engine, and aims to analyze the caller's voice when a communication terminal receives an incoming call, and also analyze the user's emotional state. This system will enable more advanced prevention of special frauds.

[1122] Hardware and software used

[1123] 1. Communication terminal:

[1124] Voice recording function: A microphone and recording app to record the user's voice.

[1125] Display: A screen for displaying the results and warning messages.

[1126] Emotion engine: Software for analyzing user emotions.

[1127] 2. Server (information processing device):

[1128] Voice recognition software: For example, Google Voice Recognition API.

[1129] Database management system: for example, MySQL.

[1130] Sentiment analysis algorithm: A model for analyzing the user's emotional state.

[1131] Process Overview

[1132] 1. Audio Recording and Transmission:

[1133] Users use their communication terminals to record the voices of their family members and friends, and the recorded voice data is sent from the communication terminals to a server.

[1134] 2. Analysis of audio data:

[1135] The server analyzes the received voice data and extracts the voice feature vector. It uses voice recognition software to analyze the voice waveform and generate the feature vector.

[1136] 3. Register feature vectors and data:

[1137] The extracted feature vectors are registered in a database along with the names and other information of the corresponding people. This registered data is managed using a database management system.

[1138] 4. Incoming call voice recognition and emotion analysis:

[1139] When a communication device receives an incoming call, it records the caller's voice in real time and sends it to the server, which analyzes the voice data, extracts feature vectors, and compares them with the registered data in a database.

[1140] At the same time, the emotion engine installed in the communication terminal analyzes the user's voice in real time and recognizes their emotional state.

[1141] 5. Sending and viewing results:

[1142] The server sends the results of the matching and emotion analysis to the communication terminal, which then displays the caller's name and a warning message.

[1143] Specific examples

[1144] Consider the case where a user registers their mother's voice. The user uses the recording function of the communication terminal to record their mother's voice and sends the voice data to the server. The server analyzes the voice data and registers the mother's voice feature vector and name in the database.

[1145] If the user then receives a call on their communication device and the caller is their mother, the caller's voice is recorded in real time and sent to the server. The server analyzes the voice feature vector, and if it matches the mother's voice feature vector in the database, the communication device displays the message "Call from mother." At the same time, the emotion engine analyzes the user's voice in real time, and if the user is expressing anxiety or caution, a warning message such as "Caution required" is displayed on the communication device.

[1146] Prompt Sentence Examples

[1147] "Please tell me the procedure for a system that registers the mother's voice, recognizes the mother's voice in real time when a call comes in, analyzes the user's emotional state, and displays a warning message."

[1148] In this way, the present invention provides a system that combines voice recognition and emotion analysis, thereby providing an environment in which telephones can be used more safely and reducing the risk of special fraud and the like.

[1149] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1150] Step 1:

[1151] A user uses a communication terminal to record the voice of a family member or friend. The user presses the record button on the communication terminal, and then presses the save button after recording is complete. The input is the user's voice, and the voice data is saved in the communication terminal as output.

[1152] Step 2:

[1153] The communication terminal sends the recorded voice data to the server. Once the recording is complete, the voice data is sent to the server via the network. The input is the voice data stored in the communication terminal, and the voice data that is transferred to the server as output is generated. During this process, the progress of the communication is displayed on the terminal.

[1154] Step 3:

[1155] The server analyzes the voice data and extracts voice features. The server analyzes the received voice data using voice recognition software (e.g., Google Voice Recognition API) and generates a voice feature vector. The input is the voice data transferred to the server, and the output is the extracted voice feature vector.

[1156] Step 4:

[1157] The server registers the extracted voice feature vector in a database. The server saves the name and information of the person corresponding to the voice feature vector in a database (e.g., MySQL). The input is the voice feature vector and the corresponding information, and the output is newly registered data in the database.

[1158] Step 5:

[1159] The communication terminal records the caller's voice in real time. When a call comes in, the communication terminal records the caller's voice as soon as the call starts. The input is the caller's voice when the call comes in, and the real-time recorded voice data is generated as the output.

[1160] Step 6:

[1161] The communication terminal sends the recorded voice data to the server. The voice data recorded in real time is sent to the server using asynchronous communication technology (e.g., WebSocket). The input is the voice data recorded in real time, and the voice data transferred to the server is generated as the output.

[1162] Step 7:

[1163] The server analyzes the incoming voice data and extracts a voice feature vector. The server analyzes the received voice data and generates a voice feature vector. The input is the voice data transferred to the server, and the output is the extracted voice feature vector.

[1164] Step 8:

[1165] The server compares the voice feature vector with the registered data in the database and determines whether they match. The server compares the extracted voice feature vector with the registered data in the database and determines whether they match. The input is the voice feature vector and the registered data in the database, and the output is the result of the determination of whether they match.

[1166] Step 9:

[1167] The server transmits the determination result to the communication terminal. The server transmits the determination result to the communication terminal. The input is the determination result, and the determination result is generated and transferred to the communication terminal as output.

[1168] Step 10:

[1169] The communication terminal displays the judgment result and begins emotion analysis. The judgment result is displayed on the communication terminal's display, and at the same time the emotion engine begins analyzing the user's voice in real time. The input is the judgment result from the server and the user's voice during the call, and the output is the result displayed on the display and emotion analysis data.

[1170] Step 11:

[1171] The communication terminal analyzes the user's emotional state and sends the results to the server. The emotion engine then sends the analyzed emotional data to the server. The input is the user's voice and the analysis results by the emotion engine, and the output is the emotional data sent to the server.

[1172] Step 12:

[1173] The server generates a warning message based on the emotion data and sends it to the communication terminal. The server generates a warning message based on the received emotion data and sends it to the communication terminal. The input is emotion data, and the warning message that is transferred to the communication terminal is generated as the output.

[1174] Step 13:

[1175] The communication terminal displays the warning message. The communication terminal displays the warning message received from the server on the display. The input is the warning message from the server, and the warning message displayed on the display is generated as the output.

[1176] (Application example 2)

[1177] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1178] Currently, fraud methods are becoming more sophisticated, and special frauds over the phone are on the rise. Many of these methods target the elderly, and the resulting losses are particularly significant, posing a problem. Conventional voice recognition systems primarily focus on identifying callers, without taking into account the user's emotional state. As a result, even if the caller is registered, fraudulent calls may not be detected. Therefore, there is a need for a system that can prevent fraud and risky calls by analyzing the user's emotional state.

[1179] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1180] In this invention, the server includes means for analyzing the emotional state of the user from the voice during the call, means for generating a warning message based on the emotion analysis result and transmitting it to the communication terminal, means for recording the voice, means for transmitting the recorded voice to the information processing device, means for the information processing device to analyze the voice data and extract voice characteristics, means for determining whether the voice data matches registered data based on the voice characteristics, means for transmitting the determination result to the communication terminal, and means for the communication terminal to display the determination result. This enables more advanced prevention of special frauds by taking into account both the caller's recognition and the user's emotional state.

[1181] "Audio recording means" means a device or software for capturing and recording audio data during a call.

[1182] The "means for transmitting the recorded voice to the information processing device" is a function or device that transfers the recorded voice data to a server or processing device via communication or a network.

[1183] The "means for an information processing device to analyze voice data and extract voice features" refers to an algorithm or device for identifying and extracting specific voice patterns or features from received voice data.

