system

Abstract

The system according to the embodiment aims to determine the possibility of fraudulent calls in real time and prevent them from occurring. [Solution] A system according to an embodiment includes a listening unit, an analysis unit, a determination unit, a recording unit, and a termination unit. The listening unit listens to the content of the call in real time. The analysis unit analyzes the content of the call collected by the listening unit. The determination unit determines the possibility of a fraudulent call based on the talk pattern analyzed by the analysis unit. The recording unit starts recording when the determination unit determines that the call is likely to be fraudulent. The termination unit ends the call when recording has been started by the recording unit.

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] Conventional technology has had the problem of making it difficult to detect fraudulent calls in real time and prevent them before they occur.

[0005] The system according to the embodiment aims to determine the possibility of fraudulent calls in real time and prevent them from occurring. [Means for solving the problem]

[0006] The system according to the embodiment includes a listening unit, an analyzing unit, a determining unit, a recording unit, and a terminating unit. The listening unit listens to the content of the call in real time. The analyzing unit analyzes the content of the call collected by the listening unit. The determining unit determines the possibility of a fraudulent call based on the talk pattern analyzed by the analyzing unit. The recording unit starts recording when the determining unit determines that the call is likely to be fraudulent. The terminating unit ends the call when recording has been started by the recording unit. [Effects of the Invention]

[0007] The system according to the embodiment can determine the possibility of a fraudulent call in real time and prevent it from occurring. [Brief explanation of the drawings]

[0008] [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. DETAILED DESCRIPTION OF THE INVENTION

[0009] 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.

[0010] First, the terms used in the following description will be explained.

[0011] 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, the 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), an APU (Accelerated Processing Unit), or a TPU (Tensor Processing Unit).

[0012] 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.

[0013] 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.

[0014] 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), and Bluetooth (registered trademark).

[0015] 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."

[0016] [First embodiment] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0017] 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.

[0018] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and 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).

[0019] 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.

[0020] The reception device 38 includes a touch panel 38A and a microphone 38B, and receives user input. The touch panel 38A detects contact with a pointer (for example, a pen or a finger) to receive user input by the touch of the pointer. 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 (see FIG. 2) acquires the data indicating the user input.

[0021] Output device 40 includes a display 40A and a speaker 40B, and presents data to a user by outputting the data in a form of expression that the user can perceive (e.g., audio and / or text). Display 40A displays visible information such as text and images in accordance with instructions from processor 46. Speaker 40B outputs audio in accordance with instructions from processor 46. 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.

[0022] 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.

[0023] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0024] 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.

[0025] 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. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0026] In the smart device 14, the specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. The specific processing program 60 is used together with the specific processing program 56 by the data processing system 10. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific processing program 60 executed on the RAM 48. Note that the smart device 14 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the specific processing unit 290 using these models.

[0027] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device (e.g., a generation server) may have the data generation model 58. In this case, the data processing device 12 obtains a processing result (prediction result, etc.) using the data generation model 58 by communicating with the server device having the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device owned by a user (e.g., a mobile phone, a robot, a home appliance, etc.). Next, an example of processing by the data processing system 10 according to the first embodiment will be described.

[0028] (Example 1) The fraudulent call prevention system according to an embodiment of the present invention is a telephone option service designed to prevent malicious fraudulent calls. In this fraudulent call prevention system, AI listens to the content of calls received from unregistered phone numbers in real time and analyzes the talk patterns. If the analyzed talk patterns share many similarities with those of fraudulent cases, the system determines the likelihood of a fraudulent call based on the number of similarities. For example, if the likelihood of a fraudulent call reaches 70%, the system informs both the caller and the callee that "This call is likely to be a fraudulent call, so we will begin recording," and asks them to hang up. This system prevents fraud before it occurs. Furthermore, because new fraudulent call methods are constantly being developed, it is necessary to train the AI ​​to learn new methods. For example, when an AI listens to the content of a call received from an unregistered phone number in real time, the AI ​​analyzes the content and extracts talk patterns. For example, the AI ​​checks whether specific keywords or phrases are included and whether the flow of the conversation resembles those of fraudulent cases. Next, if the analyzed talk patterns share many similarities with those of fraudulent cases, the system determines the likelihood of a fraudulent call based on the number of similarities. For example, if a call has many similarities to fraudulent cases, it is determined to be highly likely to be a fraudulent call. If the likelihood of a fraudulent call reaches 70%, both the caller and the recipient are notified, "This call is likely to be a fraudulent call, so we will begin recording." Furthermore, if a call is determined to be highly likely to be a fraudulent call, the call is terminated. In this case, the AI ​​automatically ends the call, preventing fraud before it occurs. For example, if a call is determined to be highly likely to be a fraudulent call, the AI ​​will inform the caller, "This call is likely to be a fraudulent call, so we will end the call," and then end the call. Furthermore, because new fraudulent call methods are constantly being developed, it is necessary to train the AI ​​to learn these new methods. For example, if new fraudulent methods are discovered, the accuracy of fraudulent call detection can be improved by training the AI ​​to learn these methods. In this way, fraudulent call prevention systems can prevent malicious fraudulent calls before they occur.

[0029] A fraudulent call prevention system according to an embodiment includes a listening unit, an analyzing unit, a determining unit, a recording unit, and a terminating unit. The listening unit listens to the content of the call in real time. For example, the listening unit may use noise canceling technology to collect the content of the call with high accuracy. The listening unit may also use voice recognition technology to convert the content of the call into text data. The listening unit is further equipped with high-speed processing technology to analyze the content of the call in real time. For example, the listening unit converts the content of the call into text data in real time and transmits it to the analyzing unit. The analyzing unit analyzes the content of the call collected by the listening unit. For example, the analyzing unit may use natural language processing technology to extract specific keywords or phrases included in the content of the call. The analyzing unit may also use a machine learning algorithm to analyze the talk pattern of the content of the call. The analyzing unit is further equipped with high-speed processing technology to analyze the content of the call in real time. For example, the analyzing unit extracts specific keywords or phrases included in the content of the call and analyzes the talk pattern. The determining unit determines the possibility of a fraudulent call based on the talk pattern analyzed by the analyzing unit. For example, the determination unit may use a machine learning algorithm to determine the possibility of a fraudulent call based on the number of points similar to fraud cases. The determination unit may also include high-speed processing technology for determining the possibility of a fraudulent call in real time. The determination unit may also set a threshold for determining the possibility of a fraudulent call. For example, the determination unit may determine a call as a fraudulent call when the possibility of a fraudulent call reaches 70%. The recording unit may start recording when the determination unit determines that the possibility of a fraudulent call is high. For example, the recording unit may use audio recording technology for recording the contents of the call with high sound quality. The recording unit may also use encryption technology for securely storing the recorded data. The recording unit may also include high-speed processing technology for saving the recorded data in real time. For example, the recording unit may record and store the contents of the call in real time. The termination unit may end the call when recording is started by the recording unit. For example, the termination unit may use call control technology for automatically ending the call.The termination unit may also use a notification technique for providing a notification after the call has ended. Furthermore, the termination unit may include a storage technique for storing the recorded data after the call has ended. For example, the termination unit may automatically end the call and store the recorded data. This allows the fraudulent call prevention system according to the embodiment to prevent malicious fraudulent calls.

[0030] The fraudulent call prevention system includes a learning unit that learns new fraudulent methods. The learning unit learns new fraudulent methods. For example, the learning unit can use a machine learning algorithm to learn new fraudulent methods based on data from past fraud cases. The learning unit also includes high-speed processing technology for learning new fraudulent methods in real time. The learning unit also includes data collection technology for learning new fraudulent methods. For example, the learning unit automatically collects and learns from fraud case data on the Internet. This allows the system to learn new fraudulent methods and improve the accuracy of identifying fraudulent calls.

[0031] The fraudulent call prevention system includes a notification unit that notifies the user of a possible fraudulent call. The notification unit notifies the user of a possible fraudulent call. For example, the notification unit can use voice synthesis technology to provide a voice notification when a fraudulent call is highly likely. The notification unit can also use message sending technology to send a text message when a fraudulent call is highly likely. The notification unit also includes alert display technology to display an alert when a fraudulent call is highly likely. For example, the notification unit can provide a voice notification when a fraudulent call is highly likely, warning both the caller and the recipient. This allows the user to be warned by notifying the user of a possible fraudulent call.

[0032] The analysis unit can analyze whether predefined keywords or phrases are included. The analysis unit can use, for example, natural language processing technology to extract specific keywords or phrases included in the call content. For example, the analysis unit extracts specific keywords or phrases included in the call content in real time and identifies factors that increase the likelihood of a fraudulent call. The analysis unit can also use a machine learning algorithm to analyze the frequency of occurrence of specific keywords or phrases. For example, the analysis unit analyzes the frequency of occurrence of specific keywords or phrases included in the call content and identifies factors that increase the likelihood of a fraudulent call. Furthermore, the analysis unit is equipped with pattern recognition technology to analyze combinations of specific keywords and phrases. For example, the analysis unit analyzes combinations of specific keywords and phrases included in the call content and identifies factors that increase the likelihood of a fraudulent call. In this way, by analyzing specific keywords and phrases, it is possible to identify factors that increase the likelihood of a fraudulent call.

[0033] The determination unit can determine the likelihood of a fraudulent call based on the number of features that match fraud cases. The determination unit can use, for example, a machine learning algorithm to determine the likelihood of a fraudulent call based on the number of similarities with fraud cases. For example, the determination unit can analyze the frequency of occurrence of specific keywords or phrases included in the call content and determine the likelihood of a fraudulent call based on the number of features that match fraud cases. The determination unit can also use pattern recognition technology to analyze combinations of features that match fraud cases. For example, the determination unit can analyze combinations of specific keywords or phrases included in the call content and determine the likelihood of a fraudulent call based on the number of features that match fraud cases. Furthermore, the determination unit can include a weighting algorithm to weight the features that match fraud cases. For example, the determination unit can evaluate the importance of specific keywords or phrases included in the call content and determine the likelihood of a fraudulent call based on the number of features that match fraud cases. In this way, by determining the likelihood of a fraudulent call based on the number of similarities with fraud cases, the accuracy of fraudulent call detection can be improved.

[0034] The recording unit can start recording when the likelihood of a fraudulent call reaches a preset threshold. The recording unit can use threshold setting technology to start recording when the likelihood of a fraudulent call reaches, for example, 70%. For example, the recording unit automatically starts recording when the likelihood of a fraudulent call reaches 70%. The recording unit can also use audio recording technology to save the recorded data in high quality. For example, the recording unit records and saves the contents of the call in high quality. Furthermore, the recording unit is equipped with encryption technology to safely save the recorded data. For example, the recording unit encrypts and saves the recorded data. In this way, by starting recording when the likelihood of a fraudulent call reaches 70%, evidence of fraud can be secured.

