Bank card application risk control method, device, electronic device and medium

Abstract

The present disclosure provides a bank card application risk control method based on voiceprint recognition, a bank card application risk control device, an electronic device, a computer-readable storage medium, and a computer program. The bank card application risk control method and device can be used in the field of artificial intelligence technology. The bank card application risk control method includes: receiving a user's bank card application request; when the user applies for a bank card, obtaining the user's first voiceprint information; when investigating the user, obtaining the user's second voiceprint information; determining a first similarity between the first voiceprint information and the second voiceprint information; when the first similarity is greater than a first threshold, determining a second similarity between the second voiceprint information and a black voiceprint in a black voiceprint library; and when the second similarity is greater than a second threshold, storing the first voiceprint information and / or the second voiceprint information in the black voiceprint library.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, and more specifically, to a bank card application risk control method based on voiceprint recognition, a bank card application risk control device, an electronic device, a computer-readable storage medium, and a computer program product. Background Art

[0002] Credit card application risk screening is an integral part of bank risk control. Relevant risk strategies can reduce the risk of bad debts and bad debts. Good risk screening and avoidance measures can enhance the bank's reputation and reduce the bank's non-performing assets. The traditional credit card application risk prevention and control mechanism is mainly based on the application information elements filled in by the applicant, and combines the bank's internal and external systems to identify the customer's identity. For example, based on the telephone number, ID number and other information filled in by the customer when applying, the Ministry of Public Security will conduct online verification and the operator's telephone information verification to ensure that the customer's application information is consistent with the registration information in the external system. In addition, manual investigation can be used to review the authenticity of the customer. Summary of the invention

[0003] In view of this, the present disclosure provides a bank card application risk control method based on voiceprint recognition, a bank card application risk control device, an electronic device, a computer-readable storage medium and a computer program product that are beneficial to risk control and reduce economic losses.

[0004] One aspect of the present disclosure provides a bank card application risk control method based on voiceprint recognition, including: receiving a bank card application request from a user; obtaining the user's first voiceprint information when the user applies for a bank card; obtaining the user's second voiceprint information when investigating the user; determining a first similarity between the first voiceprint information and the second voiceprint information; when the first similarity is greater than a first threshold, determining a second similarity between the second voiceprint information and a black voiceprint in a black voiceprint library; and when the second similarity is greater than a second threshold, storing the first voiceprint information and / or the second voiceprint information in the black voiceprint library.

[0005] According to the bank card application risk control method based on voiceprint recognition in the embodiment of the present disclosure, it is possible to accurately identify whether the actual applicant of the bank card and the investigated user are the same person based on the comparison of the first voiceprint information and the second voiceprint information; and by comparing the black voiceprint in the black voiceprint library with the second voiceprint information, it is possible to determine whether the bank card applicant has a bad record, and further measures can be taken against applicants with bad records. The process of storing the first voiceprint information and / or the second voiceprint information in the black voiceprint library can continuously improve the black voiceprint library. Therefore, the method of the present disclosure is conducive to bank risk control, thereby reducing the economic losses of banks and customers.

[0006] In some embodiments, when the first similarity is less than the first threshold, the first voiceprint information and / or the second voiceprint information is stored in the black voiceprint library.

[0007] In some embodiments, the method also includes: after obtaining the user's first voiceprint information, storing the first voiceprint information in a temporary voiceprint library, wherein the temporary voiceprint library includes at least one temporary voiceprint; determining the first similarity between the first voiceprint information and the second voiceprint information includes: matching the second voiceprint information with the temporary voiceprint in the temporary voiceprint library; obtaining a list of similarity values ​​between the second voiceprint information and the temporary voiceprint; and obtaining the first similarity between the second voiceprint information and the first voiceprint information.

[0008] In some embodiments, determining the second similarity between the second voiceprint information and the black voiceprints in the black voiceprint library includes: obtaining a list of similarity values ​​between the second voiceprint information and the black voiceprints; and obtaining the second similarity between the second voiceprint information and the black voiceprints.

[0009] In some embodiments, obtaining the first voiceprint information of the user includes: displaying text information; recording the voiceprint of the user reading the text information; and denoising the voiceprint, wherein the voiceprint after denoising is the first voiceprint information.

[0010] In some embodiments, obtaining the user's first voiceprint information includes: displaying text information; recording the user's voiceprint when reading the text information; denoising the voiceprint, wherein the denoising processing includes removing noise in the voiceprint that is not related to the reading of the text information to obtain a first effective voiceprint; and extracting the voiceprint feature of the first effective voiceprint, which is the first voiceprint information.

[0011] In some embodiments, the obtaining of the user's first voiceprint information also includes setting a first valid voiceprint progress bar, wherein the first valid voiceprint progress bar is the duration of the first valid voiceprint, and after the recorded voiceprint is processed by noise reduction, if the first valid voiceprint progress bar does not meet the threshold, then continue to record the voiceprint of the user reading the text information and process the voiceprint by noise reduction until the first valid voiceprint progress bar meets the threshold, and the extraction of the voiceprint feature of the first valid voiceprint is to extract the voiceprint feature of the first valid voiceprint after the first valid voiceprint progress bar meets the threshold.

