system

Abstract

Provide a system. When a call comes in, an answering machine first answers the call and asks for the caller's identification and requirements; A means for analyzing the content of the caller's response and determining whether the caller is trustworthy; means for blocking calls from callers determined to be untrusted; A system including means for routing calls from callers determined to be trusted to a user.

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] Special frauds and nuisance calls targeting the elderly are on the rise, causing many people to become victims of fraud and feel stressed. The elderly are particularly susceptible to fraud, so measures to prevent this are needed. There is also a need for ways to reduce the hassle of telephone calls in daily life. There is a need to solve these problems and provide an environment where the elderly and general users can use the telephone safely and with peace of mind. [Means for solving the problem]

[0005] The present invention is a system that uses AI for call handling. When a call is received, an answering machine first answers the call and asks for the caller's identification and requirements. The system includes a means for analyzing the caller's response to determine whether the caller is trustworthy, a means for blocking calls from callers determined to be untrustworthy, and a means for transferring calls from callers determined to be trustworthy to the user. Specifically, the system has a means for converting the caller's response into text using a voice recognition system and analyzing the converted text. Furthermore, the system maintains a whitelist and a blacklist, and determines the caller's trustworthiness by comparing the caller's response with these lists. In this way, users can be protected from fraud and nuisance calls, particularly those targeting the elderly.

[0006] An "answering machine" is a device that automatically answers an incoming call and asks for the caller's identification and requirements.

[0007] "Caller" refers to the person or system that placed the call.

[0008] A "voice recognition system" is a system that converts received voice data into text data.

[0009] A "whitelist" is a list of phone numbers or callers that are certified as trustworthy.

[0010] A "blacklist" is a list of phone numbers or callers that have been identified as likely to be spam or fraudulent.

[0011] The "determination means" is a device or algorithm that analyzes the contents of the sender's response and evaluates whether it is trustworthy.

[0012] A "blocking means" is a device or function that automatically disconnects a call if it is determined to be untrustworthy.

[0013] "User" refers to a person who uses the services of this system. [Brief explanation of the drawings]

[0014] [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

[0015] 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.

[0016] First, the terms used in the following description will be explained.

[0017] 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).

[0018] 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.

[0019] 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.

[0020] 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.

[0021] 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."

[0022] [First embodiment]

[0023] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0024] 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.

[0025] 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).

[0026] 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.

[0027] 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.

[0028] 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.

[0029] 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.

[0030] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] 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."

[0035] The present invention is an embodiment of a system using a call handling AI. This system is mainly composed of a server, a terminal, and a user.

[0036] The server immediately receives the signal when a call comes in and establishes a call connection. Once the connection is established, the server activates the AI ​​response module and sends instructions to the terminal.

[0037] The device receives instructions from the server and begins the AI ​​response. Specifically, the device's AI response function asks the caller questions such as "Who is this?" and "What can I do for you?"

[0038] When the caller responds, the voice is sent from the terminal to the server, which then inputs the voice data into a voice recognition system to convert the voice into text, thereby recording the caller's response in text format.

[0039] The server then analyzes the text data and extracts important keywords. This analysis is performed by matching them with whitelists and blacklists. If the caller or requirements match those on the whitelist, the server determines that the call is safe.

[0040] If the call is deemed safe, the server sends that information to the device. The device then displays a notification to the user saying, "This is a trusted caller. Would you like to transfer the call?" The user can then choose to transfer the call or hang up on the device. If the user chooses to transfer the call, the server connects the call directly to the user. Conversely, if the user chooses to hang up, the server ends the call.

[0041] On the other hand, if the caller is on a blacklist or meets other safety criteria, the server will automatically block the call, sending the blocking information to the device, which will then notify the user that the call was disconnected due to possible fraud.

[0042] As a concrete example, consider a caller calling and saying, "This is Tanaka. I'd like to talk to you about insurance." The server converts the speech to text and analyzes the caller's name and requirements. If "Tanaka" or "insurance" is included in the whitelist, the server determines that the call is safe. This information is sent to the device and notifies the user. If the user selects to proceed, the server connects the call to the user's device.

[0043] The processing flow will be explained below.

[0044] Step 1:

[0045] The server receives the signal when a call is received and establishes a telephone connection.

[0046] Step 2:

[0047] When the connection is established, the server activates the AI ​​response module and sends an instruction to the terminal to start the AI ​​response.

[0048] Step 3:

[0049] The device receives instructions from the server and activates the AI ​​response function, asking the caller questions such as "Who is this?" and "What can I do for you?"

[0050] Step 4:

[0051] The terminal receives the caller's response voice and transmits the voice data to the server.

[0052] Step 5:

[0053] The server inputs the received voice data into a voice recognition system and converts the voice into text data.

[0054] Step 6:

[0055] The server analyzes the converted text data, extracts important keywords, and compares the text data with whitelists and blacklists.

[0056] Step 7:

[0057] The server determines a call as safe if the caller or requirements match those on the whitelist, and as a scam or nuisance call if the caller or requirements match those on the blacklist or if there are other safety issues.

[0058] Step 8:

[0059] If it is determined to be safe, the server sends the information to the device and displays a notification to the user saying, "This is a trusted caller. Would you like to transfer the call?"

[0060] Step 9:

[0061] The user selects whether to transfer or disconnect the call on the terminal, and transmits the selection information from the terminal to the server.

[0062] Step 10:

[0063] If the user chooses to transfer, the server connects the call to the user's terminal, if the user chooses to disconnect, the server ends the call.

[0064] Step 11:

[0065] If the call is determined to be fraudulent or spam, the server automatically blocks the call and sends that information to the device, which then notifies the user that the call has been disconnected due to possible fraud.

[0066] Example 1

[0067] 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."

[0068] Conventional call forwarding systems have difficulty determining whether a caller is trustworthy, and lack sufficient means to prevent fraud and unwanted sales calls. Furthermore, manual intervention is required, placing a heavy burden on users. These issues need to be resolved.

[0069] 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.

[0070] In this invention, the server includes means for analyzing the caller's response and determining whether the caller is trustworthy, means for converting the caller's response into text using a voice recognition system, means for maintaining a whitelist and a blacklist and comparing the caller's response with the whitelist and the blacklist, means for connecting the call to the user if the user selects transfer, and means for terminating the call if the user selects disconnection. This makes it possible to automatically block calls from untrustworthy callers, reduce the burden on the user, and provide a safe calling environment.

[0071] An "answering device" is a device that automatically answers an incoming call and asks questions to the caller.

[0072] "Caller" refers to the person who placed the call.

[0073] "Requirements" refers to the purpose and content of the call made by the caller.

[0074] The "determination means" is a means for analyzing the contents of the response from the sender and determining whether the sender is trustworthy.

[0075] "Means for converting to text" refers to a speech recognition system for converting the caller's voice data into text information.

[0076] A "whitelist" is a list that manages information about trusted senders and messages in list format.

[0077] A "blacklist" is a list that manages information about untrustworthy senders and messages in list format.

[0078] The "checking means" is a means having a function of checking the contents of the caller's response against the whitelist and blacklist.

[0079] The "transfer means" is a means for connecting a call to a user when the caller is determined to be a trustworthy caller.

[0080] The "means for terminating a call" refers to the means for terminating a call when the call is from an untrusted caller or when the user chooses to hang up.

[0081] The present invention is a system for automating telephone transfers. This system is composed of a server, a terminal, and a user. Detailed embodiments of this system are described below.

[0082] Server role and hardware / software used

[0083] When a call comes in, the server immediately receives the signal and establishes a telephone connection. After this connection is established, the server activates an AI response module and has the function of sending instructions to the terminal. The specific hardware used is assumed to be a high-performance server. The software uses the Google (registered trademark) Cloud Speech-to-Text API as the voice recognition system.

[0084] When the server receives the caller's voice data, it uses a voice recognition system to convert it into text data. The server then analyzes the text data and extracts important keywords. At this time, the server determines the trustworthiness of the caller by comparing it with a whitelist and blacklist. A whitelist is a list in which trustworthy callers and content are registered in advance, while a blacklist is a list in which untrustworthy callers and content are registered.

[0085] Terminal role and hardware / software used

[0086] The device receives instructions from the server and begins responding with an AI. For example, the device asks the caller questions such as "Who is this?" or "What can I do for you?" This operation is performed by voice using the device's speaker.

[0087] The device also has the role of sending the caller's voice response to the server. It also has the function of receiving notifications from the server, displaying information to the user, and allowing the user to choose whether to transfer the call or disconnect based on that information. The hardware used is a smartphone or tablet, and the custom AI response app runs as software.

[0088] User Roles

[0089] The user can choose whether to transfer the caller or end the call based on the notification displayed on the device. For example, if the device displays "This is a trusted caller. Would you like to transfer the call?", the user can choose to "transfer" or "disconnect." If the user selects "transfer," the server will connect the call directly to the user. Conversely, if the user selects "disconnect," the server will end the call.

[0090] Specific examples

[0091] For example, consider the case where a caller named "Tanaka" calls and says, "I'd like to talk to you about insurance." The server converts this voice into text and extracts the keywords "Tanaka" and "insurance." This is compared with a whitelist, and if it is determined that "Tanaka" and "insurance" are trustworthy, the information is sent to the terminal. The terminal notifies the user, "This is a trustworthy caller. Would you like to transfer the call?" If the user selects "transfer," the server connects the call directly to the user's terminal.

[0092] Prompt Sentence Examples

[0093] Examples of prompts to input to a generative AI model include:

[0094] To the user: This is Tanaka. I'm calling to talk about insurance. Please help me.

[0095] server:

[0096] 1. Convert the caller's voice to text.

[0097] 2. Check "Tanaka" and "Insurance" against the whitelist.

[0098] In this way, the entire system works in coordination and automates call handling, reducing the burden on users and enabling safe and efficient call management.

[0099] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0100] Step 1:

[0101] The server receives telephone signals from the outside. The input is a telephone signal, and the output is the establishment of a telephone connection. Specifically, when a call comes in, the server detects the signal and performs the reception process. This is managed by the server's telephone line interface.

[0102] Step 2:

[0103] After the connection is established, the server activates the AI ​​response module. The input is a notification of the connection establishment, and the output is the start or running status of the AI ​​response module. In effect, the server starts the internal software module and prepares to send instructions to the terminal.

[0104] Step 3:

[0105] The device receives instructions from the server and begins its AI response. The input is the instruction from the server, and the output is the initial response to the caller. Specifically, questions such as "Who is this?" and "What can I do for you?" are asked aloud through the speaker.

[0106] Step 4:

[0107] When the caller responds, the voice data is sent from the terminal to the server. The input is the caller's voice data, and the output is the voice data transferred to the server. The terminal uses a microphone to record the voice and sends the data to the server via the network.

[0108] Step 5:

[0109] The server uses a speech recognition system to convert the received voice data into text format. The input is the caller's voice data, and the output is text data. Specifically, it uses the Google Cloud Speech-to-Text API to convert voice to text in real time.

[0110] Step 6:

[0111] The server analyzes the text data and extracts important keywords. The input is the converted text data, and the output is the extracted keywords. The analysis module in the server extracts keywords such as "name" and "requirements" from the text data.

[0112] Step 7:

[0113] The server checks the extracted keywords against whitelists and blacklists. The input is the extracted keywords, and the output is a trust decision (safe / unsafe). Specifically, the server uses a database query to match the keywords against the lists and see if the caller matches the list.

[0114] Step 8:

[0115] If the server determines that the call is safe, it sends that information to the terminal. The input is the result of the trust determination, and the output is a notification to the terminal. Specifically, a message saying "This caller is trustworthy. Would you like to take the call?" is sent to the terminal.

[0116] Step 9:

[0117] The terminal displays a notification to the user saying "This is a trusted caller. Would you like to take the call?" The input is the notification from the server and the output is what is displayed to the user. The terminal displays the message on the screen and allows the user to make a choice.

[0118] Step 10:

[0119] If the user selects transfer, the server connects the call directly to the user's terminal. The input is the user's selection (transfer), and the output is the establishment of a telephone connection. The server follows the user's instructions and connects the caller and the user directly over the telephone line.

[0120] Step 11:

[0121] If the user selects to disconnect, the server terminates the call. The input is the user's selection (disconnect), and the output is the end of the call. The server terminates the call and notifies the terminal of this information.

[0122] (Application example 1)

[0123] 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."

[0124] Food delivery services are faced with a large number of inquiries regarding delivery status and order details, placing a heavy burden on customer support. There are also many cases of fraudulent calls from unreliable callers. This has led to concerns about a decline in customer satisfaction and a deterioration in operational efficiency. Traditional manual response methods have their limitations, creating a need for an efficient and secure telephone response system.

[0125] 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.

[0126] In this invention, the server includes means for first answering an incoming call with an answering device and asking for caller identification and requirements, means for analyzing the content of the caller's response and determining whether the caller is trustworthy, means for blocking calls from callers determined to be untrustworthy, means for transferring calls from callers determined to be trustworthy to a user, means for automatically converting the delivery status into text and notifying the user if the trustworthy caller's requirements are related to food delivery, and means for determining whether a specified caller or requirements fall under a whitelist or blacklist and taking appropriate action, thereby enabling efficient and safe telephone response.

[0127] An "answering machine" is a device that initially answers an incoming call and asks for the caller's identification and requirements.

[0128] "Caller response" refers to the response given by the caller, which is usually in audio form.

[0129] The "means for analyzing" is a means for analyzing the content of the caller's response and evaluating and classifying the information.

[0130] The "means for determining trustworthiness" is a means for determining whether the sender is trustworthy based on the analyzed data.

[0131] A "blocking measure" is a measure for automatically blocking calls from callers determined to be untrustworthy.

[0132] "Transfer means" refers to a means for connecting a call from a caller determined to be trustworthy to a user.

[0133] "Food delivery related" refers to when the trusted sender requirement relates to food delivery services.

[0134] "Means for automatically converting delivery status into text and notifying the user" refers to a means for converting information such as the progress of delivery from voice to text and notifying the user of that text information.

[0135] A "whitelist" is a list of trusted senders and requirements that are registered in advance.

[0136] A "blacklist" is a list of untrustworthy senders and requirements that are registered in advance.

[0137] "Appropriate action" is the process of determining whether to route or block a call based on a whitelist or blacklist.

[0138] This invention relates to a telephone handling AI system for food delivery services. This system is composed of a server, a terminal, and a user (hereinafter referred to as "user").

[0139] Server Features

[0140] When a call comes in, the server immediately receives the signal and establishes a telephone connection. Once the connection is established, the server activates the answering machine and asks the caller for their identification and requirements. For example, the server asks questions such as "Who is this?" and "What can I do for you?" This voice response is realized using an AI module.

[0141] When the caller responds, their voice is sent through the device to the server, which uses a speech recognition system (e.g., Google Speech Recognition API) to convert this voice data into text.

[0142] The server then analyzes the text data obtained and extracts important keywords. This analysis is performed by checking against a whitelist and blacklist. If the caller or requirements match those on the whitelist, the server determines that the call is safe. If it is determined to be safe, the server sends this information to the terminal and notifies the user. At this time, a notification is displayed saying, "This is a trusted caller. Would you like to transfer the call?"

[0143] If the trusted caller requirement is related to food delivery, the server will automatically transcribe the delivery status into text and notify the user. If the specified caller or requirement is on a blacklist, the server will automatically block the call and notify the user that the call was disconnected due to possible fraud.

[0144] Device Features

[0145] The device receives instructions from the server and initiates an AI response. For example, if a user is using a food delivery app, a notification will appear within the app, providing the caller's details and requirements. The user can choose to accept or disconnect the call. If the user chooses to accept the call, the server will connect the call directly to the user's device.

[0146] Examples of use cases and generated AI prompts

[0147] As a concrete example, consider the case where a caller calls and says, "This is Sato. I'd like to confirm the arrival time of my pizza order." The server converts the speech into text and analyzes the caller's name and requirements. If "Sato" or "pizza" is included in the whitelist, the server determines that the call is safe. This information is sent to the terminal and notifies the user. If the user selects to transfer the call, the server connects the call to the user's terminal.

[0148] An example of a prompt sentence for the generative AI model input is as follows:

[0149] "This is Sato. I'd like to confirm the arrival time of the pizza I ordered."

[0150] As described above, this system enables efficient and safe telephone response in food delivery services.

[0151] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0152] Step 1:

[0153] The server receives the telephone signal: the input is the telephone signal from the caller, and the output is the establishment of a telephone connection.

[0154] Step 2:

[0155] The server activates an answering machine and asks for the caller's identification and requirements. The input is an established telephone connection and the output is a spoken question such as "Who is this?" or "What can I do for you?"

[0156] Step 3:

[0157] The terminal receives the caller's response in voice format and transmits the voice to the server. The input is the caller's voice data, and the output is the data to be transmitted to the server.

[0158] Step 4:

[0159] The server uses a speech recognition system to convert the received voice data into text. The input is voice data and the output is text data. Specifically, the server converts voice to text using the Google Speech Recognition API.

[0160] Step 5:

[0161] The server analyzes the text data and extracts important keywords. This analysis is performed by matching the data with whitelists and blacklists. The input is the converted text data, and the output is the analysis result (trustworthiness assessment).

[0162] Step 6:

[0163] If the server finds a caller or requirement that matches the whitelist, it determines that the call is safe. The input is the analysis result, and the output is the safety judgment "safe."

[0164] Step 7:

[0165] The server sends the information judged to be safe to the terminal and displays a notification to the user saying, "This is a trusted caller. Would you like to take the call?" The input is the result of the safety judgment, and the output is a notification to the user.

[0166] Step 8:

[0167] The user selects whether to transfer or disconnect. The input is the user's selection, and the output is an instruction to transfer or disconnect.

[0168] Step 9:

[0169] If the user selects transfer, the server connects the call directly to the user's terminal. The input is the user's transfer instruction, and the output is the establishment of a telephone connection.

[0170] Step 10:

[0171] If the trusted caller requirement is related to food delivery, the server automatically converts the delivery status into text and notifies the user. The input is the caller's response, and the output is the text of the delivery status and its notification.

[0172] Step 11:

[0173] On the other hand, if the specified caller or requirement is on the blacklist, the server automatically blocks the call and notifies the user, "The call was disconnected due to possible fraud." The input is the analysis result (blacklist match), and the output is the call blocking and notification to the user.

[0174] 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.

[0175] This invention is a system that uses AI to handle calls, and by combining it with an emotion engine that recognizes the user's emotions, it achieves even more advanced filtering. The system is mainly composed of a server, a terminal, and an emotion engine.

[0176] When a call comes in, the server immediately receives the signal, establishes a telephone connection, activates the AI ​​response module, and sends instructions to the terminal. The terminal receives instructions from the server, activates the AI ​​response function, and asks the caller questions such as "Who is this?" and "What can I do for you?"

[0177] The terminal receives the caller's voice response and sends it to the server. The server inputs the received voice into a voice recognition system and converts the voice into text data. The server analyzes the converted text data and compares it with whitelists and blacklists to extract important keywords. If the analysis shows that the caller is on the whitelist, the call is deemed safe.

[0178] Here, the emotion engine detects the user's emotional state in real time. The emotion engine recognizes emotions such as stress, anxiety, relief, and comfort from the user's voice and input data. The server receives the user's emotional state from the emotion engine.