[0035] The termination unit can terminate the call when recording starts. The termination unit can use, for example, call control technology for automatically terminating the call when recording starts. For example, the termination unit automatically terminates the call immediately after recording starts. The termination unit can also use notification technology for providing a notification after the call ends. For example, the termination unit notifies the caller that "This call has been terminated due to possible fraud" after the call ends. Furthermore, the termination unit includes storage technology for saving the recorded data after the call ends. For example, the termination unit automatically saves the recorded data after the call ends. In this way, by terminating the call when recording starts, fraud can be prevented before it occurs.

[0036] The listening unit may be provided with a function to automatically remove background sounds and noise when listening to the contents of a call. The listening unit may use, for example, noise filtering technology to remove background sounds and noise. For example, the listening unit may remove background sounds, such as wind noise and traffic noise, that occur during a call in real time. The listening unit may also use echo cancellation technology to remove echoes and reverberation. For example, the listening unit may automatically filter echoes and reverberation that occur during a call. Furthermore, the listening unit may include speech recognition technology to remove other people's conversations and noise. For example, the listening unit may identify and remove other people's conversations and noise that occur during a call. This may improve the accuracy of listening to the contents of a call by removing background sounds and noise.

[0037] When listening to the contents of a call, the listening unit can analyze the characteristics of the speaker's voice to identify the speaker. The listening unit can use, for example, voice analysis technology to analyze the characteristics of the speaker's voice. For example, the listening unit can analyze the tone and pitch of the speaker's voice to identify a specific speaker. The listening unit can also use voice recognition technology to analyze the rhythm and speed of the speaker's voice. For example, the listening unit can analyze the rhythm and speed of the speaker's voice to identify a specific speaker. Furthermore, the listening unit is equipped with voice processing technology to analyze the range of the speaker's voice and pronunciation characteristics. For example, the listening unit can analyze the range of the speaker's voice and pronunciation characteristics to identify a specific speaker. In this way, a specific speaker can be identified by analyzing the characteristics of the speaker's voice.

[0038] When listening to the call content, the listening unit can perform sampling at specific time intervals from the start of the call. The listening unit can use, for example, voice analysis technology to sample the call content. For example, the listening unit samples and analyzes the call content every five seconds from the start of the call. The listening unit can also use time interval setting technology to track changes in the call content. For example, the listening unit samples and analyzes the call content every ten seconds from the start of the call. Furthermore, the listening unit is equipped with high-speed processing technology to track changes in the call content over time. For example, the listening unit samples and analyzes the call content every 15 seconds from the start of the call. In this way, sampling at specific time intervals makes it easier to track changes in the call content.

[0039] The listening unit may be provided with a function for supporting multiple languages ​​when listening to call content. The listening unit may use, for example, speech recognition technology for supporting multiple languages. For example, the listening unit may simultaneously listen to and analyze call content in English and Japanese. The listening unit may also use a language model for supporting multiple languages. For example, the listening unit may simultaneously listen to and analyze call content in Spanish and French. Furthermore, the listening unit may be equipped with high-speed processing technology for supporting multiple languages. For example, the listening unit may simultaneously listen to and analyze call content in Chinese and Korean. This allows support for multiple languages, making it possible to simultaneously analyze call content in different languages.

[0040] During analysis, the analysis unit can analyze talk patterns by taking into account the context of the call content. The analysis unit can use, for example, natural language processing technology to take into account the context of the call content. For example, the analysis unit analyzes the meaning of specific keywords by taking into account the context before and after the call content. The analysis unit can also use a machine learning algorithm to analyze the speaker's intention based on the context of the call content. For example, the analysis unit analyzes the speaker's intention based on the context of the call content. Furthermore, the analysis unit is equipped with high-speed processing technology to analyze talk patterns by taking into account the context of the call content. For example, the analysis unit takes into account the context of the call content and identifies factors that increase the likelihood of fraud. As a result, by taking into account the context of the call content, the accuracy of the talk pattern analysis can be improved.

[0041] During the analysis, the analysis unit can analyze talk patterns by tracking changes in the content of the call over time. The analysis unit can use, for example, time analysis technology to track changes in the content of the call over time. For example, the analysis unit tracks changes in the content of the call over time from the start to the end of the call and analyzes the talk patterns. The analysis unit can also use a machine learning algorithm to analyze changes in the talk patterns over a specific time period. For example, the analysis unit analyzes changes in the talk patterns over a specific time period during the call. Furthermore, the analysis unit includes high-speed processing technology to track changes in the content of the call over time and analyze the talk patterns. For example, the analysis unit identifies factors that increase the likelihood of fraud based on changes in the content of the call over time. This allows for detailed analysis of changes in the talk patterns by tracking changes in the content of the call over time.

[0042] During the analysis, the analysis unit can analyze the frequency of occurrence of specific keywords and phrases in the call content. The analysis unit can use, for example, natural language processing technology to analyze the frequency of occurrence of specific keywords and phrases included in the call content. For example, the analysis unit can analyze the frequency of occurrence of specific keywords and phrases included in the call content in real time to identify factors that increase the likelihood of a fraudulent call. The analysis unit can also use a machine learning algorithm to analyze the frequency of occurrence of specific keywords and phrases. For example, the analysis unit can analyze the frequency of occurrence of specific keywords and phrases included in the call content to identify factors that increase the likelihood of a fraudulent call. Furthermore, the analysis unit is equipped with high-speed processing technology to analyze the frequency of occurrence of specific keywords and phrases. For example, the analysis unit can analyze the frequency of occurrence of specific keywords and phrases included in the call content in real time to identify factors that increase the likelihood of a fraudulent call. In this way, by analyzing the frequency of occurrence of specific keywords and phrases, it is possible to identify factors that increase the likelihood of a fraudulent call.

[0043] During the analysis, the analysis unit can analyze the emotions and intentions of the speaker of the call content. The analysis unit can use, for example, voice analysis technology to analyze the emotions and intentions of the speaker of the call content. For example, the analysis unit analyzes the tone and pitch of the speaker's voice to infer emotions. The analysis unit can also use natural language processing technology to analyze the speaker's choice of words and the flow of speech. For example, the analysis unit analyzes the speaker's choice of words and the flow of speech to infer intentions. Furthermore, the analysis unit is equipped with a machine learning algorithm to analyze the emotions and intentions of the speaker of the call content. For example, the analysis unit determines the possibility of fraud based on the speaker's emotions and intentions. In this way, by analyzing the speaker's emotions and intentions, it is possible to identify factors that increase the possibility of a fraudulent call.

[0044] The determination unit can improve the accuracy of the determination by referring to past fraud case data when making a determination. The determination unit can, for example, use database technology for referring to past fraud case data. For example, the determination unit determines the possibility of a fraudulent call based on past fraud case data. The determination unit can also use a machine learning algorithm for referring to past fraud case data. For example, the determination unit refers to past fraud case data and determines the possibility of a fraudulent call based on the number of similarities. Furthermore, the determination unit is equipped with high-speed processing technology for referring to past fraud case data in real time. For example, the determination unit analyzes past fraud case data and can also respond to new fraud methods. In this way, by referring to past fraud case data, the accuracy of determining fraudulent calls can be improved.

[0045] When making a judgment, the judgment unit can take into consideration the tone and speed of the speaker's voice in the call content. The judgment unit can use, for example, voice analysis technology for analyzing the tone and speed of the speaker's voice. For example, the judgment unit analyzes the tone of the speaker's voice to determine the possibility of a fraudulent call. The judgment unit can also use voice recognition technology for analyzing the speed of the speaker's voice. For example, the judgment unit analyzes the speed of the speaker's voice to determine the possibility of a fraudulent call. Furthermore, the judgment unit is equipped with a machine learning algorithm for determining the possibility of a fraudulent call based on the tone and speed of the speaker's voice. For example, the judgment unit determines the possibility of a fraudulent call based on the tone and speed of the speaker's voice. In this way, by taking the tone and speed of the speaker's voice into consideration, the accuracy of determining fraudulent calls can be improved.

[0046] The determination unit can make a determination by taking into consideration a combination of specific phrases in the content of the call. The determination unit can use, for example, natural language processing technology to analyze a combination of specific phrases in the content of the call. For example, the determination unit analyzes a combination of specific phrases in the content of the call and determines the possibility of a fraudulent call. The determination unit can also use a machine learning algorithm to analyze a combination of specific phrases. For example, the determination unit identifies factors that increase the possibility of a fraudulent call based on a combination of specific phrases in the content of the call. Furthermore, the determination unit is equipped with high-speed processing technology to analyze a combination of specific phrases. For example, the determination unit analyzes a combination of specific phrases in the content of the call and determines the possibility of a fraudulent call. In this way, by taking into consideration a combination of specific phrases, the accuracy of determining fraudulent calls can be improved.

[0047] When making a judgment, the judgment unit can refer to the speaker's past call history of the call content to make a judgment. The judgment unit can use, for example, database technology for referring to the speaker's past call history. For example, the judgment unit judges the possibility of a fraudulent call based on the speaker's past call history. The judgment unit can also use a machine learning algorithm for referring to the speaker's past call history. For example, the judgment unit identifies factors that increase the possibility of a fraudulent call based on the speaker's past call history. Furthermore, the judgment unit is equipped with high-speed processing technology for referring to the speaker's past call history in real time. For example, the judgment unit considers the speaker's past call history to judge the possibility of a fraudulent call. In this way, by referring to the speaker's past call history, the accuracy of fraudulent call judgment can be improved.

[0048] The recording unit may add a function to automatically highlight important parts of the call content during recording. The recording unit may use, for example, natural language processing technology to highlight important parts of the call content. For example, the recording unit may automatically highlight important keywords or phrases in the call content. The recording unit may also use a machine learning algorithm to extract important parts of the call content. For example, the recording unit may automatically extract and highlight important information in the call content. Furthermore, the recording unit may include pattern recognition technology to identify important parts of the call content. For example, the recording unit may automatically identify and highlight important parts in the call content. This allows important information to be easily checked later by highlighting the important parts of the call content.

[0049] The recording unit may add a function to optimize the sound quality of the call content during recording. The recording unit may use, for example, voice processing technology to optimize the sound quality of the call content. For example, the recording unit may analyze and optimize the sound quality of the call content in real time. The recording unit may also use noise reduction technology to remove noise from the call content. For example, the recording unit may remove noise from the call content to improve the sound quality. Furthermore, the recording unit may include volume adjustment technology to automatically adjust the volume of the call content. For example, the recording unit may automatically adjust the volume of the call content to provide optimal sound quality. This may improve the quality of the recording by optimizing the sound quality of the call content.

[0050] The recording unit may add a function to automatically tag specific portions of the call content during recording. The recording unit may, for example, use natural language processing technology to tag specific portions of the call content. For example, the recording unit may tag important keywords or phrases in the call content. The recording unit may also use machine learning algorithms to identify specific portions of the call content. For example, the recording unit may tag specific topics in the call content. Furthermore, the recording unit may include pattern recognition technology to tag specific portions of the call content. For example, the recording unit may tag specific time periods in the call content. By tagging specific portions of the call content, important information can be easily searched for later.