[1184] "Means for determining whether the voice characteristics match registered data" refers to a function or device for comparing the extracted voice characteristics with pre-registered voice data and determining whether there is a match.

[1185] The "means for transmitting the determination result to the communication terminal" is a function or device that transmits the voice feature matching result to the terminal via communication or a network.

[1186] The "means for the communication terminal to display the determination result" refers to a display device or software for visually notifying the user of the received determination result.

[1187] The "means for analyzing the user's emotional state from the voice during a call" refers to an algorithm or device for determining the user's emotions in real time from the voice data acquired during a call.

[1188] The "means for generating a warning message based on the emotion analysis result and transmitting it to the communication terminal" refers to a function or device for creating a warning message based on the recognized emotional state and transmitting it to the communication terminal.

[1189] The present invention provides a security system that prevents special frauds by analyzing the caller's voice when a communication terminal receives an incoming call and also analyzing the user's emotional state. Specific embodiments for carrying out the present invention will be described below.

[1190] System Configuration

[1191] The system includes the following major hardware and software components:

[1192] Communication device: Smartphone with voice recording and display functions

[1193] Server: Voice data analyzer, database, emotion engine

[1194] Software: Speech recognition library (e.g., speech_recognition), emotion recognition library (e.g., emotion_recognition), voice feature recognition library (e.g., voice_recognition)

[1195] Program processing and natural language explanation

[1196] 1. Voice registration:

[1197] The user uses the voice recording function of the communication device to record the voices of family members or acquaintances. After the recording is complete, the communication device sends the recorded voice data to the server.

[1198] The server analyzes the received voice data and extracts voice feature vectors. The extracted voice feature vectors and the corresponding person's name and information are registered in a database.

[1199] 2. Incoming call voice recognition and sentiment analysis:

[1200] When a communication terminal receives an incoming call, the user starts the call. The terminal records the caller's voice in real time as soon as the call starts. This recorded voice data is sent from the communication terminal to a server.

[1201] The server analyzes the received voice data and extracts the voice feature vector. The extracted feature vector is compared with the voice feature vectors already registered in the database, and the result is judged and sent to the communication terminal. The judgment result is displayed on the communication terminal.

[1202] During a call, the emotion engine installed in the communication terminal analyzes the user's voice and recognizes the user's emotional state in real time. The recognized emotional information is immediately sent to the server, which then generates a warning message based on that information and sends it to the communication terminal.

[1203] Real-world examples

[1204] For example, consider the case where a user has registered their mother's voice in advance. The user records the mother's voice using a communication terminal and sends the voice data to the server. The server analyzes the voice data and registers the mother's voice feature vector and name in a database.

[1205] Next, when the user's communication device receives a call and recognizes that the caller is the mother, the caller's voice is recorded in real time and sent to the server. The server analyzes the voice feature vector, and if it matches the voice feature vector of the mother in the database, the communication device displays the message "Call from mother." At the same time, the emotion engine analyzes the user's voice in real time, and if the user's emotion indicates anxiety or caution, a warning message such as "Caution required" is displayed.

[1206] Prompt Sentence Examples

[1207] Also, when using a generative AI model to generate an appropriate warning message, enter the following prompt:

[1208] The user's emotional state is anxious:

[1209] "We need to be extra careful as the user has expressed anxiety. Do you want to continue communicating?"

[1210] Called party is unknown:

[1211] "This caller is not registered and may be a scam. Would you like to continue the call?"

[1212] The present invention makes it possible to provide an advanced security system that analyzes both the caller's identity and the user's emotional state in real time, preventing fraud and dangerous calls before they occur.

[1213] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1214] Step 1:

[1215] Voice registration

[1216] Input: Audio data obtained using the user's recording function

[1217] How it works: A user uses the voice recording function of their communication device to record the voice of a family member or friend. After pressing the record button and finishing recording, the communication device automatically sends the voice data to the server.

[1218] Data processing: The server analyzes the received voice data and extracts voice feature vectors.

[1219] Output: The extracted speech feature vectors are registered in a database.

[1220] Step 2:

[1221] Voice recognition for incoming calls

[1222] Input: Caller's voice data recorded when receiving a call

[1223] Operation: When a communication terminal receives an incoming call, the user starts the call. At the same time, the terminal records the caller's voice in real time and sends this voice data to the server.

[1224] Data processing: The server analyzes the received voice data and extracts voice feature vectors. The extracted feature vectors are compared with previously registered voice feature vectors.

[1225] Output: The matching result is sent to the communication terminal and displayed on the screen.

[1226] Step 3:

[1227] Emotional state analysis

[1228] Input: Voice data of the user during the call

[1229] How it works: The emotion engine in the communication device analyzes the user's voice in real time during a call.

[1230] Data Processing: The emotion engine identifies the user's emotional state (e.g., anxiety, alertness, etc.).

[1231] Output: The analysis results are sent to the server.

[1232] Step 4:

[1233] Warning message generation and notification

[1234] Input: Emotion analysis results and voice feature matching results

[1235] Operation: The server generates a warning message as needed based on the emotion analysis results and voice feature matching results.

[1236] Data processing: The generated message is sent to the communication terminal.

[1237] Output: A warning message is displayed on the communication terminal.

[1238] Specific operation example

[1239] For example, when a user whose mother's voice has already been registered receives a call from her mother, the process is as follows.

[1240] In step 1, the mother's voice is pre-registered.

[1241] In step 2, the moment a call is received, the caller's voice is recorded in real time and sent to the server. The server then performs a voice analysis and compares it with pre-registered voice characteristics. If the caller is confirmed to be the mother, the message "This is from your mother" is displayed on the communication device.

[1242] In step 3, the user's voice during the call is analyzed in real time, and if the user shows any anxiety or alertness, their emotional state is sent to the server.

[1243] If the emotional state is recognized as anxious in step 4, the server generates a warning message such as "Caution required" which is displayed on the communication terminal.

[1244] Examples of prompts include:

[1245] The user's emotional state is anxious:

[1246] "We need to be extra careful as the user has expressed anxiety. Do you want to continue communicating?"

[1247] Called party is unknown:

[1248] "This caller is not registered and may be a scam. Would you like to continue the call?"

[1249] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1250] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1251] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1252] [Fourth embodiment]

[1253] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1254] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1255] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1256] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1257] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1258] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1259] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1260] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1261] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1262] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1263] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1264] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1265] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1266] The present invention is a speech recognition system that utilizes a communication terminal such as a mobile phone or a landline phone and an information processing device that processes voice data. This system is designed as an effective means of preventing damage from special frauds.

[1267] First, we will explain the process by which a user registers the voices of family members and acquaintances. The user uses the voice recording function of the communication terminal to record the voice of each person. When recording is complete, the communication terminal sends the recorded voice data to an information processing device (server). The server analyzes the received voice data and extracts a feature vector of the voice. The name and information of the person corresponding to this voice feature vector are registered in a database.

[1268] Next, we will explain the voice recognition process when an incoming call is received. When a communication terminal receives an incoming call, it records the caller's voice in real time the moment the call starts. This recorded voice data is sent from the communication terminal to a server. The server analyzes the received voice data and extracts a feature vector of the voice. The extracted feature vector is compared with voice feature vectors already registered in a database. The result of this comparison is judged by the server and sent to the communication terminal.

[1269] Based on the results of the determination received from the server, the communication device displays on its display whether the caller is a registered person. If there is a match, the communication device displays "From Mr. / Ms. XX," indicating that the caller is a family member or acquaintance. If there is no match, the communication device displays "Unregistered," warning the user to be cautious.