[0012] In some embodiments, extracting the voiceprint features of the first effective voiceprint includes extracting the voiceprint features using a neural network model.

[0013] In some embodiments, the neural network model is a Gaussian mixture model-general background model.

[0014] In some embodiments, obtaining the second voiceprint information of the user includes: the bank dialing the contact information filled in when initiating the bank card application request; and collecting audio information of the caller, and the audio information is the second voiceprint information.

[0015] In some embodiments, the acquisition of the user's second voiceprint information includes: the bank dials the contact information filled in when initiating a bank card application request; collects audio information of the caller; noise reduction processing of the audio information, the noise reduction processing of the audio information includes removing noise in the audio information that is irrelevant to the content of the conversation, to obtain a second effective voiceprint; and extracting the voiceprint feature of the second effective voiceprint, which is the second voiceprint information.

[0016] Another aspect of the present disclosure provides a bank card application risk control device based on voiceprint recognition, including: a bank card application request module, used to execute receiving a user's bank card application request; a first voiceprint acquisition module, used to execute acquiring the user's first voiceprint information when the user applies for a bank card; a second voiceprint acquisition module, used to execute acquiring the user's second voiceprint information when investigating the user; a first information matching module, used to execute determining a first similarity between the first voiceprint information and the second voiceprint information; a second information matching module, used to execute determining a second similarity between the second voiceprint information and a black voiceprint in a black voiceprint library when the first similarity is greater than a first threshold; and a black voiceprint adding module, used to execute storing the first voiceprint information and / or the second voiceprint information in the black voiceprint library when the second similarity is greater than a second threshold.

[0017] Another aspect of the present disclosure provides an electronic device, including one or more processors and one or more memories, wherein the memories are used to store executable instructions, and when the executable instructions are executed by the processors, the methods described above are implemented.

[0018] Another aspect of the present disclosure provides a computer-readable storage medium storing computer-executable instructions, which are used to implement the above method when executed.

[0019] Another aspect of the present disclosure provides a computer program product, which includes one or more executable instructions, and when the executable instructions are executed by a processor, the method described above is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The above and other objects, features and advantages of the present disclosure will become more apparent through the following description of the embodiments of the present disclosure with reference to the accompanying drawings, in which:

[0021] Figure 1An exemplary system architecture to which the method and apparatus according to an embodiment of the present disclosure can be applied is schematically shown;

[0022] Figure 2 The flowchart of the bank card application risk control method based on voiceprint recognition according to the embodiment of the present disclosure is schematically shown;

[0023] Figure 3 A flowchart of obtaining the first voiceprint information of a user according to an embodiment of the present disclosure is schematically shown;

[0024] Figure 4 A flowchart of obtaining the first voiceprint information of a user according to another embodiment of the present disclosure is schematically shown;

[0025] Figure 5 A flowchart of extracting a voiceprint feature of a first effective voiceprint according to an embodiment of the present disclosure is schematically shown;

[0026] Figure 6 A flowchart of obtaining the second voiceprint information of a user according to an embodiment of the present disclosure is schematically shown;

[0027] Figure 7 A flowchart of obtaining the second voiceprint information of a user according to another embodiment of the present disclosure is schematically shown;

[0028] Figure 8 A flowchart of extracting voiceprint features of a second effective voiceprint according to an embodiment of the present disclosure is schematically shown;

[0029] Fig. 9 A flowchart for determining a first similarity between first voiceprint information and second voiceprint information according to an embodiment of the present disclosure is schematically shown;

[0030] Fig.10 A flowchart of determining a second similarity between second voiceprint information and a black voiceprint in a black voiceprint library according to an embodiment of the present disclosure is schematically shown;

[0031] Fig.11 A block diagram of a bank card application risk control device according to an embodiment of the present disclosure is schematically shown;

[0032] Fig.12 A block diagram of an electronic device according to an embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0033] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present disclosure. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present disclosure. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present disclosure.

[0034] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. The terms "comprise", "include", etc. used herein indicate the existence of the features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0035] When using expressions such as "at least one of A, B or C, etc.", it should generally be interpreted in accordance with the meaning of the expression generally understood by those skilled in the art (for example, "a system having at least one of A, B or C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.). The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, features defined as "first" and "second" may explicitly or implicitly include one or more of the said features.

[0036] Credit card application risk screening is an indispensable part of bank risk control. Relevant risk strategies can reduce the risk of bank bad debts and non-performing loans. Good risk screening and avoidance measures can enhance the bank's reputation and reduce the bank's non-performing assets. The traditional credit card application risk prevention and control mechanism is mainly based on the application information elements filled in by the applicant, and combines the bank's internal and external systems to identify the customer's identity.

[0037] For example, according to the telephone number, ID number and other information filled in by the customer when applying, the Ministry of Public Security will conduct online verification and the operator's telephone information will be checked to ensure that the customer's application information is consistent with the registration information in the external system. In addition, manual investigation can be used to review the authenticity of the customer. However, with the development of society, the intermediary industry has begun to intervene in various agency businesses. It can circumvent the Ministry of Public Security's online verification and operator's telephone information verification by obtaining the customer's ID card and other measures. Manual investigation can only verify the authenticity of the customer based on the information displayed by the system, and cannot accurately identify whether the actual applicant and the person being investigated are the same person, which is not conducive to bank risk control.

