Voiceprint input method and device, equipment and storage medium

The voiceprint recording method categorizes users and guides them to remotely record voiceprints, addressing compliance issues by ensuring secure and convenient voiceprint storage for all users, enhancing privacy and identity confirmation in financial applications.

CN119248220APending Publication Date: 2025-01-03PING AN TECH (SHENZHEN) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411238379.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-04
Publication Date
2025-01-03

AI Technical Summary

Technical Problem

Existing financial applications face difficulties in using voiceprint recognition for identity confirmation due to users who have not completed voiceprint storage during the account opening process, leading to compliance issues in outbound call scenarios.

Method used

A voiceprint recording method that categorizes users based on their voiceprint information status, guides first-category users to record voiceprints remotely through a terminal device, and uses asynchronous thread pooling for simultaneous frame and sound extraction to ensure secure and efficient voiceprint registration.

Benefits of technology

Enables seamless and secure voiceprint storage for all users, enhancing privacy protection and identity confirmation in financial applications, reducing the risk of identity theft and improving user convenience.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119248220A_ABST
    Figure CN119248220A_ABST
Patent Text Reader

Abstract

The invention discloses a voiceprint input method, a voiceprint input device, voiceprint input equipment and a storage medium, which are used for solving the problem of compliance of voiceprint input and recognition of an existing financial application program. The method comprises the following steps: when a trigger operation for starting an application program is detected, obtaining user identity information of the application program; determining a user category to which the user belongs according to the user identity information; if the user category to which the user belongs is a first category of users, loading a preset guide page to a display interface of the application program, responding to the trigger operation and opening the display interface of the guide page; if it is detected that the user triggers the guide page, displaying a video recording interface, and sending a video recording instruction to a server; and displaying character information preset by the video recording function on the video recording interface, recording a video that the user reads the character information, and sending the video to a server so as to extract voiceprint information of the user from the video. According to the invention, the convenience and safety of remotely completing voiceprint warehousing and recognition by the user through the terminal equipment can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of biometric identification technology, and in particular to a voiceprint recording method, device, equipment and storage medium. Background Art

[0002] With the rapid development of biometric recognition technology, voiceprint recognition technology is rapidly becoming popular in financial institutions such as securities companies and banks, especially in outbound call scenarios such as compliance follow-up visits and service notifications. Compared with existing identity authentication methods such as password reporting or security questions and answers that require users to reveal information during calls, the voiceprint recognition process is more convenient and less sensitive, and can better protect user privacy. In the securities industry, the voiceprint recognition step has been integrated into the new user account opening process, requiring new users to go to the business department in person and record videos to complete the dual-factor storage of facial features and voiceprint features, so that it is possible to use voiceprint recognition technology in outbound call services in the future. However, some existing users have completed the account opening process without voiceprint storage in the early stage, and some users "don't know, don't want, and can't" go to the business department to complete the voiceprint storage. Therefore, it has brought difficulties to the use of voiceprint recognition for identity confirmation in financial industry applications (APPs) and financial outbound call scenarios. Summary of the invention

[0003] The present application discloses a voiceprint entry method, device, equipment and storage medium to solve the compliance issues of voiceprint entry and recognition in existing financial applications.

[0004] In a first aspect, the present application provides a voiceprint recording method, which is applied to a terminal device, wherein the terminal device includes an application program, including:

[0005] When a trigger operation for starting the application is detected, obtaining user identity information of the application;

[0006] Determine, according to the user identity information, a user category to which the user corresponding to the user identity information belongs, the user category including a first category of users and a second category of users, wherein the first category of users is a user whose voiceprint information has not been recorded, and the second category of users is a user whose voiceprint information has been recorded;

[0007] If the user category to which the user corresponding to the user identity information belongs is the first category of user, loading a preset guide page into the display interface of the application, responding to the trigger operation and opening the display interface including the guide page, the guide page being used to guide the user to perform voiceprint recording;

[0008] If it is detected that the user triggers the guide page, display the video recording interface;

[0009] The text information preset by the video recording function is displayed on the video recording interface, a video of the user reading the text information is recorded, and the video is sent to a server so as to extract the voiceprint information of the user from the video.

[0010] In a second aspect, the present application provides a voiceprint recording method, which is applied to a server and includes:

[0011] Receiving a video recording instruction sent by a terminal device, and starting a video recording function in response to the video recording instruction, wherein the video recording instruction is generated when it is determined that the user category is the first category of users and the user triggers a guide page displayed on a display interface of a triggering application;

[0012] Acquire a video sent by the terminal device, where the video is obtained by recording the user reading the text information aloud;

[0013] Using an asynchronous thread pool to start a first thread and a second thread; wherein the first thread extracts frames from the video and sends the frames to a human identity verification module for human identity verification; the second thread extracts sound from the video recorded by the user and sends the sound to a voiceprint registration module for voiceprint registration;

[0014] If the result of the human-evidence verification is failure, repeating the step of extracting frames from the video and sending the frames to the human-evidence verification module for human-evidence verification;

[0015] If the number of repetitions in the first thread is greater than or equal to a preset threshold, or the result of the voiceprint registration in the second thread is a failure, the first thread and the second thread are terminated, the registration is cleared, and the result of the voiceprint recording failure is sent to the terminal device;

[0016] If the result of the identity verification is passed and the result of the voiceprint registration is successful, the result of the successful voiceprint registration is sent to the terminal device.

