Payment authentication method and device, equipment, storage medium and product

Abstract

The invention belongs to the technical field of security authentication, and discloses a payment authentication method and device, equipment, a storage medium and a product. The behavior deviation degree is determined according to the payment behavior information of the user, the transaction vulnerability is determined according to the transaction information of the user transaction, then the transaction risk index is determined according to the behavior deviation degree and the transaction vulnerability, and whether the user transaction is a high-risk transaction is judged according to the transaction risk index; and if yes, determining the voiceprint of the user according to the audio clip sent by the user, and performing payment authentication according to the voiceprint similarity between the voiceprint of the user and the base voiceprint. According to the method, whether the user transaction is a high-risk transaction can be effectively evaluated, if yes, the risk of counterfeiting is remarkably reduced by using voiceprint as payment authentication, so that payment authentication is safely and conveniently performed, the reliability of payment authentication is improved, meanwhile, the voiceprint recognition process is simple and rapid, and the user experience is improved. And the user experience and the payment efficiency are greatly improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of security authentication, in particular to a payment authentication method, device, equipment, storage medium and product. BACKGROUND

[0002] Although the existing digital payment technology has made remarkable achievements in convenience and popularity, it still has several key shortcomings and limitations. For example, it adopts a single authentication method, lacks hierarchy, and has a high security risk. Traditional password-based authentication methods are vulnerable to security threats such as phishing, hacking, and password leakage. The complexity and memory burden of passwords often bring inconvenience to users. Secondly, biometric authentication methods such as fingerprint and facial recognition, although improving security, still have the risk of being imitated or deceived. The accuracy of these methods is often affected by external environmental factors such as hand moisture, facial expression changes or damage. Therefore, how to safely and conveniently conduct payment authentication and improve the reliability of payment authentication has become a problem to be solved.

[0003] SUMMARY

[0004] The main purpose of the present application is to provide a payment authentication method, device, equipment, storage medium and product, which aims to solve the technical problem of how to safely and conveniently conduct payment authentication and improve the reliability of payment authentication.

[0005] To achieve the above purpose, the present application provides a payment authentication method, which comprises the following steps:

[0006] determining a behavior deviation degree according to the payment behavior information of the user, and determining a transaction vulnerability according to the transaction information of the user transaction;

[0007] determining a transaction risk index according to the behavior deviation degree and the transaction vulnerability, and judging whether the user transaction is a high-risk transaction according to the transaction risk index;

[0008] If yes, determining a user voiceprint according to the audio segment sent by the user, and performing payment authentication according to the voiceprint similarity between the user voiceprint and a base voiceprint.

[0009] Optionally, the step of determining a behavior deviation degree according to the payment behavior information of the user, and determining a transaction vulnerability according to the transaction information of the user transaction, specifically comprises:

[0010] determining a transaction number deviation, a transaction amount deviation and a transaction frequency deviation according to the payment behavior information of the user;

[0011] determining a behavior deviation degree according to the transaction number deviation, the transaction amount deviation and the transaction frequency deviation;

[0012] determine transaction vulnerability according to transaction object information and transaction content information in transaction information of a user transaction.

[0013] Optionally, the step of determining transaction frequency deviation, transaction amount deviation, and transaction frequency deviation according to payment behavior information of the user specifically includes:

[0014] determine average daily transaction frequency, average daily transaction amount, and average transaction interval according to historical payment information of the user;

[0015] obtain payment amount, daily cumulative payment frequency, daily cumulative payment amount, this transaction time, and last transaction time in the payment behavior information of the user;

[0016] calculate transaction frequency deviation according to the average daily transaction frequency, the daily cumulative payment frequency, the average daily transaction frequency, and transaction frequency deviation tolerance;

[0017] calculate transaction amount deviation according to the average daily transaction amount, the daily cumulative payment amount, the payment amount, and transaction amount deviation tolerance;

[0018] calculate transaction frequency deviation according to the this transaction time, the last transaction time, the average transaction interval, and transaction frequency deviation tolerance.

[0019] Optionally, the step of determining user voiceprint according to the audio segment sent by the user and performing payment authentication according to voiceprint similarity between the user voiceprint and a base voiceprint if yes specifically includes:

[0020] if yes, collect the audio segment sent by the user based on a randomly generated text sequence;

[0021] determine user voiceprint according to the text sequence and the audio segment, and perform voiceprint authentication according to voiceprint similarity between the user voiceprint and a base voiceprint to obtain a voiceprint authentication result;

[0022] perform regular authentication based on authentication information of the user to obtain a regular authentication result;

[0023] perform payment authentication according to the voiceprint authentication result and the regular authentication result.

[0024] Optionally, the step of determining user voiceprint according to the text sequence and the audio segment, and performing voiceprint authentication according to voiceprint similarity between the user voiceprint and a base voiceprint to obtain a voiceprint authentication result specifically includes:

[0025] perform preprocessing on the audio segment to obtain a processed audio segment;

[0026] extract a voiceprint from the dual-network structure model, the processed audio segment, and the text sequence, to obtain a user voiceprint, the dual-network structure model including a voiceprint extraction network and an audio synthesis network;

[0027] perform voiceprint authentication according to a voiceprint similarity between the user voiceprint and a base voiceprint, to obtain a voiceprint authentication result.

[0028] Optionally, the step of preprocessing the audio segment to obtain a processed audio segment specifically includes:

[0029] determine a background noise type according to a payment scenario of a user, and determine a frequency range corresponding to the background noise type;

[0030] determine a notch center frequency according to the frequency range;

[0031] perform filtering processing on background noise in the audio segment according to the notch center frequency, a notch width, and a discretized sampling time, to obtain a processed audio segment.

[0032] In addition, to achieve the above object, the present application further provides a payment authentication device, which comprises:

[0033] an information determination module configured to determine a behavior deviation degree according to payment behavior information of a user, and determine a transaction vulnerability according to transaction information of a transaction of the user;

[0034] a transaction judgment module configured to determine a transaction risk index according to the behavior deviation degree and the transaction vulnerability, and determine whether the transaction of the user is a high-risk transaction according to the transaction risk index;

[0035] a payment authentication module configured to, when the transaction of the user is a high-risk transaction, determine a user voiceprint according to an audio segment sent by the user, and perform payment authentication according to a voiceprint similarity between the user voiceprint and a base voiceprint.