[0038] The embodiments of the present disclosure provide a bank card application risk control method based on voiceprint recognition, a bank card application risk control device, an electronic device, a computer-readable storage medium, and a computer program product. The bank card application risk control method includes receiving a user's bank card application request; when the user applies for a bank card, obtaining the user's first voiceprint information; when investigating the user, obtaining the user's second voiceprint information; determining a first similarity between the first voiceprint information and the second voiceprint information; when the first similarity is greater than a first threshold, determining a second similarity between the second voiceprint information and a black voiceprint in a black voiceprint library; when the second similarity is greater than a second threshold, storing the first voiceprint information and / or the second voiceprint information in the black voiceprint library.

[0039] It should be noted that the bank card application risk control method based on voiceprint recognition and the bank card application risk control device 10 disclosed in the present invention can be used in the field of artificial intelligence, and can also be used in any field other than the field of artificial intelligence, such as the financial field. The field of the present invention is not limited here.

[0040] Figure 1 The exemplary system architecture 100 of the bank card application risk control method based on voiceprint recognition, the bank card application risk control device, the electronic device, the computer-readable storage medium and the computer program product according to the embodiment of the present disclosure is schematically shown. It should be noted that: Figure 1 What is shown is merely an example of a system architecture to which the embodiments of the present disclosure can be applied, in order to help those skilled in the art understand the technical content of the present disclosure, but it does not mean that the embodiments of the present disclosure cannot be used in other devices, systems, environments or scenarios.

[0041] like Figure 1 As shown, the system architecture 100 according to this embodiment may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a medium for communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0042] Users can use terminal devices 101, 102, 103 to interact with server 105 through network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only examples).

[0043] The terminal devices 101 , 102 , and 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, and desktop computers.

[0044] The server 105 may be a server that provides various services, such as a background management server (only an example) that provides support for websites browsed by users using the terminal devices 101, 102, and 103. The background management server may analyze and process the received data such as user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal device.

[0045] It should be noted that the bank card application risk control method based on voiceprint recognition provided in the embodiment of the present disclosure can generally be executed by the server 105. Accordingly, the bank card application risk control device 10 provided in the embodiment of the present disclosure can generally be set in the server 105. The bank card application risk control method based on voiceprint recognition provided in the embodiment of the present disclosure can also be executed by a server or server cluster that is different from the server 105 and can communicate with the terminal devices 101, 102, 103 and / or the server 105. Accordingly, the bank card application risk control device 10 provided in the embodiment of the present disclosure can also be set in a server or server cluster that is different from the server 105 and can communicate with the terminal devices 101, 102, 103 and / or the server 105.

[0046] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0047] The following will be based on Figure 1 The scene described by Figure 2-Figure 10 A bank card application risk control method based on voiceprint recognition in a disclosed embodiment is described in detail.

[0048] Figure 2 The flowchart of the bank card application risk control method based on voiceprint recognition according to an embodiment of the present disclosure is schematically shown.

[0049] like Figure 2 As shown, the bank card application risk control method based on voiceprint recognition in this embodiment includes operations S210 to S260.

[0050] In operation S210, a user's bank card application request is received. For example, when a user wants to apply for a bank card, the user can submit a bank card application request through a mobile terminal, and the bank server receives the user's bank card application request through the network. Among them, the bank can set application information that needs to be filled in when applying for a bank card, such as the applicant's ID number, mobile phone number, name, gender, place of origin, work unit and other information.

[0051] In operation S220, when the user applies for a bank card, the first voiceprint information of the user is obtained. It should be explained that after the user submits the application information required to be filled in the bank card application request, the first voiceprint information of the user needs to be obtained. For example, the first voiceprint information can be the voice information of the user, and the way to obtain the voice information of the user can be to start the recording function of the mobile terminal or start the video recording function of the mobile terminal, etc.

[0052] Specifically, as a possible implementation method, Figure 3 As shown, operation S220 of obtaining the first voiceprint information of the user may include operation S221 of displaying text information, operation S222 of recording the voiceprint of the user reading the text information, and operation S223 of noise reduction processing of the voiceprint. Among them, the mobile terminal can display the text information to the user through the display screen, and the text information can be any text, such as a paragraph designed for a bank, famous quotes, or excerpts of beautiful articles, but is not limited to this. The user reads the text information according to the displayed content, and the mobile terminal can record the user's voiceprint at this time. The mobile terminal can also perform noise reduction processing on the recorded voiceprint, including but not limited to deleting voiceprints that are not related to the text information, and the voiceprint after noise reduction can be used as the first voiceprint information.

[0053] As another possible implementation, Figure 4 As shown, operation S220 of obtaining the first voiceprint information of the user may include operation S221 of displaying text information, operation S222 of recording the voiceprint of the user reading the text information, operation S223 of noise reduction processing of the voiceprint, and operation S224 of extracting the voiceprint feature of the first effective voiceprint. The noise reduction processing of the voiceprint may include removing noise irrelevant to the reading text information in the voiceprint to obtain the first effective voiceprint; thereby, the voiceprint feature of the first effective voiceprint may be extracted, and the voiceprint feature may be used as the first voiceprint information.