[0017] In a third aspect, the present application provides a voiceprint recording device, which is configured in a terminal device and includes:

[0018] an identity acquisition module, which acquires user identity information of the application program when a trigger operation for starting the application program is detected;

[0019] an identity classification module, which determines, according to the user identity information, a user category to which the user corresponding to the user identity information belongs, wherein the user category includes a first category of users and a second category of users, wherein the first category of users is a user whose voiceprint information has not been recorded, and the second category of users is a user whose voiceprint information has been recorded;

[0020] a guide loading module, which, if the user category to which the user corresponding to the user identity information belongs is the first category of user, loads a preset guide page to the display interface of the application, responds to the trigger operation and opens the display interface including the guide page, wherein the guide page is used to guide the user to perform voiceprint recording;

[0021] The interface display module displays the video recording interface if it is detected that the user triggers the guide page;

[0022] The video recording module displays the text information preset by the video recording function on the video recording interface, records the video of the user reading the text information, and sends the video to the server so as to extract the voiceprint information of the user from the video.

[0023] In a fourth aspect, the present application provides a voiceprint recording device, which is configured on a server and includes:

[0024] A video receiving module, used for acquiring a video sent by the terminal device, wherein the video is obtained by recording the user reading the text information aloud;

[0025] A dual-factor extraction module, configured to start a first thread and a second thread using an asynchronous thread pool; wherein the first thread extracts frames from the video and sends the frames to a person identification verification module for person identification verification; and the second thread extracts sound from the video recorded by the user and sends the sound to a voiceprint registration module for voiceprint registration;

[0026] A failure retry module, used for repeating the step of extracting frames from the video and sending the frames to the human identity verification module for human identity verification if the result of the human identity verification is failure;

[0027] A registration failure module, configured to terminate the first thread and the second thread if the number of repetitions of the first thread is greater than or equal to a preset threshold, or the result of the voiceprint registration of the second thread is a failure, and to clear the registration, and send the result of the voiceprint recording failure to the terminal device;

[0028] The registration success module sends the result of successful voiceprint registration to the terminal device if the result of the identity verification is passed and the result of the voiceprint registration is successful.

[0029] In a fifth aspect, the present application provides a computer, comprising a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein the memory stores a policy model, wherein when the computer program is executed by the processor, a voiceprint recording method as provided in any embodiment of the present application is implemented.

[0030] In a sixth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the processor enables the processor to implement a voiceprint recording method as provided in any embodiment of the present application.

[0031] The embodiments of the present application provide a voiceprint recording method, apparatus, device and storage medium, which obtains user identity information of an application when a trigger operation for opening an application is detected; determines the user category to which the user belongs based on the user identity information; if the user category to which the user belongs is a first-category user, loads a preset guide page to the display interface of the application, responds to the trigger operation and opens the display interface of the guide page; if it is detected that the user triggers the guide page, displays a video recording interface, and sends a video recording instruction to a server; displays text information preset by the video recording function on the video recording interface, records a video of the user reading the text information, and sends the video to the server so as to extract the user's voiceprint information from the video. On the one hand, users can complete voiceprint storage remotely through terminal devices, and complete dual-factor storage of face recognition and voiceprint recognition based on video recording, which improves convenience. On the other hand, by storing voiceprints, users can use voiceprint recognition to confirm their identity in financial industry applications (APPs) and financial outbound call scenarios. Compared with the existing identity authentication method that requires users to reveal information during calls, the voiceprint recognition process is more convenient and seamless, and can better protect user privacy, reducing the risk of user identity theft and impersonation, and improving the security of user accounts. Ultimately, it solves the compliance issues of voiceprint entry and recognition in existing financial applications.

[0032] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0034] Figure 1 This is a schematic flow chart of the steps of a voiceprint recording method provided in an embodiment of the present application;

[0035] Figure 2 is a schematic flow chart of steps of a method for loading a guide page provided in an embodiment of the present application;

[0036] Figure 3 is a schematic flow chart of the steps of another voiceprint recording method provided in an embodiment of the present application;

[0037] Figure 4 is a schematic flow chart of the steps of another voiceprint recording method provided by an embodiment of the present application;

[0038] Figure 5 It is a structural schematic diagram of a voiceprint recording device provided in an embodiment of the present application;

[0039] Figure 6 It is a structural schematic diagram of another voiceprint recording device provided in an embodiment of the present application;

[0040] Figure 7 It is a schematic block diagram of the structure of a computer device provided in an embodiment of the present application.

[0041] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0043] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0044] It should be understood that the terms used in this application specification are only for the purpose of describing specific embodiments and are not intended to limit the application. As used in this application specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0045] It should be understood that, in order to facilitate the clear description of the technical solutions of the embodiments of the present application, in the embodiments of the present application, words such as "first" and "second" are used to distinguish between identical or similar items with substantially identical functions and effects. For example, the first data and the second data are only used to distinguish different data, and their sequence is not limited. Those skilled in the art can understand that words such as "first" and "second" do not limit the quantity and execution order, and words such as "first" and "second" do not necessarily limit them to be different.

[0046] It should be further understood that the term “and / or” used in the specification and appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0047] To facilitate understanding of the embodiments of the present application, some words involved in the embodiments of the present application are briefly explained below.

