A method and system for combating telecom fraud based on generative artificial intelligence

Abstract

The application discloses a kind of based on generative artificial intelligence's telecom fraud countermeasure method and system, belong to telecommunication fraud prevention technical field.The method includes: receiving the countermeasure instruction that user sends actively after identifying fraud call, seamless bridging of ongoing incoming call conversation to server side generative artificial intelligence interaction module, through voice recognition, generative artificial intelligence and speech synthesis technology simulate human and incoming party carry out continuous interaction, consume its time cost.Compared with prior art, the application can effectively reduce the efficiency of fraud and strike fraudsters psychology, from the root of the business model of telecommunication fraud, operation zero threshold, countermeasure effect is good.

Description

Technical Field

[0001] This invention relates to the field of telecommunications fraud prevention technology, and in particular to a method and system for countering telecommunications fraud based on generative artificial intelligence. Background Technology

[0002] With the development of telecommunications network technology, telecommunications fraud has become increasingly rampant, causing huge losses to the property security of the people. Currently, existing telecommunications fraud prevention technologies mainly include number blocking and caller ID alerts.

[0003] However, the aforementioned existing technologies have significant drawbacks: they cannot reduce the efficiency of fraud, as scammers can still quickly screen target users by making a large number of calls. Although methods such as number blocking and caller ID do have some effect, these methods also help scammers quickly screen victims. The scammers' work efficiency is not significantly affected. Furthermore, the aforementioned existing technologies do not psychologically impact scammers.

[0004] Therefore, there is an urgent need for a method to counter telecommunications fraud by greatly reducing the efficiency of fraud and striking at the psychology of fraudsters. Summary of the Invention

[0005] The technical problem to be solved by the present invention The purpose of this invention is to overcome the shortcomings of the prior art and provide a method and system for countering telecommunications fraud based on generative artificial intelligence, so as to solve the technical problem that the prior art cannot greatly reduce the efficiency of fraud and strike at the psychology of fraudsters.

[0006] Technical solution of the present invention To achieve the above objectives, the present invention adopts the following technical solution: A method for combating telecom fraud based on generative artificial intelligence includes the following steps: S1: Receive a countermeasure instruction sent by the user terminal. The countermeasure instruction is actively triggered by the user after identifying the incoming call as a fraudulent call, and corresponds to an ongoing call. S2: Seamlessly bridge the ongoing incoming call from the user terminal to the generative artificial intelligence interaction module on the server side, interrupt the voice channel between the user and the caller, while maintaining the continuous connection of the voice channel between the caller and the server. S3: Converts the caller's real-time voice signal into text data using automatic speech recognition technology; S4: Input the text data into a pre-trained generative artificial intelligence model to generate a response text that conforms to human communication logic; S5: The reply text is converted into a natural speech signal using speech synthesis technology and sent to the caller in real time; S6: Repeat steps S3 to S5 until the caller hangs up or the preset maximum call duration is reached.

[0007] Furthermore, the method also includes step S7: recording and storing complete voice data, text transcription data, and call duration information for the entire call process.

[0008] Furthermore, the generative artificial intelligence model is configured to simulate the communication characteristics of ordinary humans, including delayed response, colloquial expression, pauses in tone, regional accents, requests for repetition, and a gradual tendency to trust the fraudulent rhetoric during the interaction, so as to make the caller expect that the fraud can be completed, thereby extending the call duration.

[0009] Furthermore, after step S7, the method further includes: sending a notification SMS to the incoming call number, informing it that the interaction target of this call is an artificial intelligence anti-fraud system, and reminding it that telecommunications fraud is an illegal and criminal act.

[0010] Furthermore, after step S7, the method further includes: identifying the type of fraud corresponding to the incoming call based on the call text data, and uploading the incoming call number and the corresponding fraud type to the shared anti-fraud database.

[0011] Furthermore, after step S7, the method further includes: performing a hash operation on the stored call data to generate an immutable electronic evidence file for use by public security organs in handling cases.

[0012] Beneficial effects of the present invention Compared with the prior art, the present invention has the following beneficial effects: 1. By using artificial intelligence to simulate continuous interaction between humans and fraudsters, the fraudsters' efficiency in committing fraud is reduced and their psychology is undermined by consuming their time. 2. It adopts a one-click transfer design, allowing users to transfer the call to AI with just one click without hanging up. The operation is extremely easy and ensures that no counter-attack opportunities are missed. 3. It does not rely on known databases of fraudulent numbers and can effectively prevent scammers from using new numbers to commit fraud; Attached Figure Description

[0013] Figure 1 This is an overall flowchart of the method described in this invention.

[0014] Figure 2 This is a structural block diagram of the system described in this invention. Detailed Implementation

[0015] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. Example 1

[0016] like Figure 1 As shown in the figure, this embodiment discloses a method for countering telecommunications fraud based on generative artificial intelligence. The specific steps are as follows: S1: The user answers the call and has a voice conversation with the caller; when the user identifies the call as a fraudulent call through the conversation, the user clicks the "One-Click Transfer AI" button on the mobile APP, and the user terminal sends a countermeasure instruction to the server. The countermeasure instruction corresponds to the ongoing call and includes the caller's number information. S2: After receiving the countermeasure instruction, the server seamlessly bridges the ongoing call from the user terminal to the generative artificial intelligence interaction module on the server side through telecommunications network signaling, immediately interrupting the voice channel between the user and the caller, while maintaining the continuous connection of the voice channel between the caller and the server, so that the caller will not be aware of the switch of the call link. S3: The server uses automatic speech recognition technology to convert the caller's real-time voice signal into text data; S4: Input the text data into a pre-trained generative artificial intelligence model to generate a response text that conforms to human communication logic; S5: The reply text is converted into a natural speech signal using speech synthesis technology and sent to the caller in real time; S6: Repeat steps S3 to S5 until the caller hangs up or the preset maximum call duration of 15 minutes is reached.