[0036] In addition, to achieve the above object, the present application further provides a payment authentication device, which comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the payment authentication method as described above.

[0037] In addition, to achieve the above object, the present application further provides a storage medium, which is a computer-readable storage medium, and has a computer program stored thereon, the computer program being executable on a processor to implement the steps of the payment authentication method as described above.

[0038] In addition, to achieve the above object, the application further provides a computer program product, which comprises a computer program, and the computer program realizes the steps of the payment authentication method when executed by a processor.

[0039] The application determines the behavior deviation according to the payment behavior information of the user, determines the transaction vulnerability according to the transaction information of the user transaction, then determines the transaction risk index according to the behavior deviation and the transaction vulnerability, and judges whether the user transaction is a high-risk transaction according to the transaction risk index. If yes, the voiceprint of the user is determined according to the audio segment sent by the user, and the payment authentication is performed according to the voiceprint similarity between the voiceprint of the user and the base voiceprint. The application determines the transaction risk index according to the behavior deviation and the transaction vulnerability, which can effectively evaluate whether the user transaction is a high-risk transaction. If yes, the payment authentication is performed according to the voiceprint similarity between the voiceprint of the user and the base voiceprint. By using the voiceprint as the payment authentication mode, the risk of being imitated is significantly reduced, so that the payment authentication is safe and convenient, the reliability of the payment authentication is improved, and at the same time, the process of voiceprint recognition is simple and fast, which greatly improves the user experience and payment efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0040] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and serve to explain the principles of the application together with the specification.

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the accompanying drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, for those skilled in the art, other drawings can also be obtained based on these drawings without creative labor.

[0042] Figure 1 The flowchart of the first embodiment of the payment authentication method of the application;

[0043] Figure 2 The schematic diagram of the product model of the first embodiment of the payment authentication method of the application;

[0044] Figure 3 The overall logic flowchart of the first embodiment of the payment authentication method of the application;

[0045] Figure 4 The flowchart of the second embodiment of the payment authentication method of the application;

[0046] Figure 5 The flowchart of the third embodiment of the payment authentication method of the application;

[0047] Figure 6A structural schematic diagram of a dual-network structure model of an embodiment of the payment authentication method of the present application;

[0048] Figure 7 A structural block diagram of a first embodiment of the payment authentication device of the present application.

[0049] Figure 8 A structural schematic diagram of a payment authentication device of a hardware running environment involved in the embodiment scheme of the present application.

[0050] The implementation, functional features and advantages of the present application will be further explained with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0051] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application, and are not used to limit the present application.

[0052] In order to better understand the technical solutions of the present application, the following will be described in detail in conjunction with the drawings and specific embodiments of the specification.

[0053] It should be noted that the execution subject of the present embodiment can be a computing service device with data processing, network communication and program running functions, such as a computer, or an electronic device, a payment authentication device, etc. capable of realizing the above functions. The present embodiment and the following embodiments will be described taking the payment authentication device as an example.

[0054] Based on this, the present embodiment provides a payment authentication method, referring to Figure 1 , Figure 1 A flowchart of a first embodiment of the payment authentication method of the present application.

[0055] In the present embodiment, the payment authentication method comprises the following steps:

[0056] Step S10: determining the behavior deviation degree according to the payment behavior information of the user, and determining the transaction vulnerability according to the transaction information of the user transaction.

[0057] It can be understood that the present embodiment can obtain the payment behavior information of the user, the user refers to the user who needs to perform payment authentication, and the payment behavior information refers to the information related to the payment behavior of the user. The payment behavior information can be automatically obtained when the user performs a transaction and before payment authentication, and can specifically include the payment amount, the cumulative payment times per day, etc. The behavior deviation degree is determined according to the payment behavior information, and the behavior deviation degree can represent the transaction frequency deviation degree, the transaction amount deviation degree and the transaction frequency deviation degree of the user.

[0058] It should be understood that the user transaction refers to a transaction generated by the user, and the transaction information of the user transaction can be acquired by the embodiment before payment authentication, and the transaction information can include the type of the opposite merchant, the transaction amount, the payment method, and the like. The transaction vulnerability is determined according to the transaction information, and the transaction vulnerability can be used to measure whether the transaction is safe and reliable. The higher the vulnerability is, the less reliable the transaction is.

[0059] Step S20: determining a transaction risk index according to the behavior deviation degree and the transaction vulnerability, and judging whether the user transaction is a high-risk transaction according to the transaction risk index.

[0060] It can be understood that the embodiment can determine the transaction risk index according to the behavior deviation degree and the transaction vulnerability. Specifically, the transaction risk index can be calculated by formula (1).

[0061] R = a · D + b · V + g · D · V (1)

[0062] In formula (1), R is the transaction risk index, D is the behavior deviation degree, V is the transaction vulnerability, a and b are weight coefficients, and need to be determined through historical data. If the behavior deviation degree is considered to be more indicative of risk than the transaction vulnerability, a should be greater than b. g is an adjustment coefficient, which can be set according to the interaction strength between D and V. If it is considered that the interaction between them has a significant impact on risk, g should be appropriately increased. The coefficients of the embodiment can be set as a = b = 0.5 and g = 0.2. The above coefficients can also be set to other values according to data and experience.

[0063] It should be understood that the transaction risk index R can be compared with a preset threshold value d. If R > d, it is determined that the user transaction is a high-risk transaction. In the embodiment, the threshold value d is dynamically changed. First, a base value is obtained by statistically analyzing the historical data of the user, and after each transaction, d is updated as formula (2).

[0064] d' = d + lr x (R - d) (2)

[0065] In formula (2), lr is the update step, and the embodiment can set lr = 0.1. d' is the updated threshold value.

[0066] Step S30: if yes, determining the user voiceprint according to the audio segment sent by the user, and performing payment authentication according to the voiceprint similarity between the user voiceprint and the base voiceprint.