[0179] Even if the server determines the caller is safe, if the emotion engine detects stress or anxiety in the user, the server will automatically block the call and send that information to the device, which will then notify the user, "The call has been disconnected due to a possible fraudulent activity."

[0180] Conversely, even if the caller is on a blacklist or it is difficult to determine whether the call is authentic or not, if the emotion engine senses a sense of security and comfort in the user, the server will transfer the call to the user. The device will display a notification to the user saying, "This is a trusted caller. Would you like to transfer?" and give the user the option to transfer or disconnect. If the user selects to transfer, the server will connect the call to the user.

[0181] As a concrete example, consider the case where a caller calls and says, "This is Tanaka. I'd like to talk to you about insurance." The server converts the speech to text and analyzes "Tanaka" and "insurance." If there are no matching keywords in the whitelist, the emotion engine detects the user's emotions. If the user expresses relief or comfort, the call is transferred. Conversely, if the user feels stressed or anxious, the call is blocked regardless of the caller's trustworthiness.

[0182] This system offers a higher level of security and user experience than conventional methods, and is expected to keep all users, including the elderly, safe from scams and nuisance calls.

[0183] The processing flow will be explained below.

[0184] Step 1:

[0185] The server receives the signal when a call is received and establishes a telephone connection.

[0186] Step 2:

[0187] When the connection is established, the server activates the AI ​​response module and sends an instruction to the terminal to start the AI ​​response.

[0188] Step 3:

[0189] The device receives instructions from the server and activates the AI ​​response function, asking the caller questions such as "Who is this?" and "What can I do for you?"

[0190] Step 4:

[0191] The terminal receives the caller's response voice and transmits the voice data to the server.

[0192] Step 5:

[0193] The server inputs the received voice data into a voice recognition system and converts the voice into text data.

[0194] Step 6:

[0195] The server analyzes the converted text data and extracts important keywords. For example, from a response such as "I'm Tanaka. I'd like to talk to you about insurance," the server extracts the keywords "Tanaka" and "insurance."

[0196] Step 7:

[0197] The server checks the extracted keywords against a whitelist and a blacklist to determine whether the sender is trustworthy.

[0198] Step 8:

[0199] The emotion engine analyzes the user's voice and input data in real time to recognize the user's emotional state (e.g., relief, excitement, stress, anxiety, etc.). The emotion engine's results are sent to the server.

[0200] Step 9:

[0201] Even if the server determines that the caller is safe, if the emotion engine detects stress or anxiety in the user, the server will automatically block the call and send that information to the device, which will then notify the user, "The call has been disconnected due to a possible fraudulent activity."

[0202] Step 10:

[0203] Conversely, even if the caller is on a blacklist or has questionable reliability, if the emotion engine makes the user feel safe and comfortable, the server will transfer the call to the user. The device will display a notification to the user saying, "This is a trusted caller. Would you like to transfer the call?"

[0204] Step 11:

[0205] The user selects whether to transfer the call or to disconnect, and the selection information is sent from the terminal to the server.

[0206] Step 12:

[0207] If the user chooses to be transferred, the server connects the call to the user's terminal.

[0208] Step 13:

[0209] If the user chooses to hang up, the server will end the call and blacklist the call if necessary.

[0210] Example 2

[0211] 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."

[0212] While conventional telephone answering systems can verify the caller and ask about their requirements, they have limitations in accurately determining the caller's trustworthiness, and in particular cannot take into account the user's emotional state, making it impossible to completely eliminate the risk of fraud or nuisance calls. Furthermore, there is a need for an answering system that gives users, including the elderly, a sense of security. Therefore, a system is needed that can evaluate both the caller's trustworthiness and the user's emotional state in real time and appropriately transfer or block calls.

[0213] 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.

[0214] In this invention, the server includes: means for first answering an incoming call with an answering device and asking for the caller's identification and requirements; means for analyzing the caller's response and determining whether the caller is trustworthy; means for blocking calls from callers determined to be untrustworthy; means for transferring calls from callers determined to be trustworthy to a user; and means for detecting the user's emotional state in real time and blocking or transferring calls based on the user's emotional state, which allows the server to evaluate not only the caller's trustworthiness but also the user's emotional state, further reducing the risk of fraud and nuisance calls and providing users with a sense of security.

[0215] An "answering device" is a device that automatically answers an incoming call and asks for the caller's identification and request.

[0216] "Means for identifying and inquiring about the caller" refers to the means for questioning the caller about their identity and purpose when receiving a call.

[0217] "Analysis" is the act of understanding and evaluating the sender's information and requirements based on the received voice and text data.

[0218] The "means for determining whether the sender is trustworthy" is a means for analyzing the contents of the sender's response and evaluating whether the contents are trustworthy.

[0219] A "call blocking measure" is a measure for automatically blocking a call if the caller is determined to be untrustworthy.

[0220] "Means for transferring calls to a user" refers to means for transferring calls from callers determined to be trustworthy to a user so that the user can answer the call.

[0221] A "voice recognition system" is a technology for converting received voice data into text data.

[0222] A "whitelist" is a list of trusted senders that is used to evaluate the trustworthiness of senders based on that list.

[0223] A "blacklist" is a list of untrustworthy senders, and is used to evaluate the trustworthiness of senders based on that list.

[0224] An "emotion recognition engine" is a technology that analyzes the user's voice and input data to detect the user's emotional state (stress, anxiety, relief, comfort, etc.) in real time.

[0225] A "means for blocking or forwarding calls based on emotional state" is a means for evaluating a user's emotional state in real time and forwarding or blocking calls based on the evaluation result.

[0226] This invention is an advanced filtering system that combines a call handling AI with an emotion recognition engine that recognizes user emotions. This system is mainly composed of a server, a terminal, and an emotion recognition engine. The specific operation and configuration of the system are explained in detail below.

[0227] Server configuration and functions

[0228] The server is equipped with a communication module for receiving telephone signals. When a telephone signal is received, the server immediately identifies the signal and establishes a telephone connection. After this, the server activates the AI ​​response module and sends instructions to the terminal.

[0229] The AI ​​response module has a script to ask the caller for confirmation and requirements, and the server sends instructions to the terminal to provide a standard response, such as "Who is this?" or "What can I do for you?"

[0230] Device configuration and functions

[0231] The terminal receives instructions from the server, asks questions to the caller according to those instructions, receives a voice response from the caller and sends it to the server. The terminal also has a display and speaker to notify the user of the server's judgment results.

[0232] Emotion recognition engine configuration and functions

[0233] The emotion recognition engine analyzes the user's voice and input data to detect emotions such as stress, anxiety, relief, comfort, etc. in real time. This engine can be built using, for example, IBM Watson's (registered trademark) Tone Analyzer API.

[0234] Server processing

[0235] The server inputs the received caller's voice into a speech recognition system within the program and converts it into text data, possibly using a system such as Google Cloud Speech-to-Text API. The converted text data is then compared against whitelists and blacklists to extract important keywords.

[0236] Based on the results of analyzing the text data, the server evaluates the trustworthiness of the sender. In addition, it obtains the user's emotional state from an emotion recognition engine and makes a final decision taking the user's emotional state into consideration.

[0237] Judgment results and notification to users

[0238] Even if the server determines that the caller is trustworthy, if the emotion recognition engine detects stress or anxiety in the user, the server will automatically block the call and send that information to the device, which will then notify the user, "The call has been disconnected due to a possible fraudulent activity."

[0239] Conversely, even if the caller is on a blacklist or it is difficult to determine whether the call is authentic or authentic, if the emotion recognition engine senses a sense of security or comfort in the user, the server will transfer the call to the user. The device will display a notification to the user saying, "This is a trusted caller. Would you like to transfer?" and give the user the option to transfer or disconnect. If the user chooses to transfer, the server will connect the call to the user.

[0240] Specific examples

[0241] As a concrete example, consider a caller calling and saying, "This is Tanaka. I'd like to talk to you about insurance." In this case, the server converts the speech to text and analyzes "Tanaka" and "insurance." If there are no matching keywords in the whitelist, the emotion recognition engine detects the user's emotions. If the user expresses relief or comfort, the call is transferred. Conversely, if the user feels stressed or anxious, the call is blocked regardless of the caller's trustworthiness.

[0242] Example prompts for generative AI models

[0243] Below are some example prompts to input to a generative AI model:

[0244] "Please explain the system that analyzes the emotional feedback from the user and the contents of the phone call to determine safety. Also, please add the following concrete example: Based on the scenario, "This is Tanaka. I'd like to talk to you about insurance," please explain the processing flow for when the user indicates safety and when they indicate anxiety."

[0245] This system offers higher levels of security and a better user experience than traditional methods, and is expected to help all users, including the elderly, stay safe from scams and spam calls.

[0246] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0247] Step 1:

[0248] The server receives the call signal and establishes a connection

[0249] Input: External telephone signal

[0250] Specific operations: The server receives an external telephone signal, analyzes the received signal, extracts identifying information, and records it in a log. Then, it establishes a telephone connection so that the call can begin.

[0251] Output: The phone connection is established and its status is logged.

[0252] Step 2:

[0253] The server activates the AI ​​response module

[0254] Input: Telephone connection establishment information

[0255] Specific operation: When a telephone connection is established, the server starts the AI ​​response module and logs the message "The response module has been started." Then, the AI ​​response module loads a script and sends instructions for the initial response to the caller to the terminal.

[0256] Output: The AI ​​response module is activated and the instructions are sent to the terminal.

[0257] Step 3:

[0258] The device receives instructions from the AI ​​response module and asks the caller a question.

[0259] Input: Indicates the initial response from the server

[0260] Specific operation: The device follows the instructions of the initial response received from the server and asks the caller questions such as "Who is this?" and "What can I do for you?" The voice response is generated based on a pre-defined script.

[0261] Output: Audio response from the caller.

[0262] Step 4:

[0263] The device receives the caller's response and sends it to the server.

[0264] Input: Voice response from the caller

[0265] Specific operation: The terminal receives a voice response from the caller and sends the voice data to the server, where it converts it into an appropriate format to prevent data degradation.

[0266] Output: The audio data sent to the server.

[0267] Step 5:

[0268] The server inputs the received voice into a voice recognition system and converts it into text data.

[0269] Input: Audio data sent from the device

[0270] Specific operation: The server inputs the received voice data into the voice recognition system, which uses the Google Cloud Speech-to-Text API to convert the voice data into text data. At this time, the voice recognition system performs phonemic analysis and converts it into corresponding text.

[0271] Output: The caller's response converted into text data.

[0272] Step 6:

[0273] The server parses the text data and checks it against the whitelist and blacklist.

[0274] Input: Text data from a speech recognition system

[0275] How it works: The server analyzes the converted text data and extracts important keywords. It then checks these keywords against a whitelist and blacklist to evaluate the trustworthiness of the sender. For example, it checks whether the text contains keywords like "Tanaka" or "insurance."

[0276] Output: The result of the evaluation of the sender's trustworthiness.

[0277] Step 7:

[0278] The server obtains the user's emotional state from the emotion recognition engine.

[0279] Input: Real-time user voice and typing data

[0280] How it works: The emotion recognition engine analyzes the user's voice and input data to detect emotional states such as stress, anxiety, relief, and comfort. The server periodically retrieves this emotional data from the emotion recognition engine, for example, using IBM Watson's Tone Analyzer API.

[0281] Output: User's emotional state data.

[0282] Step 8:

[0283] The server makes the final decision based on the sender's credibility and the user's emotional state.

[0284] Input: Evaluation results on the sender's trustworthiness and user's emotional state data

[0285] Specific operation: The server integrates the caller's trustworthiness assessment and the user's emotional state to make the final decision on whether to forward or block the call. For example, if the caller is not on the whitelist but the user feels safe, the call will be forwarded. Conversely, if the caller is on the whitelist but the user feels stressed, the call will be blocked.

[0286] Output: Instructions for transferring or blocking the call.

[0287] Step 9:

[0288] The device sends the first notification to the user

[0289] Input: Call forwarding or blocking instructions from the server

[0290] Specific behavior: The device receives instructions from the server and notifies the user of the result, for example, displaying a message such as "Call has been disconnected due to possible fraud" or "Trusted caller. Would you like to proceed?"

[0291] Output: The notification message displayed to the user.

[0292] Step 10:

[0293] User chooses to transfer or block the call

[0294] Input: Notification message displayed on the terminal

[0295] Specific operation: The user can choose to transfer or disconnect the call based on the message displayed on the device, using the buttons or touch screen on the device.

[0296] Output: Data about the user's selection.

[0297] Step 11:

[0298] The server takes the final action based on the user's selection.

[0299] Input: Data about user selections

[0300] Specific behavior: If the user selects to transfer the call, the server receives the instruction and connects the call to the user. Conversely, if the user selects to hang up the call, the server immediately hangs up the call.

[0301] Output: Calls are connected or disconnected, and the status is logged.

[0302] Through the above processing steps, the system comprehensively assesses both the caller's credibility and the user's emotional state, and then optimally routes or blocks the call.

[0303] (Application example 2)

[0304] 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."

[0305] The present invention aims to improve security and user experience in conventional telephone filtering systems. In particular, it aims to further reduce the risk of fraud and nuisance calls by considering the user's emotional state in addition to determining the trustworthiness of the caller. It also aims to provide an environment where users can receive calls with peace of mind.

[0306] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: a means for an answering device to answer an incoming call and inquire about the caller's identity and requirements; a means for analyzing the caller's response and determining whether the caller is trustworthy; a means for blocking calls from callers determined to be untrustworthy; a means for transferring calls from callers determined to be trustworthy to a user; an emotion engine for detecting the user's emotional state and filtering calls based on the emotional state; a means for transferring calls when the user indicates relief or comfort and blocking calls when the user indicates stress or anxiety; a means for converting the caller's response into text using a voice recognition system, the determination means analyzing the converted text and determining whether the caller is trustworthy; a whitelist and a blacklist maintained by the determination means for comparing the caller's response against the whitelist and blacklist; and a means for analyzing the user's emotional state using a generative AI model and generating prompt sentences to obtain an emotion analysis result. This allows calls to be appropriately filtered based on the user's emotional state, protecting users from spam calls and scams.

[0307] An "answering machine" is a device that automatically answers an incoming call and asks for the caller's identification and requirements.

[0308] "Caller identification and inquiry measures" are measures to verify the identity of the caller and the caller's intent and purpose.

[0309] The "means for analyzing the contents of the caller's response" is a means for analyzing the contents of the caller's response and determining whether the caller is trustworthy.

[0310] "Means for blocking calls from callers determined to be untrustworthy" refers to means for blocking calls from callers determined to be untrustworthy without forwarding the calls to the user.

[0311] The "means for transferring a call from a caller determined to be trustworthy to a user" refers to a means for connecting a call to a user when the caller is determined to be trustworthy.

[0312] The "emotion engine" is an engine that analyzes the user's emotional state and detects emotional data in real time.

[0313] A "means for filtering calls based on emotional state" is a means for determining whether to pass or block a call depending on the emotional state of a user.

[0314] A "voice recognition system" is a system that converts voice data into text data.

[0315] A "whitelist" is a list of trusted callers from which calls are allowed.

[0316] A "blacklist" is a list of untrusted callers from which calls are blocked.

[0317] A "generative AI model" is an artificial intelligence model that analyzes a user's emotional state, generates prompt sentences, and performs emotion analysis.

[0318] A "prompt sentence" is text data that is input into the generative AI model to perform sentiment analysis.

[0319] This invention is an advanced filtering system that combines a call handling AI system with an emotion engine. This system is mainly composed of a server, a terminal, and an emotion engine.

[0320] When the server receives a call, it immediately receives the signal and establishes a telephone connection. Next, the server activates the AI ​​response module and sends instructions to the terminal. The terminal receives the instructions from the server, activates the AI ​​response function, and asks the caller questions such as "Who is this?" and "What can I do for you?"

[0321] The terminal receives the caller's voice response and sends it to the server. The server inputs the received voice into a voice recognition system and converts the voice into text data. The server analyzes the converted text data and compares it with a whitelist and blacklist to extract important keywords. In this case, the SpeechRecognition library is used as the voice recognition system, and natural language processing is used for text analysis.

[0322] If the analysis finds that the caller is on the whitelist, the call is deemed safe. The emotion engine then detects the user's emotional state in real time. The emotion engine uses a generative AI model using Hugging Face's Transformers library. The emotion engine recognizes emotions such as stress, anxiety, relief, and comfort from the user's voice and input data.

[0323] The server receives the user's emotional state from the emotion engine, and even if the call is deemed safe, if it detects stress or anxiety in the user, it automatically blocks the call and sends that information to the device, which then notifies the user, "The call has been disconnected due to a possible fraudulent activity."

[0324] Conversely, even if the caller is on the blacklist, if the emotion engine makes the user feel safe and comfortable, the server will transfer the call to the user. The device will display a notification to the user saying "This is a trusted caller. Would you like to transfer?" and give the user the option to transfer or disconnect. If the user chooses to transfer, the server will connect the call to the user.

[0325] As a concrete example, consider a situation where a user receives keywords such as "insurance" and "important offer." At this time, the server converts the speech into text and inputs the text data into the emotion engine. If the emotion engine detects negative emotions such as "anxiety" or "stress," the call will be blocked. An example of a prompt sentence would be "I'd like to talk to you about insurance, and I have a really tailored offer for you."

[0326] In this way, the present invention is a system that realizes advanced filtering based on user sentiment and whitelists / blacklists to protect users from spam calls and scams.

[0327] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0328] Step 1:

[0329] The server receives the call and immediately receives the signal to establish a telephone connection, at which point the server activates the AI ​​response module and sends a response instruction to the terminal.

[0330] Step 2:

[0331] The device receives instructions from the server and activates the AI ​​response function, which asks the caller questions such as "Who is this?" and "What can I do for you?", and collects the caller's voice response.

[0332] Step 3:

[0333] The device sends the collected response voice of the caller to the server, which converts this voice data into text data using the SpeechRecognition library. Input: Voice data, Output: Text data.

[0334] Step 4:

[0335] The server analyzes the converted text data and extracts important keywords by matching them with the whitelist and blacklist. Input: text data, whitelist, blacklist. Output: keyword matching results.

[0336] Step 5:

[0337] The server uses a generative AI model to analyze the user's emotional state (stress, anxiety, relief, comfort, etc.) from the text data. Input: Text data, Output: Emotion analysis results.

[0338] Step 6:

[0339] The server decides whether to allow or block a call based on the emotion analysis results obtained from the emotion engine. Specifically, even if a call falls under the whitelist and is deemed safe, the call will be blocked if the user is feeling stressed or anxious. Conversely, even if a call falls under the blacklist, the call will be allowed if the user is feeling safe or comfortable.

[0340] Step 7:

[0341] The device notifies the user of the server's decision. If the user feels stressed or anxious, the device will notify them, "The call has been disconnected due to a possible scam." If the user feels safe or comfortable, the device will display a message to the user saying, "This is a trusted caller. Would you like to transfer the call?" and offer the user the option of transferring the call or disconnecting.

[0342] Step 8:

[0343] If the user selects to transfer, the server connects the call to the user. Conversely, if the user selects to hang up, the server terminates the call. Input: User selection, Output: Call connection or termination.