[0017] Furthermore, this embodiment also includes step S7: the server automatically records and stores the complete voice data, text transcription data and call duration information of the entire call process.

[0018] Furthermore, in step S5, the speech synthesis module can employ two timbre matching modes: - Mode 1 (User Authorization Mode): If the user authorizes the APP to record and store their own voice samples in advance, the voice synthesis module will use the user's own voice to generate the reply voice, achieving a completely seamless switching effect. - Mode 2 (default mode): If the user has not authorized the recording of voice, the voice synthesis module will automatically select a general natural tone similar to the user's gender and age to generate a reply voice.

[0019] Furthermore, after step S7, the server sends a notification SMS to the incoming call number: "The person who just spoke to you was the AI ​​anti-fraud system. Telecommunications fraud is an illegal and criminal activity; please do not continue to engage in fraudulent activities."

[0020] Furthermore, after step S7, the server identifies the incoming call as a "fraud impersonating public security, procuratorate, and court officials" type based on the call text data, and uploads the incoming call number and the corresponding fraud type to the national shared anti-fraud database for other users to use for early warning and interception.

[0021] Furthermore, after step S7, the server performs a hash operation on the stored call data to generate an immutable electronic evidence file, which can be directly used by public security organs in handling cases. Example 2

[0022] This embodiment discloses a telecommunications fraud countermeasure system that implements the above method, including: - User terminal: Used to answer incoming calls and send countermeasure commands to the server. - Server: Includes a speech recognition module, a generative artificial intelligence module, and a speech synthesis module; - Speech recognition module: Deployed on the server, it is used to convert the voice signal of the incoming caller into text data; - Generative AI module: Deployed on the server, it generates response text that conforms to human communication logic; - Speech synthesis module: Deployed on the server, it is used to convert reply text into natural speech signals.

[0023] It is important to note that this invention is solely for the legal countermeasure against unsolicited fraudulent calls made to users. All operations require explicit user authorization. During the AI ​​interaction, no user's real information will be disclosed, and no calls will be made proactively. After the call ends, the system automatically sends a text message to the fraudster, informing them that the caller was an AI-powered anti-fraud system and reminding them that telecommunications fraud is illegal. Those skilled in the art should understand that the technical solution of this invention must not be used for any illegal purposes, including harassing others, infringing on others' privacy, or committing fraud. All use of this invention must comply with national laws and regulations.

[0024] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for countering telecommunications fraud based on generative artificial intelligence, characterized in that, Includes the following steps: S1: Receive a countermeasure instruction sent by the user terminal. The countermeasure instruction is actively triggered by the user after identifying the incoming call as a fraudulent call, and corresponds to an ongoing call. S2: Seamlessly bridge the ongoing incoming call from the user terminal to the generative artificial intelligence interaction module on the server side, interrupt the voice channel between the user and the caller, while maintaining the continuous connection of the voice channel between the caller and the server. S3: Converts the caller's real-time voice signal into text data using automatic speech recognition technology; S4: Input the text data into a pre-trained generative artificial intelligence model to generate a response text that conforms to human communication logic; S5: The reply text is converted into a natural speech signal using speech synthesis technology and sent to the caller in real time; S6: Repeat steps S3 to S5 until the caller hangs up or the preset maximum call duration is reached.

2. The method for countering telecommunications fraud based on generative artificial intelligence according to claim 1, characterized in that, It also includes step S7: recording and storing complete voice data, text transcription data and call duration information for the entire call process.

3. The method for countering telecommunications fraud based on generative artificial intelligence according to claim 1, characterized in that, The generative artificial intelligence model is configured to simulate the communication characteristics of ordinary humans, including delayed response, colloquial expression, pauses in speech, regional accents, and requests for repetition. During the interaction, it gradually shows a tendency to trust the fraudulent rhetoric, making the caller expect that the fraud can be completed, thereby extending the call duration.

4. The method for countering telecommunications fraud based on generative artificial intelligence according to claim 2, characterized in that, After step S7, the method further includes: sending a notification SMS to the incoming call number, informing it that the interaction object of this call is an artificial intelligence anti-fraud system, and reminding it that telecommunications fraud is an illegal and criminal act.

5. The method for countering telecommunications fraud based on generative artificial intelligence according to claim 2, characterized in that, After step S7, the method further includes: identifying the type of fraud corresponding to the incoming call based on the call text data, and uploading the incoming call number and the corresponding fraud type to the shared anti-fraud database.

6. The method for countering telecommunications fraud based on generative artificial intelligence according to claim 2, characterized in that, After step S7, the method further includes: performing a hash operation on the stored call data to generate an immutable electronic evidence file for use by public security organs in handling cases.

7. A telecommunications fraud countermeasure system implementing the method of any one of claims 1 to 6, characterized in that, include: User terminal: Used to answer incoming calls and send countermeasure commands to the server; Server: Includes speech recognition module, generative artificial intelligence module, and speech synthesis module; Speech recognition module: Deployed on the server, used to convert speech signals into text data; Generative AI module: Deployed on the server, used to generate response text; Speech synthesis module: Deployed on the server, it is used to convert reply text into speech signals.

Images