[0067] Understandably, when a user's transaction is a high-risk transaction, regular authentication and voiceprint authentication can be performed. Voiceprint authentication can first determine the user's voiceprint based on the audio segment sent by the user, that is, extract the user's voiceprint from the audio segment. Each user has a unique base voiceprint. The voiceprint similarity between the user's voiceprint and the base voiceprint can be calculated, and payment authentication can be performed based on the voiceprint similarity. Specifically, if the similarity is greater than a preset value, the authentication is deemed to have failed.

[0068] It should be understood that, referring to Figure 2 , Figure 2 This is a schematic diagram of a product model representing an embodiment of the payment authentication method of this application. The payment authentication method proposed in this embodiment can... Figure 2 The product model shown consists of four modules: 1. Transaction Risk Assessment Module; 2. Smart Secure Payment Module; 3. Conventional Authentication Module; 4. Voiceprint Authentication Module. In the Transaction Risk Assessment Module, the basic information and content of the transaction are comprehensively considered to assess its risk, and the risk index is calculated quantitatively. In the Smart Secure Payment Module, the risk index of the transaction is compared with a pre-set risk threshold to determine the appropriate authentication process. For transactions below the risk threshold, conventional payment authentication methods (such as fingerprints or passwords) are used; for transactions above the risk threshold, authentication is divided into two stages: the first stage is the conventional authentication stage, which can use a password or fingerprint for authentication. Only after successful authentication in the first stage will the next stage be initiated. The second stage is the voiceprint authentication stage, which involves collecting the user's voiceprint and comparing it with the base voiceprint in the device's voiceprint library. Based on the authentication results, the Smart Secure Payment Module takes corresponding measures. Figure 3 As shown, Figure 3 This is an overall logic flowchart of an embodiment of the payment authentication method of this application.

[0069] This embodiment determines the behavioral deviation degree based on the user's payment behavior information and the transaction vulnerability based on the user's transaction information. Then, it determines the transaction risk index based on the behavioral deviation degree and transaction vulnerability, and judges whether the user's transaction is a high-risk transaction based on the transaction risk index. If so, it determines the user's voiceprint based on the audio clip sent by the user, and performs payment authentication based on the voiceprint similarity between the user's voiceprint and the base voiceprint. This embodiment determines the transaction risk index based on behavioral deviation degree and transaction vulnerability, which can effectively assess whether the user's transaction is a high-risk transaction. If so, it performs payment authentication based on the voiceprint similarity between the user's voiceprint and the base voiceprint. By using voiceprint as a payment authentication method, the risk of being counterfeited is significantly reduced, thereby enabling secure and convenient payment authentication and improving the reliability of payment authentication. At the same time, the voiceprint recognition process is simple and fast, greatly improving the user experience and payment efficiency.

[0070] refer toFigure 4 Figure 4 Flowchart of the second embodiment of the payment authentication method.

[0071] Based on the first embodiment, in the present embodiment, the step S10 comprises:

[0072] Step S101: determining the transaction frequency deviation, transaction amount deviation and transaction frequency deviation according to the payment behavior information of the user.

[0073] Further, in order to accurately determine the transaction frequency deviation, transaction amount deviation and transaction frequency deviation, in the present embodiment, the step S101 comprises: determining the average daily transaction frequency, average daily transaction amount and average transaction interval according to the historical payment information of the user; obtaining the payment amount, daily cumulative payment frequency, single-day cumulative payment amount, this transaction time and last transaction time in the payment behavior information of the user; calculating the transaction frequency deviation according to the average daily transaction frequency, daily cumulative payment frequency, average daily transaction frequency and transaction frequency deviation tolerance; calculating the transaction amount deviation according to the average daily transaction amount, daily cumulative payment amount, payment amount and transaction amount deviation tolerance; calculating the transaction frequency deviation according to the this transaction time, last transaction time, average transaction interval and transaction frequency deviation tolerance.

[0074] It can be understood that the historical payment information of the user can be collected, including identifying common transaction patterns, frequent transaction periods, and transaction frequency with specific merchants or service categories, etc., and determining the average daily transaction frequency av_ct, average daily transaction amount av_at and average transaction interval av_pd according to the historical payment information. In order to ensure the privacy and security of the user data, the above processes are all completed locally without uploading data to the Internet.

[0075] It should be understood that when the user makes a transaction, the payment behavior information of the user can be automatically obtained before payment authentication, which can include payment amount at, daily cumulative payment frequency ac_ct, single-day cumulative payment amount ac_at, this transaction time t1 and last transaction time t0. The transaction frequency deviation tolerance, transaction amount deviation tolerance and transaction frequency deviation tolerance can be set according to actual conditions. When the deviation is less than the corresponding deviation tolerance, the deviation is smaller and the growth of the deviation is slower, while when the deviation is greater than the corresponding deviation tolerance, the deviation is larger and the growth of the deviation is faster.

[0076] In a specific implementation, the transaction frequency deviation can be calculated by formula (3).

[0077]

[0078] ​In formula (3), d ct is the transaction frequency deviation, av_ct is the average daily transaction frequency, ac_ct is the cumulative payment frequency of the day, and T1 is the transaction frequency deviation tolerance.

[0079] The transaction amount deviation can be calculated by formula (4).

[0080]

[0081] In formula (4), d at is the transaction amount deviation, av_at is the average daily transaction amount, ac_at is the cumulative payment amount of the day, at is the payment amount, and T2 is the transaction amount deviation tolerance.

[0082] The transaction frequency deviation can be calculated by formula (5).

[0083]

[0084] In formula (5), d pd is the transaction frequency deviation, av_pd is the average transaction interval, t1 is the transaction time, t0 is the last transaction time, and T3 is the transaction frequency deviation tolerance.

[0085] Step S102: Determine the behavior deviation degree according to the transaction frequency deviation, the transaction amount deviation, and the transaction frequency deviation.

[0086] It can be understood that the behavior deviation degree can be calculated by formula (6).

[0087] D = w1·d ct + w2·d at + w3·d pd (6)

[0088] In formula (6), D is the behavior deviation degree, d ct is the transaction frequency deviation, d at is the transaction amount deviation, d pd is the transaction frequency deviation, and w1, w2, and w3 are weights, which can be set according to actual conditions, for example, w1 = 0.2, w2 = 0.4, and w3 = 0.4, or other values, which are not specifically limited in the embodiment.