[0054] Further, refer to Figure 4 Operation S220 of obtaining the first voiceprint information of the user may also include operation S225 of setting a first valid voiceprint progress bar, wherein the first valid voiceprint progress bar may be the duration of the first valid voiceprint. After the recorded voiceprint is processed with noise reduction, if the first valid voiceprint progress bar does not meet the threshold, the voiceprint of the user reading the text information continues to be recorded and the voiceprint is processed with noise reduction until the first valid voiceprint progress bar meets the threshold. The voiceprint feature of the first valid voiceprint is extracted as the voiceprint feature of the first valid voiceprint after the first valid voiceprint progress bar meets the threshold.

[0055] It should be noted that when a user applies for a bank card, he or she may be in a noisy environment. After the first recording of the user's voiceprint reading a text message, the voiceprint may be mixed with other noises in the environment; when the user is reading a text message, he or she may suddenly say other words unrelated to the text message to the people around him or her. Of course, the scenarios where noise appears in the user's voiceprint are not limited to this, and this is just an example.

[0056] At this time, the recorded voiceprint needs to be denoised, and the voiceprint after denoising is the first valid voiceprint. However, the duration of the first valid voiceprint will be reduced after the noise is removed, and then it is necessary to continue recording the voiceprint of the user reading the text information and denoising the voiceprint until the first valid voiceprint progress bar meets the threshold, that is, the duration of the first valid voiceprint meets the threshold. Based on this, setting the first voiceprint progress bar can make it easier for the duration of the first valid voiceprint to meet the extraction requirements, thereby facilitating the extraction of the voiceprint features of the first valid voiceprint.

[0057] Furthermore, if Figure 5 As shown, operation S224 of extracting the voiceprint feature of the first effective voiceprint may include S2241 extracting the voiceprint feature using a neural network model. It can be understood that using a neural network model can facilitate the extraction of the voiceprint feature of the first effective voiceprint.

[0058] Furthermore, the neural network model may be a Gaussian mixture model-universal background model. Of course, the neural network model may also be other models, as long as the neural network model can realize the extraction of the voiceprint features of the first effective voiceprint, it falls within the protection scope of the present disclosure.

[0059] In operation S230, when investigating the user, the second voiceprint information of the user is obtained. It should be noted that after the bank obtains the first voiceprint information of the user, it can investigate the user. Specifically, as an implementable method, Figure 6 As shown, operation S230 of obtaining the second voiceprint information of the user includes operation S231 of the bank dialing the contact information filled in when initiating the bank card application request and operation S232 of collecting the audio information of the caller, and the audio information is the second voiceprint information. Here, the contact information may include but is not limited to a mobile phone number, a WeChat number, a QQ number, etc. When the dial-up initiated by the bank is connected, the audio information of the caller can be collected.

[0060] As another possible implementation method, Figure 7 As shown, operation S230 of obtaining the second voiceprint information of the user includes operation S231 of the bank dialing the contact information filled in when initiating the bank card application request, operation S232 of collecting the audio information of the caller, operation S233 of noise reduction processing of the audio information, and operation S234 of extracting the voiceprint feature of the second effective voiceprint. Among them, the noise reduction processing of the audio information includes removing the noise irrelevant to the conversation content in the audio information to obtain the second effective voiceprint.

[0061] It should be noted that when the bank conducts an investigation, that is, when the bank initiates a call to the user, the user may be in a noisy environment, or the user may say other irrelevant words to the people around him when answering the voice. Of course, the scenario where noise appears in the user's voiceprint during the investigation is not limited to this, it is just an example here.

[0062] At this time, the recorded audio information needs to be denoised, and the voiceprint after denoising is the second effective voiceprint, thereby the voiceprint feature of the second effective voiceprint can be extracted, and the voiceprint feature can be used as the second voiceprint information.

[0063] Furthermore, if Figure 8 As shown, operation S234 of extracting the voiceprint feature of the second effective voiceprint may include operation S2341 of extracting the voiceprint feature using a neural network model. It can be understood that using a neural network model can facilitate the extraction of the voiceprint feature of the second effective voiceprint. Furthermore, the neural network model can be a Gaussian mixture model-universal background model. Of course, the neural network model can also be other models.

[0064] In operation S240, a first similarity between the first voiceprint information and the second voiceprint information is determined.

[0065] In operation S250, when the first similarity is greater than the first threshold, the second similarity between the second voiceprint information and the black voiceprint in the black voiceprint library is determined. It can be understood that after obtaining the first voiceprint information and the second voiceprint information, the first voiceprint information and the second voiceprint information can be compared to determine the first similarity between the first voiceprint information and the second voiceprint information. For example, the first similarity between the first voiceprint information and the second voiceprint information can be determined by an algorithm model, and the first threshold can be implanted in the algorithm model. When the first similarity is greater than the first threshold, it means that the first voiceprint information and the second voiceprint information are relatively consistent, and it can be determined that the person applying for the bank card is the person himself. Here, the first threshold is an arbitrary value that can prove that the first voiceprint information and the second voiceprint information are consistent, such as 99.5%.