[0048] 1. Two-factor: Two-factor authentication is a security measure that requires users to provide two different forms of identity verification to improve account security. Two-factor authentication provides a higher level of security protection than single-factor authentication (such as using only facial recognition) because even if an attacker obtains the user's facial features, they still need the user's other biometric features for identity verification. In the financial field, two factors usually include information that the user knows (such as a password) and the user's biometric features, or two different biometric features of the user, such as face and voice.

[0049] 2. Liveness detection: Liveness detection technology is a technology used to verify and distinguish between real living things and non-living things presented through photos, videos, 3D molds, etc. This technology is widely used in finance, payment, public security, social media and other fields to improve the security of identity authentication. The main methods of liveness detection include: cooperative detection: the system requires the user to complete specified actions, such as blinking, opening the mouth, nodding or shaking the head, to prove that it is a live body; RGB detection: by identifying subtle features on the image, such as moiré or imaging deformity, to distinguish whether it is a real live body, suitable for preventing deception of pictures or video screenshots; 3D structured light detection: using the principle of three-dimensional imaging, the depth image is constructed by reflecting light from the surface of the face, effectively resisting attacks from two-dimensional images and three-dimensional molds.

[0050] 3. Asynchronous thread pool: Asynchronous thread pool is a concurrent programming technology that allows multiple threads to execute tasks asynchronously without blocking the execution of the main thread or other threads. This technology is particularly suitable for scenarios where multiple tasks need to be processed simultaneously and these tasks may require different execution times. For example, using asynchronous thread pools can not only allow the two tasks of identity verification and voiceprint registration to be carried out simultaneously, but also allow the APP to not wait for the storage results. In this way, voiceprint entry will not block the execution process of the entire application, and the storage results can be prompted in time on the APP.

[0051] 4. Identity verification: Identity verification is an identity verification technology commonly used for APP user authentication. It includes using the authoritative identity verification module data or interface of the public security department to verify identity information to ensure the authenticity and validity of the user's identity. It usually involves comparing the user's name, ID number and / or facial image and other personal information with the records of the public security system to confirm the consistency of the user's identity.

[0052] 5. Boot page: A specific advertising guide interface, pop-up window or component displayed to users after they launch the APP. This design is sometimes called "Splash Ad" or "Launch Screen Ad". The boot page is also a marketing strategy that uses the initial interface when users open the APP to display advertising content, thereby quickly attracting users' attention.

[0053] 6. Excessive concentration: Excessive concentration means that the market value of a single security or securities portfolio held by an investor in his account exceeds the ratio of the total assets in his account to the standard set by the regulator or the brokerage firm. In order to avoid excessive concentration of market risks on a single security or securities portfolio and lead to potential systemic risks, the regulator or brokerage firm will proactively contact investors through telephone, Internet or other communication means to warn them of the risks.

[0054] 7. Outbound call scenario: Securities outbound call scenarios usually refer to the behavior of securities companies or financial institutions actively contacting potential or existing users through telephone, Internet or other communication means to provide them with securities market information, investment advice, risk warnings, anti-money laundering, feedback improvement and other services. Outbound call scenarios are an important user service scenario in the financial field, involving user communication, marketing promotion and business compliance.

[0055] Taking concentration exceeding the limit as an example, the securities business system will automatically call the mobile phone number reserved by the user for users whose concentration exceeds the limit. The system will automatically make an AI outbound call to the user according to the preset script, and then the system background will automatically retrieve the user's voiceprint for comparison. For those who are not the user himself, the securities company will further verify the user's position or other information. If it is still determined that it is not the user himself, a text message will be sent to remind the user of the account security information, and the business department will be required to further confirm whether the account operation is performed by the user himself.

[0056] 8. Alternative guidance projects: incentives for users who record voiceprints, including gift certificates, points, discounts, lucky draw opportunities, cash red envelopes, memberships, event tickets, data permissions and / or virtual costumes, etc.

[0057] 9. Guidance strategy: The guidance strategy consists of at least one alternative guidance item, which can be determined by the server or terminal device based on the user's user feature information, or allocated by the background.

[0058] 10. Large voice model: A pre-trained voiceprint recognition model that provides voiceprint registration, voiceprint recognition, and clean registration services. Currently, open source models include BiGE2E, 3DCNNAM, etc.

[0059] 11. Thread-safe queue: A specially designed queue. The internal synchronization mechanism of this queue ensures that when multiple threads attempt to enqueue or dequeue operations, only one thread can modify the queue at a time, thereby ensuring the integrity of the queue and the consistency of the data, thereby allowing multiple threads to access elements in the queue simultaneously without causing data corruption or race conditions.

[0060] 12.OpenCV: An open source computer vision and machine learning algorithm library that provides more than 2,500 common image processing and computer vision algorithms, covering advanced functions from basic image operations to complex 3D reconstruction, object recognition, and facial recognition, and supports C language and Python interfaces.

[0061] 13.Librosa: A Python package that focuses on audio and music signal processing and provides implementations of a variety of functions commonly used in audio processing.

[0062] 14.MySQL database: A relational database management system (RDBMS) widely used in web applications and enterprise-level database solutions. MySQL database operates based on structured query language (SQL) and can store, retrieve and manage large amounts of data.