[0089] Step S103: Determine the transaction vulnerability according to the transaction object information and the transaction content information in the transaction information of the user transaction.

[0090] It should be understood that when the user is detected to make a transaction, the transaction information corresponding to the transaction of the user can be automatically obtained before the payment authentication, and the transaction information can include transaction object information and transaction content information, the transaction object information includes: the type of the opposite party merchant, the total transaction amount t_at with the opposite party account, and the transaction content information includes: the transaction amount at, the payment method. The type of the opposite party merchant is divided into five categories: personal, catering and entertainment, retail and consumer goods, professional and financial services, medical and education. The total transaction amount with the opposite party account is obtained by accumulating historical transactions. The transaction amount is the transaction amount of this time. The payment method includes: bank card payment, credit card payment, code scanning payment, near field communication (NFC) payment. For the merchant type and the payment method, as shown in Table 1, it is a vulnerability matrix between the merchant type and the payment method.

[0091] Table 1:

[0092] Merchant Type \ Payment Method Bank Card Payment Credit Card Payment Code Scanning Payment NFC Payment Personal 4 5 5 3 Dining and Entertainment 2 3 4 2 Retail and Consumer Goods 3 4 4 3 Professional and Financial Services 2 2 2 1 Medical and Education 1 2 3 1

[0093] It can be understood that the transaction vulnerability can be calculated by formula (7).

[0094]

[0095] In formula (7), V is the transaction vulnerability, p is the merchant type, q is the payment method, v(p, q) can be obtained by querying the vulnerability matrix between the merchant type and the payment method in Table 1, at is the transaction amount, and t_at is the total transaction amount with the opposite party account.

[0096] The embodiment determines the transaction number deviation, the transaction amount deviation and the transaction frequency deviation according to the payment behavior information of the user, then determines the behavior deviation degree according to the transaction number deviation, the transaction amount deviation and the transaction frequency deviation, and then determines the transaction vulnerability according to the transaction object information and the transaction content information in the transaction information of the user transaction. The embodiment determines the behavior deviation degree according to the transaction number deviation, the transaction amount deviation and the transaction frequency deviation, can accurately obtain the deviation degree of the transaction behavior of the user, and then determines the transaction vulnerability according to the transaction object information and the transaction content information in the transaction information of the user transaction, can accurately obtain the transaction vulnerability of the user transaction, and effectively measures whether the user transaction is safe and reliable.

[0097] Reference Figure 5 , Figure 5 is a flowchart of a third embodiment of a payment authentication method of the application.

[0098] Based on the above embodiments, in the present embodiment, the step S30 comprises:

[0099] Step S301: If yes, the audio segment sent by the user is collected based on the randomly generated text sequence.

[0100] It should be understood that when the user transaction is a low-risk transaction, the user transaction can be subjected to regular authentication, if the authentication is passed, the transaction is successfully carried out, if the authentication is not passed, the transaction is forcibly terminated, and when the regular authentication fails for three times in succession, the transaction function is locked. When the user transaction is a high-risk transaction, the user transaction needs to be subjected to regular authentication and voiceprint authentication.

[0101] It can be understood that in the local device, the intelligent safe payment module randomly generates a text sequence, and in this embodiment, 6-digit numbers are taken as an example for illustration. The above-mentioned 6-digit numbers can be displayed on the screen. The system randomly generates 6-digit numbers in the following manner: the system locally maintains a number sequence generator, takes the current system time as a generation seed, and randomly generates a 6-digit number sequence. Each number sequence is displayed on the screen of the user payment terminal for 10 seconds, and if the user has not started to collect audio after 10 seconds, the number sequence is randomly generated again. The user clicks the “voiceprint authentication” button, the intelligent safe payment module will first apply to the user for recording permission, and after obtaining the permission, there will be a three-second countdown, after the countdown, the device starts recording the sound. The user reads the 6-digit numbers from left to right in normal speed, and the process cannot be paused. After the user presses the stop button, the device stops recording and saves the audio segment. The device collects the audio segment of the user, and sends the audio segment and the text sequence to the voiceprint authentication module. The intelligent safe payment module calls the voiceprint recognition module, and transmits the audio segment in the form of an audio file and the text sequence in the form of a text string to the voiceprint authentication module. The voiceprint authentication module performs voiceprint authentication according to the text sequence and the audio segment.

[0102] Step S302: determining a user voiceprint according to the text sequence and the audio segment, and performing voiceprint authentication according to a voiceprint similarity between the user voiceprint and a base voiceprint to obtain a voiceprint authentication result.

[0103] Further, in this embodiment, the step S302 includes: performing preprocessing on the audio segment to obtain a processed audio segment; performing voiceprint extraction through a double-network structure model, the processed audio segment and the text sequence to obtain a user voiceprint, the double-network structure model including a voiceprint extraction network and an audio synthesis network; and performing voiceprint authentication according to a voiceprint similarity between the user voiceprint and a base voiceprint to obtain a voiceprint authentication result.

[0104] It should be understood that the quality of the audio segment determines the effect of the voiceprint authentication. Since the transaction can occur in various scenarios, the audio segment collected by the device often contains ambient environmental sound, which interferes with voiceprint recognition. In this embodiment, the collected audio segment can be preprocessed, and the preprocessing refers to filtering out the background noise in the audio segment to obtain a processed audio segment.

[0105] In the field of voiceprint authentication payment, voice serves as a unique and individual-differentiated biometric feature, providing an effective approach to achieving a secure and convenient payment method. The core of voiceprint recognition technology lies in capturing and analyzing the unique voiceprint features of each user, which are determined by physiological factors such as vocal cord length, shape, and tension, as well as the size and shape of the throat and mouth. In addition, individual language habits, accents, and pronunciation methods add layers to the uniqueness of voice. These factors collectively influence the basic attributes of voice, such as pitch, timbre, and loudness. In voiceprint authentication payment applications, the system needs to have a high degree of sensitivity to distinguish between different users' voiceprints, while also having sufficient adaptability to accommodate voiceprint fluctuations of the same user due to different payment scenarios, payment time changes, and changes in payment terminal devices.