[0066] After proving that it is the person who applied for the bank card, the second voiceprint information can be compared with the black voiceprint in the black voiceprint library to determine the second similarity between the second voiceprint information and the black voiceprint. Here, the black voiceprint can be understood as the voiceprint information of the applicant who has a bad record. For example, the applicant with a bad record can be a user whose contact information, name, ID number and other information submitted when applying do not match each other, or the first voiceprint information and the second voiceprint information do not match, that is, the bank finds that the user is not the person who applied during the investigation. Furthermore, the black voiceprint library can be understood as a database that stores black voiceprints.

[0067] For example, the second similarity between the second voiceprint information and the black voiceprint can be determined by an algorithm model, and a second threshold can be embedded in the algorithm model.

[0068] In operation S260, when the second similarity is greater than the second threshold, the first voiceprint information and / or the second voiceprint information is stored in the black voiceprint library. It can be understood that when the second similarity is greater than the second threshold, it means that the second voiceprint information and the black voiceprint are relatively consistent, and it can be determined that the applicant of the bank card has a bad record and further measures need to be taken, such as further review, etc. To be on the safe side, the first voiceprint information can be stored in the black voiceprint library; the second voiceprint information can also be stored in the black voiceprint library; and both the first voiceprint information and the second voiceprint information can also be stored in the black voiceprint library. Here, the second threshold is any value that can prove that the second voiceprint information and the black voiceprint are consistent, such as 99.5%.

[0069] According to the bank card application risk control method based on voiceprint recognition in the embodiment of the present disclosure, it is possible to accurately identify whether the actual applicant of the bank card and the investigated user are the same person based on the comparison of the first voiceprint information and the second voiceprint information; and by comparing the black voiceprint in the black voiceprint library with the second voiceprint information, it is possible to determine whether the bank card applicant has a bad record, and further measures can be taken against applicants with bad records. The process of storing the first voiceprint information and / or the second voiceprint information in the black voiceprint library can continuously improve the black voiceprint library. Therefore, the method of the present disclosure is conducive to bank risk control, thereby reducing the economic losses of banks and customers.

[0070] In some embodiments of the present disclosure, Figure 2 As shown, when the first similarity is less than the first threshold, operation 270 stores the first voiceprint information and / or the second voiceprint information in the black voiceprint library. It can be understood that when the first similarity is less than the first threshold, it means that the first voiceprint information and the second voiceprint information do not match, and it can be determined that the person who applied for the bank card is not the person himself, and further measures can be taken, such as further manual review of the application information and applicant of the bank card. At the same time, the first voiceprint information can be stored in the black voiceprint library; the second voiceprint information can also be stored in the black voiceprint library; and both the first voiceprint information and the second voiceprint information can also be stored in the black voiceprint library.

[0071] In some embodiments of the present disclosure, Figure 2 As shown, when the second similarity is less than the second threshold, it means that the second voiceprint information does not match the black voiceprint, and it can be determined that the applicant has no bad record. At this time, the bank can give the investigation conclusion of the bank card application risk control method mentioned in this disclosure.

[0072] Fig. 9A flowchart for determining the first similarity between the first voiceprint information and the second voiceprint information according to an embodiment of the present disclosure is schematically shown. After obtaining the first voiceprint information of the user, the first voiceprint information is stored in a temporary voiceprint library, and the temporary voiceprint library includes at least one temporary voiceprint. In other words, after the first voiceprint information is stored in the temporary voiceprint library, the first voiceprint information is a temporary voiceprint in the temporary voiceprint library, and the temporary voiceprint library may also contain other voiceprints that have been stored in the temporary voiceprint library.

[0073] Operation S240 of determining the first similarity between the first voiceprint information and the second voiceprint information includes operations S241 to S243.

[0074] In operation S241, the second voiceprint information is matched with a temporary voiceprint in a temporary voiceprint library.

[0075] In operation S242, a list of similarity values ​​between the second voiceprint information and the temporary voiceprint is obtained. There may be n temporary voiceprints in the temporary voiceprint library, where n is an integer greater than or equal to 1, and the second voiceprint information may be compared with each temporary voiceprint, so n similarity values ​​may be obtained. As an example, the first m similarity values ​​may be displayed as a list in descending order, where m is an integer less than or equal to n.

[0076] In operation S243, the first similarity between the second voiceprint information and the first voiceprint information is obtained. The first similarity between the second voiceprint information and the first voiceprint information can be queried in the list, so that the first similarity between the second voiceprint information and the first voiceprint information can be easily obtained, so that the first similarity can be further compared with the first threshold.

[0077] Fig.10 The flowchart of determining the second similarity between the second voiceprint information and the black voiceprints in the black voiceprint library according to an embodiment of the present disclosure is schematically shown.

[0078] Operation S250 of determining the second similarity between the second voiceprint information and the black voiceprints in the black voiceprint library includes operations S251 and S252.

[0079] In operation S251, a similarity value list between the second voiceprint information and the black voiceprint is obtained. There may be s black voiceprints in the black voiceprint library, s is an integer greater than or equal to 1, and the second voiceprint information may be compared with each black voiceprint, so s similarity values ​​may be obtained. As an example, the first t similarity values ​​may be displayed as a list in descending order, t is an integer less than or equal to s.