[0051] The recording unit can be added with a function to automatically back up call content during recording. The recording unit can use, for example, cloud storage technology for backing up call content. For example, the recording unit automatically backs up recorded data to the cloud. The recording unit can also use local storage technology for backing up call content. For example, the recording unit automatically backs up recorded data to local storage. Furthermore, the recording unit is equipped with external device technology for backing up call content. For example, the recording unit automatically backs up recorded data to an external device. This automatically backs up call content, thereby ensuring the safety of the recorded data.

[0052] The termination unit may add a function to automatically generate a summary of the call content at the end of the call. The termination unit may use, for example, natural language processing technology to generate a summary of the call content. For example, the termination unit may automatically summarize the important points of the call content. The termination unit may also use a machine learning algorithm to generate a summary of the call content. For example, the termination unit may automatically summarize the main topics of the call content. Furthermore, the termination unit may include pattern recognition technology to generate a summary of the call content. For example, the termination unit may automatically summarize the conclusion of the call content. This allows the automatic generation of a summary of the call content to easily check important information later.

[0053] The termination unit may be added with a function of automatically evaluating the content of a call when the call ends. The termination unit may use, for example, natural language processing technology for evaluating the content of the call. For example, the termination unit may automatically evaluate the quality of the content of the call. The termination unit may also use a machine learning algorithm for evaluating the content of the call. For example, the termination unit may automatically evaluate the usefulness of the content of the call. Furthermore, the termination unit may include pattern recognition technology for evaluating the content of the call. For example, the termination unit may automatically evaluate the reliability of the content of the call. This allows the quality of the call to be improved by automatically evaluating the content of the call.

[0054] The termination unit may be added with a function to automatically save important portions of the call content when the call ends. The termination unit may, for example, use natural language processing technology to save important portions of the call content. For example, the termination unit may automatically save important keywords or phrases in the call content. The termination unit may also use a machine learning algorithm to save important portions of the call content. For example, the termination unit may automatically save important information in the call content. Furthermore, the termination unit may include pattern recognition technology to save important portions of the call content. For example, the termination unit may automatically save important portions of the call content. By automatically saving important portions of the call content, important information can be easily checked later.

[0055] The termination unit may add a function of automatically reporting the analysis results of the call content at the end of the call. The termination unit may use, for example, natural language processing technology for reporting the analysis results of the call content. For example, the termination unit may automatically report the analysis results of the call content. The termination unit may also use a machine learning algorithm for reporting the analysis results of the call content. For example, the termination unit may automatically report the important points of the call content. Furthermore, the termination unit may include pattern recognition technology for reporting the analysis results of the call content. For example, the termination unit may automatically report the conclusion of the call content. This makes it possible to improve the quality of calls by automatically reporting the analysis results of the call content.

[0056] During learning, the learning unit can optimize the learning algorithm by referring to past fraud case data. The learning unit can, for example, use database technology for referring to past fraud case data. For example, the learning unit optimizes the learning algorithm based on past fraud case data. The learning unit can also use a machine learning algorithm for referring to past fraud case data. For example, the learning unit improves the accuracy of the learning algorithm by referring to past fraud case data. Furthermore, the learning unit is equipped with high-speed processing technology for referring to past fraud case data in real time. For example, the learning unit analyzes past fraud case data and improves the learning algorithm. In this way, the accuracy of the learning algorithm can be improved by referring to past fraud case data.

[0057] The learning unit may be added with a function of automatically detecting and learning new patterns in call content during learning. The learning unit may use, for example, pattern recognition technology for detecting new patterns in call content. For example, the learning unit automatically detects and learns new patterns in call content. The learning unit may also use a machine learning algorithm for detecting new fraudulent methods. For example, the learning unit automatically detects and learns new fraudulent methods in call content. Furthermore, the learning unit is equipped with high-speed processing technology for detecting new patterns in real time. For example, the learning unit automatically detects and learns new talk patterns in call content. This makes it possible to respond to changes in fraudulent methods by automatically detecting new patterns in call content.

[0058] The learning unit may add a function to learn the frequency of occurrence of specific keywords or phrases in the call content during learning. The learning unit may use, for example, natural language processing technology to learn the frequency of occurrence of specific keywords or phrases in the call content. For example, the learning unit learns the frequency of occurrence of specific keywords in the call content. The learning unit may also use a machine learning algorithm to learn the frequency of occurrence of specific keywords or phrases. For example, the learning unit learns the frequency of occurrence of specific phrases in the call content. Furthermore, the learning unit is equipped with high-speed processing technology to learn the frequency of occurrence of specific keywords or phrases. For example, the learning unit improves the learning algorithm based on the frequency of occurrence of specific keywords or phrases in the call content. In this way, the accuracy of the learning algorithm can be improved by learning the frequency of occurrence of specific keywords or phrases.

[0059] The learning unit can be added with a function for learning the emotions and intentions of the speaker of the call content during learning. The learning unit can use, for example, a voice analysis technology for learning the emotions and intentions of the speaker of the call content. For example, the learning unit learns the tone and pitch of the speaker's voice to estimate emotions. The learning unit can also use natural language processing technology for learning the speaker's word choice and speech flow. For example, the learning unit learns the speaker's word choice and speech flow to estimate intentions. Furthermore, the learning unit is equipped with a machine learning algorithm for learning the emotions and intentions of the speaker of the call content. For example, the learning unit improves the learning algorithm based on the speaker's emotions and intentions. In this way, the accuracy of the learning algorithm can be improved by learning the speaker's emotions and intentions.

[0060] The notification unit may add a function to automatically highlight important parts of the call content when notifying. The notification unit may use, for example, natural language processing technology to highlight important parts of the call content. For example, the notification unit may automatically highlight important keywords or phrases in the call content. The notification unit may also use a machine learning algorithm to extract important parts of the call content. For example, the notification unit may automatically extract and highlight important information in the call content. Furthermore, the notification unit may include pattern recognition technology to identify important parts of the call content. For example, the notification unit may automatically identify and highlight important parts of the call content. By highlighting important parts of the call content, important information can be easily checked later.

[0061] The notification unit may add a function to automatically generate a summary of the call content at the time of notification. The notification unit may use, for example, natural language processing technology to generate a summary of the call content. For example, the notification unit may automatically summarize important points of the call content. The notification unit may also use a machine learning algorithm to generate a summary of the call content. For example, the notification unit may automatically summarize the main topics of the call content. Furthermore, the notification unit may include pattern recognition technology to generate a summary of the call content. For example, the notification unit may automatically summarize the conclusion of the call content. By automatically generating a summary of the call content, important information can be easily checked later.

[0062] The notification unit may add a function to automatically tag specific portions of the call content when notifying. The notification unit may use, for example, natural language processing technology to tag specific portions of the call content. For example, the notification unit may tag important keywords or phrases in the call content. The notification unit may also use a machine learning algorithm to identify specific portions of the call content. For example, the notification unit may tag specific topics in the call content. Furthermore, the notification unit may include pattern recognition technology to tag specific portions of the call content. For example, the notification unit may tag specific time periods in the call content. By tagging specific portions of the call content, important information can be easily searched for later.

[0063] The notification unit can add a function to automatically back up the call content at the time of notification. The notification unit can use, for example, cloud storage technology for backing up the call content. For example, the notification unit automatically backs up the notification content to the cloud. The notification unit can also use local storage technology for backing up the call content. For example, the notification unit automatically backs up the notification content to local storage. Furthermore, the notification unit is equipped with external device technology for backing up the call content. For example, the notification unit automatically backs up the notification content to an external device. By automatically backing up the call content, the safety of the notification data can be ensured.

[0064] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0065] The fraudulent call prevention system may further include a summarization unit that generates a summary of the call content. The summarization unit analyzes the call content in real time, extracts important points, and generates the summary. For example, the summarization unit may use natural language processing technology to automatically summarize the main topics and conclusions of the call content. The summarization unit may also use a machine learning algorithm to generate the summary of the call content. Furthermore, the summarization unit may include high-speed processing technology to generate the summary of the call content in real time. This allows the automatic generation of a summary of the call content to easily check important information later.

[0066] The fraudulent call prevention system may further include a translation unit that translates the content of the call. The translation unit translates the content of the call in real time and provides it to the user. For example, the translation unit may use machine translation technology to translate the content of the call into multiple languages. The translation unit may also use natural language processing technology to improve the accuracy of the translation of the content of the call. Furthermore, the translation unit may include high-speed processing technology to translate the content of the call in real time. This allows the content of the call to be translated into multiple languages, facilitating communication between users who speak different languages.

[0067] The fraudulent call prevention system may further include a topic classification unit that classifies the topics of the call content. The topic classification unit analyzes the call content and classifies the main topics. For example, the topic classification unit may use natural language processing technology to classify the call content into specific categories. The topic classification unit may also use a machine learning algorithm to classify the topics of the call content in real time. Furthermore, the topic classification unit may include high-speed processing technology to classify the topics of the call content in real time. This allows for classifying the topics of the call content, thereby improving the accuracy of detecting fraudulent calls.

[0068] The fraudulent call prevention system may further include a reliability evaluation unit that evaluates the reliability of the call content. The reliability evaluation unit analyzes the call content and evaluates its reliability. For example, the reliability evaluation unit may use natural language processing technology to evaluate the consistency and logic of the call content. The reliability evaluation unit may also use a machine learning algorithm to evaluate the reliability of the call content in real time. Furthermore, the reliability evaluation unit may include high-speed processing technology to evaluate the reliability of the call content in real time. As a result, by evaluating the reliability of the call content, it is possible to identify factors that increase the likelihood of a fraudulent call.

[0069] The fraudulent call prevention system may further include a summarization unit that generates a summary of the call content. The summarization unit analyzes the call content in real time, extracts important points, and generates the summary. For example, the summarization unit may use natural language processing technology to automatically summarize the main topics and conclusions of the call content. The summarization unit may also use a machine learning algorithm to generate the summary of the call content. Furthermore, the summarization unit may include high-speed processing technology to generate the summary of the call content in real time. This allows the automatic generation of a summary of the call content to easily check important information later.

[0070] The fraudulent call prevention system may further include a topic classification unit that classifies the topics of the call content. The topic classification unit analyzes the call content and classifies the main topics. For example, the topic classification unit may use natural language processing technology to classify the call content into specific categories. The topic classification unit may also use a machine learning algorithm to classify the topics of the call content in real time. Furthermore, the topic classification unit may include high-speed processing technology to classify the topics of the call content in real time. This allows for classifying the topics of the call content, thereby improving the accuracy of detecting fraudulent calls.

[0071] The fraudulent call prevention system may further include a reliability evaluation unit that evaluates the reliability of the call content. The reliability evaluation unit analyzes the call content and evaluates its reliability. For example, the reliability evaluation unit may use natural language processing technology to evaluate the consistency and logic of the call content. The reliability evaluation unit may also use a machine learning algorithm to evaluate the reliability of the call content in real time. Furthermore, the reliability evaluation unit may include high-speed processing technology to evaluate the reliability of the call content in real time. As a result, by evaluating the reliability of the call content, it is possible to identify factors that increase the likelihood of a fraudulent call.

[0072] The processing flow of the first embodiment will be briefly explained below.