[0106] It can be understood that the present embodiment can extract user voiceprints through a dual-network structure model, which aims to improve recognition accuracy and ensure the security of the payment process, while also considering a certain fault tolerance for natural changes in user voiceprints, thereby ensuring the convenience and user experience of the payment process. Referring to Figure 6 , Figure 6 The dual-network structure model of the voiceprint authentication method embodiment of the present application is shown in the structural diagram as Figure 6 In the process of extracting user voiceprints, the processed audio segments can be first subjected to feature transformation, which includes pre-emphasis, framing, and calculation of Mel-frequency cepstral coefficients (MFCC). The dual-network structure model uses user audio segments and text sequences corresponding to the audio as input, and uses two fully convolutional neural networks (voiceprint extraction network E and audio synthesis network G) to adapt to audio segments of different lengths. A joint loss function is used as the training loss function of the dual-network structure model, which consists of four parts: content recognition loss L1, voiceprint recognition loss L2, voiceprint distinction loss L3, and voiceprint confusion loss L4. The above four losses are represented by equations (8), (9), (10), and (11).

[0107]

[0108]

[0109]

[0110] Wherein, the processed audio segment is converted into V through feature conversion, s is a text sequence, L1 loss is used to measure the audio content recognition ability of the voiceprint extraction network, only accurate recognition of the audio content can exclude the influence of the content on the voiceprint, and better extract the voiceprint feature; L2 loss is used to measure the voiceprint extraction effect of the model, an additional audio generation network is used to assist training, and the feature extraction ability and generalization ability of the voiceprint extraction network are better trained through multi-task learning; L3 and L4 losses are used to measure the distinguishing and mixing abilities of the model, through the way of contrast learning, the inner product between all voiceprints in a batch of data is calculated, and it is hoped that the inner product of the voiceprints of the same sounder is 1, and the inner product of the voiceprints of different sounders is 0. The four loss functions are combined, that is, L = λ1·L1 + λ2·L2 + λ3·L3 + λ4·L4, and the model parameters of the two networks are optimized and minimized in reverse, and the coefficients in the embodiment can be: λ1 = 0.2, λ2 = 0.4, λ3 = 0.2, λ4 = 0.2, or other coefficients, which are not limited in the embodiment.

[0111] In a specific implementation, a voiceprint library is maintained in the local device of the user, the voiceprint library is composed of a base voiceprint V b and a voiceprint similarity tolerance τ. The base voiceprint is a 256-dimensional feature vector, and each legitimate user has a unique base voiceprint. When the user makes a payment, the voiceprint of the user is extracted through the transaction terminal, and the similarity is calculated by comparing the base voiceprint of the user in the voiceprint library. When the voiceprint similarity is within the similarity tolerance (s ≤ τ), the voiceprint authentication is successful, and a voiceprint authentication success identifier is returned to the intelligent secure payment module. The user has three opportunities for voiceprint authentication, and when all three voiceprint authentications fail, the voiceprint authentication module returns a voiceprint authentication failure identifier to the intelligent secure payment module.

[0112] In the embodiment, the DTW distance between the user voiceprint during the transaction and the base voiceprint in the voiceprint library is calculated based on the dynamic time warping (DTW) algorithm, and then the voiceprint similarity is measured, that is, s = d DTw The advantage of using DTW to calculate the voiceprint distance is that the optimal matching between the user voiceprint and the base voiceprint can be found, even if they are distorted or stretched in time. The process of calculating the DTW distance is as follows:

[0113] For voiceprints P = {p0, p1, p2, …, p 256} and voiceprints Q = {q0, q1, q2, …, q 256}, a 256 × 256 matrix D is defined to calculate and store the similarity matching between the two voiceprints. Wherein, D i,j = |p i -q j | + min(D i-1,j , Di,j-1 D i-1,j-1 ), using dynamic programming to solve DTW: Minimize (D·W), Subject to: w(i,j) = {0,1}, w(1,1) = 0, w(256,256) = 1, w(i,j) = [w(i-1,j) + w(i-1,j-1) + w(i,j-1)] x w(i,j). The DTW distance is defined as equation (12).

[0114]

[0115] In addition, the voiceprint similarity tolerance τ is the maximum DTW distance of the voiceprint authentication module for legal voiceprint authentication. Since the user's voiceprint fluctuates with different transaction scenarios and transaction times, an initial similarity tolerance τ = τ0 is set in this embodiment, and the tolerance is updated with each successful voiceprint authentication transaction. After the kth voiceprint authentication, the similarity tolerance τ is updated to τ = τk-1 + r, where r is the update rate. k , as shown in equation (13).

[0116]

[0117] where s1, s2…s k is the similarity of the kth voiceprint authentication; is the update threshold; r is the update rate. When τ > τ sup after updating, the user needs to re-collect and update the base voiceprint V b , and reset the similarity tolerance τ = τ0. The parameters in this embodiment can be: τ0 = 0.03, τ sup = 0.06, r = 0.05.

[0118] Further, the background noise in the audio segment is effectively filtered out. In this embodiment, the step of pre-processing the audio segment to obtain a processed audio segment specifically includes: determining the type of background noise according to the user's payment scenario, and determining the frequency range corresponding to the type of background noise; determining the notch center frequency according to the frequency range; and filtering the background noise in the audio segment according to the notch center frequency, notch width, and discrete sampling time to obtain a processed audio segment.

[0119] It can be understood that the pre-processing method of the audio segment in this embodiment is to use a 16 kHz sampling rate and a 16-bit depth to resample the audio segment, and convert the audio data into a discrete time series, i.e., a processed audio segment.

[0120] It should be understood that the noise process that can exist in different payment scenarios is that the user selects the payment scenario preset by the system when performing voiceprint authentication, and the system determines the type and frequency range of the background noise that can exist in the payment scenario according to the payment scenario. In this embodiment, the background noise is divided into 6 types, which are low-frequency human voice noise lv, high-frequency human voice noise hv, low-frequency mechanical noise lm, high-frequency mechanical noise hm, music broadcast noise mb, and electronic device running noise en. The type of background noise and the corresponding frequency range are shown in Table 2.