[0344] 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.

[0345] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0346] 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.

[0347] [Second embodiment]

[0348] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0349] 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.

[0350] 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).

[0351] 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.

[0352] 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.

[0353] 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).

[0354] 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.

[0355] 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.

[0356] 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.

[0357] 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.

[0358] 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.

[0359] 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."

[0360] The present invention is an embodiment of a system using a call handling AI. This system is mainly composed of a server, a terminal, and a user.

[0361] The server immediately receives the signal when a call comes in and establishes a call connection. Once the connection is established, the server activates the AI ​​response module and sends instructions to the terminal.

[0362] The device receives instructions from the server and begins the AI ​​response. Specifically, the device's AI response function asks the caller questions such as "Who is this?" and "What can I do for you?"

[0363] When the caller responds, the voice is sent from the terminal to the server, which then inputs the voice data into a voice recognition system to convert the voice into text, thereby recording the caller's response in text format.

[0364] The server then analyzes the text data and extracts important keywords. This analysis is performed by matching them with whitelists and blacklists. If the caller or requirements match those on the whitelist, the server determines that the call is safe.

[0365] If the call is deemed safe, the server sends that information to the device. The device then displays a notification to the user saying, "This is a trusted caller. Would you like to transfer the call?" The user can then choose to transfer the call or hang up on the device. If the user chooses to transfer the call, the server connects the call directly to the user. Conversely, if the user chooses to hang up, the server ends the call.

[0366] On the other hand, if the caller is on a blacklist or meets other safety criteria, the server will automatically block the call, sending the blocking information to the device, which will then notify the user that the call was disconnected due to possible fraud.

[0367] As a concrete example, consider a caller calling and saying, "This is Tanaka. I'd like to talk to you about insurance." The server converts the speech to text and analyzes the caller's name and requirements. If "Tanaka" or "insurance" is included in the whitelist, the server determines that the call is safe. This information is sent to the device and notifies the user. If the user selects to proceed, the server connects the call to the user's device.

[0368] The processing flow will be explained below.

[0369] Step 1:

[0370] The server receives the signal when a call is received and establishes a telephone connection.

[0371] Step 2:

[0372] When the connection is established, the server activates the AI ​​response module and sends an instruction to the terminal to start the AI ​​response.

[0373] Step 3:

[0374] The device receives instructions from the server and activates the AI ​​response function, asking the caller questions such as "Who is this?" and "What can I do for you?"

[0375] Step 4:

[0376] The terminal receives the caller's response voice and transmits the voice data to the server.

[0377] Step 5:

[0378] The server inputs the received voice data into a voice recognition system and converts the voice into text data.

[0379] Step 6:

[0380] The server analyzes the converted text data, extracts important keywords, and compares the text data with whitelists and blacklists.

[0381] Step 7:

[0382] The server determines a call as safe if the caller or requirements match those on the whitelist, and as a scam or nuisance call if the caller or requirements match those on the blacklist or if there are other safety issues.

[0383] Step 8:

[0384] If it is determined to be safe, the server sends the information to the device and displays a notification to the user saying, "This is a trusted caller. Would you like to transfer the call?"

[0385] Step 9:

[0386] The user selects whether to transfer or disconnect the call on the terminal, and transmits the selection information from the terminal to the server.

[0387] Step 10:

[0388] If the user chooses to transfer, the server connects the call to the user's terminal, if the user chooses to disconnect, the server ends the call.

[0389] Step 11:

[0390] If the call is determined to be fraudulent or spam, the server automatically blocks the call and sends that information to the device, which then notifies the user that the call has been disconnected due to possible fraud.

[0391] Example 1

[0392] 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."

[0393] Conventional call forwarding systems have difficulty determining whether a caller is trustworthy, and lack sufficient means to prevent fraud and unwanted sales calls. Furthermore, manual intervention is required, placing a heavy burden on users. These issues need to be resolved.

[0394] 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.

[0395] In this invention, the server includes means for analyzing the caller's response and determining whether the caller is trustworthy, means for converting the caller's response into text using a voice recognition system, means for maintaining a whitelist and a blacklist and comparing the caller's response with the whitelist and the blacklist, means for connecting the call to the user if the user selects transfer, and means for terminating the call if the user selects disconnection. This makes it possible to automatically block calls from untrustworthy callers, reduce the burden on the user, and provide a safe calling environment.

[0396] An "answering device" is a device that automatically answers an incoming call and asks questions to the caller.

[0397] "Caller" refers to the person who placed the call.

[0398] "Requirements" refers to the purpose and content of the call made by the caller.

[0399] The "determination means" is a means for analyzing the contents of the response from the sender and determining whether the sender is trustworthy.

[0400] "Means for converting to text" refers to a speech recognition system for converting the caller's voice data into text information.

[0401] A "whitelist" is a list that manages information about trusted senders and messages in list format.

[0402] A "blacklist" is a list that manages information about untrustworthy senders and messages in list format.

[0403] The "checking means" is a means having a function of checking the contents of the caller's response against the whitelist and blacklist.

[0404] The "transfer means" is a means for connecting a call to a user when the caller is determined to be a trustworthy caller.

[0405] The "means for terminating a call" refers to the means for terminating a call when the call is from an untrusted caller or when the user chooses to hang up.

[0406] The present invention is a system for automating telephone transfers. This system is composed of a server, a terminal, and a user. Detailed embodiments of this system are described below.

[0407] Server role and hardware / software used

[0408] When a call comes in, the server immediately receives the signal and establishes a telephone connection. After this connection is established, the server activates the AI ​​response module and has the function of sending instructions to the terminal. The specific hardware used is a high-performance server. The software uses the Google Cloud Speech-to-Text API as the voice recognition system.

[0409] When the server receives the caller's voice data, it uses a voice recognition system to convert it into text data. The server then analyzes the text data and extracts important keywords. At this time, the server determines the trustworthiness of the caller by comparing it with a whitelist and blacklist. A whitelist is a list in which trustworthy callers and content are registered in advance, while a blacklist is a list in which untrustworthy callers and content are registered.

[0410] Terminal role and hardware / software used

[0411] The device receives instructions from the server and begins responding with an AI. For example, the device asks the caller questions such as "Who is this?" or "What can I do for you?" This operation is performed by voice using the device's speaker.

[0412] The device also has the role of sending the caller's voice response to the server. It also has the function of receiving notifications from the server, displaying information to the user, and allowing the user to choose whether to transfer the call or disconnect based on that information. The hardware used is a smartphone or tablet, and the custom AI response app runs as software.

[0413] User Roles

[0414] The user can choose whether to transfer the caller or end the call based on the notification displayed on the device. For example, if the device displays "This is a trusted caller. Would you like to transfer the call?", the user can choose to "transfer" or "disconnect." If the user selects "transfer," the server will connect the call directly to the user. Conversely, if the user selects "disconnect," the server will end the call.

[0415] Specific examples

[0416] For example, consider the case where a caller named "Tanaka" calls and says, "I'd like to talk to you about insurance." The server converts this voice into text and extracts the keywords "Tanaka" and "insurance." This is compared with a whitelist, and if it is determined that "Tanaka" and "insurance" are trustworthy, the information is sent to the terminal. The terminal notifies the user, "This is a trustworthy caller. Would you like to transfer the call?" If the user selects "transfer," the server connects the call directly to the user's terminal.

[0417] Prompt Sentence Examples

[0418] Examples of prompts to input to a generative AI model include:

[0419] To the user: This is Tanaka. I'm calling to talk about insurance. Please help me.

[0420] server:

[0421] 1. Convert the caller's voice to text.

[0422] 2. Check "Tanaka" and "Insurance" against the whitelist.

[0423] In this way, the entire system works in coordination and automates call handling, reducing the burden on users and enabling safe and efficient call management.

[0424] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0425] Step 1:

[0426] The server receives telephone signals from the outside. The input is a telephone signal, and the output is the establishment of a telephone connection. Specifically, when a call comes in, the server detects the signal and performs the reception process. This is managed by the server's telephone line interface.

[0427] Step 2:

[0428] After the connection is established, the server activates the AI ​​response module. The input is a notification of the connection establishment, and the output is the start or running status of the AI ​​response module. In effect, the server starts the internal software module and prepares to send instructions to the terminal.

[0429] Step 3:

[0430] The device receives instructions from the server and begins its AI response. The input is the instruction from the server, and the output is the initial response to the caller. Specifically, questions such as "Who is this?" and "What can I do for you?" are asked aloud through the speaker.

[0431] Step 4:

[0432] When the caller responds, the voice data is sent from the terminal to the server. The input is the caller's voice data, and the output is the voice data transferred to the server. The terminal uses a microphone to record the voice and sends the data to the server via the network.

[0433] Step 5:

[0434] The server uses a speech recognition system to convert the received voice data into text format. The input is the caller's voice data, and the output is text data. Specifically, it uses the Google Cloud Speech-to-Text API to convert voice to text in real time.

[0435] Step 6:

[0436] The server analyzes the text data and extracts important keywords. The input is the converted text data, and the output is the extracted keywords. The analysis module in the server extracts keywords such as "name" and "requirements" from the text data.

[0437] Step 7:

[0438] The server checks the extracted keywords against whitelists and blacklists. The input is the extracted keywords, and the output is a trust decision (safe / unsafe). Specifically, the server uses a database query to match the keywords against the lists and see if the caller matches the list.

[0439] Step 8:

[0440] If the server determines that the call is safe, it sends that information to the terminal. The input is the result of the trust determination, and the output is a notification to the terminal. Specifically, a message saying "This caller is trustworthy. Would you like to take the call?" is sent to the terminal.

[0441] Step 9:

[0442] The terminal displays a notification to the user saying "This is a trusted caller. Would you like to take the call?" The input is the notification from the server and the output is what is displayed to the user. The terminal displays the message on the screen and allows the user to make a choice.

[0443] Step 10:

[0444] If the user selects transfer, the server connects the call directly to the user's terminal. The input is the user's selection (transfer), and the output is the establishment of a telephone connection. The server follows the user's instructions and connects the caller and the user directly over the telephone line.

[0445] Step 11:

[0446] If the user selects to disconnect, the server terminates the call. The input is the user's selection (disconnect), and the output is the end of the call. The server terminates the call and notifies the terminal of this information.

[0447] (Application example 1)

[0448] 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."

[0449] Food delivery services are faced with a large number of inquiries regarding delivery status and order details, placing a heavy burden on customer support. There are also many cases of fraudulent calls from unreliable callers. This has led to concerns about a decline in customer satisfaction and a deterioration in operational efficiency. Traditional manual response methods have their limitations, creating a need for an efficient and secure telephone response system.

[0450] 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.

[0451] In this invention, the server includes means for first answering an incoming call with an answering device and asking for caller identification and requirements, means for analyzing the content of the caller's response and determining whether the caller is trustworthy, means for blocking calls from callers determined to be untrustworthy, means for transferring calls from callers determined to be trustworthy to a user, means for automatically converting the delivery status into text and notifying the user if the trustworthy caller's requirements are related to food delivery, and means for determining whether a specified caller or requirements fall under a whitelist or blacklist and taking appropriate action, thereby enabling efficient and safe telephone response.

[0452] An "answering machine" is a device that initially answers an incoming call and asks for the caller's identification and requirements.

[0453] "Caller response" refers to the response given by the caller, which is usually in audio form.

[0454] The "means for analyzing" is a means for analyzing the content of the caller's response and evaluating and classifying the information.

[0455] The "means for determining trustworthiness" is a means for determining whether the sender is trustworthy based on the analyzed data.

[0456] A "blocking measure" is a measure for automatically blocking calls from callers determined to be untrustworthy.

[0457] "Transfer means" refers to a means for connecting a call from a caller determined to be trustworthy to a user.

[0458] "Food delivery related" refers to when the trusted sender requirement relates to food delivery services.

[0459] "Means for automatically converting delivery status into text and notifying the user" refers to a means for converting information such as the progress of delivery from voice to text and notifying the user of that text information.

[0460] A "whitelist" is a list of trusted senders and requirements that are registered in advance.

[0461] A "blacklist" is a list of untrustworthy senders and requirements that are registered in advance.

[0462] "Appropriate action" is the process of determining whether to route or block a call based on a whitelist or blacklist.

[0463] This invention relates to a telephone handling AI system for food delivery services. This system is composed of a server, a terminal, and a user (hereinafter referred to as "user").

[0464] Server Features

[0465] When a call comes in, the server immediately receives the signal and establishes a telephone connection. Once the connection is established, the server activates the answering machine and asks the caller for their identification and requirements. For example, the server asks questions such as "Who is this?" and "What can I do for you?" This voice response is realized using an AI module.

[0466] When the caller responds, their voice is sent through the device to the server, which uses a speech recognition system (e.g., Google Speech Recognition API) to convert this voice data into text.

[0467] The server then analyzes the text data obtained and extracts important keywords. This analysis is performed by checking against a whitelist and blacklist. If the caller or requirements match those on the whitelist, the server determines that the call is safe. If it is determined to be safe, the server sends this information to the terminal and notifies the user. At this time, a notification is displayed saying, "This is a trusted caller. Would you like to transfer the call?"

[0468] If the trusted caller requirement is related to food delivery, the server will automatically transcribe the delivery status into text and notify the user. If the specified caller or requirement is on a blacklist, the server will automatically block the call and notify the user that the call was disconnected due to possible fraud.

[0469] Device Features

[0470] The device receives instructions from the server and initiates an AI response. For example, if a user is using a food delivery app, a notification will appear within the app, providing the caller's details and requirements. The user can choose to accept or disconnect the call. If the user chooses to accept the call, the server will connect the call directly to the user's device.

[0471] Examples of use cases and generated AI prompts

[0472] As a concrete example, consider the case where a caller calls and says, "This is Sato. I'd like to confirm the arrival time of my pizza order." The server converts the speech into text and analyzes the caller's name and requirements. If "Sato" or "pizza" is included in the whitelist, the server determines that the call is safe. This information is sent to the terminal and notifies the user. If the user selects to transfer the call, the server connects the call to the user's terminal.

[0473] An example of a prompt sentence for the generative AI model input is as follows:

[0474] "This is Sato. I'd like to confirm the arrival time of the pizza I ordered."

[0475] As described above, this system enables efficient and safe telephone response in food delivery services.

[0476] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0477] Step 1:

[0478] The server receives the telephone signal: the input is the telephone signal from the caller, and the output is the establishment of a telephone connection.

[0479] Step 2:

[0480] The server activates an answering machine and asks for the caller's identification and requirements. The input is an established telephone connection and the output is a spoken question such as "Who is this?" or "What can I do for you?"

[0481] Step 3:

[0482] The terminal receives the caller's response in voice format and transmits the voice to the server. The input is the caller's voice data, and the output is the data to be transmitted to the server.

[0483] Step 4:

[0484] The server uses a speech recognition system to convert the received voice data into text. The input is voice data and the output is text data. Specifically, the server converts voice to text using the Google Speech Recognition API.

[0485] Step 5:

[0486] The server analyzes the text data and extracts important keywords. This analysis is performed by matching the data with whitelists and blacklists. The input is the converted text data, and the output is the analysis result (trustworthiness assessment).

[0487] Step 6:

[0488] If the server finds a caller or requirement that matches the whitelist, it determines that the call is safe. The input is the analysis result, and the output is the safety judgment "safe."

[0489] Step 7:

[0490] The server sends the information judged to be safe to the terminal and displays a notification to the user saying, "This is a trusted caller. Would you like to take the call?" The input is the result of the safety judgment, and the output is a notification to the user.

[0491] Step 8:

[0492] The user selects whether to transfer or disconnect. The input is the user's selection, and the output is an instruction to transfer or disconnect.

[0493] Step 9:

[0494] If the user selects transfer, the server connects the call directly to the user's terminal. The input is the user's transfer instruction, and the output is the establishment of a telephone connection.

[0495] Step 10:

[0496] If the trusted caller requirement is related to food delivery, the server automatically converts the delivery status into text and notifies the user. The input is the caller's response, and the output is the text of the delivery status and its notification.

[0497] Step 11:

[0498] On the other hand, if the specified caller or requirement is on the blacklist, the server automatically blocks the call and notifies the user, "The call was disconnected due to possible fraud." The input is the analysis result (blacklist match), and the output is the call blocking and notification to the user.

[0499] 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.

[0500] This invention is a system that uses AI to handle calls, and by combining it with an emotion engine that recognizes the user's emotions, it achieves even more advanced filtering. The system is mainly composed of a server, a terminal, and an emotion engine.

[0501] When a call comes in, the server immediately receives the signal, establishes a telephone connection, activates the AI ​​response module, and sends instructions to the terminal. The terminal receives instructions from the server, activates the AI ​​response function, and asks the caller questions such as "Who is this?" and "What can I do for you?"

[0502] The terminal receives the caller's voice response and sends it to the server. The server inputs the received voice into a voice recognition system and converts the voice into text data. The server analyzes the converted text data and compares it with whitelists and blacklists to extract important keywords. If the analysis shows that the caller is on the whitelist, the call is deemed safe.

[0503] Here, the emotion engine detects the user's emotional state in real time. The emotion engine recognizes emotions such as stress, anxiety, relief, and comfort from the user's voice and input data. The server receives the user's emotional state from the emotion engine.

[0504] Even if the server determines the caller is safe, if the emotion engine detects stress or anxiety in the user, the server will automatically block the call and send that information to the device, which will then notify the user, "The call has been disconnected due to a possible fraudulent activity."

[0505] Conversely, even if the caller is on a blacklist or it is difficult to determine whether the call is authentic or not, if the emotion engine senses a sense of security and comfort in the user, the server will transfer the call to the user. The device will display a notification to the user saying, "This is a trusted caller. Would you like to transfer?" and give the user the option to transfer or disconnect. If the user selects to transfer, the server will connect the call to the user.

[0506] As a concrete example, consider the case where a caller calls and says, "This is Tanaka. I'd like to talk to you about insurance." The server converts the speech to text and analyzes "Tanaka" and "insurance." If there are no matching keywords in the whitelist, the emotion engine detects the user's emotions. If the user expresses relief or comfort, the call is transferred. Conversely, if the user feels stressed or anxious, the call is blocked regardless of the caller's trustworthiness.

[0507] This system offers a higher level of security and user experience than conventional methods, and is expected to keep all users, including the elderly, safe from scams and nuisance calls.

[0508] The processing flow will be explained below.

[0509] Step 1:

[0510] The server receives the signal when a call is received and establishes a telephone connection.

[0511] Step 2:

[0512] When the connection is established, the server activates the AI ​​response module and sends an instruction to the terminal to start the AI ​​response.

[0513] Step 3:

[0514] The device receives instructions from the server and activates the AI ​​response function, asking the caller questions such as "Who is this?" and "What can I do for you?"

[0515] Step 4:

[0516] The terminal receives the caller's response voice and transmits the voice data to the server.

[0517] Step 5:

[0518] The server inputs the received voice data into a voice recognition system and converts the voice into text data.

[0519] Step 6:

[0520] The server analyzes the converted text data and extracts important keywords. For example, from a response such as "I'm Tanaka. I'd like to talk to you about insurance," the server extracts the keywords "Tanaka" and "insurance."