[0073] Step 1: The listening unit listens to the call content in real time. For example, the listening unit can use noise cancellation technology to collect the call content with high accuracy. It also has voice recognition technology to convert the call content into text data and high-speed processing technology to analyze it in real time. Step 2: The analysis unit analyzes the call content collected by the listening unit. For example, the analysis unit can use natural language processing technology to extract specific keywords and phrases contained in the call content, or machine learning algorithms to analyze talk patterns. It also has high-speed processing technology for real-time analysis. Step 3: The determination unit determines the possibility of a fraudulent call based on the talk pattern analyzed by the analysis unit. For example, a machine learning algorithm can be used to determine the possibility of a fraudulent call based on the number of similarities with fraud cases, or high-speed processing technology can be used to make a determination in real time. A threshold can also be set to determine the possibility of a fraudulent call. Step 4: The recording unit starts recording if the judgment unit determines that the call is likely to be fraudulent. For example, the recording unit can use voice recording technology to record the call contents in high quality and encryption technology to safely store the recorded data. It also has high-speed processing technology to store the data in real time. Step 5: The termination unit terminates the call when recording is initiated by the recording unit. For example, the termination unit may use a call control technique for automatically terminating the call or a notification technique for notifying the user after the call has ended. The termination unit may also include a storage technique for saving the recorded data after the call has ended.

[0074] (Example 2) The fraudulent call prevention system according to an embodiment of the present invention is a telephone option service designed to prevent malicious fraudulent calls. In this fraudulent call prevention system, AI listens to the content of calls received from unregistered phone numbers in real time and analyzes the talk patterns. If the analyzed talk patterns share many similarities with those of fraudulent cases, the system determines the likelihood of a fraudulent call based on the number of similarities. For example, if the likelihood of a fraudulent call reaches 70%, the system informs both the caller and the callee that "This call is likely to be a fraudulent call, so we will begin recording," and asks them to hang up. This system prevents fraud before it occurs. Furthermore, because new fraudulent call methods are constantly being developed, it is necessary to train the AI ​​to learn new methods. For example, when an AI listens to the content of a call received from an unregistered phone number in real time, the AI ​​analyzes the content and extracts talk patterns. For example, the AI ​​checks whether specific keywords or phrases are included and whether the flow of the conversation resembles those of fraudulent cases. Next, if the analyzed talk patterns share many similarities with those of fraudulent cases, the system determines the likelihood of a fraudulent call based on the number of similarities. For example, if a call has many similarities to fraudulent cases, it is determined to be highly likely to be a fraudulent call. If the likelihood of a fraudulent call reaches 70%, both the caller and the recipient are notified, "This call is likely to be a fraudulent call, so we will begin recording." Furthermore, if a call is determined to be highly likely to be a fraudulent call, the call is terminated. In this case, the AI ​​automatically ends the call, preventing fraud before it occurs. For example, if a call is determined to be highly likely to be a fraudulent call, the AI ​​will inform the caller, "This call is likely to be a fraudulent call, so we will end the call," and then end the call. Furthermore, because new fraudulent call methods are constantly being developed, it is necessary to train the AI ​​to learn these new methods. For example, if new fraudulent methods are discovered, the accuracy of fraudulent call detection can be improved by training the AI ​​to learn these methods. In this way, fraudulent call prevention systems can prevent malicious fraudulent calls before they occur.

[0075] A fraudulent call prevention system according to an embodiment includes a listening unit, an analyzing unit, a determining unit, a recording unit, and a terminating unit. The listening unit listens to the content of the call in real time. For example, the listening unit may use noise canceling technology to collect the content of the call with high accuracy. The listening unit may also use voice recognition technology to convert the content of the call into text data. The listening unit is further equipped with high-speed processing technology to analyze the content of the call in real time. For example, the listening unit converts the content of the call into text data in real time and transmits it to the analyzing unit. The analyzing unit analyzes the content of the call collected by the listening unit. For example, the analyzing unit may use natural language processing technology to extract specific keywords or phrases included in the content of the call. The analyzing unit may also use a machine learning algorithm to analyze the talk pattern of the content of the call. The analyzing unit is further equipped with high-speed processing technology to analyze the content of the call in real time. For example, the analyzing unit extracts specific keywords or phrases included in the content of the call and analyzes the talk pattern. The determining unit determines the possibility of a fraudulent call based on the talk pattern analyzed by the analyzing unit. For example, the determination unit may use a machine learning algorithm to determine the possibility of a fraudulent call based on the number of points similar to fraud cases. The determination unit may also include high-speed processing technology for determining the possibility of a fraudulent call in real time. The determination unit may also set a threshold for determining the possibility of a fraudulent call. For example, the determination unit may determine a call as a fraudulent call when the possibility of a fraudulent call reaches 70%. The recording unit may start recording when the determination unit determines that the possibility of a fraudulent call is high. For example, the recording unit may use audio recording technology for recording the contents of the call with high sound quality. The recording unit may also use encryption technology for securely storing the recorded data. The recording unit may also include high-speed processing technology for saving the recorded data in real time. For example, the recording unit may record and store the contents of the call in real time. The termination unit may end the call when recording is started by the recording unit. For example, the termination unit may use call control technology for automatically ending the call.The termination unit may also use a notification technique for providing a notification after the call has ended. Furthermore, the termination unit may include a storage technique for storing the recorded data after the call has ended. For example, the termination unit may automatically end the call and store the recorded data. This allows the fraudulent call prevention system according to the embodiment to prevent malicious fraudulent calls.

[0076] The fraudulent call prevention system includes a learning unit that learns new fraudulent methods. The learning unit learns new fraudulent methods. For example, the learning unit can use a machine learning algorithm to learn new fraudulent methods based on data from past fraud cases. The learning unit also includes high-speed processing technology for learning new fraudulent methods in real time. The learning unit also includes data collection technology for learning new fraudulent methods. For example, the learning unit automatically collects and learns from fraud case data on the Internet. This allows the system to learn new fraudulent methods and improve the accuracy of identifying fraudulent calls.

[0077] The fraudulent call prevention system includes a notification unit that notifies the user of a possible fraudulent call. The notification unit notifies the user of a possible fraudulent call. For example, the notification unit can use voice synthesis technology to provide a voice notification when a fraudulent call is highly likely. The notification unit can also use message sending technology to send a text message when a fraudulent call is highly likely. The notification unit also includes alert display technology to display an alert when a fraudulent call is highly likely. For example, the notification unit can provide a voice notification when a fraudulent call is highly likely, warning both the caller and the recipient. This allows the user to be warned by notifying the user of a possible fraudulent call.

[0078] The analysis unit can analyze whether predefined keywords or phrases are included. For example, the analysis unit can use natural language processing techniques to extract specific keywords or phrases from the call content. For instance, the analysis unit can extract specific keywords or phrases from the call content in real time and identify elements that increase the likelihood of a fraudulent call. The analysis unit can also use machine learning algorithms to analyze the frequency of occurrence of specific keywords or phrases. For example, the analysis unit can analyze the frequency of occurrence of specific keywords or phrases from the call content and identify elements that increase the likelihood of a fraudulent call. Furthermore, the analysis unit is equipped with pattern recognition techniques to analyze combinations of specific keywords or phrases. For example, the analysis unit can analyze combinations of specific keywords or phrases from the call content and identify elements that increase the likelihood of a fraudulent call. In this way, by analyzing specific keywords or phrases, elements that increase the likelihood of a fraudulent call can be identified.

[0079] The detection unit can determine the likelihood of a call being a scam based on the number of features that match a scam case. For example, the detection unit can use a machine learning algorithm to determine the likelihood of a call being a scam based on the number of similarities to a scam case. For example, the detection unit can analyze the frequency of occurrence of specific keywords or phrases in the call content and determine the likelihood of a call being a scam based on the number of features that match a scam case. The detection unit can also use pattern recognition technology to analyze combinations of features that match a scam case. For example, the detection unit can analyze combinations of specific keywords or phrases in the call content and determine the likelihood of a call being a scam based on the number of features that match a scam case. Furthermore, the detection unit is equipped with a weighting algorithm for weighting features that match a scam case. For example, the detection unit can evaluate the importance of specific keywords or phrases in the call content and determine the likelihood of a call being a scam based on the number of features that match a scam case. This improves the accuracy of scam call detection by determining the likelihood of a call being a scam based on the number of similarities to a scam case.

[0080] The recording unit can start recording when the probability of a call being a scam reaches a preset threshold. For example, the recording unit can use threshold setting technology to start recording when the probability of a call being a scam reaches 70%. For example, the recording unit can automatically start recording when the probability of a call being a scam reaches 70%. The recording unit can also use voice recording technology to save the recorded data in high quality. For example, the recording unit can record and save the content of the call in high quality. Furthermore, the recording unit is equipped with encryption technology to securely store the recorded data. For example, the recording unit can encrypt and save the recorded data. This allows for the securing of evidence of fraud by starting recording when the probability of a call being a scam reaches 70%.

[0081] The termination unit can terminate a call when recording begins. The termination unit can use call control technology to automatically terminate a call when recording begins. For example, the termination unit can automatically terminate the call immediately after recording begins. The termination unit can also use notification technology to provide a notification after the call ends. For example, the termination unit can notify the user that "This call has been terminated because it may be a scam" after the call ends. Furthermore, the termination unit is equipped with storage technology to save the recording data after the call ends. For example, the termination unit can automatically save the recording data after the call ends. This allows for the prevention of fraud by terminating the call when recording begins.

[0082] The listening unit can estimate the user's emotions and adjust the listening accuracy of the call content based on the estimated user's emotions. The listening unit can use, for example, voice analysis technology to estimate the user's emotions. For example, the listening unit can analyze the tone and pitch of the user's voice to estimate the emotion. The listening unit can also use voice processing technology to adjust the listening accuracy of the call content based on the user's emotions. For example, the listening unit can increase the listening accuracy of the call content when the user is nervous. Furthermore, the listening unit is equipped with high-speed processing technology to adjust the listening accuracy of the call content in real time based on the user's emotions. For example, the listening unit can set the listening accuracy of the call content to normal when the user is relaxed. In this way, by adjusting the listening accuracy of the call content according to the user's emotions, important information can be heard without missing anything.

[0083] The listening unit may be provided with a function to automatically remove background sounds and noise when listening to the contents of a call. The listening unit may use, for example, noise filtering technology to remove background sounds and noise. For example, the listening unit may remove background sounds, such as wind noise and traffic noise, that occur during a call in real time. The listening unit may also use echo cancellation technology to remove echoes and reverberation. For example, the listening unit may automatically filter echoes and reverberation that occur during a call. Furthermore, the listening unit may include speech recognition technology to remove other people's conversations and noise. For example, the listening unit may identify and remove other people's conversations and noise that occur during a call. This may improve the accuracy of listening to the contents of a call by removing background sounds and noise.