[0121] Table 2:

[0122] Background Noise Type Frequency Range (Hz) Low Frequency Human Voice Noise 100-250 High Frequency Human Voice Noise 1000-1500 Low Frequency Mechanical Noise 60-300 High Frequency Mechanical Noise 2000-5000 Music Broadcast Noise 800-1500 Electronic Device Running Noise 800-3000

[0123] In addition, 5 payment scenarios are preset in this embodiment, and each payment scenario includes different types and different amounts of background noise. The user selects the most similar preset environment for noise cancellation during voiceprint authentication according to the surrounding environment when the transaction occurs. The 5 preset payment scenarios are: shopping mall / supermarket, restaurant / entertainment venue, street / outdoor, public transportation facility, and small indoor space. The types of background noise included in different payment scenarios are shown in Table 3.

[0124] Table 3:

[0125]

[0126]

[0127] In a specific implementation, the type of background noise can be determined according to the payment scenario of the user, which can be determined based on Table 3, and the frequency range corresponding to the type of background noise can be determined, which can be determined based on Table 2. This embodiment can be filtered by a notch filter, which is a filter that can rapidly attenuate an input signal at a certain frequency point to achieve the filtering effect of hindering the passage of signals at this frequency. In this proposal, the bilinear transformation method is used to discretize the notch filter, so that it can filter and eliminate noise at a specific frequency in the user's voice segment (time sequence). The difference equation of the notch filter after bilinear transformation discretization is formula (14).

[0128]

[0129] In formula (14), ω bω is the notch width; ω c is the notch center frequency, which can be determined according to the frequency range corresponding to the type of background noise; T s is the discretized sampling time, which is 1 / 16000 seconds, x(k) is the kth input signal, and y(k) is the kth output signal, i.e. the audio segment after filtering processing.

[0130] Step S303: Perform a regular authentication based on the authentication information of the user, and obtain a regular authentication result.

[0131] It should be understood that the regular authentication operation in the embodiment includes fingerprint authentication, face authentication, and password authentication. In the local device of the user, a fingerprint feature library, a face feature library, and a password library are saved, and a regular authentication module randomly selects an authentication method, and the device collects authentication information of the user and compares and authenticates the data in the library. If the authentication fails, the module will randomly select an authentication method again. The user has three opportunities to perform regular authentication. If the three times of user authentication all fail, the module will return a regular authentication failure identifier. Otherwise, through any one authentication, a regular authentication success identifier is returned. The regular authentication result can include regular authentication success and regular authentication failure. The method of randomly selecting an authentication method is as follows: obtaining a timestamp of this transaction, adding all the digits of the timestamp, and taking the remainder of 3 to obtain a random authentication method 0, 1, and 2, wherein 0 represents fingerprint authentication, 1 represents face authentication, and 2 represents password authentication.

[0132] Step S304: Perform a payment authentication according to the voiceprint authentication result and the regular authentication result.

[0133] It should be understood that the voiceprint authentication result can include voiceprint authentication success and voiceprint authentication failure, and the regular authentication result can include regular authentication success and regular authentication failure. Table 4 is a corresponding relationship between the regular authentication result, the voiceprint authentication result, and the response operation.

[0134] Table 4:

[0135] Regular Authentication Result Voiceprint Authentication Result Response Operation 0 0 A+B 0 1 B 1 0 B 1 1 C

[0136] In Table 4, ‘0’ indicates authentication failure, and ‘1’ indicates authentication success. The response operation A is to lock the transaction, and the user cannot initiate any transaction again within 30 minutes. The response operation B is to immediately terminate the current transaction. The response operation C is to allow the transaction. The above response operations can be the final payment authentication result.

[0137] The embodiment collects an audio segment sent by a user based on a randomly generated text sequence when the user transaction is a high-risk transaction, determines a user voiceprint according to the text sequence and the audio segment, performs voiceprint authentication according to a voiceprint similarity between the user voiceprint and a base voiceprint, obtains a voiceprint authentication result, performs regular authentication based on authentication information of the user, obtains a regular authentication result, and performs payment authentication according to the voiceprint authentication result and the regular authentication result. When the user transaction is a high-risk transaction, the voiceprint authentication according to the voiceprint similarity between the user voiceprint and the base voiceprint can significantly reduce the risk of being imitated, and the user only needs to speak or read a specified text to complete the voiceprint authentication, greatly improving the user experience and payment efficiency, and the combination of regular authentication for payment authentication improves the accuracy of payment authentication.

[0138] Reference Figure 7 , Figure 7 FIG. 1 is a structural block diagram of a payment authentication device according to a first embodiment of the present application.

[0139] As shown in FIG. 1, the payment authentication device according to the first embodiment of the present application comprises: Figure 7

[0140] An information determination module 10 is configured to determine a behavior deviation degree according to payment behavior information of a user, and determine a transaction vulnerability according to transaction information of a user transaction;

[0141] A transaction judgment module 20 is configured to determine a transaction risk index according to the behavior deviation degree and the transaction vulnerability, and determine whether the user transaction is a high-risk transaction according to the transaction risk index;

[0142] A payment authentication module 30 is configured to determine a user voiceprint according to an audio segment sent by the user when the user transaction is a high-risk transaction, and perform payment authentication according to a voiceprint similarity between the user voiceprint and a base voiceprint.

[0143] The embodiment determines a behavior deviation degree according to payment behavior information of a user, and determines a transaction vulnerability according to transaction information of a user transaction, then determines a transaction risk index according to the behavior deviation degree and the transaction vulnerability, and determines whether the user transaction is a high-risk transaction according to the transaction risk index, if yes, determines a user voiceprint according to an audio segment sent by the user, and performs payment authentication according to a voiceprint similarity between the user voiceprint and a base voiceprint. The embodiment determines a transaction risk index according to the behavior deviation degree and the transaction vulnerability, which can effectively evaluate whether the user transaction is a high-risk transaction, if yes, performs payment authentication according to a voiceprint similarity between the user voiceprint and a base voiceprint, which significantly reduces the risk of being imitated, thereby safely and conveniently performing payment authentication and improving the reliability of payment authentication. At the same time, the process of voiceprint recognition is simple and fast, which greatly improves the user experience and payment efficiency.​

[0144] It should be noted that the above-described workflow is merely illustrative and does not limit the scope of protection of the present application. In actual applications, a person skilled in the art can select part or all of the above-described workflow to achieve the purpose of the present embodiment, and the present application does not limit the selection.