[0063] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0064] In outbound call scenarios such as compliance follow-up and service notifications in the financial industry, existing identity authentication methods such as password reporting and security questions and answers require users to disclose sensitive information during calls, which is more likely to be known by others or illegal applications during conversations, increasing the risk of privacy leakage. In contrast, voiceprint recognition technology, as an advanced identity authentication method, provides a more convenient and concealed authentication process, especially in noisy environments or outbound call services, which can more effectively protect user privacy.

[0065] In the securities industry, the current compliance management system has been adjusted accordingly for the application of voiceprint recognition technology. In particular, the voiceprint recognition step has been integrated into the account opening process for new users, requiring new users to visit business outlets in person and complete the storage of two-factor identity information including facial and voiceprint features, so as to ensure that voiceprint recognition technology can be safely applied in future outbound call services. However, there are also some existing users whose early account opening processes no longer include voiceprint storage, and some users "do not know, are unwilling, and cannot" go to the business department to complete the voiceprint storage. This has brought difficulties to the use of voiceprint recognition for identity confirmation in financial industry applications (APPs) and financial outbound call scenarios.

[0066] Completing the collection and storage of user voiceprint information is an important step in the subsequent use of voiceprint recognition technology and reducing the risk of data leakage. However, since some users "do not know, are unwilling, and are unable" to go to the business department to complete the voiceprint storage, it has caused difficulties in the use of voiceprint recognition for identity authentication in financial industry applications (APPs) and financial outbound call scenarios.

[0067] To solve the above problems, the present application embodiment proposes a voiceprint recording method, which is applied to a terminal device, and the terminal device includes an application. Figure 1 , Figure 1 This is a schematic flowchart of the steps of the first voiceprint recording method provided in an embodiment of the present application.

[0068] like Figure 1 As shown, the voiceprint recording method specifically includes steps S101 to S106.

[0069] S101. When a trigger operation for starting an application is detected, obtain user identity information of the application.

[0070] It should be noted that the user identity information of the application can be obtained from the terminal or the server.

[0071] Exemplarily, when the terminal device detects a trigger operation for starting an application, it sends an instruction for obtaining user identity information to the server, and obtains the user identity information fed back by the server.

[0072] Exemplarily, when the terminal device detects a trigger operation for starting an application, the terminal device obtains the user identity information stored on the terminal device.

[0073] Specifically, the trigger operation for opening the application program may be a variety of trigger operations, such as a slide trigger, a click trigger, a voice trigger, a gesture trigger, a timer trigger, and the like.

[0074] Specifically, the user identity information of the application may include account status information, such as whether the user has completed the entry of voiceprint information, whether the user is a new user or an existing user, the account activity level, whether the account is sealed, etc.

[0075] Furthermore, the user identity information of the application may also include basic user information, such as the user's name, phone number, account name, user ID, etc.

[0076] Furthermore, the user identity information of the application may also include user transaction information, such as the user's transaction record, asset status, compliance status, credit status, risk preference, etc.

[0077] By using the obtained user identity information to distinguish the user category, it is possible to ensure that the subsequent loading of the guide page is accurate and effective, and will not affect the software use of other users who have recorded voiceprint information.

[0078] By utilizing the acquired user identity information, it is possible to ensure that the generation of subsequent guidance strategies can better meet the actual needs of users and improve the user's cooperation and enthusiasm for voiceprint recording.

[0079] S102. Determine, based on the user identity information, a user category to which the user corresponding to the user identity information belongs, where the user categories include first category users and second category users, wherein the first category users are users whose voiceprint information has not been recorded, and the second category users are users whose voiceprint information has been recorded.

[0080] Exemplarily, based on the account status information in the user identity information, it is determined that the user is a first type of user, that is, a user who has not entered voiceprint information.

[0081] Through screening, it is possible to avoid pushing or loading the guide page to the user account that has already entered the voiceprint information, which would cause a waste of resources.

[0082] Exemplarily, screening is performed based on the acquired user identity information to obtain existing users who meet the requirements of active accounts, compliant behaviors, and good credit, and then a secondary screening is performed based on the user identity information to determine the user category to which the user belongs.

[0083] Through secondary screening, we can avoid pushing or loading guide pages to sealed user accounts, avoid compliance risks, and reduce resource waste.

[0084] It should be noted that if the user category to which the user belongs is the first category of users, steps S103 to S105 are executed; if the user category to which the user belongs is the second category of users, step S106 is executed.

[0085] S103. If the user category to which the user corresponding to the user identity information belongs is the first category of user, a preset guide page is loaded into the display interface of the application, and a display interface including the guide page is opened in response to a trigger operation, where the guide page is used to guide the user to record a voiceprint.

[0086] Specifically, the guide page may display a guide message, a component that helps the user to record a video, and words and pictures of the optional guide items. The component that helps the user to record a video may include, for example, a start or end button, a video recording button, and / or a jump interface of a local video recording application.

[0087] For example, if the user category to which the user corresponding to the user identity information belongs is the first category of user, the preset guide page is loaded into the display interface of the application. The guide page displays guidance words such as "Enter your voiceprint to ensure account security" and displays words related to alternative guide items such as "Enter your voiceprint to win advanced data permissions". By displaying the words and pictures of the alternative guide items on the guide page, the user is guided and motivated, which can enhance the user's enthusiasm and cooperation in participating in voiceprint entry, and help reduce the user's resistance or impatience to voiceprint collection.

[0088] S104: If it is detected that the user triggers the guide page, a video recording interface is displayed.