[0521] Step 7:

[0522] The server checks the extracted keywords against a whitelist and a blacklist to determine whether the sender is trustworthy.

[0523] Step 8:

[0524] The emotion engine analyzes the user's voice and input data in real time to recognize the user's emotional state (e.g., relief, excitement, stress, anxiety, etc.). The emotion engine's results are sent to the server.

[0525] Step 9:

[0526] Even if the server determines that the caller is safe, if the emotion engine detects stress or anxiety in the user, the server will automatically block the call and send that information to the device, which will then notify the user, "The call has been disconnected due to a possible fraudulent activity."

[0527] Step 10:

[0528] Conversely, even if the caller is on a blacklist or has questionable reliability, if the emotion engine makes the user feel safe and comfortable, the server will transfer the call to the user. The device will display a notification to the user saying, "This is a trusted caller. Would you like to transfer the call?"

[0529] Step 11:

[0530] The user selects whether to transfer the call or to disconnect, and the selection information is sent from the terminal to the server.

[0531] Step 12:

[0532] If the user chooses to be transferred, the server connects the call to the user's terminal.

[0533] Step 13:

[0534] If the user chooses to hang up, the server will end the call and blacklist the call if necessary.

[0535] Example 2

[0536] 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."

[0537] While conventional telephone answering systems can verify the caller and ask about their requirements, they have limitations in accurately determining the caller's trustworthiness, and in particular cannot take into account the user's emotional state, making it impossible to completely eliminate the risk of fraud or nuisance calls. Furthermore, there is a need for an answering system that gives users, including the elderly, a sense of security. Therefore, a system is needed that can evaluate both the caller's trustworthiness and the user's emotional state in real time and appropriately transfer or block calls.

[0538] 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.

[0539] In this invention, the server includes: means for first answering an incoming call with an answering device and asking for the caller's identification and requirements; means for analyzing the caller's response and determining whether the caller is trustworthy; means for blocking calls from callers determined to be untrustworthy; means for transferring calls from callers determined to be trustworthy to a user; and means for detecting the user's emotional state in real time and blocking or transferring calls based on the user's emotional state, which allows the server to evaluate not only the caller's trustworthiness but also the user's emotional state, further reducing the risk of fraud and nuisance calls and providing users with a sense of security.

[0540] An "answering device" is a device that automatically answers an incoming call and asks for the caller's identification and request.

[0541] "Means for identifying and inquiring about the caller" refers to the means for questioning the caller about their identity and purpose when receiving a call.

[0542] "Analysis" is the act of understanding and evaluating the sender's information and requirements based on the received voice and text data.

[0543] The "means for determining whether the sender is trustworthy" is a means for analyzing the contents of the sender's response and evaluating whether the contents are trustworthy.

[0544] A "call blocking measure" is a measure for automatically blocking a call if the caller is determined to be untrustworthy.

[0545] "Means for transferring calls to a user" refers to means for transferring calls from callers determined to be trustworthy to a user so that the user can answer the call.

[0546] A "voice recognition system" is a technology for converting received voice data into text data.

[0547] A "whitelist" is a list of trusted senders that is used to evaluate the trustworthiness of senders based on that list.

[0548] A "blacklist" is a list of untrustworthy senders, and is used to evaluate the trustworthiness of senders based on that list.

[0549] An "emotion recognition engine" is a technology that analyzes the user's voice and input data to detect the user's emotional state (stress, anxiety, relief, comfort, etc.) in real time.

[0550] A "means for blocking or forwarding calls based on emotional state" is a means for evaluating a user's emotional state in real time and forwarding or blocking calls based on the evaluation result.

[0551] This invention is an advanced filtering system that combines a call handling AI with an emotion recognition engine that recognizes user emotions. This system is mainly composed of a server, a terminal, and an emotion recognition engine. The specific operation and configuration of the system are explained in detail below.

[0552] Server configuration and functions

[0553] The server is equipped with a communication module for receiving telephone signals. When a telephone signal is received, the server immediately identifies the signal and establishes a telephone connection. After this, the server activates the AI ​​response module and sends instructions to the terminal.

[0554] The AI ​​response module has a script to ask the caller for confirmation and requirements, and the server sends instructions to the terminal to provide a standard response, such as "Who is this?" or "What can I do for you?"

[0555] Device configuration and functions

[0556] The terminal receives instructions from the server, asks questions to the caller according to those instructions, receives a voice response from the caller and sends it to the server. The terminal also has a display and speaker to notify the user of the server's judgment results.

[0557] Emotion recognition engine configuration and functions

[0558] The emotion recognition engine analyzes user voice and input data to detect emotions such as stress, anxiety, relief, comfort, etc. in real time. This engine can be built using, for example, IBM Watson's Tone Analyzer API.

[0559] Server processing

[0560] The server inputs the received caller's voice into a speech recognition system within the program and converts it into text data, possibly using a system such as Google Cloud Speech-to-Text API. The converted text data is then compared against whitelists and blacklists to extract important keywords.

[0561] Based on the results of analyzing the text data, the server evaluates the trustworthiness of the sender. In addition, it obtains the user's emotional state from an emotion recognition engine and makes a final decision taking the user's emotional state into consideration.

[0562] Judgment results and notification to users

[0563] Even if the server determines that the caller is trustworthy, if the emotion recognition engine detects stress or anxiety in the user, the server will automatically block the call and send that information to the device, which will then notify the user, "The call has been disconnected due to a possible fraudulent activity."

[0564] Conversely, even if the caller is on a blacklist or it is difficult to determine whether the call is authentic or authentic, if the emotion recognition engine senses a sense of security or comfort in the user, the server will transfer the call to the user. The device will display a notification to the user saying, "This is a trusted caller. Would you like to transfer?" and give the user the option to transfer or disconnect. If the user chooses to transfer, the server will connect the call to the user.

[0565] Specific examples

[0566] As a concrete example, consider a caller calling and saying, "This is Tanaka. I'd like to talk to you about insurance." In this case, the server converts the speech to text and analyzes "Tanaka" and "insurance." If there are no matching keywords in the whitelist, the emotion recognition engine detects the user's emotions. If the user expresses relief or comfort, the call is transferred. Conversely, if the user feels stressed or anxious, the call is blocked regardless of the caller's trustworthiness.

[0567] Example prompts for generative AI models

[0568] Below are some example prompts to input to a generative AI model:

[0569] "Please explain the system that analyzes the emotional feedback from the user and the contents of the phone call to determine safety. Also, please add the following concrete example: Based on the scenario, "This is Tanaka. I'd like to talk to you about insurance," please explain the processing flow for when the user indicates safety and when they indicate anxiety."

[0570] This system offers higher levels of security and a better user experience than traditional methods, and is expected to help all users, including the elderly, stay safe from scams and spam calls.

[0571] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0572] Step 1:

[0573] The server receives the call signal and establishes a connection

[0574] Input: External telephone signal

[0575] Specific operations: The server receives an external telephone signal, analyzes the received signal, extracts identifying information, and records it in a log. Then, it establishes a telephone connection so that the call can begin.

[0576] Output: The phone connection is established and its status is logged.

[0577] Step 2:

[0578] The server activates the AI ​​response module

[0579] Input: Telephone connection establishment information

[0580] Specific operation: When a telephone connection is established, the server starts the AI ​​response module and logs the message "The response module has been started." Then, the AI ​​response module loads a script and sends instructions for the initial response to the caller to the terminal.

[0581] Output: The AI ​​response module is activated and the instructions are sent to the terminal.

[0582] Step 3:

[0583] The device receives instructions from the AI ​​response module and asks the caller a question.

[0584] Input: Indicates the initial response from the server

[0585] Specific operation: The device follows the instructions of the initial response received from the server and asks the caller questions such as "Who is this?" and "What can I do for you?" The voice response is generated based on a pre-defined script.

[0586] Output: Audio response from the caller.

[0587] Step 4:

[0588] The device receives the caller's response and sends it to the server.

[0589] Input: Voice response from the caller

[0590] Specific operation: The terminal receives a voice response from the caller and sends the voice data to the server, where it converts it into an appropriate format to prevent data degradation.

[0591] Output: The audio data sent to the server.

[0592] Step 5:

[0593] The server inputs the received voice into a voice recognition system and converts it into text data.

[0594] Input: Audio data sent from the device

[0595] Specific operation: The server inputs the received voice data into the voice recognition system, which uses the Google Cloud Speech-to-Text API to convert the voice data into text data. At this time, the voice recognition system performs phonemic analysis and converts it into corresponding text.

[0596] Output: The caller's response converted into text data.

[0597] Step 6:

[0598] The server parses the text data and checks it against the whitelist and blacklist.

[0599] Input: Text data from a speech recognition system

[0600] How it works: The server analyzes the converted text data and extracts important keywords. It then checks these keywords against a whitelist and blacklist to evaluate the trustworthiness of the sender. For example, it checks whether the text contains keywords like "Tanaka" or "insurance."

[0601] Output: The result of the evaluation of the sender's trustworthiness.

[0602] Step 7:

[0603] The server obtains the user's emotional state from the emotion recognition engine.

[0604] Input: Real-time user voice and typing data

[0605] How it works: The emotion recognition engine analyzes the user's voice and input data to detect emotional states such as stress, anxiety, relief, and comfort. The server periodically retrieves this emotional data from the emotion recognition engine, for example, using IBM Watson's Tone Analyzer API.

[0606] Output: User's emotional state data.

[0607] Step 8:

[0608] The server makes the final decision based on the sender's credibility and the user's emotional state.

[0609] Input: Evaluation results on the sender's trustworthiness and user's emotional state data

[0610] Specific operation: The server integrates the caller's trustworthiness assessment and the user's emotional state to make the final decision on whether to forward or block the call. For example, if the caller is not on the whitelist but the user feels safe, the call will be forwarded. Conversely, if the caller is on the whitelist but the user feels stressed, the call will be blocked.

[0611] Output: Instructions for transferring or blocking the call.

[0612] Step 9:

[0613] The device sends the first notification to the user

[0614] Input: Call forwarding or blocking instructions from the server

[0615] Specific behavior: The device receives instructions from the server and notifies the user of the result, for example, displaying a message such as "Call has been disconnected due to possible fraud" or "Trusted caller. Would you like to proceed?"

[0616] Output: The notification message displayed to the user.

[0617] Step 10:

[0618] User chooses to transfer or block the call

[0619] Input: Notification message displayed on the terminal

[0620] Specific operation: The user can choose to transfer or disconnect the call based on the message displayed on the device, using the buttons or touch screen on the device.

[0621] Output: Data about the user's selection.

[0622] Step 11:

[0623] The server takes the final action based on the user's selection.

[0624] Input: Data about user selections

[0625] Specific behavior: If the user selects to transfer the call, the server receives the instruction and connects the call to the user. Conversely, if the user selects to hang up the call, the server immediately hangs up the call.

[0626] Output: Calls are connected or disconnected, and the status is logged.

[0627] Through the above processing steps, the system comprehensively assesses both the caller's credibility and the user's emotional state, and then optimally routes or blocks the call.

[0628] (Application example 2)

[0629] 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."

[0630] The present invention aims to improve security and user experience in conventional telephone filtering systems. In particular, it aims to further reduce the risk of fraud and nuisance calls by considering the user's emotional state in addition to determining the trustworthiness of the caller. It also aims to provide an environment where users can receive calls with peace of mind.

[0631] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: a means for an answering device to answer an incoming call and inquire about the caller's identity and requirements; a means for analyzing the caller's response and determining whether the caller is trustworthy; a means for blocking calls from callers determined to be untrustworthy; a means for transferring calls from callers determined to be trustworthy to a user; an emotion engine for detecting the user's emotional state and filtering calls based on the emotional state; a means for transferring calls when the user indicates relief or comfort and blocking calls when the user indicates stress or anxiety; a means for converting the caller's response into text using a voice recognition system, the determination means analyzing the converted text and determining whether the caller is trustworthy; a whitelist and a blacklist maintained by the determination means for comparing the caller's response against the whitelist and blacklist; and a means for analyzing the user's emotional state using a generative AI model and generating prompt sentences to obtain an emotion analysis result. This allows calls to be appropriately filtered based on the user's emotional state, protecting users from spam calls and scams.

[0632] An "answering machine" is a device that automatically answers an incoming call and asks for the caller's identification and requirements.

[0633] "Caller identification and inquiry measures" are measures to verify the identity of the caller and the caller's intent and purpose.

[0634] The "means for analyzing the contents of the caller's response" is a means for analyzing the contents of the caller's response and determining whether the caller is trustworthy.

[0635] "Means for blocking calls from callers determined to be untrustworthy" refers to means for blocking calls from callers determined to be untrustworthy without forwarding the calls to the user.

[0636] The "means for transferring a call from a caller determined to be trustworthy to a user" refers to a means for connecting a call to a user when the caller is determined to be trustworthy.

[0637] The "emotion engine" is an engine that analyzes the user's emotional state and detects emotional data in real time.

[0638] A "means for filtering calls based on emotional state" is a means for determining whether to pass or block a call depending on the emotional state of a user.

[0639] A "voice recognition system" is a system that converts voice data into text data.

[0640] A "whitelist" is a list of trusted callers from which calls are allowed.

[0641] A "blacklist" is a list of untrusted callers from which calls are blocked.

[0642] A "generative AI model" is an artificial intelligence model that analyzes a user's emotional state, generates prompt sentences, and performs emotion analysis.

[0643] A "prompt sentence" is text data that is input into the generative AI model to perform sentiment analysis.

[0644] This invention is an advanced filtering system that combines a call handling AI system with an emotion engine. This system is mainly composed of a server, a terminal, and an emotion engine.

[0645] When the server receives a call, it immediately receives the signal and establishes a telephone connection. Next, the server activates the AI ​​response module and sends instructions to the terminal. The terminal receives the instructions from the server, activates the AI ​​response function, and asks the caller questions such as "Who is this?" and "What can I do for you?"

[0646] The terminal receives the caller's voice response and sends it to the server. The server inputs the received voice into a voice recognition system and converts the voice into text data. The server analyzes the converted text data and compares it with a whitelist and blacklist to extract important keywords. In this case, the SpeechRecognition library is used as the voice recognition system, and natural language processing is used for text analysis.

[0647] If the analysis finds that the caller is on the whitelist, the call is deemed safe. The emotion engine then detects the user's emotional state in real time. The emotion engine uses a generative AI model using Hugging Face's Transformers library. The emotion engine recognizes emotions such as stress, anxiety, relief, and comfort from the user's voice and input data.

[0648] The server receives the user's emotional state from the emotion engine, and even if the call is deemed safe, if it detects stress or anxiety in the user, it automatically blocks the call and sends that information to the device, which then notifies the user, "The call has been disconnected due to a possible fraudulent activity."

[0649] Conversely, even if the caller is on the blacklist, if the emotion engine makes the user feel safe and comfortable, the server will transfer the call to the user. The device will display a notification to the user saying "This is a trusted caller. Would you like to transfer?" and give the user the option to transfer or disconnect. If the user chooses to transfer, the server will connect the call to the user.

[0650] As a concrete example, consider a situation where a user receives keywords such as "insurance" and "important offer." At this time, the server converts the speech into text and inputs the text data into the emotion engine. If the emotion engine detects negative emotions such as "anxiety" or "stress," the call will be blocked. An example of a prompt sentence would be "I'd like to talk to you about insurance, and I have a really tailored offer for you."

[0651] In this way, the present invention is a system that realizes advanced filtering based on user sentiment and whitelists / blacklists to protect users from spam calls and scams.

[0652] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0653] Step 1:

[0654] The server receives the call and immediately receives the signal to establish a telephone connection, at which point the server activates the AI ​​response module and sends a response instruction to the terminal.

[0655] Step 2:

[0656] The device receives instructions from the server and activates the AI ​​response function, which asks the caller questions such as "Who is this?" and "What can I do for you?", and collects the caller's voice response.

[0657] Step 3:

[0658] The device sends the collected response voice of the caller to the server, which converts this voice data into text data using the SpeechRecognition library. Input: Voice data, Output: Text data.

[0659] Step 4:

[0660] The server analyzes the converted text data and extracts important keywords by matching them with the whitelist and blacklist. Input: text data, whitelist, blacklist. Output: keyword matching results.

[0661] Step 5:

[0662] The server uses a generative AI model to analyze the user's emotional state (stress, anxiety, relief, comfort, etc.) from the text data. Input: Text data, Output: Emotion analysis results.

[0663] Step 6:

[0664] The server decides whether to allow or block a call based on the emotion analysis results obtained from the emotion engine. Specifically, even if a call falls under the whitelist and is deemed safe, the call will be blocked if the user is feeling stressed or anxious. Conversely, even if a call falls under the blacklist, the call will be allowed if the user is feeling safe or comfortable.

[0665] Step 7:

[0666] The device notifies the user of the server's decision. If the user feels stressed or anxious, the device will notify them, "The call has been disconnected due to a possible scam." If the user feels safe or comfortable, the device will display a message to the user saying, "This is a trusted caller. Would you like to transfer the call?" and offer the user the option of transferring the call or disconnecting.

[0667] Step 8:

[0668] If the user selects to transfer, the server connects the call to the user. Conversely, if the user selects to hang up, the server terminates the call. Input: User selection, Output: Call connection or termination.

[0669] 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.

[0670] 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.

[0671] 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.

[0672] [Third embodiment]

[0673] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0674] 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.

[0675] 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).

[0676] 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.

[0677] 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.

[0678] 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).

[0679] 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.

[0680] 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.

[0681] 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.

[0682] 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.

[0683] 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.

[0684] 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."

[0685] The present invention is an embodiment of a system using a call handling AI. This system is mainly composed of a server, a terminal, and a user.

[0686] The server immediately receives the signal when a call comes in and establishes a call connection. Once the connection is established, the server activates the AI ​​response module and sends instructions to the terminal.

[0687] The device receives instructions from the server and begins the AI ​​response. Specifically, the device's AI response function asks the caller questions such as "Who is this?" and "What can I do for you?"

[0688] When the caller responds, the voice is sent from the terminal to the server, which then inputs the voice data into a voice recognition system to convert the voice into text, thereby recording the caller's response in text format.

[0689] The server then analyzes the text data and extracts important keywords. This analysis is performed by matching them with whitelists and blacklists. If the caller or requirements match those on the whitelist, the server determines that the call is safe.

[0690] If the call is deemed safe, the server sends that information to the device. The device then displays a notification to the user saying, "This is a trusted caller. Would you like to transfer the call?" The user can then choose to transfer the call or hang up on the device. If the user chooses to transfer the call, the server connects the call directly to the user. Conversely, if the user chooses to hang up, the server ends the call.

[0691] On the other hand, if the caller is on a blacklist or meets other safety criteria, the server will automatically block the call, sending the blocking information to the device, which will then notify the user that the call was disconnected due to possible fraud.

[0692] As a concrete example, consider a caller calling and saying, "This is Tanaka. I'd like to talk to you about insurance." The server converts the speech to text and analyzes the caller's name and requirements. If "Tanaka" or "insurance" is included in the whitelist, the server determines that the call is safe. This information is sent to the device and notifies the user. If the user selects to proceed, the server connects the call to the user's device.