[1270] As a specific example, consider the case where a user registers the voice of their son. The user uses the recording function of their communication device to record their son's voice and sends the voice data to the server. The server analyzes the voice data and registers the son's voice feature vector and name in a database. After that, when the user receives a call on their communication device and the caller is their son, the caller's voice is recorded in real time and sent to the server. The server analyzes the voice feature vector and, if it matches the son's voice feature vector in the database, the communication device displays the message "Call from your son." On the other hand, if the caller is not registered, the communication device displays the message "Unregistered caller."

[1271] In this way, the system based on the present invention allows users to receive calls with peace of mind and is effective in preventing damage from special frauds.

[1272] The processing flow will be explained below.

[1273] Step 1:

[1274] The user starts the application on the communication terminal, selects the voice registration menu of the phone book, and then enters the name of the person to be registered.

[1275] Step 2:

[1276] The user records the voice of the person to be registered into the communication terminal. When the user presses the recording completion button, the recorded voice data is saved in the communication terminal.

[1277] Step 3:

[1278] The device then transmits the stored audio data to a server, typically via the Internet.

[1279] Step 4:

[1280] The server analyzes the received voice data and extracts a voice feature vector using a voice recognition algorithm.

[1281] Step 5:

[1282] The server registers the extracted voice feature vectors together with the corresponding person's name in a database, which stores the voice feature vectors and person information.

[1283] Step 6:

[1284] When the communication terminal receives an incoming call, the user starts the call, and the terminal records the caller's voice in real time.

[1285] Step 7:

[1286] The device sends the recorded voice data of the caller to the server, where it is immediately analyzed.

[1287] Step 8:

[1288] The server analyzes the received voice data and extracts the caller's voice feature vector, which is then compared with the registered data in the database.

[1289] Step 9:

[1290] The server judges the result of matching the voice feature vector. If the caller's voice matches the registered data, the judgement result is recorded as "Registered Person." If not, it is recorded as "Unregistered Person."

[1291] Step 10:

[1292] The server transmits the result of the determination to the communication terminal, which then determines the content to display based on the received result of the determination.

[1293] Step 11:

[1294] The device displays the result on the screen. If the caller's voice matches the registered voice, the name of the registered person is displayed. If they do not match, the device displays "Unregistered."

[1295] Step 12:

[1296] The user checks the display and determines whether the caller is a family member or acquaintance. If the message "Unregistered" appears, the user becomes more vigilant.

[1297] Example 1

[1298] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1299] The purpose of this invention is to provide a communication system that protects users from crimes such as special fraud and allows them to make calls with peace of mind. In particular, there is a need to prevent fraudulent acts by recognizing the voice of the other party in real time and comparing it with pre-registered voices. This system provides a method for registering and recognizing the voices of family members and acquaintances, and comprehensively identifying suspicious callers.

[1300] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1301] In this invention, the server includes a means for analyzing voice data and extracting a voice feature vector, a means for determining whether the voice feature vector matches registered data in a database, and a means for transmitting the determination result to the communication terminal. This allows users to analyze the caller's voice in real time when a call starts and compare it with pre-registered voices. Furthermore, the determination result is transmitted to the communication terminal and displayed on the display, allowing users to immediately confirm whether the caller is a registered person. This prevents crimes such as special fraud and allows users to make calls with peace of mind.

[1302] A "communication terminal" is a device that allows a user to record voice and transmit the voice data to an information processing device.

[1303] "Recording means" refers to the function of recording audio within the communication terminal.

[1304] The "information processing device" is a device that analyzes transmitted voice data and extracts voice feature vectors.

[1305] "Audio data" refers to a recording of audio that is recorded by a communication terminal and transmitted to an information processing device.

[1306] A "voice feature vector" is a collection of data that is extracted from voice data and that quantifies individual voice features.

[1307] A "database" is a digital data structure for storing speech feature vectors and related information.

[1308] "Matching means" refers to a function that compares the extracted voice feature vector with existing data in the database and determines whether they match.

[1309] The "determination result" is information indicating the result of matching or mismatching obtained by matching the speech feature vectors.

[1310] The "display means" is a function that enables the communication terminal to visually convey the determination result to the user.

[1311] This invention is a voice recognition system using a communication terminal and an information processing device (server). The main purpose of this system is to prevent damage from special frauds, specifically by recognizing the caller's voice in real time and comparing it with pre-registered voices to detect suspicious callers.

[1312] Hardware and software used

[1313] The communication terminals used are mobile phones and landlines, which have the functions of recording voice and transmitting data via the Internet.

[1314] The server is an information processing device required to analyze voice data. Voice analysis software, such as OpenAI's Whisper or Google's Speech-to-Text, is installed on the server.

[1315] Processing flow

[1316] 1. Recording Audio

[1317] The user uses the recording function of the communication device to record the voices of family members and acquaintances, and the recorded voice data is saved in the communication device.

[1318] 2. Sending audio data

[1319] The terminal transmits the recorded voice data to an information processing device (server) via the Internet.

[1320] 3. Analysis and registration of audio data

[1321] The server analyzes the received voice data and extracts a voice feature vector. The extracted voice feature vector is then registered in a database. This saves the registrant's voice data in the database.

[1322] 4. Processing when an incoming call arrives

[1323] When the device receives an incoming call, it records the caller's voice in real time as soon as the call begins, and the recorded voice data is sent to the server.

[1324] 5. Real-time analysis and matching of voice data

[1325] The server analyzes the received voice data in real time, extracts voice feature vectors, and then compares the extracted feature vectors with existing data in the database to determine whether they match.

[1326] 6. Sending and displaying the results

[1327] The server sends the result of the judgment to the communication terminal, which then displays the result to the user. If the caller is a registered person, it will display "From Mr. / Ms. XX," and if not, it will display "Unregistered." This allows the user to easily check whether the caller is trustworthy.

[1328] Specific examples

[1329] When a user registers his son's voice

[1330] The user uses the recording function of the communication device to have their son say, "Hello, I'm your son." Once the recording is complete, the voice data is sent to the server, which performs voice analysis, extracts the voice feature vector, and registers it in a database.

[1331] When a call comes in

[1332] The device receives an incoming call and displays "Incoming call." When the call begins, the caller's voice is recorded in real time and sent to the server. The server analyzes the voice data in real time and compares it with the son's voice feature vector in the database. As a result, the message "Call from your son" is displayed.

[1333] Prompt Sentence Examples

[1334] "We will record and register your mother's voice. We will start recording, so please say, 'Hello, I'm your mother.'"

[1335] "You have an incoming call. We are verifying that the caller is a registered user."

[1336] "(For example, my son) I'll record my son's voice and say, 'Hi, I'm your son,' and we'll save it."

[1337] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1338] Step 1: Record your audio

[1339] A user uses the voice recording function of a communication terminal to record the voices of family members or acquaintances.

[1340] Input: Voice of family or friends

[1341] Specific behavior:

[1342] 1. Select "Audio Recording" from the menu on your communication device.

[1343] 2. Press the start recording button and have the subject speak.

[1344] 3. Press the Stop Recording button to save the recording.

[1345] Output: Recorded audio data

[1346] Step 2: Sending audio data

[1347] The terminal transmits the recorded voice data to an information processing device (server).