[0080] In operation S252, the second similarity between the second voiceprint information and the black voiceprint is obtained. The second similarity between the second voiceprint information and the black voiceprint can be queried in the list, so that the second similarity between the second voiceprint information and the black voiceprint can be easily obtained, so that the second similarity can be further compared with the second threshold.

[0081] The following is a detailed description of a bank card application risk control method based on voiceprint recognition according to an embodiment of the present disclosure. It is worth noting that the following description is only an exemplary description and is not a specific limitation of the present disclosure.

[0082] The bank card application risk control method based on voiceprint recognition disclosed in the present invention is that the user initiates a bank card application from a mobile terminal, undergoes multiple hardware voiceprint collection and noise reduction processing, as well as voiceprint model training, matching and other neural network algorithms, and is combined with the bank user application bank card risk screening mechanism to supplement the management of the black voiceprint library.

[0083] Step 1: The user initiates a bank card application based on a mobile terminal. The mobile terminal displays text information for the user to read aloud and obtains the user's audio information. Because the environment in which the user applies is relatively complex, noise reduction processing is required to obtain effective audio, obtain a clear and usable voiceprint stream, and return to the front end to obtain a valid voiceprint progress bar. When the progress bar is full, the voiceprint recording can be interrupted, and the effective voiceprint determination time can be flexibly configured through parameters.

[0084] Step 2: Extract voiceprint features based on voiceprint collection equipment and neural network model, and train the neural network model to generate model data. Train Gaussian mixture model-universal background model (GMM-UBM), obtain user voiceprint features and register them to the temporary voiceprint library. Gaussian mixture model-universal background model (GMM-UBM) is a voiceprint model trained based on massive background data, and the extracted voiceprint features are registered to the temporary voiceprint library.

[0085] Step 3: The bank investigator initiates the investigation, dials the application phone number, initiates the investigation, repeats step 1 to obtain the voiceprint information of the person under investigation, and uses the neural network algorithm to initiate the first voiceprint match, matches the investigated voiceprint with the temporary voiceprint library for consistency, and returns the first N similarity score voiceprint list (the N value and similarity threshold of the first N can be adjusted through parameters), N is an integer greater than or equal to 1, if the person under investigation has a low match with the temporary voiceprint library, the authenticity of the investigated user and the applicant user will be manually consulted, and the voiceprint features registered in the temporary voiceprint library and / or the investigated voiceprint will be added to the black voiceprint library. The system provides a front-end function for the business to add the voiceprint blacklist. If the person under investigation has a high match with the temporary voiceprint library, execute step 4.

[0086] Step 4: Further use the voiceprint information of the person under investigation to search in the black voiceprint database, and return the top N similarity score list to display to the front desk, and migrate the voiceprint features and / or the investigated voiceprint registered in the temporary voiceprint database to the black voiceprint database.

[0087] Step 5: People can flexibly add, delete or modify the blacklist through the system front desk, and give the investigation conclusion to continue the subsequent investigation steps.

[0088] Based on the above-mentioned bank card application risk control method based on voiceprint recognition, the present disclosure also provides a bank card application risk control device 10 based on voiceprint recognition. Fig.11 The bank card application risk control device 10 is described in detail.

[0089] Fig.11 The structural block diagram of the bank card application risk control device 10 according to an embodiment of the present disclosure is schematically shown.

[0090] The bank card application risk control device 10 includes a bank card application request module 1, a first voiceprint acquisition module 2, a second voiceprint acquisition module 3, a first information matching module 4, a second information matching module 5 and a black voiceprint adding module 6.

[0091] The bank card application request module 1 is used to execute operation S210 to receive a user's bank card application request.

[0092] The first voiceprint obtaining module 2 is used to perform operation S220 to obtain the first voiceprint information of the user when the user applies for a bank card.

[0093] The second voiceprint obtaining module 3 is used to perform operation S230 to obtain the second voiceprint information of the user when investigating the user.

[0094] The first information matching module 4 is used to perform operation S240 to determine a first similarity between the first voiceprint information and the second voiceprint information.

[0095] The second information matching module 5 is configured to perform operation S250 to determine a second similarity between the second voiceprint information and the black voiceprints in the black voiceprint database when the first similarity is greater than the first threshold.

[0096] The black voiceprint adding module 6 is used to perform operation S260 to store the first voiceprint information and / or the second voiceprint information into the black voiceprint library when the second similarity is greater than the second threshold.

[0097] Since the above-mentioned bank card application risk control device 10 is set based on the bank card application risk control method, the beneficial effects of the above-mentioned bank card application risk control device 10 are the same as those of the bank card application risk control method, which will not be repeated here.

[0098] In addition, according to an embodiment of the present disclosure, any multiple modules among the bank card application request module 1, the first voiceprint acquisition module 2, the second voiceprint acquisition module 3, the first information matching module 4, the second information matching module 5 and the black voiceprint adding module 6 can be combined into one module for implementation, or any one of the modules can be split into multiple modules. Alternatively, at least part of the functions of one or more of these modules can be combined with at least part of the functions of other modules and implemented in one module.