[0693] The processing flow will be explained below.

[0694] Step 1:

[0695] The server receives the signal when a call is received and establishes a telephone connection.

[0696] Step 2:

[0697] When the connection is established, the server activates the AI ​​response module and sends an instruction to the terminal to start the AI ​​response.

[0698] Step 3:

[0699] The device receives instructions from the server and activates the AI ​​response function, asking the caller questions such as "Who is this?" and "What can I do for you?"

[0700] Step 4:

[0701] The terminal receives the caller's response voice and transmits the voice data to the server.

[0702] Step 5:

[0703] The server inputs the received voice data into a voice recognition system and converts the voice into text data.

[0704] Step 6:

[0705] The server analyzes the converted text data, extracts important keywords, and compares the text data with whitelists and blacklists.

[0706] Step 7:

[0707] The server determines a call as safe if the caller or requirements match those on the whitelist, and as a scam or nuisance call if the caller or requirements match those on the blacklist or if there are other safety issues.

[0708] Step 8:

[0709] If it is determined to be safe, the server sends the information to the device and displays a notification to the user saying, "This is a trusted caller. Would you like to transfer the call?"

[0710] Step 9:

[0711] The user selects whether to transfer or disconnect the call on the terminal, and transmits the selection information from the terminal to the server.

[0712] Step 10:

[0713] If the user chooses to transfer, the server connects the call to the user's terminal, if the user chooses to disconnect, the server ends the call.

[0714] Step 11:

[0715] If the call is determined to be fraudulent or spam, the server automatically blocks the call and sends that information to the device, which then notifies the user that the call has been disconnected due to possible fraud.

[0716] Example 1

[0717] 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."

[0718] Conventional call forwarding systems have difficulty determining whether a caller is trustworthy, and lack sufficient means to prevent fraud and unwanted sales calls. Furthermore, manual intervention is required, placing a heavy burden on users. These issues need to be resolved.

[0719] 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.

[0720] In this invention, the server includes means for analyzing the caller's response and determining whether the caller is trustworthy, means for converting the caller's response into text using a voice recognition system, means for maintaining a whitelist and a blacklist and comparing the caller's response with the whitelist and the blacklist, means for connecting the call to the user if the user selects transfer, and means for terminating the call if the user selects disconnection. This makes it possible to automatically block calls from untrustworthy callers, reduce the burden on the user, and provide a safe calling environment.

[0721] An "answering device" is a device that automatically answers an incoming call and asks questions to the caller.

[0722] "Caller" refers to the person who placed the call.

[0723] "Requirements" refers to the purpose and content of the call made by the caller.

[0724] The "determination means" is a means for analyzing the contents of the response from the sender and determining whether the sender is trustworthy.

[0725] "Means for converting to text" refers to a speech recognition system for converting the caller's voice data into text information.

[0726] A "whitelist" is a list that manages information about trusted senders and messages in list format.

[0727] A "blacklist" is a list that manages information about untrustworthy senders and messages in list format.

[0728] The "checking means" is a means having a function of checking the contents of the caller's response against the whitelist and blacklist.

[0729] The "transfer means" is a means for connecting a call to a user when the caller is determined to be a trustworthy caller.

[0730] The "means for terminating a call" refers to the means for terminating a call when the call is from an untrusted caller or when the user chooses to hang up.

[0731] The present invention is a system for automating telephone transfers. This system is composed of a server, a terminal, and a user. Detailed embodiments of this system are described below.

[0732] Server role and hardware / software used

[0733] When a call comes in, the server immediately receives the signal and establishes a telephone connection. After this connection is established, the server activates the AI ​​response module and has the function of sending instructions to the terminal. The specific hardware used is a high-performance server. The software uses the Google Cloud Speech-to-Text API as the voice recognition system.

[0734] When the server receives the caller's voice data, it uses a voice recognition system to convert it into text data. The server then analyzes the text data and extracts important keywords. At this time, the server determines the trustworthiness of the caller by comparing it with a whitelist and blacklist. A whitelist is a list in which trustworthy callers and content are registered in advance, while a blacklist is a list in which untrustworthy callers and content are registered.

[0735] Terminal role and hardware / software used

[0736] The device receives instructions from the server and begins responding with an AI. For example, the device asks the caller questions such as "Who is this?" or "What can I do for you?" This operation is performed by voice using the device's speaker.

[0737] The device also has the role of sending the caller's voice response to the server. It also has the function of receiving notifications from the server, displaying information to the user, and allowing the user to choose whether to transfer the call or disconnect based on that information. The hardware used is a smartphone or tablet, and the custom AI response app runs as software.

[0738] User Roles

[0739] The user can choose whether to transfer the caller or end the call based on the notification displayed on the device. For example, if the device displays "This is a trusted caller. Would you like to transfer the call?", the user can choose to "transfer" or "disconnect." If the user selects "transfer," the server will connect the call directly to the user. Conversely, if the user selects "disconnect," the server will end the call.

[0740] Specific examples

[0741] For example, consider the case where a caller named "Tanaka" calls and says, "I'd like to talk to you about insurance." The server converts this voice into text and extracts the keywords "Tanaka" and "insurance." This is compared with a whitelist, and if it is determined that "Tanaka" and "insurance" are trustworthy, the information is sent to the terminal. The terminal notifies the user, "This is a trustworthy caller. Would you like to transfer the call?" If the user selects "transfer," the server connects the call directly to the user's terminal.

[0742] Prompt Sentence Examples

[0743] Examples of prompts to input to a generative AI model include:

[0744] To the user: This is Tanaka. I'm calling to talk about insurance. Please help me.

[0745] server:

[0746] 1. Convert the caller's voice to text.

[0747] 2. Check "Tanaka" and "Insurance" against the whitelist.

[0748] In this way, the entire system works in coordination and automates call handling, reducing the burden on users and enabling safe and efficient call management.

[0749] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0750] Step 1:

[0751] The server receives telephone signals from the outside. The input is a telephone signal, and the output is the establishment of a telephone connection. Specifically, when a call comes in, the server detects the signal and performs the reception process. This is managed by the server's telephone line interface.

[0752] Step 2:

[0753] After the connection is established, the server activates the AI ​​response module. The input is a notification of the connection establishment, and the output is the start or running status of the AI ​​response module. In effect, the server starts the internal software module and prepares to send instructions to the terminal.

[0754] Step 3:

[0755] The device receives instructions from the server and begins its AI response. The input is the instruction from the server, and the output is the initial response to the caller. Specifically, questions such as "Who is this?" and "What can I do for you?" are asked aloud through the speaker.

[0756] Step 4:

[0757] When the caller responds, the voice data is sent from the terminal to the server. The input is the caller's voice data, and the output is the voice data transferred to the server. The terminal uses a microphone to record the voice and sends the data to the server via the network.

[0758] Step 5:

[0759] The server uses a speech recognition system to convert the received voice data into text format. The input is the caller's voice data, and the output is text data. Specifically, it uses the Google Cloud Speech-to-Text API to convert voice to text in real time.

[0760] Step 6:

[0761] The server analyzes the text data and extracts important keywords. The input is the converted text data, and the output is the extracted keywords. The analysis module in the server extracts keywords such as "name" and "requirements" from the text data.

[0762] Step 7:

[0763] The server checks the extracted keywords against whitelists and blacklists. The input is the extracted keywords, and the output is a trust decision (safe / unsafe). Specifically, the server uses a database query to match the keywords against the lists and see if the caller matches the list.

[0764] Step 8:

[0765] If the server determines that the call is safe, it sends that information to the terminal. The input is the result of the trust determination, and the output is a notification to the terminal. Specifically, a message saying "This caller is trustworthy. Would you like to take the call?" is sent to the terminal.

[0766] Step 9:

[0767] The terminal displays a notification to the user saying "This is a trusted caller. Would you like to take the call?" The input is the notification from the server and the output is what is displayed to the user. The terminal displays the message on the screen and allows the user to make a choice.

[0768] Step 10:

[0769] If the user selects transfer, the server connects the call directly to the user's terminal. The input is the user's selection (transfer), and the output is the establishment of a telephone connection. The server follows the user's instructions and connects the caller and the user directly over the telephone line.

[0770] Step 11:

[0771] If the user selects to disconnect, the server terminates the call. The input is the user's selection (disconnect), and the output is the end of the call. The server terminates the call and notifies the terminal of this information.

[0772] (Application example 1)

[0773] 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."

[0774] Food delivery services are faced with a large number of inquiries regarding delivery status and order details, placing a heavy burden on customer support. There are also many cases of fraudulent calls from unreliable callers. This has led to concerns about a decline in customer satisfaction and a deterioration in operational efficiency. Traditional manual response methods have their limitations, creating a need for an efficient and secure telephone response system.

[0775] 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.

[0776] In this invention, the server includes means for first answering an incoming call with an answering device and asking for caller identification and requirements, means for analyzing the content of the caller's response and determining whether the caller is trustworthy, means for blocking calls from callers determined to be untrustworthy, means for transferring calls from callers determined to be trustworthy to a user, means for automatically converting the delivery status into text and notifying the user if the trustworthy caller's requirements are related to food delivery, and means for determining whether a specified caller or requirements fall under a whitelist or blacklist and taking appropriate action, thereby enabling efficient and safe telephone response.

[0777] An "answering machine" is a device that initially answers an incoming call and asks for the caller's identification and requirements.

[0778] "Caller response" refers to the response given by the caller, which is usually in audio form.

[0779] The "means for analyzing" is a means for analyzing the content of the caller's response and evaluating and classifying the information.

[0780] The "means for determining trustworthiness" is a means for determining whether the sender is trustworthy based on the analyzed data.

[0781] A "blocking measure" is a measure for automatically blocking calls from callers determined to be untrustworthy.

[0782] "Transfer means" refers to a means for connecting a call from a caller determined to be trustworthy to a user.

[0783] "Food delivery related" refers to when the trusted sender requirement relates to food delivery services.

[0784] "Means for automatically converting delivery status into text and notifying the user" refers to a means for converting information such as the progress of delivery from voice to text and notifying the user of that text information.

[0785] A "whitelist" is a list of trusted senders and requirements that are registered in advance.

[0786] A "blacklist" is a list of untrustworthy senders and requirements that are registered in advance.

[0787] "Appropriate action" is the process of determining whether to route or block a call based on a whitelist or blacklist.

[0788] This invention relates to a telephone handling AI system for food delivery services. This system is composed of a server, a terminal, and a user (hereinafter referred to as "user").

[0789] Server Features

[0790] When a call comes in, the server immediately receives the signal and establishes a telephone connection. Once the connection is established, the server activates the answering machine and asks the caller for their identification and requirements. For example, the server asks questions such as "Who is this?" and "What can I do for you?" This voice response is realized using an AI module.

[0791] When the caller responds, their voice is sent through the device to the server, which uses a speech recognition system (e.g., Google Speech Recognition API) to convert this voice data into text.

[0792] The server then analyzes the text data obtained and extracts important keywords. This analysis is performed by checking against a whitelist and blacklist. If the caller or requirements match those on the whitelist, the server determines that the call is safe. If it is determined to be safe, the server sends this information to the terminal and notifies the user. At this time, a notification is displayed saying, "This is a trusted caller. Would you like to transfer the call?"

[0793] If the trusted caller requirement is related to food delivery, the server will automatically transcribe the delivery status into text and notify the user. If the specified caller or requirement is on a blacklist, the server will automatically block the call and notify the user that the call was disconnected due to possible fraud.

[0794] Device Features

[0795] The device receives instructions from the server and initiates an AI response. For example, if a user is using a food delivery app, a notification will appear within the app, providing the caller's details and requirements. The user can choose to accept or disconnect the call. If the user chooses to accept the call, the server will connect the call directly to the user's device.

[0796] Examples of use cases and generated AI prompts

[0797] As a concrete example, consider the case where a caller calls and says, "This is Sato. I'd like to confirm the arrival time of my pizza order." The server converts the speech into text and analyzes the caller's name and requirements. If "Sato" or "pizza" is included in the whitelist, the server determines that the call is safe. This information is sent to the terminal and notifies the user. If the user selects to transfer the call, the server connects the call to the user's terminal.

[0798] An example of a prompt sentence for the generative AI model input is as follows:

[0799] "This is Sato. I'd like to confirm the arrival time of the pizza I ordered."

[0800] As described above, this system enables efficient and safe telephone response in food delivery services.

[0801] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0802] Step 1:

[0803] The server receives the telephone signal: the input is the telephone signal from the caller, and the output is the establishment of a telephone connection.

[0804] Step 2:

[0805] The server activates an answering machine and asks for the caller's identification and requirements. The input is an established telephone connection and the output is a spoken question such as "Who is this?" or "What can I do for you?"

[0806] Step 3:

[0807] The terminal receives the caller's response in voice format and transmits the voice to the server. The input is the caller's voice data, and the output is the data to be transmitted to the server.

[0808] Step 4:

[0809] The server uses a speech recognition system to convert the received voice data into text. The input is voice data and the output is text data. Specifically, the server converts voice to text using the Google Speech Recognition API.

[0810] Step 5:

[0811] The server analyzes the text data and extracts important keywords. This analysis is performed by matching the data with whitelists and blacklists. The input is the converted text data, and the output is the analysis result (trustworthiness assessment).

[0812] Step 6:

[0813] If the server finds a caller or requirement that matches the whitelist, it determines that the call is safe. The input is the analysis result, and the output is the safety judgment "safe."

[0814] Step 7:

[0815] The server sends the information judged to be safe to the terminal and displays a notification to the user saying, "This is a trusted caller. Would you like to take the call?" The input is the result of the safety judgment, and the output is a notification to the user.

[0816] Step 8:

[0817] The user selects whether to transfer or disconnect. The input is the user's selection, and the output is an instruction to transfer or disconnect.

[0818] Step 9:

[0819] If the user selects transfer, the server connects the call directly to the user's terminal. The input is the user's transfer instruction, and the output is the establishment of a telephone connection.

[0820] Step 10:

[0821] If the trusted caller requirement is related to food delivery, the server automatically converts the delivery status into text and notifies the user. The input is the caller's response, and the output is the text of the delivery status and its notification.

[0822] Step 11:

[0823] On the other hand, if the specified caller or requirement is on the blacklist, the server automatically blocks the call and notifies the user, "The call was disconnected due to possible fraud." The input is the analysis result (blacklist match), and the output is the call blocking and notification to the user.

[0824] 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.

[0825] This invention is a system that uses AI to handle calls, and by combining it with an emotion engine that recognizes the user's emotions, it achieves even more advanced filtering. The system is mainly composed of a server, a terminal, and an emotion engine.

[0826] When a call comes in, the server immediately receives the signal, establishes a telephone connection, activates the AI ​​response module, and sends instructions to the terminal. The terminal receives instructions from the server, activates the AI ​​response function, and asks the caller questions such as "Who is this?" and "What can I do for you?"

[0827] The terminal receives the caller's voice response and sends it to the server. The server inputs the received voice into a voice recognition system and converts the voice into text data. The server analyzes the converted text data and compares it with whitelists and blacklists to extract important keywords. If the analysis shows that the caller is on the whitelist, the call is deemed safe.

[0828] Here, the emotion engine detects the user's emotional state in real time. The emotion engine recognizes emotions such as stress, anxiety, relief, and comfort from the user's voice and input data. The server receives the user's emotional state from the emotion engine.

[0829] Even if the server determines the caller is safe, if the emotion engine detects stress or anxiety in the user, the server will automatically block the call and send that information to the device, which will then notify the user, "The call has been disconnected due to a possible fraudulent activity."

[0830] Conversely, even if the caller is on a blacklist or it is difficult to determine whether the call is authentic or not, if the emotion engine senses a sense of security and comfort in the user, the server will transfer the call to the user. The device will display a notification to the user saying, "This is a trusted caller. Would you like to transfer?" and give the user the option to transfer or disconnect. If the user selects to transfer, the server will connect the call to the user.

[0831] As a concrete example, consider the case where a caller calls and says, "This is Tanaka. I'd like to talk to you about insurance." The server converts the speech to text and analyzes "Tanaka" and "insurance." If there are no matching keywords in the whitelist, the emotion engine detects the user's emotions. If the user expresses relief or comfort, the call is transferred. Conversely, if the user feels stressed or anxious, the call is blocked regardless of the caller's trustworthiness.

[0832] This system offers a higher level of security and user experience than conventional methods, and is expected to keep all users, including the elderly, safe from scams and nuisance calls.

[0833] The processing flow will be explained below.

[0834] Step 1:

[0835] The server receives the signal when a call is received and establishes a telephone connection.

[0836] Step 2:

[0837] When the connection is established, the server activates the AI ​​response module and sends an instruction to the terminal to start the AI ​​response.

[0838] Step 3:

[0839] The device receives instructions from the server and activates the AI ​​response function, asking the caller questions such as "Who is this?" and "What can I do for you?"

[0840] Step 4:

[0841] The terminal receives the caller's response voice and transmits the voice data to the server.

[0842] Step 5:

[0843] The server inputs the received voice data into a voice recognition system and converts the voice into text data.

[0844] Step 6:

[0845] The server analyzes the converted text data and extracts important keywords. For example, from a response such as "I'm Tanaka. I'd like to talk to you about insurance," the server extracts the keywords "Tanaka" and "insurance."

[0846] Step 7:

[0847] The server checks the extracted keywords against a whitelist and a blacklist to determine whether the sender is trustworthy.

[0848] Step 8:

[0849] The emotion engine analyzes the user's voice and input data in real time to recognize the user's emotional state (e.g., relief, excitement, stress, anxiety, etc.). The emotion engine's results are sent to the server.

[0850] Step 9:

[0851] Even if the server determines that the caller is safe, if the emotion engine detects stress or anxiety in the user, the server will automatically block the call and send that information to the device, which will then notify the user, "The call has been disconnected due to a possible fraudulent activity."

[0852] Step 10:

[0853] Conversely, even if the caller is on a blacklist or has questionable reliability, if the emotion engine makes the user feel safe and comfortable, the server will transfer the call to the user. The device will display a notification to the user saying, "This is a trusted caller. Would you like to transfer the call?"

[0854] Step 11:

[0855] The user selects whether to transfer the call or to disconnect, and the selection information is sent from the terminal to the server.

[0856] Step 12:

[0857] If the user chooses to be transferred, the server connects the call to the user's terminal.

[0858] Step 13:

[0859] If the user chooses to hang up, the server will end the call and blacklist the call if necessary.

[0860] Example 2

[0861] 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."

[0862] While conventional telephone answering systems can verify the caller and ask about their requirements, they have limitations in accurately determining the caller's trustworthiness, and in particular cannot take into account the user's emotional state, making it impossible to completely eliminate the risk of fraud or nuisance calls. Furthermore, there is a need for an answering system that gives users, including the elderly, a sense of security. Therefore, a system is needed that can evaluate both the caller's trustworthiness and the user's emotional state in real time and appropriately transfer or block calls.

[0863] 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.

[0864] In this invention, the server includes: means for first answering an incoming call with an answering device and asking for the caller's identification and requirements; means for analyzing the caller's response and determining whether the caller is trustworthy; means for blocking calls from callers determined to be untrustworthy; means for transferring calls from callers determined to be trustworthy to a user; and means for detecting the user's emotional state in real time and blocking or transferring calls based on the user's emotional state, which allows the server to evaluate not only the caller's trustworthiness but also the user's emotional state, further reducing the risk of fraud and nuisance calls and providing users with a sense of security.