[0084] When listening to the contents of a call, the listening unit can analyze the characteristics of the speaker's voice to identify the speaker. The listening unit can use, for example, voice analysis technology to analyze the characteristics of the speaker's voice. For example, the listening unit can analyze the tone and pitch of the speaker's voice to identify a specific speaker. The listening unit can also use voice recognition technology to analyze the rhythm and speed of the speaker's voice. For example, the listening unit can analyze the rhythm and speed of the speaker's voice to identify a specific speaker. Furthermore, the listening unit is equipped with voice processing technology to analyze the range of the speaker's voice and pronunciation characteristics. For example, the listening unit can analyze the range of the speaker's voice and pronunciation characteristics to identify a specific speaker. In this way, a specific speaker can be identified by analyzing the characteristics of the speaker's voice.

[0085] The listening unit can estimate the user's emotions and determine listening priorities based on the estimated user's emotions. The listening unit can use, for example, voice analysis technology to estimate the user's emotions. For example, the listening unit can analyze the tone and pitch of the user's voice to estimate the emotions. The listening unit can also use voice processing technology to determine listening priorities based on the user's emotions. For example, the listening unit prioritizes listening for important information when the user is nervous. Furthermore, the listening unit has high-speed processing technology to adjust listening priorities in real time based on the user's emotions. For example, the listening unit listens to the entire conversation evenly when the user is relaxed. In this way, by determining listening priorities according to the user's emotions, important information can be prioritized.

[0086] When listening to the call content, the listening unit can perform sampling at specific time intervals from the start of the call. The listening unit can use, for example, voice analysis technology to sample the call content. For example, the listening unit samples and analyzes the call content every five seconds from the start of the call. The listening unit can also use time interval setting technology to track changes in the call content. For example, the listening unit samples and analyzes the call content every ten seconds from the start of the call. Furthermore, the listening unit is equipped with high-speed processing technology to track changes in the call content over time. For example, the listening unit samples and analyzes the call content every 15 seconds from the start of the call. In this way, sampling at specific time intervals makes it easier to track changes in the call content.

[0087] The listening unit may be provided with a function for supporting multiple languages ​​when listening to call content. The listening unit may use, for example, speech recognition technology for supporting multiple languages. For example, the listening unit may simultaneously listen to and analyze call content in English and Japanese. The listening unit may also use a language model for supporting multiple languages. For example, the listening unit may simultaneously listen to and analyze call content in Spanish and French. Furthermore, the listening unit may be equipped with high-speed processing technology for supporting multiple languages. For example, the listening unit may simultaneously listen to and analyze call content in Chinese and Korean. This allows support for multiple languages, making it possible to simultaneously analyze call content in different languages.

[0088] The analysis unit can estimate the user's emotions and adjust the level of analysis detail based on the estimated user's emotions. The analysis unit can use, for example, voice analysis technology to estimate the user's emotions. For example, the analysis unit analyzes the tone and pitch of the user's voice to estimate the emotions. The analysis unit can also use voice processing technology to adjust the level of analysis detail based on the user's emotions. For example, the analysis unit performs a detailed analysis when the user is nervous and extracts important information. Furthermore, the analysis unit is equipped with high-speed processing technology to adjust the level of analysis detail in real time based on the user's emotions. For example, the analysis unit performs a normal analysis when the user is relaxed and grasps the overall talk pattern. As a result, important information can be analyzed in detail by adjusting the level of analysis detail according to the user's emotions.

[0089] During analysis, the analysis unit can analyze talk patterns by taking into account the context of the call content. The analysis unit can use, for example, natural language processing technology to take into account the context of the call content. For example, the analysis unit analyzes the meaning of specific keywords by taking into account the context before and after the call content. The analysis unit can also use a machine learning algorithm to analyze the speaker's intention based on the context of the call content. For example, the analysis unit analyzes the speaker's intention based on the context of the call content. Furthermore, the analysis unit is equipped with high-speed processing technology to analyze talk patterns by taking into account the context of the call content. For example, the analysis unit takes into account the context of the call content and identifies factors that increase the likelihood of fraud. As a result, by taking into account the context of the call content, the accuracy of the talk pattern analysis can be improved.

[0090] During the analysis, the analysis unit can analyze talk patterns by tracking changes in the content of the call over time. The analysis unit can use, for example, time analysis technology to track changes in the content of the call over time. For example, the analysis unit tracks changes in the content of the call over time from the start to the end of the call and analyzes the talk patterns. The analysis unit can also use a machine learning algorithm to analyze changes in the talk patterns over a specific time period. For example, the analysis unit analyzes changes in the talk patterns over a specific time period during the call. Furthermore, the analysis unit includes high-speed processing technology to track changes in the content of the call over time and analyze the talk patterns. For example, the analysis unit identifies factors that increase the likelihood of fraud based on changes in the content of the call over time. This allows for detailed analysis of changes in the talk patterns by tracking changes in the content of the call over time.

[0091] The analysis unit can estimate the user's emotions and adjust the display method of the analysis results based on the estimated user's emotions. The analysis unit can use, for example, voice analysis technology to estimate the user's emotions. For example, the analysis unit analyzes the tone and pitch of the user's voice to estimate the emotions. The analysis unit can also use display technology to adjust the display method of the analysis results based on the user's emotions. For example, the analysis unit displays concise, highly visible analysis results when the user is nervous. Furthermore, the analysis unit is equipped with high-speed processing technology to adjust the display method of the analysis results in real time based on the user's emotions. For example, the analysis unit displays detailed analysis results when the user is relaxed. In this way, by adjusting the display method of the analysis results according to the user's emotions, it is possible to provide highly visible analysis results.

[0092] During the analysis, the analysis unit can analyze the frequency of occurrence of specific keywords and phrases in the call content. The analysis unit can use, for example, natural language processing technology to analyze the frequency of occurrence of specific keywords and phrases included in the call content. For example, the analysis unit can analyze the frequency of occurrence of specific keywords and phrases included in the call content in real time to identify factors that increase the likelihood of a fraudulent call. The analysis unit can also use a machine learning algorithm to analyze the frequency of occurrence of specific keywords and phrases. For example, the analysis unit can analyze the frequency of occurrence of specific keywords and phrases included in the call content to identify factors that increase the likelihood of a fraudulent call. Furthermore, the analysis unit is equipped with high-speed processing technology to analyze the frequency of occurrence of specific keywords and phrases. For example, the analysis unit can analyze the frequency of occurrence of specific keywords and phrases included in the call content in real time to identify factors that increase the likelihood of a fraudulent call. In this way, by analyzing the frequency of occurrence of specific keywords and phrases, it is possible to identify factors that increase the likelihood of a fraudulent call.

[0093] During the analysis, the analysis unit can analyze the emotions and intentions of the speaker of the call content. The analysis unit can use, for example, voice analysis technology to analyze the emotions and intentions of the speaker of the call content. For example, the analysis unit analyzes the tone and pitch of the speaker's voice to infer emotions. The analysis unit can also use natural language processing technology to analyze the speaker's choice of words and the flow of speech. For example, the analysis unit analyzes the speaker's choice of words and the flow of speech to infer intentions. Furthermore, the analysis unit is equipped with a machine learning algorithm to analyze the emotions and intentions of the speaker of the call content. For example, the analysis unit determines the possibility of fraud based on the speaker's emotions and intentions. In this way, by analyzing the speaker's emotions and intentions, it is possible to identify factors that increase the possibility of a fraudulent call.

[0094] The detection unit can estimate the user's emotions and adjust the criteria for determining the likelihood of a fraudulent call based on the estimated emotions. For example, the detection unit can use voice analysis technology to estimate the user's emotions. For instance, it can analyze the tone and pitch of the user's voice to estimate their emotions. The detection unit can also use machine learning algorithms to adjust the criteria for determining the likelihood of a fraudulent call based on the user's emotions. For example, the detection unit can tighten the criteria for determining the likelihood of a fraudulent call when the user is tense. Furthermore, the detection unit is equipped with high-speed processing technology to adjust the criteria for determining the likelihood of a fraudulent call in real time based on the user's emotions. For example, the detection unit can set the criteria for determining the likelihood of a fraudulent call to normal when the user is relaxed. This improves the accuracy of fraudulent call detection by adjusting the criteria according to the user's emotions.

[0095] The detection unit can improve its detection accuracy by referring to past fraud case data during the detection process. For example, the detection unit can use database technology to refer to past fraud case data. For example, the detection unit can determine the likelihood of a fraudulent call based on past fraud case data. The detection unit can also use machine learning algorithms to refer to past fraud case data. For example, the detection unit can refer to past fraud case data and determine the likelihood of a fraudulent call based on the number of similarities. Furthermore, the detection unit is equipped with high-speed processing technology to refer to past fraud case data in real time. For example, the detection unit can analyze past fraud case data and respond to new fraud methods. As a result, the accuracy of detecting fraudulent calls can be improved by referring to past fraud case data.

[0096] When making a judgment, the judgment unit can take into consideration the tone and speed of the speaker's voice in the call content. The judgment unit can use, for example, voice analysis technology for analyzing the tone and speed of the speaker's voice. For example, the judgment unit analyzes the tone of the speaker's voice to determine the possibility of a fraudulent call. The judgment unit can also use voice recognition technology for analyzing the speed of the speaker's voice. For example, the judgment unit analyzes the speed of the speaker's voice to determine the possibility of a fraudulent call. Furthermore, the judgment unit is equipped with a machine learning algorithm for determining the possibility of a fraudulent call based on the tone and speed of the speaker's voice. For example, the judgment unit determines the possibility of a fraudulent call based on the tone and speed of the speaker's voice. In this way, by taking the tone and speed of the speaker's voice into consideration, the accuracy of determining fraudulent calls can be improved.

[0097] The determination unit can estimate the user's emotion and adjust the notification method of the determination result based on the estimated user's emotion. The determination unit can use, for example, voice analysis technology to estimate the user's emotion. For example, the determination unit can analyze the tone and pitch of the user's voice to estimate the emotion. The determination unit can also use notification technology to adjust the notification method of the determination result based on the user's emotion. For example, the determination unit provides a concise and highly visible notification method when the user is nervous. Furthermore, the determination unit includes high-speed processing technology for adjusting the notification method of the determination result in real time based on the user's emotion. For example, the determination unit provides a detailed notification method when the user is relaxed. In this way, by adjusting the notification method according to the user's emotion, a highly visible notification can be provided.

[0098] The determination unit can make a determination by considering specific combinations of phrases in the call content. For example, the determination unit can use natural language processing techniques to analyze specific combinations of phrases in the call content. For example, the determination unit analyzes specific combinations of phrases in the call content and determines the possibility of a fraudulent call. The determination unit can also use machine learning algorithms to analyze specific combinations of phrases. For example, the determination unit identifies elements that increase the possibility of a fraudulent call based on specific combinations of phrases in the call content. Furthermore, the determination unit is equipped with high-speed processing techniques for analyzing specific combinations of phrases. For example, the determination unit analyzes specific combinations of phrases in the call content and determines the possibility of a fraudulent call. As a result, the accuracy of fraudulent call detection can be improved by considering specific combinations of phrases.