[0089] For example, a jump button for calling the user's local video recording application is displayed on the guide page. If it is detected that the user triggers the guide page, a pop-up window requesting authorization to call the camera is displayed. If it is detected that the user agrees to authorize the camera call, the video recording interface is displayed.

[0090] S105: Displaying text information preset by the video recording function on the video recording interface, recording a video of the user reading the text information, and sending the video to a server so as to extract the user's voiceprint information from the video.

[0091] The text information preset by the video recording function can be displayed on the video recording interface. The text information can be Chinese characters, letters or numbers, or a combination of the three. By recording a video of a user reading a text message, the user's facial information and voiceprint information can be retained in the same video, avoiding the waste of terminal equipment and server resources caused by uploading facial information and voiceprint information twice. In some embodiments, based on the matching relationship between the dynamic facial information and the dynamic voice information in the video of reading the text message, deep learning technology can be used to verify the authenticity of the video and ensure the security of the user account.

[0092] S106: If the user category to which the user corresponding to the user identity information belongs is a second category user, opening a display interface of the application in response to a triggering operation.

[0093] Exemplarily, if the user category to which the user corresponding to the user identity information belongs is a second category user, when a trigger operation for opening the application is detected, the display interface of the application is opened. It should be noted that the difference between this display interface and the display interface including the guide page in step S103 is that the display interface in step S106 may be a display interface without the guide page, and the user can access and use the standard functions provided by the application through the display interface in step S106.

[0094] In some embodiments, see Figure 2 , Figure 21 is a schematic flow chart of the steps of a method for loading a guide page provided by an embodiment of the present application. Figure 2 As shown, the method for loading the guide page specifically includes steps S103a to S103d.

[0095] S103a. Obtain a guidance strategy corresponding to the user, where the guidance strategy is determined according to user characteristic information of the user.

[0096] Exemplarily, the terminal device sends an acquisition instruction for acquiring a guidance strategy to the server, so that the server feeds back a guidance strategy corresponding to the user according to the acquisition instruction. The server determines at least one corresponding alternative guidance item according to the user characteristic information of the user, and determines the guidance strategy corresponding to the user according to the at least one alternative guidance item, the user characteristic information includes login frequency, activity level, service level, transaction frequency, geographic location, service items and / or risk preference, and the alternative guidance items include gift certificates, points, discounts, lottery opportunities, cash red envelopes, memberships, event tickets, data permissions and / or virtual costumes.

[0097] Exemplarily, a guidance strategy corresponding to the user is obtained, and the guidance strategy includes three alternative guidance items, such as food and daily necessities exchange coupons, grain, oil and condiment exchange coupons, and physical examination coupons of a certain e-commerce platform. The guidance strategy is determined based on the user's user feature information, such as the number of daily logins greater than or equal to 10.

[0098] For example, based on user feature information, such as the number of logins per trading day being stable at three to five times for a long time, a guidance strategy corresponding to the user is determined. The guidance strategy includes an alternative guidance item, such as points of a certain airline. By generating multiple alternative guidance items, the utilization efficiency of existing resources is improved, the needs or interests of users are adapted as much as possible, and the user's willingness to participate in voiceprint recording is increased.

[0099] S103b: Generate at least one candidate guide page according to the guide strategy.

[0100] Exemplarily, in step S103a, if the obtained guidance strategy includes three alternative guidance items, a guidance page template is designed using a template engine or component library; and three alternative guidance pages are generated using the guidance page template according to the patterns and words of the three alternative guidance items.

[0101] S103c: Select one from at least one candidate guide page as a guide page according to a preset period, where the preset period includes a preset duration or a number of login times.

[0102] Specifically, the preset period can be calculated in units of natural days, trading days, hours or login times. For example, if the preset period is 1 trading day and three alternative guide pages are generated in step S103a, the terminal device can randomly select one of the three guide pages as the guide page of the day on the next trading day. By generating multiple guide pages and automatically switching them, it is possible to improve the utilization efficiency of existing resources and adapt to user interests or needs as much as possible, so that customers can choose the alternative guide items that best meet their needs instead of obtaining all the alternative guide items, thereby achieving the effect of saving resources.

[0103] S103d, switching the alternative guide pages to obtain the guide page based on time or the number of logins.

[0104] For example, if the preset period is ten logins and three candidate boot pages are generated in step S103a, the terminal device can accumulate the count after each user logs in. When the user logs in ten times, a different candidate boot page is selected from the three candidate boot pages in order as the boot page.

[0105] In some embodiments, see Figure 3 , Figure 3 is a schematic flow chart of the steps of another voiceprint recording method provided in an embodiment of the present application. Figure 1 The method shown differs in that Figure 3 The method shown applies to the server.

[0106] like Figure 3 As shown, the voiceprint recording method specifically includes steps S201 to S205.

[0107] S201. Obtain the video sent by the terminal device.

[0108] The video sent by the terminal device may be obtained by recording a user reading a text message.

[0109] S202, using the asynchronous thread pool to start the first thread and the second thread; wherein the first thread extracts frames from the video and sends the frames to the identity verification module for identity verification; the second thread extracts the sound from the video recorded by the user and sends it to the voiceprint registration module for voiceprint registration.

[0110] It should be noted that this application does not limit the tools for video processing, voiceprint extraction, voiceprint registration, and clean-up registration. The following only uses OpenCV as a video processing tool, librosa as a voiceprint extraction tool, and a pre-trained large human voice model as a tool for voiceprint registration and clean-up registration as an example.