[0865] An "answering device" is a device that automatically answers an incoming call and asks for the caller's identification and request.

[0866] "Means for identifying and inquiring about the caller" refers to the means for questioning the caller about their identity and purpose when receiving a call.

[0867] "Analysis" is the act of understanding and evaluating the sender's information and requirements based on the received voice and text data.

[0868] The "means for determining whether the sender is trustworthy" is a means for analyzing the contents of the sender's response and evaluating whether the contents are trustworthy.

[0869] A "call blocking measure" is a measure for automatically blocking a call if the caller is determined to be untrustworthy.

[0870] "Means for transferring calls to a user" refers to means for transferring calls from callers determined to be trustworthy to a user so that the user can answer the call.

[0871] A "voice recognition system" is a technology for converting received voice data into text data.

[0872] A "whitelist" is a list of trusted senders that is used to evaluate the trustworthiness of senders based on that list.

[0873] A "blacklist" is a list of untrustworthy senders, and is used to evaluate the trustworthiness of senders based on that list.

[0874] An "emotion recognition engine" is a technology that analyzes the user's voice and input data to detect the user's emotional state (stress, anxiety, relief, comfort, etc.) in real time.

[0875] A "means for blocking or forwarding calls based on emotional state" is a means for evaluating a user's emotional state in real time and forwarding or blocking calls based on the evaluation result.

[0876] This invention is an advanced filtering system that combines a call handling AI with an emotion recognition engine that recognizes user emotions. This system is mainly composed of a server, a terminal, and an emotion recognition engine. The specific operation and configuration of the system are explained in detail below.

[0877] Server configuration and functions

[0878] The server is equipped with a communication module for receiving telephone signals. When a telephone signal is received, the server immediately identifies the signal and establishes a telephone connection. After this, the server activates the AI ​​response module and sends instructions to the terminal.

[0879] The AI ​​response module has a script to ask the caller for confirmation and requirements, and the server sends instructions to the terminal to provide a standard response, such as "Who is this?" or "What can I do for you?"

[0880] Device configuration and functions

[0881] The terminal receives instructions from the server, asks questions to the caller according to those instructions, receives a voice response from the caller and sends it to the server. The terminal also has a display and speaker to notify the user of the server's judgment results.

[0882] Emotion recognition engine configuration and functions

[0883] The emotion recognition engine analyzes user voice and input data to detect emotions such as stress, anxiety, relief, comfort, etc. in real time. This engine can be built using, for example, IBM Watson's Tone Analyzer API.

[0884] Server processing

[0885] The server inputs the received caller's voice into a speech recognition system within the program and converts it into text data, possibly using a system such as Google Cloud Speech-to-Text API. The converted text data is then compared against whitelists and blacklists to extract important keywords.

[0886] Based on the results of analyzing the text data, the server evaluates the trustworthiness of the sender. In addition, it obtains the user's emotional state from an emotion recognition engine and makes a final decision taking the user's emotional state into consideration.

[0887] Judgment results and notification to users

[0888] Even if the server determines that the caller is trustworthy, if the emotion recognition engine detects stress or anxiety in the user, the server will automatically block the call and send that information to the device, which will then notify the user, "The call has been disconnected due to a possible fraudulent activity."

[0889] Conversely, even if the caller is on a blacklist or it is difficult to determine whether the call is authentic or authentic, if the emotion recognition engine senses a sense of security or comfort in the user, the server will transfer the call to the user. The device will display a notification to the user saying, "This is a trusted caller. Would you like to transfer?" and give the user the option to transfer or disconnect. If the user chooses to transfer, the server will connect the call to the user.

[0890] Specific examples

[0891] As a concrete example, consider a caller calling and saying, "This is Tanaka. I'd like to talk to you about insurance." In this case, the server converts the speech to text and analyzes "Tanaka" and "insurance." If there are no matching keywords in the whitelist, the emotion recognition engine detects the user's emotions. If the user expresses relief or comfort, the call is transferred. Conversely, if the user feels stressed or anxious, the call is blocked regardless of the caller's trustworthiness.

[0892] Example prompts for generative AI models

[0893] Below are some example prompts to input to a generative AI model:

[0894] "Please explain the system that analyzes the emotional feedback from the user and the contents of the phone call to determine safety. Also, please add the following concrete example: Based on the scenario, "This is Tanaka. I'd like to talk to you about insurance," please explain the processing flow for when the user indicates safety and when they indicate anxiety."

[0895] This system offers higher levels of security and a better user experience than traditional methods, and is expected to help all users, including the elderly, stay safe from scams and spam calls.

[0896] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0897] Step 1:

[0898] The server receives the call signal and establishes a connection

[0899] Input: External telephone signal

[0900] Specific operations: The server receives an external telephone signal, analyzes the received signal, extracts identifying information, and records it in a log. Then, it establishes a telephone connection so that the call can begin.

[0901] Output: The phone connection is established and its status is logged.

[0902] Step 2:

[0903] The server activates the AI ​​response module

[0904] Input: Telephone connection establishment information

[0905] Specific operation: When a telephone connection is established, the server starts the AI ​​response module and logs the message "The response module has been started." Then, the AI ​​response module loads a script and sends instructions for the initial response to the caller to the terminal.

[0906] Output: The AI ​​response module is activated and the instructions are sent to the terminal.

[0907] Step 3:

[0908] The device receives instructions from the AI ​​response module and asks the caller a question.

[0909] Input: Indicates the initial response from the server

[0910] Specific operation: The device follows the instructions of the initial response received from the server and asks the caller questions such as "Who is this?" and "What can I do for you?" The voice response is generated based on a pre-defined script.

[0911] Output: Audio response from the caller.

[0912] Step 4:

[0913] The device receives the caller's response and sends it to the server.

[0914] Input: Voice response from the caller

[0915] Specific operation: The terminal receives a voice response from the caller and sends the voice data to the server, where it converts it into an appropriate format to prevent data degradation.

[0916] Output: The audio data sent to the server.

[0917] Step 5:

[0918] The server inputs the received voice into a voice recognition system and converts it into text data.

[0919] Input: Audio data sent from the device

[0920] Specific operation: The server inputs the received voice data into the voice recognition system, which uses the Google Cloud Speech-to-Text API to convert the voice data into text data. At this time, the voice recognition system performs phonemic analysis and converts it into corresponding text.

[0921] Output: The caller's response converted into text data.

[0922] Step 6:

[0923] The server parses the text data and checks it against the whitelist and blacklist.

[0924] Input: Text data from a speech recognition system

[0925] How it works: The server analyzes the converted text data and extracts important keywords. It then checks these keywords against a whitelist and blacklist to evaluate the trustworthiness of the sender. For example, it checks whether the text contains keywords like "Tanaka" or "insurance."

[0926] Output: The result of the evaluation of the sender's trustworthiness.

[0927] Step 7:

[0928] The server obtains the user's emotional state from the emotion recognition engine.

[0929] Input: Real-time user voice and typing data

[0930] How it works: The emotion recognition engine analyzes the user's voice and input data to detect emotional states such as stress, anxiety, relief, and comfort. The server periodically retrieves this emotional data from the emotion recognition engine, for example, using IBM Watson's Tone Analyzer API.

[0931] Output: User's emotional state data.

[0932] Step 8:

[0933] The server makes the final decision based on the sender's credibility and the user's emotional state.

[0934] Input: Evaluation results on the sender's trustworthiness and user's emotional state data

[0935] Specific operation: The server integrates the caller's trustworthiness assessment and the user's emotional state to make the final decision on whether to forward or block the call. For example, if the caller is not on the whitelist but the user feels safe, the call will be forwarded. Conversely, if the caller is on the whitelist but the user feels stressed, the call will be blocked.

[0936] Output: Instructions for transferring or blocking the call.

[0937] Step 9:

[0938] The device sends the first notification to the user

[0939] Input: Call forwarding or blocking instructions from the server

[0940] Specific behavior: The device receives instructions from the server and notifies the user of the result, for example, displaying a message such as "Call has been disconnected due to possible fraud" or "Trusted caller. Would you like to proceed?"

[0941] Output: The notification message displayed to the user.

[0942] Step 10:

[0943] User chooses to transfer or block the call

[0944] Input: Notification message displayed on the terminal

[0945] Specific operation: The user can choose to transfer or disconnect the call based on the message displayed on the device, using the buttons or touch screen on the device.

[0946] Output: Data about the user's selection.

[0947] Step 11:

[0948] The server takes the final action based on the user's selection.

[0949] Input: Data about user selections

[0950] Specific behavior: If the user selects to transfer the call, the server receives the instruction and connects the call to the user. Conversely, if the user selects to hang up the call, the server immediately hangs up the call.

[0951] Output: Calls are connected or disconnected, and the status is logged.

[0952] Through the above processing steps, the system comprehensively assesses both the caller's credibility and the user's emotional state, and then optimally routes or blocks the call.

[0953] (Application example 2)

[0954] 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."

[0955] The present invention aims to improve security and user experience in conventional telephone filtering systems. In particular, it aims to further reduce the risk of fraud and nuisance calls by considering the user's emotional state in addition to determining the trustworthiness of the caller. It also aims to provide an environment where users can receive calls with peace of mind.

[0956] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: a means for an answering device to answer an incoming call and inquire about the caller's identity and requirements; a means for analyzing the caller's response and determining whether the caller is trustworthy; a means for blocking calls from callers determined to be untrustworthy; a means for transferring calls from callers determined to be trustworthy to a user; an emotion engine for detecting the user's emotional state and filtering calls based on the emotional state; a means for transferring calls when the user indicates relief or comfort and blocking calls when the user indicates stress or anxiety; a means for converting the caller's response into text using a voice recognition system, the determination means analyzing the converted text and determining whether the caller is trustworthy; a whitelist and a blacklist maintained by the determination means for comparing the caller's response against the whitelist and blacklist; and a means for analyzing the user's emotional state using a generative AI model and generating prompt sentences to obtain an emotion analysis result. This allows calls to be appropriately filtered based on the user's emotional state, protecting users from spam calls and scams.

[0957] An "answering machine" is a device that automatically answers an incoming call and asks for the caller's identification and requirements.

[0958] "Caller identification and inquiry measures" are measures to verify the identity of the caller and the caller's intent and purpose.

[0959] The "means for analyzing the contents of the caller's response" is a means for analyzing the contents of the caller's response and determining whether the caller is trustworthy.

[0960] "Means for blocking calls from callers determined to be untrustworthy" refers to means for blocking calls from callers determined to be untrustworthy without forwarding the calls to the user.

[0961] The "means for transferring a call from a caller determined to be trustworthy to a user" refers to a means for connecting a call to a user when the caller is determined to be trustworthy.

[0962] The "emotion engine" is an engine that analyzes the user's emotional state and detects emotional data in real time.

[0963] A "means for filtering calls based on emotional state" is a means for determining whether to pass or block a call depending on the emotional state of a user.

[0964] A "voice recognition system" is a system that converts voice data into text data.

[0965] A "whitelist" is a list of trusted callers from which calls are allowed.

[0966] A "blacklist" is a list of untrusted callers from which calls are blocked.

[0967] A "generative AI model" is an artificial intelligence model that analyzes a user's emotional state, generates prompt sentences, and performs emotion analysis.

[0968] A "prompt sentence" is text data that is input into the generative AI model to perform sentiment analysis.

[0969] This invention is an advanced filtering system that combines a call handling AI system with an emotion engine. This system is mainly composed of a server, a terminal, and an emotion engine.

[0970] When the server receives a call, it immediately receives the signal and establishes a telephone connection. Next, the server activates the AI ​​response module and sends instructions to the terminal. The terminal receives the instructions from the server, activates the AI ​​response function, and asks the caller questions such as "Who is this?" and "What can I do for you?"

[0971] The terminal receives the caller's voice response and sends it to the server. The server inputs the received voice into a voice recognition system and converts the voice into text data. The server analyzes the converted text data and compares it with a whitelist and blacklist to extract important keywords. In this case, the SpeechRecognition library is used as the voice recognition system, and natural language processing is used for text analysis.

[0972] If the analysis finds that the caller is on the whitelist, the call is deemed safe. The emotion engine then detects the user's emotional state in real time. The emotion engine uses a generative AI model using Hugging Face's Transformers library. The emotion engine recognizes emotions such as stress, anxiety, relief, and comfort from the user's voice and input data.

[0973] The server receives the user's emotional state from the emotion engine, and even if the call is deemed safe, if it detects stress or anxiety in the user, it automatically blocks the call and sends that information to the device, which then notifies the user, "The call has been disconnected due to a possible fraudulent activity."

[0974] Conversely, even if the caller is on the blacklist, if the emotion engine makes the user feel safe and comfortable, the server will transfer the call to the user. The device will display a notification to the user saying "This is a trusted caller. Would you like to transfer?" and give the user the option to transfer or disconnect. If the user chooses to transfer, the server will connect the call to the user.

[0975] As a concrete example, consider a situation where a user receives keywords such as "insurance" and "important offer." At this time, the server converts the speech into text and inputs the text data into the emotion engine. If the emotion engine detects negative emotions such as "anxiety" or "stress," the call will be blocked. An example of a prompt sentence would be "I'd like to talk to you about insurance, and I have a really tailored offer for you."

[0976] In this way, the present invention is a system that realizes advanced filtering based on user sentiment and whitelists / blacklists to protect users from spam calls and scams.

[0977] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0978] Step 1:

[0979] The server receives the call and immediately receives the signal to establish a telephone connection, at which point the server activates the AI ​​response module and sends a response instruction to the terminal.

[0980] Step 2:

[0981] The device receives instructions from the server and activates the AI ​​response function, which asks the caller questions such as "Who is this?" and "What can I do for you?", and collects the caller's voice response.

[0982] Step 3:

[0983] The device sends the collected response voice of the caller to the server, which converts this voice data into text data using the SpeechRecognition library. Input: Voice data, Output: Text data.

[0984] Step 4:

[0985] The server analyzes the converted text data and extracts important keywords by matching them with the whitelist and blacklist. Input: text data, whitelist, blacklist. Output: keyword matching results.

[0986] Step 5:

[0987] The server uses a generative AI model to analyze the user's emotional state (stress, anxiety, relief, comfort, etc.) from the text data. Input: Text data, Output: Emotion analysis results.

[0988] Step 6:

[0989] The server decides whether to allow or block a call based on the emotion analysis results obtained from the emotion engine. Specifically, even if a call falls under the whitelist and is deemed safe, the call will be blocked if the user is feeling stressed or anxious. Conversely, even if a call falls under the blacklist, the call will be allowed if the user is feeling safe or comfortable.

[0990] Step 7:

[0991] The device notifies the user of the server's decision. If the user feels stressed or anxious, the device will notify them, "The call has been disconnected due to a possible scam." If the user feels safe or comfortable, the device will display a message to the user saying, "This is a trusted caller. Would you like to transfer the call?" and offer the user the option of transferring the call or disconnecting.

[0992] Step 8:

[0993] If the user selects to transfer, the server connects the call to the user. Conversely, if the user selects to hang up, the server terminates the call. Input: User selection, Output: Call connection or termination.

[0994] 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.

[0995] 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.

[0996] 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.

[0997] [Fourth embodiment]

[0998] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0999] 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.

[1000] 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).

[1001] 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.

[1002] 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.

[1003] 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).

[1004] 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.

[1005] 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.

[1006] 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.

[1007] 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.

[1008] 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.

[1009] 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.

[1010] 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."

[1011] The present invention is an embodiment of a system using a call handling AI. This system is mainly composed of a server, a terminal, and a user.

[1012] The server immediately receives the signal when a call comes in and establishes a call connection. Once the connection is established, the server activates the AI ​​response module and sends instructions to the terminal.

[1013] The device receives instructions from the server and begins the AI ​​response. Specifically, the device's AI response function asks the caller questions such as "Who is this?" and "What can I do for you?"

[1014] When the caller responds, the voice is sent from the terminal to the server, which then inputs the voice data into a voice recognition system to convert the voice into text, thereby recording the caller's response in text format.

[1015] The server then analyzes the text data and extracts important keywords. This analysis is performed by matching them with whitelists and blacklists. If the caller or requirements match those on the whitelist, the server determines that the call is safe.

[1016] If the call is deemed safe, the server sends that information to the device. The device then displays a notification to the user saying, "This is a trusted caller. Would you like to transfer the call?" The user can then choose to transfer the call or hang up on the device. If the user chooses to transfer the call, the server connects the call directly to the user. Conversely, if the user chooses to hang up, the server ends the call.

[1017] On the other hand, if the caller is on a blacklist or meets other safety criteria, the server will automatically block the call, sending the blocking information to the device, which will then notify the user that the call was disconnected due to possible fraud.

[1018] As a concrete example, consider a caller calling and saying, "This is Tanaka. I'd like to talk to you about insurance." The server converts the speech to text and analyzes the caller's name and requirements. If "Tanaka" or "insurance" is included in the whitelist, the server determines that the call is safe. This information is sent to the device and notifies the user. If the user selects to proceed, the server connects the call to the user's device.

[1019] The processing flow will be explained below.

[1020] Step 1:

[1021] The server receives the signal when a call is received and establishes a telephone connection.

[1022] Step 2:

[1023] When the connection is established, the server activates the AI ​​response module and sends an instruction to the terminal to start the AI ​​response.

[1024] Step 3:

[1025] The device receives instructions from the server and activates the AI ​​response function, asking the caller questions such as "Who is this?" and "What can I do for you?"

[1026] Step 4:

[1027] The terminal receives the caller's response voice and transmits the voice data to the server.

[1028] Step 5:

[1029] The server inputs the received voice data into a voice recognition system and converts the voice into text data.

[1030] Step 6:

[1031] The server analyzes the converted text data, extracts important keywords, and compares the text data with whitelists and blacklists.

[1032] Step 7:

[1033] The server determines a call as safe if the caller or requirements match those on the whitelist, and as a scam or nuisance call if the caller or requirements match those on the blacklist or if there are other safety issues.

[1034] Step 8:

[1035] If it is determined to be safe, the server sends the information to the device and displays a notification to the user saying, "This is a trusted caller. Would you like to transfer the call?"

[1036] Step 9:

[1037] The user selects whether to transfer or disconnect the call on the terminal, and transmits the selection information from the terminal to the server.

[1038] Step 10:

[1039] If the user chooses to transfer, the server connects the call to the user's terminal, if the user chooses to disconnect, the server ends the call.

[1040] Step 11:

[1041] If the call is determined to be fraudulent or spam, the server automatically blocks the call and sends that information to the device, which then notifies the user that the call has been disconnected due to possible fraud.

[1042] Example 1

[1043] 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."

[1044] Conventional call forwarding systems have difficulty determining whether a caller is trustworthy, and lack sufficient means to prevent fraud and unwanted sales calls. Furthermore, manual intervention is required, placing a heavy burden on users. These issues need to be resolved.

[1045] 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.