[0145] In addition, technical details not described in detail in the present embodiment can be found in the payment authentication method provided by any embodiment of the present application, which will not be described here.

[0146] Based on the first embodiment of the payment authentication device described above, the second embodiment of the payment authentication device of the present application is proposed.

[0147] In the present embodiment, the information determination module 10 is further configured to determine transaction frequency deviation, transaction amount deviation and transaction frequency deviation according to the payment behavior information of the user; determine behavior deviation degree according to the transaction frequency deviation, the transaction amount deviation and the transaction frequency deviation; and determine transaction vulnerability according to the transaction object information and the transaction content information in the transaction information of the user transaction.

[0148] Further, the information determination module 10 is further configured to determine average daily transaction frequency, average daily transaction amount and average transaction interval according to the historical payment information of the user; obtain the payment amount, the daily cumulative payment frequency, the single-day cumulative payment amount, the current transaction time and the last transaction time in the payment behavior information of the user; calculate the transaction frequency deviation according to the average daily transaction frequency, the daily cumulative payment frequency, the average daily transaction frequency and the transaction frequency deviation tolerance; calculate the transaction amount deviation according to the average daily transaction amount, the daily cumulative payment amount, the payment amount and the transaction amount deviation tolerance; and calculate the transaction frequency deviation according to the current transaction time, the last transaction time, the average transaction interval and the transaction frequency deviation tolerance.

[0149] Further, the payment authentication module 30 is further configured to, when the user transaction is a high-risk transaction, collect an audio clip sent by the user based on a randomly generated text sequence; determine a voiceprint of the user according to the text sequence and the audio clip, and perform voiceprint authentication according to the voiceprint similarity between the voiceprint of the user and a base voiceprint to obtain a voiceprint authentication result; perform regular authentication based on the authentication information of the user to obtain a regular authentication result; and perform payment authentication according to the voiceprint authentication result and the regular authentication result.

[0150] Further, the payment authentication module 30 is further configured to preprocess the audio segment to obtain a processed audio segment, extract a voiceprint from the processed audio segment and the text sequence by using a double-network structure model, the processed audio segment and the text sequence, the double-network structure model including a voiceprint extraction network and an audio synthesis network, and perform voiceprint authentication according to a voiceprint similarity between the user voiceprint and a base voiceprint to obtain a voiceprint authentication result.

[0151] Further, the payment authentication module 30 is further configured to determine a background noise type according to a payment scenario of a user, determine a frequency range corresponding to the background noise type, determine a notch center frequency according to the frequency range, and perform filtering processing on the background noise in the audio segment according to the notch center frequency, a notch width and a discretization sampling time to obtain the processed audio segment.

[0152] Other embodiments or specific implementations of the payment authentication apparatus can refer to the above-mentioned method embodiments, and will not be described here.

[0153] The present application provides a payment authentication device, which includes at least one processor and a memory connected with the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the payment authentication method in the above-mentioned embodiment one.

[0154] Reference will be made to the following description Figure 8 which shows a structural schematic diagram of a payment authentication device suitable for implementing the embodiments of the present application. The payment authentication device in the embodiments of the present application can include, but is not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (Personal Digital Assistant), PADs (Portable Application Description), PMPs (Portable Media Player), vehicle-mounted terminals (such as vehicle-mounted navigation terminals) and the like, and fixed terminals such as digital TVs, desktop computers and the like. Figure 8 The payment authentication device shown is only an example, and should not bring any limitation to the functions and use range of the embodiments of the present application.

[0155] As Figure 8As shown, the payment authentication device can include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.) that can perform various appropriate actions and processes according to programs stored in a read only memory (ROM) 1002 or loaded from a storage device 1003 into a random access memory (RAM) 1004. Various programs and data required for the payment authentication device operation are also stored in the RAM 1004. The processing device 1001, the ROM 1002, and the RAM 1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; the storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the payment authentication device to communicate with other devices wirelessly or by wire to exchange data. Although the payment authentication device with various systems is shown in the figure, it should be understood that all the shown systems are not required to be implemented or possessed. More or less systems can be alternatively implemented or possessed.

[0156] In particular, according to the embodiments disclosed in the present application, the processes described above with reference to the flowcharts can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product comprising a computer program carried on a computer readable medium, the computer program containing program codes for executing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network through the communication device, or installed from the storage device 1003, or installed from the ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the methods of the embodiments disclosed in the present application are performed.

[0157] The payment authentication device provided by the present application adopts the payment authentication method in the above-mentioned embodiments, and can solve the technical problems of payment authentication. Compared with the prior art, the payment authentication device provided by the present application has the same beneficial effects as the payment authentication method provided by the above-mentioned embodiments, and other technical features in the payment authentication device are the same as the features disclosed in the previous embodiment method, which will not be repeated here.

[0158] It should be understood that portions of the application disclosed can be implemented in hardware, software, firmware, or combinations thereof. In the description of the embodiments above, specific features, structures, materials or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0159] The above description is merely illustrative of the application and is not intended to limit the scope of the application. Any modifications or equivalents of the application should be construed as falling within the scope of the application. The scope of the application should be determined by the appended claims.

[0160] The application provides a computer readable storage medium having stored thereon computer readable program instructions (i.e., a computer program) for performing the payment authentication method in the above embodiments.

[0161] The computer readable storage medium provided by the application may, for example, be a U disk, but is not limited to an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, system, or device, or any combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection having one or more conductive wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present embodiment, the computer readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer readable storage medium can be transmitted in any suitable medium, including but not limited to electrical wire, optical cable, RF (Radio Frequency), etc., or any suitable combination of the above.

[0162] The above computer readable storage medium can be included in the payment authentication device; or can exist separately and not be assembled into the payment authentication device.

[0163] The above computer readable storage medium carries one or more programs, which, when executed by the payment authentication device, cause the payment authentication device to perform payment authentication.

[0164] Computer program code for carrying out operations of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0165] The flow diagrams and the block diagrams in the drawings are illustrations of architectures, functionalities, and operations of possible implementations of systems, methods, and computer program products according to various embodiments of the present application. In this regard, each block in the flow diagrams or block diagrams can represent a module, a segment, or a portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that in some alternative implementations, the functions noted in the block can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or the blocks may

[0166] The modules involved in the embodiments of the present application can be implemented in the form of software or in the form of hardware. In some cases, the name of the module does not constitute a limitation on the module itself.