[0111] Specifically, the voiceprint registration module can be used to achieve: using the API interface provided by the voiceprint registration module to convert voice data to obtain voiceprint features; saving the features to the voiceprint library, and feeding back the results of the voiceprint registration; among which, voiceprint registration, cleanup registration and other services are provided by the relevant interfaces of the pre-trained human voice model.

[0112] Specifically, the first thread and the second thread can be run on the same server, and the server runs a voiceprint service module, and the voiceprint service module is used to implement Figure 3 The voiceprint recording method shown.

[0113] Furthermore, the voiceprint registration module may be run on a server that runs the voiceprint service module, or may be run on another server.

[0114] Exemplarily, an asynchronous thread pool is used to start the first thread and the second thread; wherein the first thread is used to implement: using OpenCV to decode the video and obtain a video stream; extracting key frames from the video stream using preset conditions, for example, extracting one frame every few frames or extracting one frame when a face is detected, to obtain at least one key frame; performing pre-processing such as scaling, cropping, and format conversion on the key frames to adapt to the input requirements of the identity verification module to obtain processed key frames; sending the processed key frames to the identity verification module through a thread-safe queue for identity verification.

[0115] Exemplarily, an asynchronous thread pool is used to start the first thread and the second thread; wherein the second thread is used to implement: using the audio processing library librosa to decode the video and extract the audio stream; performing pre-processing operations such as noise reduction, resampling, and segmentation on the audio stream to meet the input requirements of the voiceprint registration module to obtain processed audio data; sending the processed audio data to the voiceprint registration module through a thread-safe queue to perform voiceprint registration.

[0116] S203: If the result of the person-witness verification is failure, repeatedly extract frames from the video and send the frames to the person-witness verification module for the step of person-witness verification.

[0117] It should be noted that the input of the witness verification module can be a video frame. If the first witness verification fails, other frames can be re-extracted for verification instead of re-recording the video, which improves the convenience of the user in recording the voiceprint. Exemplarily, if the result of the witness verification is failure, the first thread is used to implement: re-extracting frames in the video and sending the frames to the witness verification module for witness verification.

[0118] S204: If the number of repetitions of the first thread is greater than or equal to a preset threshold, or the result of voiceprint registration of the second thread is failure, terminate the first thread and the second thread, clear the registration, and send the result of voiceprint recording failure to the terminal device.

[0119] It should be noted that the input of the voiceprint registration module can be the entire audio data of the text information read in the video obtained when the user reads the text information. If the first voiceprint registration fails, other audio data cannot be re-extracted for verification. It is necessary to terminate the first thread and the second thread and clean up the registration, and send the result of the voiceprint recording failure to the terminal device so that the terminal device can guide the user to re-record the video.

[0120] Exemplarily, if the result of the first thread identity verification is passed, but the result of the second thread voiceprint registration is failed, the first thread and the second thread are terminated, the registration is cleared, and the result of the voiceprint recording failure is sent to the terminal device.

[0121] Exemplarily, if the result of voiceprint registration of the second thread is successful, but the number of repetitions of the first thread is greater than or equal to a preset threshold, the first thread and the second thread are terminated, the registration is cleared, and the result of voiceprint recording failure is sent to the terminal device.

[0122] S205. If the result of the identity verification is passed and the result of the voiceprint registration is successful, the result of the successful voiceprint registration is sent to the terminal device.

[0123] Exemplarily, the voiceprint registration result or the person-verification result can be a number and a text describing the number, such as 100000 success or 207001 quality inspection failed. Exemplarily, the voiceprint registration result and the person-verification result can be text, such as success, quality inspection failed, insufficient valid frames. Exemplarily, the voiceprint registration result or the person-verification result can be a number, such as 100000, 207001, etc.

[0124] In some embodiments, see Figure 4 , Figure 4 This is a flowchart showing the steps of another voiceprint recording method provided in an embodiment of the present application.

[0125] like Figure 4 As shown, the voiceprint recording method specifically includes steps S301 to S317.

[0126] S301: When a trigger operation for starting an application is detected, obtain user identity information of the application.

[0127] The user identity information of the application program may be obtained from the terminal or the server.

[0128] Specifically, the user identity information may include the user category to which the user belongs, wherein the first category of users are users whose voiceprint information has not been recorded, and the second category of users are users whose voiceprint information has been recorded.

[0129] S302: If the user category to which the user corresponding to the user identity information belongs is a first category user, a preset guide page is loaded into a display interface of the application program, and the display interface including the guide page is opened in response to a trigger operation.

[0130] Specifically, before step S302, if the user category to which the user corresponding to the user identity information belongs is the second category of user, step S106 is executed, and the display interface of the application is opened in response to the trigger operation. The display interface is different from the display interface including the guide page in S302 in that the display interface may be a display interface that does not include a guide page, and the user can access and use the standard functions provided by the application through the display interface of step S106.

[0131] S303: If it is detected that the user triggers the guide page, the video recording interface is displayed and the liveness detection function is enabled.

[0132] The trigger operation may be a variety of trigger operations, such as sliding trigger, clicking trigger, voice trigger, gesture trigger, timing trigger, etc.