[1046] In this invention, the server includes means for analyzing the caller's response and determining whether the caller is trustworthy, means for converting the caller's response into text using a voice recognition system, means for maintaining a whitelist and a blacklist and comparing the caller's response with the whitelist and the blacklist, means for connecting the call to the user if the user selects transfer, and means for terminating the call if the user selects disconnection. This makes it possible to automatically block calls from untrustworthy callers, reduce the burden on the user, and provide a safe calling environment.

[1047] An "answering device" is a device that automatically answers an incoming call and asks questions to the caller.

[1048] "Caller" refers to the person who placed the call.

[1049] "Requirements" refers to the purpose and content of the call made by the caller.

[1050] The "determination means" is a means for analyzing the contents of the response from the sender and determining whether the sender is trustworthy.

[1051] "Means for converting to text" refers to a speech recognition system for converting the caller's voice data into text information.

[1052] A "whitelist" is a list that manages information about trusted senders and messages in list format.

[1053] A "blacklist" is a list that manages information about untrustworthy senders and messages in list format.

[1054] The "checking means" is a means having a function of checking the contents of the caller's response against the whitelist and blacklist.

[1055] The "transfer means" is a means for connecting a call to a user when the caller is determined to be a trustworthy caller.

[1056] The "means for terminating a call" refers to the means for terminating a call when the call is from an untrusted caller or when the user chooses to hang up.

[1057] The present invention is a system for automating telephone transfers. This system is composed of a server, a terminal, and a user. Detailed embodiments of this system are described below.

[1058] Server role and hardware / software used

[1059] When a call comes in, the server immediately receives the signal and establishes a telephone connection. After this connection is established, the server activates the AI ​​response module and has the function of sending instructions to the terminal. The specific hardware used is a high-performance server. The software uses the Google Cloud Speech-to-Text API as the voice recognition system.

[1060] When the server receives the caller's voice data, it uses a voice recognition system to convert it into text data. The server then analyzes the text data and extracts important keywords. At this time, the server determines the trustworthiness of the caller by comparing it with a whitelist and blacklist. A whitelist is a list in which trustworthy callers and content are registered in advance, while a blacklist is a list in which untrustworthy callers and content are registered.

[1061] Terminal role and hardware / software used

[1062] The device receives instructions from the server and begins responding with an AI. For example, the device asks the caller questions such as "Who is this?" or "What can I do for you?" This operation is performed by voice using the device's speaker.

[1063] The device also has the role of sending the caller's voice response to the server. It also has the function of receiving notifications from the server, displaying information to the user, and allowing the user to choose whether to transfer the call or disconnect based on that information. The hardware used is a smartphone or tablet, and the custom AI response app runs as software.

[1064] User Roles

[1065] The user can choose whether to transfer the caller or end the call based on the notification displayed on the device. For example, if the device displays "This is a trusted caller. Would you like to transfer the call?", the user can choose to "transfer" or "disconnect." If the user selects "transfer," the server will connect the call directly to the user. Conversely, if the user selects "disconnect," the server will end the call.

[1066] Specific examples

[1067] For example, consider the case where a caller named "Tanaka" calls and says, "I'd like to talk to you about insurance." The server converts this voice into text and extracts the keywords "Tanaka" and "insurance." This is compared with a whitelist, and if it is determined that "Tanaka" and "insurance" are trustworthy, the information is sent to the terminal. The terminal notifies the user, "This is a trustworthy caller. Would you like to transfer the call?" If the user selects "transfer," the server connects the call directly to the user's terminal.

[1068] Prompt Sentence Examples

[1069] Examples of prompts to input to a generative AI model include:

[1070] To the user: This is Tanaka. I'm calling to talk about insurance. Please help me.

[1071] server:

[1072] 1. Convert the caller's voice to text.

[1073] 2. Check "Tanaka" and "Insurance" against the whitelist.

[1074] In this way, the entire system works in coordination and automates call handling, reducing the burden on users and enabling safe and efficient call management.

[1075] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1076] Step 1:

[1077] The server receives telephone signals from the outside. The input is a telephone signal, and the output is the establishment of a telephone connection. Specifically, when a call comes in, the server detects the signal and performs the reception process. This is managed by the server's telephone line interface.

[1078] Step 2:

[1079] After the connection is established, the server activates the AI ​​response module. The input is a notification of the connection establishment, and the output is the start or running status of the AI ​​response module. In effect, the server starts the internal software module and prepares to send instructions to the terminal.

[1080] Step 3:

[1081] The device receives instructions from the server and begins its AI response. The input is the instruction from the server, and the output is the initial response to the caller. Specifically, questions such as "Who is this?" and "What can I do for you?" are asked aloud through the speaker.

[1082] Step 4:

[1083] When the caller responds, the voice data is sent from the terminal to the server. The input is the caller's voice data, and the output is the voice data transferred to the server. The terminal uses a microphone to record the voice and sends the data to the server via the network.

[1084] Step 5:

[1085] The server uses a speech recognition system to convert the received voice data into text format. The input is the caller's voice data, and the output is text data. Specifically, it uses the Google Cloud Speech-to-Text API to convert voice to text in real time.

[1086] Step 6:

[1087] The server analyzes the text data and extracts important keywords. The input is the converted text data, and the output is the extracted keywords. The analysis module in the server extracts keywords such as "name" and "requirements" from the text data.

[1088] Step 7:

[1089] The server checks the extracted keywords against whitelists and blacklists. The input is the extracted keywords, and the output is a trust decision (safe / unsafe). Specifically, the server uses a database query to match the keywords against the lists and see if the caller matches the list.

[1090] Step 8:

[1091] If the server determines that the call is safe, it sends that information to the terminal. The input is the result of the trust determination, and the output is a notification to the terminal. Specifically, a message saying "This caller is trustworthy. Would you like to take the call?" is sent to the terminal.

[1092] Step 9:

[1093] The terminal displays a notification to the user saying "This is a trusted caller. Would you like to take the call?" The input is the notification from the server and the output is what is displayed to the user. The terminal displays the message on the screen and allows the user to make a choice.

[1094] Step 10:

[1095] If the user selects transfer, the server connects the call directly to the user's terminal. The input is the user's selection (transfer), and the output is the establishment of a telephone connection. The server follows the user's instructions and connects the caller and the user directly over the telephone line.

[1096] Step 11:

[1097] If the user selects to disconnect, the server terminates the call. The input is the user's selection (disconnect), and the output is the end of the call. The server terminates the call and notifies the terminal of this information.

[1098] (Application example 1)

[1099] 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."

[1100] Food delivery services are faced with a large number of inquiries regarding delivery status and order details, placing a heavy burden on customer support. There are also many cases of fraudulent calls from unreliable callers. This has led to concerns about a decline in customer satisfaction and a deterioration in operational efficiency. Traditional manual response methods have their limitations, creating a need for an efficient and secure telephone response system.

[1101] 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.

[1102] In this invention, the server includes means for first answering an incoming call with an answering device and asking for caller identification and requirements, means for analyzing the content of the caller's response and determining whether the caller is trustworthy, means for blocking calls from callers determined to be untrustworthy, means for transferring calls from callers determined to be trustworthy to a user, means for automatically converting the delivery status into text and notifying the user if the trustworthy caller's requirements are related to food delivery, and means for determining whether a specified caller or requirements fall under a whitelist or blacklist and taking appropriate action, thereby enabling efficient and safe telephone response.

[1103] An "answering machine" is a device that initially answers an incoming call and asks for the caller's identification and requirements.

[1104] "Caller response" refers to the response given by the caller, which is usually in audio form.

[1105] The "means for analyzing" is a means for analyzing the content of the caller's response and evaluating and classifying the information.

[1106] The "means for determining trustworthiness" is a means for determining whether the sender is trustworthy based on the analyzed data.

[1107] A "blocking measure" is a measure for automatically blocking calls from callers determined to be untrustworthy.

[1108] "Transfer means" refers to a means for connecting a call from a caller determined to be trustworthy to a user.

[1109] "Food delivery related" refers to when the trusted sender requirement relates to food delivery services.

[1110] "Means for automatically converting delivery status into text and notifying the user" refers to a means for converting information such as the progress of delivery from voice to text and notifying the user of that text information.

[1111] A "whitelist" is a list of trusted senders and requirements that are registered in advance.

[1112] A "blacklist" is a list of untrustworthy senders and requirements that are registered in advance.

[1113] "Appropriate action" is the process of determining whether to route or block a call based on a whitelist or blacklist.

[1114] This invention relates to a telephone handling AI system for food delivery services. This system is composed of a server, a terminal, and a user (hereinafter referred to as "user").

[1115] Server Features

[1116] When a call comes in, the server immediately receives the signal and establishes a telephone connection. Once the connection is established, the server activates the answering machine and asks the caller for their identification and requirements. For example, the server asks questions such as "Who is this?" and "What can I do for you?" This voice response is realized using an AI module.

[1117] When the caller responds, their voice is sent through the device to the server, which uses a speech recognition system (e.g., Google Speech Recognition API) to convert this voice data into text.

[1118] The server then analyzes the text data obtained and extracts important keywords. This analysis is performed by checking against a whitelist and blacklist. If the caller or requirements match those on the whitelist, the server determines that the call is safe. If it is determined to be safe, the server sends this information to the terminal and notifies the user. At this time, a notification is displayed saying, "This is a trusted caller. Would you like to transfer the call?"

[1119] If the trusted caller requirement is related to food delivery, the server will automatically transcribe the delivery status into text and notify the user. If the specified caller or requirement is on a blacklist, the server will automatically block the call and notify the user that the call was disconnected due to possible fraud.

[1120] Device Features

[1121] The device receives instructions from the server and initiates an AI response. For example, if a user is using a food delivery app, a notification will appear within the app, providing the caller's details and requirements. The user can choose to accept or disconnect the call. If the user chooses to accept the call, the server will connect the call directly to the user's device.

[1122] Examples of use cases and generated AI prompts

[1123] As a concrete example, consider the case where a caller calls and says, "This is Sato. I'd like to confirm the arrival time of my pizza order." The server converts the speech into text and analyzes the caller's name and requirements. If "Sato" or "pizza" is included in the whitelist, the server determines that the call is safe. This information is sent to the terminal and notifies the user. If the user selects to transfer the call, the server connects the call to the user's terminal.

[1124] An example of a prompt sentence for the generative AI model input is as follows:

[1125] "This is Sato. I'd like to confirm the arrival time of the pizza I ordered."

[1126] As described above, this system enables efficient and safe telephone response in food delivery services.

[1127] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1128] Step 1:

[1129] The server receives the telephone signal: the input is the telephone signal from the caller, and the output is the establishment of a telephone connection.

[1130] Step 2:

[1131] The server activates an answering machine and asks for the caller's identification and requirements. The input is an established telephone connection and the output is a spoken question such as "Who is this?" or "What can I do for you?"

[1132] Step 3:

[1133] The terminal receives the caller's response in voice format and transmits the voice to the server. The input is the caller's voice data, and the output is the data to be transmitted to the server.

[1134] Step 4:

[1135] The server uses a speech recognition system to convert the received voice data into text. The input is voice data and the output is text data. Specifically, the server converts voice to text using the Google Speech Recognition API.

[1136] Step 5:

[1137] The server analyzes the text data and extracts important keywords. This analysis is performed by matching the data with whitelists and blacklists. The input is the converted text data, and the output is the analysis result (trustworthiness assessment).

[1138] Step 6:

[1139] If the server finds a caller or requirement that matches the whitelist, it determines that the call is safe. The input is the analysis result, and the output is the safety judgment "safe."

[1140] Step 7:

[1141] The server sends the information judged to be safe to the terminal and displays a notification to the user saying, "This is a trusted caller. Would you like to take the call?" The input is the result of the safety judgment, and the output is a notification to the user.

[1142] Step 8:

[1143] The user selects whether to transfer or disconnect. The input is the user's selection, and the output is an instruction to transfer or disconnect.

[1144] Step 9:

[1145] If the user selects transfer, the server connects the call directly to the user's terminal. The input is the user's transfer instruction, and the output is the establishment of a telephone connection.

[1146] Step 10:

[1147] If the trusted caller requirement is related to food delivery, the server automatically converts the delivery status into text and notifies the user. The input is the caller's response, and the output is the text of the delivery status and its notification.

[1148] Step 11:

[1149] On the other hand, if the specified caller or requirement is on the blacklist, the server automatically blocks the call and notifies the user, "The call was disconnected due to possible fraud." The input is the analysis result (blacklist match), and the output is the call blocking and notification to the user.

[1150] 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.

[1151] This invention is a system that uses AI to handle calls, and by combining it with an emotion engine that recognizes the user's emotions, it achieves even more advanced filtering. The system is mainly composed of a server, a terminal, and an emotion engine.

[1152] When a call comes in, the server immediately receives the signal, establishes a telephone connection, activates the AI ​​response module, and sends instructions to the terminal. The terminal receives instructions from the server, activates the AI ​​response function, and asks the caller questions such as "Who is this?" and "What can I do for you?"

[1153] The terminal receives the caller's voice response and sends it to the server. The server inputs the received voice into a voice recognition system and converts the voice into text data. The server analyzes the converted text data and compares it with whitelists and blacklists to extract important keywords. If the analysis shows that the caller is on the whitelist, the call is deemed safe.

[1154] Here, the emotion engine detects the user's emotional state in real time. The emotion engine recognizes emotions such as stress, anxiety, relief, and comfort from the user's voice and input data. The server receives the user's emotional state from the emotion engine.

[1155] Even if the server determines the caller is safe, if the emotion engine detects stress or anxiety in the user, the server will automatically block the call and send that information to the device, which will then notify the user, "The call has been disconnected due to a possible fraudulent activity."

[1156] Conversely, even if the caller is on a blacklist or it is difficult to determine whether the call is authentic or not, if the emotion engine senses a sense of security and comfort in the user, the server will transfer the call to the user. The device will display a notification to the user saying, "This is a trusted caller. Would you like to transfer?" and give the user the option to transfer or disconnect. If the user selects to transfer, the server will connect the call to the user.

[1157] As a concrete example, consider the case where a caller calls and says, "This is Tanaka. I'd like to talk to you about insurance." The server converts the speech to text and analyzes "Tanaka" and "insurance." If there are no matching keywords in the whitelist, the emotion engine detects the user's emotions. If the user expresses relief or comfort, the call is transferred. Conversely, if the user feels stressed or anxious, the call is blocked regardless of the caller's trustworthiness.

[1158] This system offers a higher level of security and user experience than conventional methods, and is expected to keep all users, including the elderly, safe from scams and nuisance calls.

[1159] The processing flow will be explained below.

[1160] Step 1:

[1161] The server receives the signal when a call is received and establishes a telephone connection.

[1162] Step 2:

[1163] When the connection is established, the server activates the AI ​​response module and sends an instruction to the terminal to start the AI ​​response.

[1164] Step 3:

[1165] The device receives instructions from the server and activates the AI ​​response function, asking the caller questions such as "Who is this?" and "What can I do for you?"

[1166] Step 4:

[1167] The terminal receives the caller's response voice and transmits the voice data to the server.

[1168] Step 5:

[1169] The server inputs the received voice data into a voice recognition system and converts the voice into text data.

[1170] Step 6:

[1171] The server analyzes the converted text data and extracts important keywords. For example, from a response such as "I'm Tanaka. I'd like to talk to you about insurance," the server extracts the keywords "Tanaka" and "insurance."

[1172] Step 7:

[1173] The server checks the extracted keywords against a whitelist and a blacklist to determine whether the sender is trustworthy.

[1174] Step 8:

[1175] The emotion engine analyzes the user's voice and input data in real time to recognize the user's emotional state (e.g., relief, excitement, stress, anxiety, etc.). The emotion engine's results are sent to the server.

[1176] Step 9:

[1177] Even if the server determines that the caller is safe, if the emotion engine detects stress or anxiety in the user, the server will automatically block the call and send that information to the device, which will then notify the user, "The call has been disconnected due to a possible fraudulent activity."

[1178] Step 10:

[1179] Conversely, even if the caller is on a blacklist or has questionable reliability, if the emotion engine makes the user feel safe and comfortable, the server will transfer the call to the user. The device will display a notification to the user saying, "This is a trusted caller. Would you like to transfer the call?"

[1180] Step 11:

[1181] The user selects whether to transfer the call or to disconnect, and the selection information is sent from the terminal to the server.

[1182] Step 12:

[1183] If the user chooses to be transferred, the server connects the call to the user's terminal.

[1184] Step 13:

[1185] If the user chooses to hang up, the server will end the call and blacklist the call if necessary.

[1186] Example 2

[1187] 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."

[1188] While conventional telephone answering systems can verify the caller and ask about their requirements, they have limitations in accurately determining the caller's trustworthiness, and in particular cannot take into account the user's emotional state, making it impossible to completely eliminate the risk of fraud or nuisance calls. Furthermore, there is a need for an answering system that gives users, including the elderly, a sense of security. Therefore, a system is needed that can evaluate both the caller's trustworthiness and the user's emotional state in real time and appropriately transfer or block calls.

[1189] 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.

[1190] In this invention, the server includes: means for first answering an incoming call with an answering device and asking for the caller's identification and requirements; means for analyzing the caller's response and determining whether the caller is trustworthy; means for blocking calls from callers determined to be untrustworthy; means for transferring calls from callers determined to be trustworthy to a user; and means for detecting the user's emotional state in real time and blocking or transferring calls based on the user's emotional state, which allows the server to evaluate not only the caller's trustworthiness but also the user's emotional state, further reducing the risk of fraud and nuisance calls and providing users with a sense of security.

[1191] An "answering device" is a device that automatically answers an incoming call and asks for the caller's identification and request.

[1192] "Means for identifying and inquiring about the caller" refers to the means for questioning the caller about their identity and purpose when receiving a call.

[1193] "Analysis" is the act of understanding and evaluating the sender's information and requirements based on the received voice and text data.

[1194] The "means for determining whether the sender is trustworthy" is a means for analyzing the contents of the sender's response and evaluating whether the contents are trustworthy.

[1195] A "call blocking measure" is a measure for automatically blocking a call if the caller is determined to be untrustworthy.

[1196] "Means for transferring calls to a user" refers to means for transferring calls from callers determined to be trustworthy to a user so that the user can answer the call.

[1197] A "voice recognition system" is a technology for converting received voice data into text data.

[1198] A "whitelist" is a list of trusted senders that is used to evaluate the trustworthiness of senders based on that list.

[1199] A "blacklist" is a list of untrustworthy senders, and is used to evaluate the trustworthiness of senders based on that list.

[1200] An "emotion recognition engine" is a technology that analyzes the user's voice and input data to detect the user's emotional state (stress, anxiety, relief, comfort, etc.) in real time.

[1201] A "means for blocking or forwarding calls based on emotional state" is a means for evaluating a user's emotional state in real time and forwarding or blocking calls based on the evaluation result.

[1202] This invention is an advanced filtering system that combines a call handling AI with an emotion recognition engine that recognizes user emotions. This system is mainly composed of a server, a terminal, and an emotion recognition engine. The specific operation and configuration of the system are explained in detail below.

[1203] Server configuration and functions

[1204] The server is equipped with a communication module for receiving telephone signals. When a telephone signal is received, the server immediately identifies the signal and establishes a telephone connection. After this, the server activates the AI ​​response module and sends instructions to the terminal.