[1348] Input: Recorded audio data

[1349] Specific behavior:

[1350] 1. Select the saved audio data.

[1351] 2. Send the data to the server using your internet connection.

[1352] Output: Audio data sent to the server

[1353] Step 3: Analyze and register audio data

[1354] The server receives the audio data and analyzes it.

[1355] Input: Transmitted audio data

[1356] Specific behavior:

[1357] 1. The server analyzes the voice data using voice analysis software (e.g., OpenAI's Whisper or Google's Speech-to-Text).

[1358] 2. Extract the voice feature vector and register it in a database along with the subject's name and information.

[1359] Output: Audio feature vectors and correspondence information registered in the database

[1360] Step 4: Receiving a call

[1361] The device receives an incoming call.

[1362] Input: Incoming call signal from the caller

[1363] Specific behavior:

[1364] 1. The call will ring and the display will show "Incoming call."

[1365] 2. The user receives the call and starts the conversation.

[1366] Output: Call Start

[1367] Step 5: Record the caller's voice

[1368] The device records the caller's voice in real time.

[1369] Input: Caller's voice

[1370] Specific behavior:

[1371] 1. Start recording as soon as the call begins.

[1372] 2. Prepare to send the recording data to the server.

[1373] Output: Recorded caller voice data

[1374] Step 6: Sending audio data

[1375] The terminal transmits the recorded voice data of the caller to the server.

[1376] Input: Recorded caller voice data

[1377] Specific behavior:

[1378] 1. Send real-time audio data to a server via an internet connection.

[1379] Output: Caller's voice data sent to the server

[1380] Step 7: Analyze and match the audio data

[1381] The server analyzes the caller's voice data received and compares it with existing data in the database.

[1382] Input: Caller's voice data sent to the server

[1383] Specific behavior:

[1384] 1. The server uses speech analysis software to extract the caller's voice feature vector.

[1385] 2. Compare the extracted speech feature vector with existing speech feature vectors in the database.

[1386] 3. Determine the matching result.

[1387] Output: Matching result (determining whether the caller is a registered person)

[1388] Step 8: Receive and display the results

[1389] The terminal displays the judgment result received from the server.

[1390] Input: Matching result sent from the server

[1391] Specific behavior:

[1392] 1. The terminal displays the received judgment result.

[1393] 2. If the caller is a registered person, it will display "From Mr. / Ms. XX."

[1394] 3. If there is no match, the message "Unregistered" will be displayed to warn the user.

[1395] Output: Display of judgment result (registered person or unregistered person)

[1396] This series of processes allows the user to check in real time whether the caller is trustworthy.

[1397] (Application example 1)

[1398] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1399] In recent years, voice-based fraud has become more prevalent, requiring users to be vigilant against calls from unknown callers. However, conventional voice recognition systems are inadequate, particularly in their inability to display real-time results, making it difficult for users to make a quick decision. This has led to a growing demand for effective means to prevent fraud. Furthermore, there is a need for a system that can instantly display a warning on smart devices, such as smart glasses and head-mounted displays, that are linked to communication terminals.

[1400] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1401] In this invention, the server includes a means for recording voice, a means for transmitting the recorded voice to an information processing device, a means for the information processing device to analyze the voice data and extract voice characteristics, a means for determining whether the voice data matches registered data based on the voice characteristics, a means for transmitting the determination result to a communication terminal, a means for the communication terminal to display the determination result, and a means for linking with a smart device to display the determination result in real time. This allows users to immediately be alert to calls from unknown callers and prevent fraud. Furthermore, by using smart glasses or a head-mounted display, a warning is displayed directly in the user's field of vision, allowing for a quick and effective response.

[1402] "Audio recording means" means equipment or software for recording audio signals as digital data.

[1403] The "means for transmitting the recorded voice to the information processing device" is a function or protocol for transferring the recorded voice data to the information processing device through a communication line.

[1404] "Means for an information processing device to analyze voice data and extract voice features" refers to a process in which an information processing device analyzes input voice data using an algorithm and extracts features.

[1405] The "means for determining whether or not the data matches the registered data based on the voice characteristics" is a process for comparing pre-registered voice characteristic data with newly extracted voice characteristic data and determining whether or not there is a match.

[1406] The "means for transmitting the determination result to the communication terminal" is a communication means or protocol for transmitting the determination result to the communication terminal in real time.

[1407] The "means for the communication terminal to display the determination result" is a function that visually presents the determination result to the user using the display or interface of the communication terminal.

[1408] "Means for linking with a smart device and displaying the judgment results in real time" refers to a system that links with a smart device such as smart glasses or a head-mounted display and presents the judgment results to the user in real time.

[1409] "Smart glasses" are glasses-type devices that run specific applications and provide augmented reality and notification information through a display.

[1410] A "head-mounted display" is a device worn on the head and positioned within the user's field of vision to provide visual information to the user.

[1411] This invention is a voice recognition system that allows users to receive calls with peace of mind. In order to prevent damage from special fraud, the system analyzes the caller's voice in real time and compares it with pre-registered voice data to determine whether the caller is a registered person.

[1412] The system of the present invention includes the following hardware and software.

[1413] Hardware: Smart glasses (e.g., Google Glass, Vuzix Blade), smartphones

[1414] Software: Audio processing libraries (e.g., pyaudio, librosa), dedicated APIs for smart glasses (e.g., Google Glass API), local databases (e.g., SQLite)

[1415] When a user uses the system for the first time, they record the voice data they want to register in advance using a communication terminal and send it to the server. The server analyzes the received voice data and extracts a voice feature vector. This extracted voice feature vector is registered in the database. Through this process, the voices of family members and acquaintances are registered in the system.

[1416] When a call comes in, the communication terminal records the caller's voice in real time and sends the recorded voice data to the server. The server analyzes the voice data and extracts a voice feature vector. It then compares it with voice feature vectors already registered in a database and sends the result to the communication terminal. The communication terminal displays the result on smart glasses or a head-mounted display. This allows the user to confirm whether the caller is a registered person before starting the call.

[1417] As a specific example, consider the case where a user registers the voice of their son. The user uses the recording function of their communication device to record their son's voice and sends the voice data to the server. The server analyzes the voice data and registers the son's voice feature vector in a database. If the user then receives a call while cooking and the caller is their son, the communication device records the caller's voice in real time and sends it to the server. The server analyzes the voice feature vector, and if it matches the voice feature vector of the son that has already been registered, the smart glasses display will say, "Call from your son." On the other hand, if the caller is an unregistered person, the message will say, "This call is from your son. Please be careful."

[1418] This system allows users to immediately become alert to calls from unknown callers, preventing fraud before they occur. In addition, by using smart glasses or a head-mounted display, a warning can be displayed directly in the user's field of vision, enabling a quick and effective response.

[1419] An example of a prompt is as follows:

[1420] "This is a demo scene of a voice recognition security guard. It registers the voices of family members, identifies suspicious callers, and displays a warning. It compares them in real time with a recording of the son's voice. This system significantly reduces the risk of special fraud."

[1421] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1422] Step 1:

[1423] A user uses the recording function of a communication terminal to record the voices of family members or acquaintances.

[1424] Input: Audio signal (voice of user's family or acquaintance)

[1425] Processing: Use the recording function of the communication device to record the audio as digital data.

[1426] Output: Recorded audio data (digital format)

[1427] Step 2:

[1428] The communication terminal transmits the recorded voice data to the server.