[0133] Exemplarily, if it is detected that the user triggers the guide page, the video recording interface is displayed, and the liveness detection function is turned on; wherein the video recording interface includes at least: a video display area for displaying a real-time video stream so that the user can see his or her real-time picture; video recording controls, such as recording start and end buttons, for allowing the user to control the recording process; a liveness detection light source area for serving as a color-changing light source so that the terminal device can capture a facial image corresponding to the face of the user illuminated by the color-changing light; and a text display area for displaying preset text information so that the user can read it aloud when recording the video.

[0134] Through the principle of 3D structured light, it can resist attacks from two-dimensional images and three-dimensional molds and ensure the security of users' accounts.

[0135] S304: Control the video recording interface to change color based on the liveness detection function, and collect a face image corresponding to the face of the user illuminated by the color change.

[0136] Exemplarily, the terminal device controls the video recording interface or the liveness detection light source area of ​​the video recording interface to change color based on the liveness detection function, and collects a facial image corresponding to the face of the user illuminated by the color change.

[0137] S305: Send the face image to the server so that the server can perform liveness detection.

[0138] Exemplarily, the face image is preprocessed by segmentation, cropping, scaling, resampling, etc. to obtain a preprocessed face image; the preprocessed face image is sent to a server so that the server performs liveness detection.

[0139] S306: Receive the face image sent by the terminal device, perform liveness detection, and obtain and feed back the detection result.

[0140] Exemplarily, a facial image sent by a terminal device is received, and the phase change of the pixel points is calculated using the structured light pattern reflected by the user's facial skin collected in the facial image, and a three-dimensional model of the face is reconstructed; the obtained three-dimensional face model is used to further perform skin feature analysis to obtain a liveness detection result; and the liveness detection result is fed back to the terminal device.

[0141] Furthermore, the server may feed back the liveness detection result to the terminal device through step S306. If the detection result is liveness, preset text information may be generated and fed back. The preset text information is used to be displayed in the text display area of ​​the video recording interface in step S309.

[0142] S307: Obtain the detection result fed back by the server.

[0143] Specifically, the detection result fed back by the server may be a live object or a non-live object.

[0144] Exemplarily, after the terminal device obtains the detection result fed back by the server, it can also be used to achieve: if the detection result is a live body, obtain the preset text information generated and fed back by the server, and the preset text information is used to display in the text display area of ​​the video recording interface in step S309.

[0145] S308: If the detection result fed back by the server is that the body is alive, the step of displaying text information preset by the video recording function is executed.

[0146] For example, if the detection result fed back by the server is that the body is alive, the text display area of ​​the video recording interface is used to obtain and display preset text information such as "Please read aloud the numbers on the screen: 1234".

[0147] S309: displaying text information preset by the video recording function on the video recording interface, recording a video of the user reading the text information, and sending the video to a server so as to extract the user's voiceprint information from the video.

[0148] In one embodiment, the terminal device uses the text display area of ​​the video recording interface to display preset text information, and records a video of the user reading the text information; the video is sent to the server to extract the user's voiceprint information from the video.

[0149] It should be noted that, in step S309, before the terminal device sends the video to the server, it may pre-process the video recorded when the user reads the text information, so as to adapt to the input requirements of the server.

[0150] In another embodiment, the terminal device uses the text display area of ​​the video recording interface to display preset text information, and records a video of the user reading the text information; uses OpenCV to perform pre-processing such as cropping, segmentation, and resampling on the video of the user reading the text information to obtain a processed video; and sends the pre-processed video to the server to extract the user's voiceprint information from the video.

[0151] The server for receiving the video of the user reading the text message can run a voiceprint service module, which is used to implement the voiceprint service module. Figure 3 The voiceprint recording method shown, Figure 4 The corresponding steps include: step S311, step S312, step S313, step S315 and step S316.

[0152] Furthermore, the server for receiving the video of the user reading the text information and the server for receiving the face image sent by the terminal device and performing liveness detection in step S306 may be the same server or different servers.

[0153] S310, start the scheduled polling task.

[0154] Exemplarily, before the voiceprint registration result is returned, the terminal device repeatedly queries the server for the voiceprint registration result according to a preset time interval; if no result is received when the preset repetition threshold or the preset polling duration threshold is reached, the terminal device terminates the scheduled polling task and displays words such as "voiceprint registration failed" or "server not responding" on the display interface of the application; on the contrary, if the registration result is received before the repetition threshold or the polling duration threshold is reached, the terminal device displays the corresponding result information on the display interface of the application.

[0155] S311. Obtain the video sent by the terminal device.

[0156] Exemplarily, the server obtains the video sent by the terminal device, wherein the video sent by the terminal device is the video recorded in step S309 when the user reads the text information.

[0157] It should be noted that the server can run a voiceprint service module, which is used to implement the voiceprint service module. Figure 3 The voiceprint recording method shown, Figure 4The corresponding steps include: step S311, step S312, step S313, step S315 and step S316.

[0158] S312, using the asynchronous thread pool to start the first thread and the second thread; wherein the first thread extracts frames from the video and sends the frames to the identity verification module for identity verification; the second thread extracts the sound from the video recorded by the user and sends it to the voiceprint registration module for voiceprint registration.

[0159] Exemplarily, the server uses an asynchronous thread pool to start a first thread and a second thread; wherein the first thread extracts frames from the video and sends the frames to a person identification verification module for person identification verification; the second thread extracts sound from the video recorded by the user and sends it to a voiceprint registration module for voiceprint registration.