[1205] The AI ​​response module has a script to ask the caller for confirmation and requirements, and the server sends instructions to the terminal to provide a standard response, such as "Who is this?" or "What can I do for you?"

[1206] Device configuration and functions

[1207] The terminal receives instructions from the server, asks questions to the caller according to those instructions, receives a voice response from the caller and sends it to the server. The terminal also has a display and speaker to notify the user of the server's judgment results.

[1208] Emotion recognition engine configuration and functions

[1209] The emotion recognition engine analyzes user voice and input data to detect emotions such as stress, anxiety, relief, comfort, etc. in real time. This engine can be built using, for example, IBM Watson's Tone Analyzer API.

[1210] Server processing

[1211] The server inputs the received caller's voice into a speech recognition system within the program and converts it into text data, possibly using a system such as Google Cloud Speech-to-Text API. The converted text data is then compared against whitelists and blacklists to extract important keywords.

[1212] Based on the results of analyzing the text data, the server evaluates the trustworthiness of the sender. In addition, it obtains the user's emotional state from an emotion recognition engine and makes a final decision taking the user's emotional state into consideration.

[1213] Judgment results and notification to users

[1214] Even if the server determines that the caller is trustworthy, if the emotion recognition engine detects stress or anxiety in the user, the server will automatically block the call and send that information to the device, which will then notify the user, "The call has been disconnected due to a possible fraudulent activity."

[1215] Conversely, even if the caller is on a blacklist or it is difficult to determine whether the call is authentic or authentic, if the emotion recognition engine senses a sense of security or comfort in the user, the server will transfer the call to the user. The device will display a notification to the user saying, "This is a trusted caller. Would you like to transfer?" and give the user the option to transfer or disconnect. If the user chooses to transfer, the server will connect the call to the user.

[1216] Specific examples

[1217] As a concrete example, consider a caller calling and saying, "This is Tanaka. I'd like to talk to you about insurance." In this case, the server converts the speech to text and analyzes "Tanaka" and "insurance." If there are no matching keywords in the whitelist, the emotion recognition engine detects the user's emotions. If the user expresses relief or comfort, the call is transferred. Conversely, if the user feels stressed or anxious, the call is blocked regardless of the caller's trustworthiness.

[1218] Example prompts for generative AI models

[1219] Below are some example prompts to input to a generative AI model:

[1220] "Please explain the system that analyzes the emotional feedback from the user and the contents of the phone call to determine safety. Also, please add the following concrete example: Based on the scenario, "This is Tanaka. I'd like to talk to you about insurance," please explain the processing flow for when the user indicates safety and when they indicate anxiety."

[1221] This system offers higher levels of security and a better user experience than traditional methods, and is expected to help all users, including the elderly, stay safe from scams and spam calls.

[1222] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1223] Step 1:

[1224] The server receives the call signal and establishes a connection

[1225] Input: External telephone signal

[1226] Specific operations: The server receives an external telephone signal, analyzes the received signal, extracts identifying information, and records it in a log. Then, it establishes a telephone connection so that the call can begin.

[1227] Output: The phone connection is established and its status is logged.

[1228] Step 2:

[1229] The server activates the AI ​​response module

[1230] Input: Telephone connection establishment information

[1231] Specific operation: When a telephone connection is established, the server starts the AI ​​response module and logs the message "The response module has been started." Then, the AI ​​response module loads a script and sends instructions for the initial response to the caller to the terminal.

[1232] Output: The AI ​​response module is activated and the instructions are sent to the terminal.

[1233] Step 3:

[1234] The device receives instructions from the AI ​​response module and asks the caller a question.

[1235] Input: Indicates the initial response from the server

[1236] Specific operation: The device follows the instructions of the initial response received from the server and asks the caller questions such as "Who is this?" and "What can I do for you?" The voice response is generated based on a pre-defined script.

[1237] Output: Audio response from the caller.

[1238] Step 4:

[1239] The device receives the caller's response and sends it to the server.

[1240] Input: Voice response from the caller

[1241] Specific operation: The terminal receives a voice response from the caller and sends the voice data to the server, where it converts it into an appropriate format to prevent data degradation.

[1242] Output: The audio data sent to the server.

[1243] Step 5:

[1244] The server inputs the received voice into a voice recognition system and converts it into text data.

[1245] Input: Audio data sent from the device

[1246] Specific operation: The server inputs the received voice data into the voice recognition system, which uses the Google Cloud Speech-to-Text API to convert the voice data into text data. At this time, the voice recognition system performs phonemic analysis and converts it into corresponding text.

[1247] Output: The caller's response converted into text data.

[1248] Step 6:

[1249] The server parses the text data and checks it against the whitelist and blacklist.

[1250] Input: Text data from a speech recognition system

[1251] How it works: The server analyzes the converted text data and extracts important keywords. It then checks these keywords against a whitelist and blacklist to evaluate the trustworthiness of the sender. For example, it checks whether the text contains keywords like "Tanaka" or "insurance."

[1252] Output: The result of the evaluation of the sender's trustworthiness.

[1253] Step 7:

[1254] The server obtains the user's emotional state from the emotion recognition engine.

[1255] Input: Real-time user voice and typing data

[1256] How it works: The emotion recognition engine analyzes the user's voice and input data to detect emotional states such as stress, anxiety, relief, and comfort. The server periodically retrieves this emotional data from the emotion recognition engine, for example, using IBM Watson's Tone Analyzer API.

[1257] Output: User's emotional state data.

[1258] Step 8:

[1259] The server makes the final decision based on the sender's credibility and the user's emotional state.

[1260] Input: Evaluation results on the sender's trustworthiness and user's emotional state data

[1261] Specific operation: The server integrates the caller's trustworthiness assessment and the user's emotional state to make the final decision on whether to forward or block the call. For example, if the caller is not on the whitelist but the user feels safe, the call will be forwarded. Conversely, if the caller is on the whitelist but the user feels stressed, the call will be blocked.

[1262] Output: Instructions for transferring or blocking the call.

[1263] Step 9:

[1264] The device sends the first notification to the user

[1265] Input: Call forwarding or blocking instructions from the server

[1266] Specific behavior: The device receives instructions from the server and notifies the user of the result, for example, displaying a message such as "Call has been disconnected due to possible fraud" or "Trusted caller. Would you like to proceed?"

[1267] Output: The notification message displayed to the user.

[1268] Step 10:

[1269] User chooses to transfer or block the call

[1270] Input: Notification message displayed on the terminal

[1271] Specific operation: The user can choose to transfer or disconnect the call based on the message displayed on the device, using the buttons or touch screen on the device.

[1272] Output: Data about the user's selection.

[1273] Step 11:

[1274] The server takes the final action based on the user's selection.

[1275] Input: Data about user selections

[1276] Specific behavior: If the user selects to transfer the call, the server receives the instruction and connects the call to the user. Conversely, if the user selects to hang up the call, the server immediately hangs up the call.

[1277] Output: Calls are connected or disconnected, and the status is logged.

[1278] Through the above processing steps, the system comprehensively assesses both the caller's credibility and the user's emotional state, and then optimally routes or blocks the call.

[1279] (Application example 2)

[1280] 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."

[1281] The present invention aims to improve security and user experience in conventional telephone filtering systems. In particular, it aims to further reduce the risk of fraud and nuisance calls by considering the user's emotional state in addition to determining the trustworthiness of the caller. It also aims to provide an environment where users can receive calls with peace of mind.

[1282] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: a means for an answering device to answer an incoming call and inquire about the caller's identity and requirements; a means for analyzing the caller's response and determining whether the caller is trustworthy; a means for blocking calls from callers determined to be untrustworthy; a means for transferring calls from callers determined to be trustworthy to a user; an emotion engine for detecting the user's emotional state and filtering calls based on the emotional state; a means for transferring calls when the user indicates relief or comfort and blocking calls when the user indicates stress or anxiety; a means for converting the caller's response into text using a voice recognition system, the determination means analyzing the converted text and determining whether the caller is trustworthy; a whitelist and a blacklist maintained by the determination means for comparing the caller's response against the whitelist and blacklist; and a means for analyzing the user's emotional state using a generative AI model and generating prompt sentences to obtain an emotion analysis result. This allows calls to be appropriately filtered based on the user's emotional state, protecting users from spam calls and scams.

[1283] An "answering machine" is a device that automatically answers an incoming call and asks for the caller's identification and requirements.

[1284] "Caller identification and inquiry measures" are measures to verify the identity of the caller and the caller's intent and purpose.

[1285] The "means for analyzing the contents of the caller's response" is a means for analyzing the contents of the caller's response and determining whether the caller is trustworthy.

[1286] "Means for blocking calls from callers determined to be untrustworthy" refers to means for blocking calls from callers determined to be untrustworthy without forwarding the calls to the user.

[1287] The "means for transferring a call from a caller determined to be trustworthy to a user" refers to a means for connecting a call to a user when the caller is determined to be trustworthy.

[1288] The "emotion engine" is an engine that analyzes the user's emotional state and detects emotional data in real time.

[1289] A "means for filtering calls based on emotional state" is a means for determining whether to pass or block a call depending on the emotional state of a user.

[1290] A "voice recognition system" is a system that converts voice data into text data.

[1291] A "whitelist" is a list of trusted callers from which calls are allowed.

[1292] A "blacklist" is a list of untrusted callers from which calls are blocked.

[1293] A "generative AI model" is an artificial intelligence model that analyzes a user's emotional state, generates prompt sentences, and performs emotion analysis.

[1294] A "prompt sentence" is text data that is input into the generative AI model to perform sentiment analysis.

[1295] This invention is an advanced filtering system that combines a call handling AI system with an emotion engine. This system is mainly composed of a server, a terminal, and an emotion engine.

[1296] When the server receives a call, it immediately receives the signal and establishes a telephone connection. Next, the server activates the AI ​​response module and sends instructions to the terminal. The terminal receives the instructions from the server, activates the AI ​​response function, and asks the caller questions such as "Who is this?" and "What can I do for you?"

[1297] The terminal receives the caller's voice response and sends it to the server. The server inputs the received voice into a voice recognition system and converts the voice into text data. The server analyzes the converted text data and compares it with a whitelist and blacklist to extract important keywords. In this case, the SpeechRecognition library is used as the voice recognition system, and natural language processing is used for text analysis.

[1298] If the analysis finds that the caller is on the whitelist, the call is deemed safe. The emotion engine then detects the user's emotional state in real time. The emotion engine uses a generative AI model using Hugging Face's Transformers library. The emotion engine recognizes emotions such as stress, anxiety, relief, and comfort from the user's voice and input data.

[1299] The server receives the user's emotional state from the emotion engine, and even if the call is deemed safe, if it detects stress or anxiety in the user, it automatically blocks the call and sends that information to the device, which then notifies the user, "The call has been disconnected due to a possible fraudulent activity."

[1300] Conversely, even if the caller is on the blacklist, if the emotion engine makes the user feel safe and comfortable, the server will transfer the call to the user. The device will display a notification to the user saying "This is a trusted caller. Would you like to transfer?" and give the user the option to transfer or disconnect. If the user chooses to transfer, the server will connect the call to the user.

[1301] As a concrete example, consider a situation where a user receives keywords such as "insurance" and "important offer." At this time, the server converts the speech into text and inputs the text data into the emotion engine. If the emotion engine detects negative emotions such as "anxiety" or "stress," the call will be blocked. An example of a prompt sentence would be "I'd like to talk to you about insurance, and I have a really tailored offer for you."

[1302] In this way, the present invention is a system that realizes advanced filtering based on user sentiment and whitelists / blacklists to protect users from spam calls and scams.

[1303] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1304] Step 1:

[1305] The server receives the call and immediately receives the signal to establish a telephone connection, at which point the server activates the AI ​​response module and sends a response instruction to the terminal.

[1306] Step 2:

[1307] The device receives instructions from the server and activates the AI ​​response function, which asks the caller questions such as "Who is this?" and "What can I do for you?", and collects the caller's voice response.

[1308] Step 3:

[1309] The device sends the collected response voice of the caller to the server, which converts this voice data into text data using the SpeechRecognition library. Input: Voice data, Output: Text data.

[1310] Step 4:

[1311] The server analyzes the converted text data and extracts important keywords by matching them with the whitelist and blacklist. Input: text data, whitelist, blacklist. Output: keyword matching results.

[1312] Step 5:

[1313] The server uses a generative AI model to analyze the user's emotional state (stress, anxiety, relief, comfort, etc.) from the text data. Input: Text data, Output: Emotion analysis results.

[1314] Step 6:

[1315] The server decides whether to allow or block a call based on the emotion analysis results obtained from the emotion engine. Specifically, even if a call falls under the whitelist and is deemed safe, the call will be blocked if the user is feeling stressed or anxious. Conversely, even if a call falls under the blacklist, the call will be allowed if the user is feeling safe or comfortable.

[1316] Step 7:

[1317] The device notifies the user of the server's decision. If the user feels stressed or anxious, the device will notify them, "The call has been disconnected due to a possible scam." If the user feels safe or comfortable, the device will display a message to the user saying, "This is a trusted caller. Would you like to transfer the call?" and offer the user the option of transferring the call or disconnecting.

[1318] Step 8:

[1319] If the user selects to transfer, the server connects the call to the user. Conversely, if the user selects to hang up, the server terminates the call. Input: User selection, Output: Call connection or termination.

[1320] 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.

[1321] 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.

[1322] 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.

[1323] 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.

[1324] 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.

[1325] 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.

[1326] 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).

[1327] 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.

[1328] 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."

[1329] 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.

[1330] 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).

[1331] 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.

[1332] 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.

[1333] 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.

[1334] 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.

[1335] 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.

[1336] 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.

[1337] 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.

[1338] 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.

[1339] 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.

[1340] 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.

[1341] The following is further disclosed regarding the above embodiment.

[1342] (Claim 1)

[1343] When a call comes in, an answering machine first answers the call and asks for the caller's identification and requirements;

[1344] A means for analyzing the content of the caller's response and determining whether the caller is trustworthy;

[1345] means for blocking calls from callers determined to be untrusted;

[1346] A system including means for routing calls from callers determined to be trusted to a user.

[1347] (Claim 2)

[1348] a means for converting the caller's response into text using a voice recognition system;

[1349] 2. The system according to claim 1, wherein the determining means analyzes the converted text and determines whether it is trustworthy.

[1350] (Claim 3)

[1351] 2. The system of claim 1, wherein a white list and a black list are maintained, and said determining means includes means for checking the contents of the caller's response against the white list and the black list.

[1352] "Example 1"

[1353] (Claim 1)

[1354] When a call comes in, an answering machine first answers the call and asks for the caller's identification and requirements;

[1355] A means for analyzing the content of the caller's response and determining whether the caller is trustworthy;

[1356] means for blocking calls from callers determined to be untrusted;

[1357] a means for routing calls from callers determined to be trusted to a user;

[1358] a means for converting the caller's response to text using a speech recognition system;

[1359] A means for analyzing the converted text and determining whether it is trustworthy; and

[1360] means for maintaining whitelists and blacklists and for matching caller responses against the whitelists and blacklists;

[1361] means for connecting the call to the user if the user selects transfer;

[1362] The system includes a means for terminating a call if the user chooses to hang up.

[1363] (Claim 2)

[1364] 2. The system according to claim 1, further comprising means for converting the contents of the caller's response into text using a voice recognition system, and said determination means for analyzing the converted text and determining whether it is trustworthy.

[1365] (Claim 3)

[1366] 2. The system of claim 1, wherein a white list and a black list are maintained, and said determining means includes means for checking the contents of the caller's response against the white list and the black list.

[1367] "Application Example 1"

[1368] (Claim 1)

[1369] When a call comes in, an answering machine first answers the call and asks for the caller's identification and requirements;

[1370] A means for analyzing the content of the caller's response and determining whether the caller is trustworthy;

[1371] means for blocking calls from callers determined to be untrusted;

[1372] a means for routing calls from callers determined to be trusted to a user;

[1373] If the requirement for a trusted sender is related to food delivery, a means to automatically convert delivery status into text and notify users,

[1374] A means to determine whether a given caller or requirement falls under a whitelist or blacklist and take appropriate action; and

[1375] A system including:

[1376] (Claim 2)

[1377] a means for converting the caller's response into text using a voice recognition system;

[1378] 2. The system according to claim 1, wherein the determining means analyzes the converted text and determines whether it is trustworthy.

[1379] (Claim 3)

[1380] 2. The system of claim 1, wherein a white list and a black list are maintained, and said determining means includes means for checking the contents of the caller's response against the white list and the black list.

[1381] "Example 2: Combining Emotion Engines"

[1382] (Claim 1)

[1383] When a call comes in, an answering machine first answers the call and asks for the caller's identification and requirements;

[1384] A means for analyzing the content of the caller's response and determining whether the caller is trustworthy;

[1385] means for blocking calls from callers determined to be untrusted;

[1386] a means for routing calls from callers determined to be trusted to a user;

[1387] Equipped with an emotion recognition engine that detects the user's emotional state in real time,

[1388] means for blocking or routing calls based on the user's emotional state;

[1389] A system including:

[1390] (Claim 2)

[1391] 2. The system according to claim 1, further comprising means for converting the contents of the caller's response into text using a voice recognition system, and said determination means for analyzing the converted text and determining whether it is trustworthy.

[1392] (Claim 3)

[1393] 2. The system of claim 1, wherein a white list and a black list are maintained, and said determining means includes means for checking the contents of the caller's response against the white list and the black list.

[1394] "Application example 2 when combining emotion engines"

[1395] (Claim 1)

[1396] When a call comes in, an answering machine first answers the call and asks for the caller's identification and requirements;

[1397] A means for analyzing the content of the caller's response and determining whether the caller is trustworthy;

[1398] means for blocking calls from callers determined to be untrusted;

[1399] a means for routing calls from callers determined to be trusted to a user;

[1400] an emotion engine for detecting an emotional state of a user and means for filtering calls based on the emotional state;

[1401] A system including means for transferring calls when a user indicates relief or comfort and for blocking calls when a user indicates stress or anxiety.

[1402] (Claim 2)

[1403] a means for converting the caller's response into text using a voice recognition system;

[1404] 2. The system according to claim 1, wherein the determining means analyzes the converted text and determines whether it is trustworthy.

[1405] (Claim 3)

[1406] 2. The system of claim 1, wherein a white list and a black list are maintained, and said determining means includes means for checking the contents of the caller's response against the white list and the black list.

[1407] (Claim 4)

[1408] 2. The system of claim 1, further comprising means for analyzing the user's emotional state using a generative AI model and generating prompt sentences to obtain an emotional analysis result. [Explanation of symbols]

[1409] 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. When a call comes in, an answering machine first answers the call and asks for the caller's identification and requirements; A means for analyzing the content of the caller's response and determining whether the caller is trustworthy; means for blocking calls from callers determined to be untrusted; A system including means for routing calls from callers determined to be trusted to a user.

2. a means for converting the caller's response into text using a voice recognition system; 2. The system according to claim 1, wherein the determining means analyzes the converted text and determines whether it is reliable.

3. 2. The system of claim 1, wherein a white list and a black list are maintained, and said determining means includes means for checking the contents of the caller's response against the white list and the black list.

Images