[1429] Input: Recorded audio data

[1430] Processing: Upload the audio data to the server via a communication line.

[1431] Output: Audio data sent to the server

[1432] Step 3:

[1433] The server analyzes the received voice data and extracts a voice feature vector.

[1434] Input: Received audio data

[1435] Processing: The server uses an audio processing library (e.g., librosa) to analyze the audio data and generate an audio feature vector using techniques such as FFT (Fast Fourier Transform).

[1436] Output: Audio feature vector

[1437] Step 4:

[1438] The server registers the voice feature vector in a database.

[1439] Input: Audio feature vector, corresponding person name and information

[1440] Processing: The server uses a local database (e.g., SQLite) to store the voice feature vectors and person information in the database.

[1441] Output: Trained speech feature data

[1442] Step 5:

[1443] When a call is received, the communication terminal records the caller's voice in real time.

[1444] Input: Incoming voice signal (caller's voice)

[1445] Processing: Record the audio as digital data in real time using the recording function of the communication device.

[1446] Output: Recorded real-time audio data

[1447] Step 6:

[1448] The communication terminal transmits the recorded real-time voice data to the server.

[1449] Input: Recorded real-time audio data

[1450] Processing: Upload the audio data to the server via a communication line.

[1451] Output: Real-time audio data sent to the server

[1452] Step 7:

[1453] The server analyzes the real-time voice data and extracts voice feature vectors.

[1454] Input: Received real-time audio data

[1455] Processing: The server uses a speech processing library to analyze the speech data and generate a speech feature vector.

[1456] Output: Real-time audio feature vectors

[1457] Step 8:

[1458] The server matches the real-time voice feature vector with the registered voice feature vectors in the database.

[1459] Input: Real-time speech feature vector, registered speech feature vector in database

[1460] Processing: The server calculates the cosine similarity or Euclidean distance to assess the degree of match between the two vectors.

[1461] Output: Matching result (match / no match)

[1462] Step 9:

[1463] The server transmits the determination result to the communication terminal.

[1464] Input: Judgment result

[1465] Processing: The judgment result is sent to the communication terminal via the communication line.

[1466] Output: The judgment result sent to the communication terminal

[1467] Step 10:

[1468] The communication terminal displays the judgment results on smart glasses or a head-mounted display.

[1469] Input: Received verdict

[1470] Processing: Using the smart glasses' dedicated API, the judgment result is displayed on the screen.

[1471] Output: The result displayed on the screen ("This is from ____" or "This is an unregistered user")

[1472] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1473] The present invention is a system that aims to prevent more sophisticated special frauds by combining a voice recognition system with an emotion engine, and by analyzing the caller's voice when a communication terminal receives an incoming call, and also analyzing the user's emotional state.

[1474] First, we will explain the process by which a user registers the voices of family members and acquaintances. The user uses the voice recording function of the communication device to record the voice of each person. When recording is complete, the communication device sends the recorded voice data to the server. The server analyzes the received voice data and extracts a feature vector of the voice. The name and information of the person corresponding to this voice feature vector are registered in a database.

[1475] Next, we will explain the voice recognition and emotion analysis process when an incoming call is received. When a communication terminal receives an incoming call, the user starts the call. As soon as the call starts, the terminal records the caller's voice in real time. This recorded voice data is sent from the communication terminal to a server. The server analyzes the received voice data and extracts a feature vector of the voice. The extracted feature vector is compared with voice feature vectors already registered in a database, and the result of this comparison is determined by the server. The result is sent to the communication terminal and displayed on a display.

[1476] In addition, the communication terminal is equipped with an emotion engine that analyzes the user's voice during a call and recognizes the user's emotional state in real time. The recognized emotion information is immediately sent to the server, which then generates a warning message based on that information and sends it to the communication terminal.

[1477] As a specific example, consider the case where a user registers their mother's voice. The user uses the recording function of their communication device to record their mother's voice and sends the voice data to the server. The server analyzes the voice data and registers the mother's voice feature vector and name in a database. If the user's communication device then receives a call from the mother, the caller's voice is recorded in real time and sent to the server. The server analyzes the voice feature vector and, if it matches the mother's voice feature vector in the database, the communication device displays the message "It's from your mother." At the same time, the emotion engine analyzes the user's voice in real time, and if the user's emotions indicate anxiety or caution, a warning message such as "Caution required" is displayed on the communication device.

[1478] In this way, the system based on the present invention can provide more accurate measures to prevent special frauds by combining the results of analyzing the user's emotional state using an emotion engine in addition to caller recognition using voice feature vectors. It also provides an environment where users can receive calls with peace of mind, which is extremely effective in preventing fraud damage before it happens.

[1479] The processing flow will be explained below.

[1480] This invention is a system that combines a voice recognition system and an emotion engine, and when a communication terminal receives an incoming call, it analyzes the caller's voice and also analyzes the user's emotional state, thereby preventing damage from special fraud. The specific processing flow is explained below step by step.

[1481] User Voice Enrollment Process

[1482] Step 1:

[1483] The user starts the application on the communication terminal, selects the voice registration menu of the phone book, and then enters the name of the person to be registered.

[1484] Step 2:

[1485] The user records the voice of the person to be registered into the communication terminal, and when the recording is complete, the user presses the recording completion button.

[1486] Step 3:

[1487] The device sends the recorded audio data to a server via the Internet.

[1488] Step 4:

[1489] The server analyzes the received voice data and extracts a voice feature vector using a voice recognition algorithm.

[1490] Step 5:

[1491] The server registers the extracted voice feature vectors together with the names of the corresponding people in a database.

[1492] Incoming voice recognition and sentiment analysis process

[1493] Step 1:

[1494] When the communication terminal receives an incoming call, the user starts the call, and the terminal records the caller's voice in real time.

[1495] Step 2:

[1496] The device sends the recorded voice data of the caller to the server, where it is immediately analyzed.

[1497] Step 3:

[1498] The server analyzes the received voice data and extracts the caller's voice feature vector, which is then compared with the registered data in the database.

[1499] Step 4:

[1500] The server judges the result of matching the voice feature vector. If the caller's voice matches the registered data, the judgement result is recorded as "Registered Person." If not, it is recorded as "Unregistered Person."

[1501] Step 5:

[1502] The server transmits the result of the determination to the communication terminal, which then determines the content to display based on the received result of the determination.

[1503] Step 6:

[1504] The device displays the result on the screen. If the caller's voice matches the registered voice, the name of the registered person is displayed. If they do not match, the device displays "Unregistered."

[1505] Emotion recognition process by emotion engine

[1506] Step 1:

[1507] During a call, the device records the user's voice in real time and sends it to the emotion engine.

[1508] Step 2:

[1509] The emotion engine analyzes the received user voice data and extracts the user's emotion feature vector using emotion recognition algorithms.

[1510] Step 3:

[1511] The emotion engine determines the user's emotional state, such as anxiety, alertness, or calmness, based on the extracted emotion feature vector.

[1512] Step 4:

[1513] The emotion engine sends the judgment result to the server, which generates a warning message as needed based on the received emotion information.

[1514] Step 5:

[1515] The server transmits the generated warning message to the communication terminal, and the terminal displays the received warning message on its display.

[1516] This allows users to not only confirm whether the caller is a registered person, but also receive warning messages tailored to their own emotional state, making it possible to prevent damage from special frauds.