[0167] The readable storage medium provided by the present application is a computer readable storage medium, which stores computer readable program instructions (i.e. computer programs) for performing the above-mentioned payment authentication method, and can solve the technical problem of how to safely and conveniently perform payment authentication and improve the reliability of payment authentication. Compared with the prior art, the computer readable storage medium provided by the present application has the same beneficial effects as the payment authentication method provided by the above-mentioned embodiments, which will not be described here.

[0168] The application further provides a computer program product comprising a computer program which, when executed by a processor, implements the steps of the payment authentication method as described above.

[0169] The computer program product provided by the application can solve the technical problem of how to safely and conveniently perform payment authentication and improve the reliability of payment authentication. Compared with the prior art, the beneficial effects of the computer program product provided by the application are the same as those of the payment authentication method provided by the above-described embodiments, and are not described here.

[0170] The above only describes some embodiments of the application, and does not limit the patent scope of the application. Any equivalent structural transformation made by using the content of the specification and drawings, or direct / indirect application in other related technical fields under the technical concept of the application is included in the patent protection scope of the application.

Claims

1. A method of payment authentication, characterized by The payment authentication method comprises the following steps: According to the payment behavior information of the user, the behavior deviation is determined, and according to the transaction information of the user transaction, the transaction vulnerability is determined; According to the behavior deviation and the transaction vulnerability, the transaction risk index is determined, and whether the user transaction is a high-risk transaction is judged according to the transaction risk index; If yes, the user voiceprint is determined according to the audio segment sent by the user, and the payment authentication is carried out according to the voiceprint similarity between the user voiceprint and the base voiceprint.

2. The payment authentication method of claim 1, wherein, The step of determining the behavior deviation according to the payment behavior information of the user and determining the transaction vulnerability according to the transaction information of the user transaction, specifically comprises: According to the payment behavior information of the user, the transaction frequency deviation, the transaction amount deviation and the transaction frequency deviation are determined; According to the transaction frequency deviation, the transaction amount deviation and the transaction frequency deviation, the behavior deviation is determined; According to the transaction object information and the transaction content information in the transaction information of the user transaction, the transaction vulnerability is determined.

3. The payment authentication method of claim 2, wherein, The step of determining the transaction frequency deviation, the transaction amount deviation and the transaction frequency deviation according to the payment behavior information of the user, specifically comprises: According to the historical payment information of the user, the average daily transaction frequency, the average daily transaction amount and the average transaction interval are determined; The payment amount, the daily cumulative payment frequency, the single-day cumulative payment amount, the transaction time and the last transaction time in the payment behavior information of the user are obtained; According to the average daily transaction frequency, the daily cumulative payment frequency, the average daily transaction frequency and the transaction frequency deviation tolerance, the transaction frequency deviation is calculated; According to the average daily transaction amount, the daily cumulative payment amount, the payment amount and the transaction amount deviation tolerance, the transaction amount deviation is calculated; According to the transaction time, the last transaction time, the average transaction interval and the transaction frequency deviation tolerance, the transaction frequency deviation is calculated.

4. The payment authentication method of claim 1, wherein, The step of determining the user voiceprint according to the audio segment sent by the user and carrying out payment authentication according to the voiceprint similarity between the user voiceprint and the base voiceprint if yes, specifically comprises: If yes, the audio segment sent by the user is collected based on the randomly generated text sequence; According to the text sequence and the audio segment, the user voiceprint is determined, and the voiceprint authentication is carried out according to the voiceprint similarity between the user voiceprint and the base voiceprint, to obtain the voiceprint authentication result; Based on the authentication information of the user, the regular authentication is carried out, to obtain the regular authentication result; According to the voiceprint authentication result and the regular authentication result, the payment authentication is carried out.

5. The payment authentication method of claim 4, wherein, The step of determining the user voiceprint according to the text sequence and the audio segment, and carrying out voiceprint authentication according to the voiceprint similarity between the user voiceprint and the base voiceprint to obtain the voiceprint authentication result, specifically comprises: The audio segment is preprocessed to obtain a processed audio segment; Through a double-network structure model, the processed audio segment and the text sequence, voiceprint extraction is carried out to obtain the user voiceprint, the double-network structure model comprising a voiceprint extraction network and an audio synthesis network; According to the voiceprint similarity between the user voiceprint and the base voiceprint, the voiceprint authentication is carried out to obtain the voiceprint authentication result.

6. The payment authentication method of claim 5, wherein, The step of preprocessing the audio segment to obtain a processed audio segment specifically comprises: determining a background noise type according to a payment scenario of the user, and determining a frequency range corresponding to the background noise type; determining a notch center frequency according to the frequency range; filtering the background noise in the audio segment according to the notch center frequency, a notch width, and a discretized sampling time to obtain a processed audio segment.

7. A payment authentication apparatus characterized by comprising: The payment authentication device comprises: an information determination module configured to determine a behavior deviation degree according to payment behavior information of the user, and determine a transaction vulnerability according to transaction information of a transaction of the user; a transaction judgment module configured to determine a transaction risk index according to the behavior deviation degree and the transaction vulnerability, and determine whether the transaction of the user is a high-risk transaction according to the transaction risk index; a payment authentication module configured to determine a user voiceprint according to an audio segment sent by the user when the transaction of the user is a high-risk transaction, and perform payment authentication according to a voiceprint similarity between the user voiceprint and a base voiceprint.

8. A payment authentication device, characterized by The device comprises a memory, a processor, and a computer program stored on the memory and executable on the processor, and the computer program is configured to implement the steps of the payment authentication method according to any one of claims 1 to 6.

9. A storage medium, characterized by The storage medium is a computer readable storage medium, and the storage medium stores a computer program, and the computer program is executed by the processor to implement the steps of the payment authentication method according to any one of claims 1 to 6.

10. A computer program product, characterised in that, The computer program product comprises a computer program, and the computer program is executed by the processor to implement the steps of the payment authentication method according to any one of claims 1 to 6.

Images