[0099] According to an embodiment of the present disclosure, at least one of the bank card application request module 1, the first voiceprint acquisition module 2, the second voiceprint acquisition module 3, the first information matching module 4, the second information matching module 5 and the black voiceprint adding module 6 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application-specific integrated circuit (ASIC), or can be implemented by hardware or firmware such as any other reasonable way of integrating or packaging the circuit, or can be implemented in any one of the three implementation methods of software, hardware and firmware, or in an appropriate combination of any of them.

[0100] Alternatively, at least one of the bank card application request module 1, the first voiceprint acquisition module 2, the second voiceprint acquisition module 3, the first information matching module 4, the second information matching module 5 and the black voiceprint adding module 6 can be at least partially implemented as a computer program module, and when the computer program module is run, the corresponding function can be executed.

[0101] Fig.12 A block diagram of an electronic device suitable for implementing a bank card application risk control method based on voiceprint recognition according to an embodiment of the present disclosure is schematically shown.

[0102] like Fig.12 As shown, the electronic device 900 according to an embodiment of the present disclosure includes a processor 901, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 902 or a program loaded from a storage part 908 into a random access memory (RAM) 903. The processor 901 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or a related chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 901 may also include an onboard memory for caching purposes. The processor 901 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.

[0103] In RAM 903, various programs and data required for the operation of electronic device 900 are stored. Processor 901, ROM 902 and RAM 903 are connected to each other via bus 904. Processor 901 performs various operations of the method flow according to the embodiment of the present disclosure by executing the program in ROM 902 and / or RAM 903. It should be noted that the program can also be stored in one or more memories other than ROM 902 and RAM 903. Processor 901 can also perform various operations of the method flow according to the embodiment of the present disclosure by executing the program stored in the one or more memories.

[0104] According to an embodiment of the present disclosure, the electronic device 900 may further include an input / output (I / O) interface 905, which is also connected to the bus 904. The electronic device 900 may further include one or more of the following components connected to the I / O interface 905: an input portion 906 including a keyboard, a mouse, etc.; an output portion 907 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage portion 908 including a hard disk, etc.; and a communication portion 909 including a network interface card such as a LAN card, a modem, etc. The communication portion 909 performs communication processing via a network such as the Internet. A drive 910 is also connected to the input / output (I / O) interface 905 as needed. A removable medium 911, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 910 as needed, so that a computer program read therefrom is installed into the storage portion 908 as needed.

[0105] The present disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or may exist independently without being assembled into the device / apparatus / system. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the method according to the embodiment of the present disclosure is implemented.

[0106] According to an embodiment of the present disclosure, a computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, may include but is not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, an apparatus, or a device. For example, according to an embodiment of the present disclosure, a computer-readable storage medium may include the ROM 902 and / or RAM 903 described above and / or one or more memories other than ROM 902 and RAM 903.

[0107] The embodiment of the present disclosure also includes a computer program product, which includes a computer program, and the computer program contains program code for executing the method shown in the flowchart. When the computer program product is run in a computer system, the program code is used to enable the computer system to implement the method of the embodiment of the present disclosure.

[0108] The above functions defined in the system / device of the embodiment of the present disclosure are performed when the computer program is executed by the processor 901. According to the embodiment of the present disclosure, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0109] In one embodiment, the computer program may be based on a tangible storage medium such as an optical storage device, a magnetic storage device, etc. In another embodiment, the computer program may also be transmitted and distributed in the form of a signal on a network medium, and downloaded and installed through the communication part 909, and / or installed from a removable medium 911. The program code contained in the computer program may be transmitted using any appropriate network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0110] In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 909, and / or installed from the removable medium 911. When the computer program is executed by the processor 901, the above functions defined in the system of the embodiment of the present disclosure are performed. According to the embodiment of the present disclosure, the system, device, means, module, unit, etc. described above can be implemented by a computer program module.

[0111] According to an embodiment of the present disclosure, the program code for executing the computer program provided by the embodiment of the present disclosure can be written in any combination of one or more programming languages. Specifically, these computing programs can be implemented using high-level process and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, Java, C++, python, "C" language or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on the remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect through the Internet).

[0112] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0113] It will be appreciated by those skilled in the art that the features described in the various embodiments and / or claims of the present disclosure may be combined and / or combined in a variety of ways, even if such combinations and / or combinations are not explicitly described in the present disclosure. In particular, the features described in the various embodiments and / or claims of the present disclosure may be combined and / or combined in a variety of ways without departing from the spirit and teachings of the present disclosure. All of these combinations and / or combinations fall within the scope of the present disclosure.

[0114] The embodiments of the present disclosure are described above. However, these embodiments are only for illustrative purposes and are not intended to limit the scope of the present disclosure. Although the embodiments are described above separately, this does not mean that the measures in the various embodiments cannot be used in combination to advantage. The scope of the present disclosure is defined by the attached claims and their equivalents. Without departing from the scope of the present disclosure, those skilled in the art may make a variety of substitutions and modifications, which should all fall within the scope of the present disclosure.