[1517] Example 2

[1518] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1519] Conventional voice recognition systems only analyze and recognize the caller's voice, so they cannot completely prevent risks such as special fraud. Also, because they do not take into account the user's emotional state, they have the problem of not being able to issue appropriate warnings even if the user is feeling anxious or cautious.

[1520] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1521] In this invention, the server includes means for analyzing voice data and extracting voice characteristics, means for determining whether the voice characteristics match registered data, and means for generating a warning message based on emotion data and transmitting it to the communication terminal, thereby enabling the user's emotional state to be analyzed in real time and a warning message to be displayed as necessary.

[1522] The "means for recording voice" refers to a function or device that allows a user to record voice using a communication terminal.

[1523] The "means for transmitting recorded voice to the information processing device" refers to a function or device for transmitting voice data recorded by a communication terminal to the information processing device via a network.

[1524] The "means for an information processing device to analyze voice data and extract voice features" refers to software or algorithms that allow the information processing device to analyze received voice data and generate voice feature vectors.

[1525] "Means for determining whether the voice characteristics match the registered data" refers to a function or software for comparing the analyzed voice characteristics with the registered data in the database and determining whether they match.

[1526] The "means for transmitting the determination result to the communication terminal" refers to a communication function or protocol for transmitting the determination result from the information processing device to the communication terminal.

[1527] The "means for the communication terminal to display the determination result" refers to a display or display function for visually presenting the determination result received by the communication terminal to the user.

[1528] The "means for the communication terminal to analyze the emotional state of the user" refers to an emotion engine or software that allows the communication terminal to analyze the user's voice during a call and recognize the user's emotional state in real time.

[1529] The "means for transmitting analyzed emotion data to the information processing device" refers to a communication function or protocol that allows the communication terminal to transmit analyzed user emotion data to the information processing device.

[1530] "Means for an information processing device to generate an alert message based on emotion data and transmit it to a communication terminal" refers to a function or software that allows an information processing device to create an alert message based on emotion data received and transmit it to a communication terminal.

[1531] This invention combines a voice recognition system with an emotion engine, and aims to analyze the caller's voice when a communication terminal receives an incoming call, and also analyze the user's emotional state. This system will enable more advanced prevention of special frauds.

[1532] Hardware and software used

[1533] 1. Communication terminal:

[1534] Voice recording function: A microphone and recording app to record the user's voice.

[1535] Display: A screen for displaying the results and warning messages.

[1536] Emotion engine: Software for analyzing user emotions.

[1537] 2. Server (information processing device):

[1538] Voice recognition software: For example, Google Voice Recognition API.

[1539] Database management system: for example, MySQL.

[1540] Sentiment analysis algorithm: A model for analyzing the user's emotional state.

[1541] Process Overview

[1542] 1. Audio Recording and Transmission:

[1543] Users use their communication terminals to record the voices of their family members and friends, and the recorded voice data is sent from the communication terminals to a server.

[1544] 2. Analysis of audio data:

[1545] The server analyzes the received voice data and extracts the voice feature vector. It uses voice recognition software to analyze the voice waveform and generate the feature vector.

[1546] 3. Register feature vectors and data:

[1547] The extracted feature vectors are registered in a database along with the names and other information of the corresponding people. This registered data is managed using a database management system.

[1548] 4. Incoming call voice recognition and emotion analysis:

[1549] When a communication device receives an incoming call, it records the caller's voice in real time and sends it to the server, which analyzes the voice data, extracts feature vectors, and compares them with the registered data in a database.

[1550] At the same time, the emotion engine installed in the communication terminal analyzes the user's voice in real time and recognizes their emotional state.

[1551] 5. Sending and viewing results:

[1552] The server sends the results of the matching and emotion analysis to the communication terminal, which then displays the caller's name and a warning message.

[1553] Specific examples

[1554] Consider the case where a user registers their mother's voice. The user uses the recording function of the communication terminal to record their mother's voice and sends the voice data to the server. The server analyzes the voice data and registers the mother's voice feature vector and name in the database.

[1555] If the user then receives a call on their communication device and the caller is their mother, the caller's voice is recorded in real time and sent to the server. The server analyzes the voice feature vector, and if it matches the mother's voice feature vector in the database, the communication device displays the message "Call from mother." At the same time, the emotion engine analyzes the user's voice in real time, and if the user is expressing anxiety or caution, a warning message such as "Caution required" is displayed on the communication device.

[1556] Prompt Sentence Examples

[1557] "Please tell me the procedure for a system that registers the mother's voice, recognizes the mother's voice in real time when a call comes in, analyzes the user's emotional state, and displays a warning message."

[1558] In this way, the present invention provides a system that combines voice recognition and emotion analysis, thereby providing an environment in which telephones can be used more safely and reducing the risk of special fraud and the like.

[1559] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1560] Step 1:

[1561] A user uses a communication terminal to record the voice of a family member or friend. The user presses the record button on the communication terminal, and then presses the save button after recording is complete. The input is the user's voice, and the voice data is saved in the communication terminal as output.

[1562] Step 2:

[1563] The communication terminal sends the recorded voice data to the server. Once the recording is complete, the voice data is sent to the server via the network. The input is the voice data stored in the communication terminal, and the voice data that is transferred to the server as output is generated. During this process, the progress of the communication is displayed on the terminal.

[1564] Step 3:

[1565] The server analyzes the voice data and extracts voice features. The server analyzes the received voice data using voice recognition software (e.g., Google Voice Recognition API) and generates a voice feature vector. The input is the voice data transferred to the server, and the output is the extracted voice feature vector.

[1566] Step 4:

[1567] The server registers the extracted voice feature vector in a database. The server saves the name and information of the person corresponding to the voice feature vector in a database (e.g., MySQL). The input is the voice feature vector and the corresponding information, and the output is newly registered data in the database.

[1568] Step 5:

[1569] The communication terminal records the caller's voice in real time. When a call comes in, the communication terminal records the caller's voice as soon as the call starts. The input is the caller's voice when the call comes in, and the real-time recorded voice data is generated as the output.

[1570] Step 6:

[1571] The communication terminal sends the recorded voice data to the server. The voice data recorded in real time is sent to the server using asynchronous communication technology (e.g., WebSocket). The input is the voice data recorded in real time, and the voice data transferred to the server is generated as the output.

[1572] Step 7:

[1573] The server analyzes the incoming voice data and extracts a voice feature vector. The server analyzes the received voice data and generates a voice feature vector. The input is the voice data transferred to the server, and the output is the extracted voice feature vector.

[1574] Step 8:

[1575] The server compares the voice feature vector with the registered data in the database and determines whether they match. The server compares the extracted voice feature vector with the registered data in the database and determines whether they match. The input is the voice feature vector and the registered data in the database, and the output is the result of the determination of whether they match.

[1576] Step 9:

[1577] The server transmits the determination result to the communication terminal. The server transmits the determination result to the communication terminal. The input is the determination result, and the determination result is generated and transferred to the communication terminal as output.

[1578] Step 10:

[1579] The communication terminal displays the judgment result and begins emotion analysis. The judgment result is displayed on the communication terminal's display, and at the same time the emotion engine begins analyzing the user's voice in real time. The input is the judgment result from the server and the user's voice during the call, and the output is the result displayed on the display and emotion analysis data.

[1580] Step 11:

[1581] The communication terminal analyzes the user's emotional state and sends the results to the server. The emotion engine then sends the analyzed emotional data to the server. The input is the user's voice and the analysis results by the emotion engine, and the output is the emotional data sent to the server.