[0099] The determination unit can make a determination by referring to the speaker's past call history during the determination process. For example, the determination unit can use database technology to refer to the speaker's past call history. For example, the determination unit can determine the possibility of a fraudulent call based on the speaker's past call history. The determination unit can also use machine learning algorithms to refer to the speaker's past call history. For example, the determination unit can identify factors that increase the likelihood of a fraudulent call based on the speaker's past call history. Furthermore, the determination unit is equipped with high-speed processing technology to refer to the speaker's past call history in real time. For example, the determination unit considers the speaker's past call history to determine the possibility of a fraudulent call. This improves the accuracy of fraudulent call detection by referring to the speaker's past call history.

[0100] The recording unit can estimate the user's emotions and adjust the start timing of recording based on the estimated user's emotions. The recording unit can use, for example, voice analysis technology to estimate the user's emotions. For example, the recording unit can analyze the tone and pitch of the user's voice to estimate the emotions. The recording unit can also use a machine learning algorithm to adjust the start timing of recording based on the user's emotions. For example, the recording unit can start recording earlier if the user is nervous. Furthermore, the recording unit is equipped with high-speed processing technology to adjust the start timing of recording in real time based on the user's emotions. For example, the recording unit can start recording at a normal timing if the user is relaxed. In this way, by adjusting the start timing of recording according to the user's emotions, important information can be recorded without missing anything.

[0101] The recording unit may add a function to automatically highlight important parts of the call content during recording. The recording unit may use, for example, natural language processing technology to highlight important parts of the call content. For example, the recording unit may automatically highlight important keywords or phrases in the call content. The recording unit may also use a machine learning algorithm to extract important parts of the call content. For example, the recording unit may automatically extract and highlight important information in the call content. Furthermore, the recording unit may include pattern recognition technology to identify important parts of the call content. For example, the recording unit may automatically identify and highlight important parts in the call content. This allows important information to be easily checked later by highlighting the important parts of the call content.

[0102] The recording unit may add a function to optimize the sound quality of the call content during recording. The recording unit may use, for example, voice processing technology to optimize the sound quality of the call content. For example, the recording unit may analyze and optimize the sound quality of the call content in real time. The recording unit may also use noise reduction technology to remove noise from the call content. For example, the recording unit may remove noise from the call content to improve the sound quality. Furthermore, the recording unit may include volume adjustment technology to automatically adjust the volume of the call content. For example, the recording unit may automatically adjust the volume of the call content to provide optimal sound quality. This may improve the quality of the recording by optimizing the sound quality of the call content.

[0103] The recording unit can estimate the user's emotions and adjust the recording storage method based on the estimated user's emotions. The recording unit can use, for example, voice analysis technology to estimate the user's emotions. For example, the recording unit can analyze the tone and pitch of the user's voice to estimate the emotions. The recording unit can also use a machine learning algorithm to adjust the recording storage method based on the user's emotions. For example, the recording unit selects a method to safely store the recording data when the user is nervous. Furthermore, the recording unit is equipped with high-speed processing technology to adjust the recording storage method in real time based on the user's emotions. For example, the recording unit selects a normal storage method when the user is relaxed. In this way, the safety of the recording data can be ensured by adjusting the recording storage method according to the user's emotions.

[0104] The recording unit may add a function to automatically tag specific portions of the call content during recording. The recording unit may, for example, use natural language processing technology to tag specific portions of the call content. For example, the recording unit may tag important keywords or phrases in the call content. The recording unit may also use machine learning algorithms to identify specific portions of the call content. For example, the recording unit may tag specific topics in the call content. Furthermore, the recording unit may include pattern recognition technology to tag specific portions of the call content. For example, the recording unit may tag specific time periods in the call content. By tagging specific portions of the call content, important information can be easily searched for later.

[0105] The recording unit can be added with a function to automatically back up call content during recording. The recording unit can use, for example, cloud storage technology for backing up call content. For example, the recording unit automatically backs up recorded data to the cloud. The recording unit can also use local storage technology for backing up call content. For example, the recording unit automatically backs up recorded data to local storage. Furthermore, the recording unit is equipped with external device technology for backing up call content. For example, the recording unit automatically backs up recorded data to an external device. This automatically backs up call content, thereby ensuring the safety of the recorded data.

[0106] The termination unit can estimate the user's emotions and adjust the timing of ending the call based on the estimated user's emotions. The termination unit can use, for example, voice analysis technology to estimate the user's emotions. For example, the termination unit can analyze the tone and pitch of the user's voice to estimate the emotions. The termination unit can also use a machine learning algorithm to adjust the timing of ending the call based on the user's emotions. For example, the termination unit can end the call early if the user is nervous. Furthermore, the termination unit includes high-speed processing technology to adjust the timing of ending the call in real time based on the user's emotions. For example, the termination unit can end the call at a normal timing if the user is relaxed. In this way, the timing of ending the call can be adjusted according to the user's emotions, thereby ending the call at an appropriate time.

[0107] The termination unit may add a function to automatically generate a summary of the call content at the end of the call. The termination unit may use, for example, natural language processing technology to generate a summary of the call content. For example, the termination unit may automatically summarize the important points of the call content. The termination unit may also use a machine learning algorithm to generate a summary of the call content. For example, the termination unit may automatically summarize the main topics of the call content. Furthermore, the termination unit may include pattern recognition technology to generate a summary of the call content. For example, the termination unit may automatically summarize the conclusion of the call content. This allows the automatic generation of a summary of the call content to easily check important information later.

[0108] The termination unit can be enhanced with a function to automatically evaluate the call content at the end of the call. For example, the termination unit can use natural language processing techniques to evaluate the call content. For instance, it can automatically evaluate the quality of the call content. The termination unit can also use machine learning algorithms to evaluate the call content. For example, it can automatically evaluate the usefulness of the call content. Furthermore, the termination unit can incorporate pattern recognition techniques for evaluating the call content. For example, it can automatically evaluate the reliability of the call content. This allows for improved call quality by automatically evaluating the call content.

[0109] The termination unit can estimate the user's emotions and adjust the notification method after the call ends based on the estimated emotions. For example, the termination unit can use voice analysis technology to estimate the user's emotions. For instance, it can analyze the tone and pitch of the user's voice to estimate their emotions. The termination unit can also use notification technology to adjust the notification method after the call ends based on the user's emotions. For example, it can provide a concise and easily visible notification method when the user is tense. Furthermore, the termination unit is equipped with high-speed processing technology to adjust the notification method after the call ends in real time based on the user's emotions. For example, it can provide a detailed notification method when the user is relaxed. This allows for the provision of easily visible notifications by adjusting the notification method according to the user's emotions.

[0110] The termination section can be enhanced to automatically save important parts of a call at the end of the call. For example, the termination section can use natural language processing techniques to save important parts of the call. For instance, it can automatically save important keywords or phrases from the call. The termination section can also use machine learning algorithms to save important parts of the call. For example, it can automatically save important information from the call. Furthermore, the termination section can incorporate pattern recognition techniques to save important parts of the call. For example, it can automatically save important parts of the call. This allows for easy later review of important information by automatically saving the important parts of the call.

[0111] The termination unit can be enhanced to automatically report the results of a call analysis at the end of the call. For example, the termination unit can use natural language processing techniques to report the call analysis results. For instance, it can automatically report the call analysis results. Alternatively, the termination unit can use machine learning algorithms to report the call analysis results. For example, it can automatically report the key points of the call. Furthermore, the termination unit can incorporate pattern recognition techniques to report the call analysis results. For example, it can automatically report the conclusion of the call. This allows for improved call quality by automatically reporting the results of the call analysis.

[0112] The learning unit can estimate the user's emotions and select training data based on those estimated emotions. For example, the learning unit can use speech analysis technology to estimate the user's emotions. For instance, it can analyze the tone and pitch of the user's voice to estimate their emotions. The learning unit can also use machine learning algorithms to select training data based on the user's emotions. For example, it prioritizes learning important fraud case data when the user is feeling stressed. Furthermore, the learning unit is equipped with high-speed processing technology to select training data in real time based on the user's emotions. For example, it learns normal fraud case data when the user is relaxed. This allows for improved learning efficiency by selecting training data according to the user's emotions.

[0113] The learning unit can optimize its learning algorithm by referencing past fraud case data during training. For example, the learning unit can use database technology to reference past fraud case data. For instance, the learning unit optimizes the learning algorithm based on past fraud case data. The learning unit can also use machine learning algorithms to reference past fraud case data. For example, the learning unit improves the accuracy of the learning algorithm by referencing past fraud case data. Furthermore, the learning unit is equipped with high-speed processing technology for real-time referencing of past fraud case data. For example, the learning unit analyzes past fraud case data to improve the learning algorithm. This allows the accuracy of the learning algorithm to be improved by referencing past fraud case data.

[0114] The learning unit may be added with a function of automatically detecting and learning new patterns in call content during learning. The learning unit may use, for example, pattern recognition technology for detecting new patterns in call content. For example, the learning unit automatically detects and learns new patterns in call content. The learning unit may also use a machine learning algorithm for detecting new fraudulent methods. For example, the learning unit automatically detects and learns new fraudulent methods in call content. Furthermore, the learning unit is equipped with high-speed processing technology for detecting new patterns in real time. For example, the learning unit automatically detects and learns new talk patterns in call content. This makes it possible to respond to changes in fraudulent methods by automatically detecting new patterns in call content.

[0115] The learning unit can estimate the user's emotions and adjust the frequency of learning based on the estimated user's emotions. The learning unit can use, for example, voice analysis technology to estimate the user's emotions. For example, the learning unit can analyze the tone and pitch of the user's voice to estimate the emotions. The learning unit can also use a machine learning algorithm to adjust the frequency of learning based on the user's emotions. For example, the learning unit increases the frequency of learning when the user is nervous. Furthermore, the learning unit is equipped with high-speed processing technology to adjust the frequency of learning in real time based on the user's emotions. For example, the learning unit performs learning at a normal frequency when the user is relaxed. In this way, the efficiency of learning can be improved by adjusting the frequency of learning according to the user's emotions.

[0116] The learning unit may add a function to learn the frequency of occurrence of specific keywords or phrases in the call content during learning. The learning unit may use, for example, natural language processing technology to learn the frequency of occurrence of specific keywords or phrases in the call content. For example, the learning unit learns the frequency of occurrence of specific keywords in the call content. The learning unit may also use a machine learning algorithm to learn the frequency of occurrence of specific keywords or phrases. For example, the learning unit learns the frequency of occurrence of specific phrases in the call content. Furthermore, the learning unit is equipped with high-speed processing technology to learn the frequency of occurrence of specific keywords or phrases. For example, the learning unit improves the learning algorithm based on the frequency of occurrence of specific keywords or phrases in the call content. In this way, the accuracy of the learning algorithm can be improved by learning the frequency of occurrence of specific keywords or phrases.

[0117] The learning unit can be added with a function for learning the emotions and intentions of the speaker of the call content during learning. The learning unit can use, for example, a voice analysis technology for learning the emotions and intentions of the speaker of the call content. For example, the learning unit learns the tone and pitch of the speaker's voice to estimate emotions. The learning unit can also use natural language processing technology for learning the speaker's word choice and speech flow. For example, the learning unit learns the speaker's word choice and speech flow to estimate intentions. Furthermore, the learning unit is equipped with a machine learning algorithm for learning the emotions and intentions of the speaker of the call content. For example, the learning unit improves the learning algorithm based on the speaker's emotions and intentions. In this way, the accuracy of the learning algorithm can be improved by learning the speaker's emotions and intentions.