Claims

1. A bank card application risk control method based on voiceprint recognition, characterized in that: include: Receive bank card application requests from users; When a user applies for a bank card, obtain the user's first voiceprint information; Storing the first voiceprint information in a temporary voiceprint database, wherein the temporary voiceprint database includes at least one temporary voiceprint; When investigating a user, obtain the user's second voiceprint information; Determining a first similarity between the first voiceprint information and the second voiceprint information; When the first similarity is greater than a first threshold, determining a second similarity between the second voiceprint information and a black voiceprint in a black voiceprint library; and When the second similarity is greater than a second threshold, storing the first voiceprint information and / or the second voiceprint information in the black voiceprint library; The determining of the first similarity between the first voiceprint information and the second voiceprint information includes: Matching the second voiceprint information with the temporary voiceprint in the temporary voiceprint library; Obtaining a list of similarity values ​​between the second voiceprint information and the temporary voiceprint; and The first similarity between the second voiceprint information and the first voiceprint information is obtained.

2. The method according to claim 1, characterized in that When the first similarity is less than the first threshold, the first voiceprint information and / or the second voiceprint information is stored in the black voiceprint library.

3. The method according to claim 1, characterized in that: Determining the second similarity between the second voiceprint information and the black voiceprint in the black voiceprint database includes: Obtaining a list of similarity values ​​between the second voiceprint information and the black voiceprint; and The second similarity between the second voiceprint information and the black voiceprint is obtained.

4. The method according to claim 1, characterized in that The obtaining of the first voiceprint information of the user includes: Display text information; Recording a voiceprint of a user reading the text message; and The voiceprint is subjected to noise reduction processing, and the voiceprint after noise reduction is the first voiceprint information.

5. The method according to claim 1, characterized in that The obtaining of the first voiceprint information of the user includes: Display text information; Recording the voiceprint of the user reading the text message; Denoising the voiceprint, wherein the denoising the voiceprint comprises removing noise in the voiceprint that is irrelevant to the reading of the text information, to obtain a first valid voiceprint; and A voiceprint feature of the first valid voiceprint is extracted, where the voiceprint feature is the first voiceprint information.

6. The method according to claim 5, characterized in that The obtaining of the user's first voiceprint information further includes: A first valid voiceprint progress bar is set, wherein the first valid voiceprint progress bar is the duration of the first valid voiceprint. After the recorded voiceprint is processed with noise reduction, if the first valid voiceprint progress bar does not meet the threshold, continue to record the voiceprint of the user reading the text information and process the voiceprint with noise reduction until the first valid voiceprint progress bar meets the threshold. The voiceprint feature of extracting the first valid voiceprint is the voiceprint feature of extracting the first valid voiceprint after the first valid voiceprint progress bar meets the threshold.

7. The method according to claim 5, characterized in that The extracting of the voiceprint features of the first effective voiceprint includes extracting the voiceprint features using a neural network model.

8. The method according to claim 7, characterized in that The neural network model is a Gaussian mixture model-universal background model.

9. The method according to any one of claims 1 to 8, characterized in that The obtaining of the second voiceprint information of the user comprises: The bank calls the contact information provided when initiating the bank card application request; and The audio information of the caller is collected, where the audio information is the second voiceprint information.

10. The method according to any one of claims 1 to 8, characterized in that The obtaining of the second voiceprint information of the user comprises: The bank calls the contact information filled in when initiating the bank card application request; Collecting audio information of the receiver; Noise reduction processing is performed on the audio information, wherein the noise reduction processing includes removing noise irrelevant to the conversation content in the audio information to obtain a second effective voiceprint; and A voiceprint feature of the second valid voiceprint is extracted, where the voiceprint feature is the second voiceprint information.

11. A bank card application risk control device based on voiceprint recognition, characterized in that: include: A bank card application request module, used to execute and receive a user's bank card application request; A first voiceprint acquisition module, when a user applies for a bank card, is used to acquire the user's first voiceprint information and store the first voiceprint information in a temporary voiceprint database, wherein the temporary voiceprint database includes at least one temporary voiceprint; A second voiceprint acquisition module, used to acquire the second voiceprint information of the user when investigating the user; A first information matching module, configured to determine a first similarity between the first voiceprint information and the second voiceprint information; A second information matching module, when the first similarity is greater than a first threshold, for determining a second similarity between the second voiceprint information and a black voiceprint in a black voiceprint library; and a black voiceprint adding module, for executing storage of the first voiceprint information and / or the second voiceprint information into the black voiceprint library when the second similarity is greater than a second threshold; The determining of the first similarity between the first voiceprint information and the second voiceprint information includes: Matching the second voiceprint information with the temporary voiceprint in the temporary voiceprint library; Obtaining a list of similarity values ​​between the second voiceprint information and the temporary voiceprint; and The first similarity between the second voiceprint information and the first voiceprint information is obtained.

12. An electronic device, characterized in that: include: one or more processors; One or more memories, used to store executable instructions, which, when executed by the processor, implement the method according to any one of claims 1 to 10.

13. A computer-readable storage medium, characterized in that: The storage medium stores executable instructions, which, when executed by a processor, implement the method according to any one of claims 1 to 10.

14. A computer program product, characterized in that The computer program product comprises one or more executable instructions, and when the executable instructions are executed by a processor, the method according to any one of claims 1 to 10 is implemented.

Images