[1582] Step 12:

[1583] The server generates a warning message based on the emotion data and sends it to the communication terminal. The server generates a warning message based on the received emotion data and sends it to the communication terminal. The input is emotion data, and the warning message that is transferred to the communication terminal is generated as the output.

[1584] Step 13:

[1585] The communication terminal displays the warning message. The communication terminal displays the warning message received from the server on the display. The input is the warning message from the server, and the warning message displayed on the display is generated as the output.

[1586] (Application example 2)

[1587] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1588] Currently, fraud methods are becoming more sophisticated, and special frauds over the phone are on the rise. Many of these methods target the elderly, and the resulting losses are particularly significant, posing a problem. Conventional voice recognition systems primarily focus on identifying callers, without taking into account the user's emotional state. As a result, even if the caller is registered, fraudulent calls may not be detected. Therefore, there is a need for a system that can prevent fraud and risky calls by analyzing the user's emotional state.

[1589] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1590] In this invention, the server includes means for analyzing the emotional state of the user from the voice during the call, means for generating a warning message based on the emotion analysis result and transmitting it to the communication terminal, means for recording the voice, means for transmitting the recorded voice to the information processing device, means for the information processing device to analyze the voice data and extract voice characteristics, means for determining whether the voice data matches registered data based on the voice characteristics, means for transmitting the determination result to the communication terminal, and means for the communication terminal to display the determination result. This enables more advanced prevention of special frauds by taking into account both the caller's recognition and the user's emotional state.

[1591] "Audio recording means" means a device or software for capturing and recording audio data during a call.

[1592] The "means for transmitting the recorded voice to the information processing device" is a function or device that transfers the recorded voice data to a server or processing device via communication or a network.

[1593] The "means for an information processing device to analyze voice data and extract voice features" refers to an algorithm or device for identifying and extracting specific voice patterns or features from received voice data.

[1594] "Means for determining whether the voice characteristics match registered data" refers to a function or device for comparing the extracted voice characteristics with pre-registered voice data and determining whether there is a match.

[1595] The "means for transmitting the determination result to the communication terminal" is a function or device that transmits the voice feature matching result to the terminal via communication or a network.

[1596] The "means for the communication terminal to display the determination result" refers to a display device or software for visually notifying the user of the received determination result.

[1597] The "means for analyzing the user's emotional state from the voice during a call" refers to an algorithm or device for determining the user's emotions in real time from the voice data acquired during a call.

[1598] The "means for generating a warning message based on the emotion analysis result and transmitting it to the communication terminal" refers to a function or device for creating a warning message based on the recognized emotional state and transmitting it to the communication terminal.

[1599] The present invention provides a security system that prevents special frauds by analyzing the caller's voice when a communication terminal receives an incoming call and also analyzing the user's emotional state. Specific embodiments for carrying out the present invention will be described below.

[1600] System Configuration

[1601] The system includes the following major hardware and software components:

[1602] Communication device: Smartphone with voice recording and display functions

[1603] Server: Voice data analyzer, database, emotion engine

[1604] Software: Speech recognition library (e.g., speech_recognition), emotion recognition library (e.g., emotion_recognition), voice feature recognition library (e.g., voice_recognition)

[1605] Program processing and natural language explanation

[1606] 1. Voice registration:

[1607] The user uses the voice recording function of the communication device to record the voices of family members or acquaintances. After the recording is complete, the communication device sends the recorded voice data to the server.

[1608] The server analyzes the received voice data and extracts voice feature vectors. The extracted voice feature vectors and the corresponding person's name and information are registered in a database.

[1609] 2. Incoming call voice recognition and sentiment analysis:

[1610] When a communication terminal receives an incoming call, the user starts the call. The terminal records the caller's voice in real time as soon as the call starts. This recorded voice data is sent from the communication terminal to a server.

[1611] The server analyzes the received voice data and extracts the voice feature vector. The extracted feature vector is compared with the voice feature vectors already registered in the database, and the result is judged and sent to the communication terminal. The judgment result is displayed on the communication terminal.

[1612] During a call, the emotion engine installed in the communication terminal analyzes the user's voice and recognizes the user's emotional state in real time. The recognized emotional information is immediately sent to the server, which then generates a warning message based on that information and sends it to the communication terminal.

[1613] Real-world examples

[1614] For example, consider the case where a user has registered their mother's voice in advance. The user records the mother's voice using a communication terminal and sends the voice data to the server. The server analyzes the voice data and registers the mother's voice feature vector and name in a database.

[1615] Next, when the user's communication device receives a call and recognizes that the caller is the mother, the caller's voice is recorded in real time and sent to the server. The server analyzes the voice feature vector, and if it matches the voice feature vector of the mother in the database, the communication device displays the message "Call from mother." At the same time, the emotion engine analyzes the user's voice in real time, and if the user's emotion indicates anxiety or caution, a warning message such as "Caution required" is displayed.

[1616] Prompt Sentence Examples

[1617] Also, when using a generative AI model to generate an appropriate warning message, enter the following prompt:

[1618] The user's emotional state is anxious:

[1619] "We need to be extra careful as the user has expressed anxiety. Do you want to continue communicating?"

[1620] Called party is unknown:

[1621] "This caller is not registered and may be a scam. Would you like to continue the call?"

[1622] The present invention makes it possible to provide an advanced security system that analyzes both the caller's identity and the user's emotional state in real time, preventing fraud and dangerous calls before they occur.

[1623] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1624] Step 1:

[1625] Voice registration

[1626] Input: Audio data obtained using the user's recording function

[1627] How it works: A user uses the voice recording function of their communication device to record the voice of a family member or friend. After pressing the record button and finishing recording, the communication device automatically sends the voice data to the server.

[1628] Data processing: The server analyzes the received voice data and extracts voice feature vectors.

[1629] Output: The extracted speech feature vectors are registered in a database.

[1630] Step 2:

[1631] Voice recognition for incoming calls

[1632] Input: Caller's voice data recorded when receiving a call

[1633] Operation: When a communication terminal receives an incoming call, the user starts the call. At the same time, the terminal records the caller's voice in real time and sends this voice data to the server.

[1634] Data processing: The server analyzes the received voice data and extracts voice feature vectors. The extracted feature vectors are compared with previously registered voice feature vectors.

[1635] Output: The matching result is sent to the communication terminal and displayed on the screen.

[1636] Step 3:

[1637] Emotional state analysis

[1638] Input: Voice data of the user during the call

[1639] How it works: The emotion engine in the communication device analyzes the user's voice in real time during a call.

[1640] Data Processing: The emotion engine identifies the user's emotional state (e.g., anxiety, alertness, etc.).

[1641] Output: The analysis results are sent to the server.

[1642] Step 4:

[1643] Warning message generation and notification

[1644] Input: Emotion analysis results and voice feature matching results

[1645] Operation: The server generates a warning message as needed based on the emotion analysis results and voice feature matching results.

[1646] Data processing: The generated message is sent to the communication terminal.

[1647] Output: A warning message is displayed on the communication terminal.

[1648] Specific operation example

[1649] For example, when a user whose mother's voice has already been registered receives a call from her mother, the process is as follows.

[1650] In step 1, the mother's voice is pre-registered.