[0118] The notification unit can estimate the user's emotion and adjust the content of the notification based on the estimated user's emotion. The notification unit can use, for example, voice analysis technology to estimate the user's emotion. For example, the notification unit can analyze the tone and pitch of the user's voice to estimate the emotion. The notification unit can also use notification technology to adjust the content of the notification based on the user's emotion. For example, the notification unit provides concise and highly visible notification content when the user is nervous. Furthermore, the notification unit includes high-speed processing technology to adjust the content of the notification in real time based on the user's emotion. For example, the notification unit provides detailed notification content when the user is relaxed. In this way, by adjusting the content of the notification according to the user's emotion, it is possible to provide a highly visible notification.

[0119] The notification unit may add a function to automatically highlight important parts of the call content when notifying. The notification unit may use, for example, natural language processing technology to highlight important parts of the call content. For example, the notification unit may automatically highlight important keywords or phrases in the call content. The notification unit may also use a machine learning algorithm to extract important parts of the call content. For example, the notification unit may automatically extract and highlight important information in the call content. Furthermore, the notification unit may include pattern recognition technology to identify important parts of the call content. For example, the notification unit may automatically identify and highlight important parts of the call content. By highlighting important parts of the call content, important information can be easily checked later.

[0120] The notification unit may add a function to automatically generate a summary of the call content at the time of notification. The notification unit may use, for example, natural language processing technology to generate a summary of the call content. For example, the notification unit may automatically summarize important points of the call content. The notification unit may also use a machine learning algorithm to generate a summary of the call content. For example, the notification unit may automatically summarize the main topics of the call content. Furthermore, the notification unit may include pattern recognition technology to generate a summary of the call content. For example, the notification unit may automatically summarize the conclusion of the call content. By automatically generating a summary of the call content, important information can be easily checked later.

[0121] The notification unit can estimate the user's emotion and adjust the timing of the notification based on the estimated user's emotion. The notification unit can use, for example, voice analysis technology to estimate the user's emotion. For example, the notification unit can analyze the tone and pitch of the user's voice to estimate the emotion. The notification unit can also use notification technology to adjust the timing of the notification based on the user's emotion. For example, the notification unit can provide an early notification when the user is nervous. Furthermore, the notification unit includes high-speed processing technology to adjust the timing of the notification in real time based on the user's emotion. For example, the notification unit can provide a notification at a normal timing when the user is relaxed. In this way, the timing of the notification can be adjusted according to the user's emotion, thereby providing the notification at an appropriate timing.

[0122] The notification unit may add a function to automatically tag specific portions of the call content when notifying. The notification unit may use, for example, natural language processing technology to tag specific portions of the call content. For example, the notification unit may tag important keywords or phrases in the call content. The notification unit may also use a machine learning algorithm to identify specific portions of the call content. For example, the notification unit may tag specific topics in the call content. Furthermore, the notification unit may include pattern recognition technology to tag specific portions of the call content. For example, the notification unit may tag specific time periods in the call content. By tagging specific portions of the call content, important information can be easily searched for later.

[0123] The notification unit can add a function to automatically back up the call content at the time of notification. The notification unit can use, for example, cloud storage technology for backing up the call content. For example, the notification unit automatically backs up the notification content to the cloud. The notification unit can also use local storage technology for backing up the call content. For example, the notification unit automatically backs up the notification content to local storage. Furthermore, the notification unit is equipped with external device technology for backing up the call content. For example, the notification unit automatically backs up the notification content to an external device. By automatically backing up the call content, the safety of the notification data can be ensured. === Hard Collateral 1-1 === Each of the multiple elements, including the listening unit, analysis unit, determination unit, recording unit, termination unit, learning unit, and notification unit, is realized, for example, by at least one of the smart device 14 and the data processing device 12. For example, the listening unit listens to the contents of a call in real time using the microphone 38B of the smart device 14 and converts the contents into text data using the control unit 46A. The analysis unit analyzes the contents of the call using the specific processing unit 290 of the data processing device 12 and extracts specific keywords or phrases. The determination unit determines the possibility of a fraudulent call using the specific processing unit 290 of the data processing device 12. The recording unit records and saves the contents of the call using the control unit 46A of the smart device 14. The termination unit automatically terminates the call using the control unit 46A of the smart device 14. The learning unit learns new fraudulent methods using the specific processing unit 290 of the data processing device 12. The notification unit notifies the possibility of a fraudulent call using the control unit 46A of the smart device 14. === Hard Collateral 1-2 === Each of the multiple elements, including the listening unit, analysis unit, determination unit, recording unit, termination unit, learning unit, and notification unit, described above, is realized, for example, in at least one of the smart glasses 214 and the data processing device 12. For example, the listening unit listens to the contents of a call in real time using the microphone 238 of the smart glasses 214 and converts the contents into text data using the control unit 46A. For example, the analysis unit analyzes the contents of the call using the identification processing unit 290 of the data processing device 12 and extracts specific keywords or phrases. For example, the determination unit determines the possibility of a fraudulent call using the identification processing unit 290 of the data processing device 12. For example, the recording unit records and saves the contents of the call using the control unit 46A of the smart glasses 214. For example, the termination unit automatically terminates the call using the control unit 46A of the smart glasses 214. For example, the learning unit learns new fraudulent methods using the identification processing unit 290 of the data processing device 12. For example, the notification unit notifies the user of the possibility of a fraudulent call using the control unit 46A of the smart glasses 214. === Hard Collateral 1-3 === Each of the multiple elements including the above-mentioned listening unit, analysis unit, determination unit, recording unit, termination unit, learning unit, and notification unit is realized, for example, in at least one of the headset-type terminal 314 and the data processing device 12. For example, the listening unit listens to the contents of a call in real time using the microphone 238 of the headset-type terminal 314 and converts it into text data using the control unit 46A. The analysis unit analyzes the contents of the call using the identification processing unit 290 of the data processing device 12, for example, and extracts specific keywords or phrases. The determination unit determines the possibility of a fraudulent call using the identification processing unit 290 of the data processing device 12, for example. The recording unit records and saves the contents of the call using the control unit 46A of the headset-type terminal 314, for example. The termination unit automatically terminates the call using the control unit 46A of the headset-type terminal 314, for example. The learning unit learns new fraudulent methods using the identification processing unit 290 of the data processing device 12, for example. The notification unit notifies the possibility of a fraudulent call using the control unit 46A of the headset-type terminal 314, for example. === Hard Collateral 1-4 === Each of the multiple elements, including the listening unit, analysis unit, determination unit, recording unit, termination unit, learning unit, and notification unit, is realized, for example, by at least one of the robot 414 and the data processing device 12. For example, the listening unit listens to the contents of a call in real time using the microphone 238 of the robot 414 and converts the contents into text data by the control unit 46A. The analysis unit analyzes the contents of the call and extracts specific keywords or phrases by the identification processing unit 290 of the data processing device 12, for example. The determination unit determines the possibility of a fraudulent call by the identification processing unit 290 of the data processing device 12, for example. The recording unit records and saves the contents of the call by the control unit 46A of the robot 414, for example. The termination unit automatically terminates the call by the control unit 46A of the robot 414, for example. The learning unit learns new fraudulent methods by the identification processing unit 290 of the data processing device 12, for example. The notification unit notifies the possibility of a fraudulent call by the control unit 46A of the robot 414, for example.

[0124] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0125] The fraudulent call prevention system may further include a summarization unit that generates a summary of the call content. The summarization unit analyzes the call content in real time, extracts important points, and generates the summary. For example, the summarization unit may use natural language processing technology to automatically summarize the main topics and conclusions of the call content. The summarization unit may also use a machine learning algorithm to generate the summary of the call content. Furthermore, the summarization unit may include high-speed processing technology to generate the summary of the call content in real time. This allows the automatic generation of a summary of the call content to easily check important information later.

[0126] The fraudulent call prevention system may further include a translation unit that translates the content of the call. The translation unit translates the content of the call in real time and provides it to the user. For example, the translation unit may use machine translation technology to translate the content of the call into multiple languages. The translation unit may also use natural language processing technology to improve the accuracy of the translation of the content of the call. Furthermore, the translation unit may include high-speed processing technology to translate the content of the call in real time. This allows the content of the call to be translated into multiple languages, facilitating communication between users who speak different languages.

[0127] The fraudulent call prevention system may further include an emotion analysis unit that performs emotion analysis of the call content. The emotion analysis unit analyzes the call content and estimates the speaker's emotion. For example, the emotion analysis unit may use voice analysis technology to analyze the speaker's tone and pitch. The emotion analysis unit may also use natural language processing technology to analyze the speaker's choice of words and the flow of speech. Furthermore, the emotion analysis unit may include a machine learning algorithm for estimating the speaker's emotion in real time. By analyzing the speaker's emotion, it is possible to identify factors that increase the likelihood of a fraudulent call.

[0128] The fraudulent call prevention system may further include a topic classification unit that classifies the topics of the call content. The topic classification unit analyzes the call content and classifies the main topics. For example, the topic classification unit may use natural language processing technology to classify the call content into specific categories. The topic classification unit may also use a machine learning algorithm to classify the topics of the call content in real time. Furthermore, the topic classification unit may include high-speed processing technology to classify the topics of the call content in real time. This allows for classifying the topics of the call content, thereby improving the accuracy of detecting fraudulent calls.

[0129] The fraudulent call prevention system may further include a reliability evaluation unit that evaluates the reliability of the call content. The reliability evaluation unit analyzes the call content and evaluates its reliability. For example, the reliability evaluation unit may use natural language processing technology to evaluate the consistency and logic of the call content. The reliability evaluation unit may also use a machine learning algorithm to evaluate the reliability of the call content in real time. Furthermore, the reliability evaluation unit may include high-speed processing technology to evaluate the reliability of the call content in real time. As a result, by evaluating the reliability of the call content, it is possible to identify factors that increase the likelihood of a fraudulent call.

[0130] The fraudulent call prevention system may further include an emotion tracking unit that tracks emotional changes in the content of the call. The emotion tracking unit analyzes the content of the call and tracks changes in the speaker's emotion in real time. For example, the emotion tracking unit may use voice analysis technology to analyze changes in the speaker's tone and pitch. The emotion tracking unit may also use natural language processing technology to analyze the speaker's choice of words and changes in the flow of speech. Furthermore, the emotion tracking unit may include a machine learning algorithm that tracks changes in the speaker's emotion in real time. By tracking changes in the speaker's emotion, factors that increase the likelihood of a fraudulent call can be identified.

[0131] The fraudulent call prevention system may further include a summarization unit that generates a summary of the call content. The summarization unit analyzes the call content in real time, extracts important points, and generates the summary. For example, the summarization unit may use natural language processing technology to automatically summarize the main topics and conclusions of the call content. The summarization unit may also use a machine learning algorithm to generate the summary of the call content. Furthermore, the summarization unit may include high-speed processing technology to generate the summary of the call content in real time. This allows the automatic generation of a summary of the call content to easily check important information later.