[1651] In step 2, the moment a call is received, the caller's voice is recorded in real time and sent to the server. The server then performs a voice analysis and compares it with pre-registered voice characteristics. If the caller is confirmed to be the mother, the message "This is from your mother" is displayed on the communication device.

[1652] In step 3, the user's voice during the call is analyzed in real time, and if the user shows any anxiety or alertness, their emotional state is sent to the server.

[1653] If the emotional state is recognized as anxious in step 4, the server generates a warning message such as "Caution required" which is displayed on the communication terminal.

[1654] Examples of prompts include:

[1655] The user's emotional state is anxious:

[1656] "We need to be extra careful as the user has expressed anxiety. Do you want to continue communicating?"

[1657] Called party is unknown:

[1658] "This caller is not registered and may be a scam. Would you like to continue the call?"

[1659] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1660] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1661] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1662] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1663] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1664] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1665] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1666] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1667] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1668] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1669] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1670] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1671] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1672] 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.

[1673] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1674] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1675] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1676] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1677] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1678] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1679] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1680] The following is further disclosed regarding the above embodiment.

[1681] (Claim 1)

[1682] A means for recording audio;

[1683] means for transmitting the recorded voice to an information processing device;

[1684] An information processing device includes means for analyzing voice data and extracting voice features;

[1685] A means for determining whether the voice characteristics match the enrollment data;

[1686] means for transmitting the determination result to the communication terminal;

[1687] A means for displaying the determination result in the communication terminal;

[1688] A system including:

[1689] (Claim 2)

[1690] The information processing device includes a means for allowing a user to record a voice when registering the telephone directory, for transmitting the recorded voice data to the information processing device, and for extracting voice characteristics from the recorded voice data and registering the voice characteristics in the database.

[1691] 10. The system of claim 1.

[1692] (Claim 3)

[1693] The communication terminal records the caller's voice in real time when an incoming call is received, and transmits the recorded voice data to the information processing device, and the information processing device extracts the caller's voice characteristics and compares them with registered voice data to determine whether they match.

[1694] 10. The system of claim 1.

[1695] "Example 1"

[1696] (Claim 1)

[1697] a means for recording audio by the communication terminal;

[1698] means for transmitting the recorded voice to an information processing device;

[1699] A means for an information processing device to analyze voice data and extract a voice feature vector;

[1700] A means for determining whether the voice feature vector matches registered data in a database;

[1701] means for transmitting the determination result to the communication terminal;

[1702] A means for displaying the determination result in the communication terminal;

[1703] A system including:

[1704] (Claim 2)

[1705] The system includes a means for a user to record the voice of a specific person when registering the person in the telephone directory, the communication terminal transmitting the recorded voice data to the information processing device, and the information processing device extracting a voice feature vector from the recorded voice data and registering the voice feature vector in the database.

[1706] 10. The system of claim 1.

[1707] (Claim 3)

[1708] The communication terminal records the caller's voice in real time when a call is received, and transmits the recorded voice data to the information processing device, and the information processing device extracts the caller's voice feature vector and compares it with registered voice data to determine whether it matches,

[1709] 10. The system of claim 1.

[1710] "Application Example 1"

[1711] (Claim 1)

[1712] A means for recording audio;

[1713] means for transmitting the recorded voice to an information processing device;

[1714] An information processing device includes means for analyzing voice data and extracting voice features;

[1715] A means for determining whether the voice characteristics match the enrollment data;

[1716] means for transmitting the determination result to the communication terminal;

[1717] A means for displaying the determination result in the communication terminal;

[1718] A means to link with smart devices and display the results in real time,

[1719] A system including:

[1720] (Claim 2)

[1721] The information processing device includes a means for allowing a user to record a voice when registering the telephone directory, for transmitting the recorded voice data to the information processing device, and for extracting voice characteristics from the recorded voice data and registering the voice characteristics in the database.

[1722] 10. The system of claim 1.

[1723] (Claim 3)

[1724] The communication terminal records the caller's voice in real time when an incoming call is received, and transmits the recorded voice data to the information processing device, and the information processing device extracts the caller's voice characteristics and compares them with registered voice data to determine whether they match.

[1725] 10. The system of claim 1.

[1726] (Claim 4)

[1727] and means for displaying the determination result on smart glasses or a head-mounted display.

[1728] 10. The system of claim 1.

[1729] "Example 2: Combining Emotion Engines"

[1730] (Claim 1)

[1731] A means for recording audio;

[1732] means for transmitting the recorded voice to an information processing device;

[1733] An information processing device includes means for analyzing voice data and extracting voice features;

[1734] A means for determining whether the voice characteristics match the enrollment data;

[1735] means for transmitting the determination result to the communication terminal;

[1736] A means for displaying the determination result in the communication terminal;

[1737] means for analyzing an emotional state of a user in the communication terminal;

[1738] means for transmitting the analyzed emotion data to an information processing device;

[1739] means for generating a warning message based on emotion data and transmitting the message to a communication terminal;

[1740] A system including:

[1741] (Claim 2)

[1742] The information processing device includes a means for allowing a user to record a voice when registering the telephone directory, for transmitting the recorded voice data to the information processing device, and for extracting voice characteristics from the recorded voice data and registering the voice characteristics in the database.

[1743] 10. The system of claim 1.

[1744] (Claim 3)

[1745] The communication terminal records the caller's voice in real time when an incoming call is received, and transmits the recorded voice data to the information processing device, and the information processing device extracts the caller's voice characteristics and compares them with registered voice data to determine whether they match.

[1746] 10. The system of claim 1.

[1747] "Application example 2 when combining emotion engines"

[1748] (Claim 1)

[1749] A means for recording audio;

[1750] means for transmitting the recorded voice to an information processing device;

[1751] An information processing device includes means for analyzing voice data and extracting voice features;

[1752] A means for determining whether the voice characteristics match the enrollment data;

[1753] means for transmitting the determination result to the communication terminal;

[1754] A means for displaying the determination result in the communication terminal;

[1755] A means for analyzing the emotional state of a user from the voice during a call;

[1756] means for generating a warning message based on the emotion analysis result and transmitting the warning message to the communication terminal;

[1757] A system including:

[1758] (Claim 2)

[1759] The information processing device includes a means for allowing a user to record a voice when registering the telephone directory, for transmitting the recorded voice data to the information processing device, and for extracting voice characteristics from the recorded voice data and registering the voice characteristics in the database.

[1760] 10. The system of claim 1.

[1761] (Claim 3)

[1762] The communication terminal records the caller's voice in real time when an incoming call is received, and transmits the recorded voice data to the information processing device, and the information processing device extracts the caller's voice characteristics and compares them with registered voice data to determine whether they match.

[1763] 10. The system of claim 1. [Explanation of symbols]

[1764] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for recording audio; means for transmitting the recorded voice to an information processing device; An information processing device includes means for analyzing voice data and extracting voice features; A means for determining whether the voice characteristics match the enrollment data; means for transmitting the determination result to the communication terminal; A means for displaying the determination result in the communication terminal; A system including:

2. The information processing device includes a means for allowing a user to record a voice when registering the telephone directory, for transmitting the recorded voice data to the information processing device, and for extracting voice characteristics from the recorded voice data and registering the voice characteristics in the database. The system of claim 1 .

3. The communication terminal records the caller's voice in real time when an incoming call is received, and transmits the recorded voice data to the information processing device, and the information processing device extracts the caller's voice characteristics and compares them with registered voice data to determine whether they match. The system of claim 1 .

Images