[0132] The fraudulent call prevention system may further include an emotion analysis unit that performs emotion analysis of the call content. The emotion analysis unit analyzes the call content and estimates the speaker's emotion. For example, the emotion analysis unit may use voice analysis technology to analyze the speaker's tone and pitch. The emotion analysis unit may also use natural language processing technology to analyze the speaker's choice of words and the flow of speech. Furthermore, the emotion analysis unit may include a machine learning algorithm for estimating the speaker's emotion in real time. By analyzing the speaker's emotion, it is possible to identify factors that increase the likelihood of a fraudulent call.

[0133] The fraudulent call prevention system may further include a topic classification unit that classifies the topics of the call content. The topic classification unit analyzes the call content and classifies the main topics. For example, the topic classification unit may use natural language processing technology to classify the call content into specific categories. The topic classification unit may also use a machine learning algorithm to classify the topics of the call content in real time. Furthermore, the topic classification unit may include high-speed processing technology to classify the topics of the call content in real time. This allows for classifying the topics of the call content, thereby improving the accuracy of detecting fraudulent calls.

[0134] The fraudulent call prevention system may further include a reliability evaluation unit that evaluates the reliability of the call content. The reliability evaluation unit analyzes the call content and evaluates its reliability. For example, the reliability evaluation unit may use natural language processing technology to evaluate the consistency and logic of the call content. The reliability evaluation unit may also use a machine learning algorithm to evaluate the reliability of the call content in real time. Furthermore, the reliability evaluation unit may include high-speed processing technology to evaluate the reliability of the call content in real time. As a result, by evaluating the reliability of the call content, it is possible to identify factors that increase the likelihood of a fraudulent call.

[0135] The processing flow of the second embodiment will be briefly explained below.

[0136] Step 1: The listening unit listens to the call content in real time. For example, the listening unit can use noise cancellation technology to collect the call content with high accuracy. It also has voice recognition technology to convert the call content into text data and high-speed processing technology to analyze it in real time. Step 2: The analysis unit analyzes the call content collected by the listening unit. For example, the analysis unit can use natural language processing technology to extract specific keywords and phrases contained in the call content, or machine learning algorithms to analyze talk patterns. It also has high-speed processing technology for real-time analysis. Step 3: The determination unit determines the possibility of a fraudulent call based on the talk pattern analyzed by the analysis unit. For example, a machine learning algorithm can be used to determine the possibility of a fraudulent call based on the number of similarities with fraud cases, or high-speed processing technology can be used to make a determination in real time. A threshold can also be set to determine the possibility of a fraudulent call. Step 4: The recording unit starts recording if the judgment unit determines that the call is likely to be fraudulent. For example, the recording unit can use voice recording technology to record the call contents in high quality and encryption technology to safely store the recorded data. It also has high-speed processing technology to store the data in real time. Step 5: The termination unit terminates the call when recording is initiated by the recording unit. For example, the termination unit may use a call control technique for automatically terminating the call or a notification technique for notifying the user after the call has ended. The termination unit may also include a storage technique for saving the recorded data after the call has ended.

[0137] 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.

[0138] 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> Examples of the generative AI include a neural network (NN) and a neural network (NN). The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt containing an instruction, as well as inference data such as voice data representing speech, text data representing text, and image data representing an image (e.g., still image data or video data). 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 one or more data formats of voice data, text data, image data, etc. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AIs other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and may perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processing of each of the above-mentioned parts is performed by an AI, the processing may be performed in part or entirely by the AI, but is not limited to these examples. The processing performed by an AI including the generative AI may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by an AI including the generative AI.

[0139] Furthermore, the processing by the data processing system 10 described above is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart device 14, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart device 14. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart device 14 or an external device, and the smart device 14 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0140] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0141] [Second embodiment] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0142] 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.

[0143] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and 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 and / or a LAN.

[0144] 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.

[0145] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0146] 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 user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0147] 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.

[0148] 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.

[0149] 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.

[0150] 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. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0151] In the smart glasses 214, the specific processing is performed by the processor 46. A specific processing program 60 is stored in the storage 50. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific processing program 60 executed on the RAM 48. The smart glasses 214 also have a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0152] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0153] 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.

[0154] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt including an instruction, as well as inference data such as audio data indicating speech, text data indicating text, and image data indicating an image (e.g., still image data or video data). 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 one or more data formats, such as audio data, text data, and image data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation models 58 include AIs other than the generation AI. Examples of AIs other than the generation AI include, but are not limited to, linear regression, logistic regression, decision trees, random forests, support vector machines (SVMs), k-means clustering, convolutional neural networks (CNNs), recurrent neural networks (RNNs), generative adversarial networks (GANs), and naive Bayes. These AIs can perform various types of processing, but are not limited to these examples. The AI ​​may also be an AI agent. When the processing of each of the above-described parts is performed by an AI, the processing may be performed in part or entirely by the AI, but is not limited to these examples. Processing performed by an AI, including the generation AI, may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by an AI, including the generation AI.

[0155] The data processing system 210 according to the second embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 210 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart glasses 214, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart glasses 214. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the smart glasses 214 or an external device, etc., and the smart glasses 214 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0156] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0157] [Third embodiment] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0158] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0159] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and 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 and / or a LAN.

[0160] 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.

[0161] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0162] 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 user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0163] 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.

[0164] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset 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.

[0165] 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.

[0166] 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. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0167] In the headset type terminal 314, the identification process is performed by the processor 46. A identification program 60 is stored in the storage 50. The processor 46 reads the identification program 60 from the storage 50 and executes the read identification program 60 on the RAM 48. The identification process is realized by the processor 46 operating as a control unit 46A in accordance with the identification program 60 executed on the RAM 48. Note that the headset type terminal 314 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the identification processing unit 290 using these models.

[0168] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0169] 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.

[0170] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt including an instruction, as well as inference data such as audio data indicating speech, text data indicating text, and image data indicating an image (e.g., still image data or video data). 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 one or more data formats, such as audio data, text data, and image data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation models 58 include AIs other than the generation AI. Examples of AIs other than the generation AI include, but are not limited to, linear regression, logistic regression, decision trees, random forests, support vector machines (SVMs), k-means clustering, convolutional neural networks (CNNs), recurrent neural networks (RNNs), generative adversarial networks (GANs), and naive Bayes. These AIs can perform various types of processing, but are not limited to these examples. The AI ​​may also be an AI agent. When the processing of each of the above-described parts is performed by an AI, the processing may be performed in part or entirely by the AI, but is not limited to these examples. Processing performed by an AI, including the generation AI, may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by an AI, including the generation AI.

[0171] The data processing system 310 according to the third embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 310 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the headset type terminal 314, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the headset type terminal 314. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the headset type terminal 314 or an external device, etc., and the headset type terminal 314 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0172] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0173] [Fourth embodiment] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0174] 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.

[0175] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and 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 and / or a LAN.

[0176] 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.

[0177] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0178] 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 image sensor or a CCD image sensor, and captures images of the user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0179] 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.

[0180] The control object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the 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.

[0181] 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.

[0182] 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.

[0183] 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. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0184] In the robot 414, the processor 46 performs the identification process. The storage 50 stores the identification program 60. The processor 46 reads the identification program 60 from the storage 50 and executes the read identification program 60 on the RAM 48. The identification process is realized by the processor 46 operating as the control unit 46A in accordance with the identification program 60 executed on the RAM 48. The robot 414 also has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can perform the same process as the identification processing unit 290 using these models.

[0185] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0186] 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.

[0187] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt including an instruction, as well as inference data such as audio data indicating speech, text data indicating text, and image data indicating an image (e.g., still image data or video data). 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 one or more data formats, such as audio data, text data, and image data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation models 58 include AIs other than the generation AI. Examples of AIs other than the generation AI include, but are not limited to, linear regression, logistic regression, decision trees, random forests, support vector machines (SVMs), k-means clustering, convolutional neural networks (CNNs), recurrent neural networks (RNNs), generative adversarial networks (GANs), and naive Bayes. These AIs can perform various types of processing, but are not limited to these examples. The AI ​​may also be an AI agent. When the processing of each of the above-described parts is performed by an AI, the processing may be performed in part or entirely by the AI, but is not limited to these examples. Processing performed by an AI, including the generation AI, may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by an AI, including the generation AI.

[0188] The data processing system 410 according to the fourth embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 410 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the robot 414, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the robot 414. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the robot 414 or an external device, etc., and the robot 414 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0189] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0190] 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.

[0191] FIG. 9 illustrates an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion encompasses both emotions 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.

[0192] 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.

[0193] 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).

[0194] 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 expressed, and when they approach the ideal, a state of pleasure is expressed. Emotions can also be created for robots, cars, 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 expressed, and when they approach the ideal, a state of pleasure is expressed. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on speech emotion recognition and brain physiological signal analysis systems for emotions, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the area called "reaction," where sensation is dominant. The right half of the emotion map lists emotions belonging to the area called "situation," where situational awareness is dominant.

[0195] 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."

[0196] 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.

[0197] In the above embodiment, an example was given in which a specific process is performed by one computer 22, but the technology disclosed herein is not limited to this, and distributed processing of the specific process may be performed by multiple computers including computer 22.

[0198] 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.

[0199] 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.

[0200] 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.

[0201] The hardware resource for executing a specific process can be any of the following types of processors: A processor, for example, 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. A processor also includes 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.

[0202] The hardware resource that executes the specific process 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 process may be a single processor.

[0203] 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.

[0204] 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.

[0205] In the above example, the first to fourth embodiments have been described separately, but some or all of these embodiments may be combined. The smart device 14, smart glasses 214, headset terminal 314, and robot 414 are merely examples, and they may be combined, or other devices may be used. In the above example, the first and second embodiments have been described separately, but they may be combined.

[0206] 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.

[0207] 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.

[0208] [Explanation of symbols]

[0209] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot

Claims

1. A listening section that listens to the contents of the call in real time, an analysis unit that analyzes the call content collected by the listening unit; a determination unit that determines whether a call is fraudulent based on the talk pattern analyzed by the analysis unit; a recording unit that starts recording when the determination unit determines that the call is likely to be a fraudulent call; a termination unit that terminates the call when recording is started by the recording unit; Equipped with A system characterized by:

2. Equipping a learning department to study new fraud methods The system of claim 1 .

3. Equipped with a notification section that notifies you of possible fraudulent calls The system of claim 1 .

4. The analysis unit Analyzes for predefined keywords and phrases The system of claim 1 .

5. The determination unit Determine the likelihood of a fraudulent call based on the number of features that match the fraud case The system of claim 1 .

6. The recording unit Start recording when the likelihood of a fraudulent call reaches a preset threshold The system of claim 1 .

7. The end portion is End the call if recording starts The system of claim 1 .

8. The listening unit Estimates the user's emotions and adjusts the accuracy of listening to the content of the call based on the estimated user emotions. The system of claim 1 .

9. The listening unit Add a feature to automatically remove background sounds and noise when listening to phone calls. The system of claim 1 .

Images