[0160] It should be noted that the identity verification module and the voiceprint registration module can run on the same server as the voiceprint service module, or on other servers.

[0161] Exemplarily, a server running a voiceprint service module uses an asynchronous thread pool to start a first thread and a second thread; wherein the first thread extracts frames from a video and sends the frames to a server running a person verification module for person verification; the second thread extracts sound from a video recorded by a user and sends it to a server running a voiceprint registration module for voiceprint registration.

[0162] S313: If the result of the person-witness verification is failure, repeatedly extract frames from the video and send the frames to the person-witness verification module for the step of person-witness verification.

[0163] In some embodiments, the voiceprint service module server sends information to the terminal device. If the result of the identity verification is passed and the result of the voiceprint registration is successful, the result of the successful voiceprint entry is sent to the terminal device.

[0164] S314: Send a request to obtain the voiceprint recording result.

[0165] Exemplarily, after starting the scheduled polling task in step S310, the terminal device repeatedly sends a voiceprint recording result acquisition request to the server according to a preset time interval to query the voiceprint recording result.

[0166] S315: If the number of repetitions of the first thread is greater than or equal to a preset threshold, or the result of the voiceprint registration of the second thread is a failure, terminate the first thread and the second thread, clear the registration, and send the result of the voiceprint recording failure to the terminal device.

[0167] Exemplarily, the server obtains the number of repetitions of the first thread and the result of the voiceprint registration of the second thread. If the number of repetitions of the first thread is greater than or equal to a preset threshold, or the result of the voiceprint registration of the second thread is a failure, the first thread and the second thread are terminated, and the registration is cleared, and the result of the voiceprint recording failure is sent to the terminal device, so that the terminal device redirects the user to record the voiceprint. By clearing the registration in time, the waste of concurrent capacity resources of the server can be reduced, and the maintainability and robustness of the system can be improved.

[0168] S316. If the result of the identity verification is passed and the result of the voiceprint registration is successful, the result of the successful voiceprint registration is sent to the terminal device.

[0169] Exemplarily, the server obtains the result of the first thread identity verification and the result of the second thread voiceprint registration. If the result of the identity verification is passed and the result of the voiceprint registration is successful, the result of successful voiceprint recording is sent to the terminal device.

[0170] Furthermore, after step S315 or step S316, the server can also be used to: save the voiceprint recording result to the database for subsequent query; update the user identity information on the server and / or terminal device according to the voiceprint recording result, for example, update the user's unfinished voiceprint recording to the user's completed voiceprint recording. The database used to save the voiceprint recording result can be a MySQL database.

[0171] S317: Receive the voiceprint recording result sent by the server, and display the corresponding result words or patterns on the display interface of the application.

[0172] Exemplarily, the terminal device receives the result of the voiceprint entry sent by the server; if the voiceprint entry result is successful, the words or pictures such as "Voiceprint entry is successful, please receive the reward" are displayed; if the voiceprint entry result is failed, the words or pictures such as "Voiceprint entry failed, please re-enter" are displayed, and a display interface including a guide page is opened.

[0173] The embodiments of the present application provide a voiceprint entry method, device, equipment and storage medium. When the application is triggered, the terminal device obtains the user identity information and determines whether the voiceprint has been entered. The user who has not entered the voiceprint will load the guide page, and after the trigger, the video recording interface will be displayed, and the reading video will be recorded and uploaded to the server. The server uses an asynchronous thread pool to start the identity verification and voiceprint registration. If the identity verification fails, it can be repeated until the preset threshold is reached; if the identity verification and voiceprint registration are both successful, the terminal device receives feedback that the voiceprint entry is successful, etc. First, the user can complete the voiceprint entry remotely through the terminal device. Second, the asynchronous thread pool is used for identity verification and voiceprint registration to improve the system operation efficiency and server resource utilization. Third, the user feature information is used to determine the guidance strategy, which increases the user's willingness to enter the voiceprint. Finally, the compliance issues of voiceprint entry and recognition of existing financial applications are solved.

[0174] like Figure 5 As shown, Figure 5 1 is a schematic diagram of the structure of a voiceprint recording device provided in an embodiment of the present application, wherein the voiceprint recording device is used to execute the aforementioned voiceprint recording method. The voiceprint recording device can be configured in a terminal or a server.

[0175] like Figure 5 As shown, the voiceprint recording device 100 includes an identity acquisition module 101, an identity classification module 102, a boot loading module 103, an interface display module 104 and a video recording module 105.

[0176] The identity acquisition module 101 is used to acquire the user identity information of the application when a trigger operation for opening the application is detected;

[0177] The identity classification module 102 is used to determine the user category to which the user corresponding to the user identity information belongs according to the user identity information, where the user category includes a first category of users and a second category of users, wherein the first category of users are users who have not recorded voiceprint information, and the second category of users are users who have recorded voiceprint information;

[0178] A guide loading module 103, configured to load a preset guide page to the display interface of the application program if the user category to which the user corresponding to the user identity information belongs is a first category user, and respond to a triggering operation and open the display interface of the guide page, wherein the guide page is used to guide the user to perform voiceprint recording;

[0179] The interface display module 104 is used to display the video recording interface if it is detected that the user triggers the guide page, and send a video recording instruction to the server, so that the server responds to the video recording instruction to start the video recording function;

[0180] The video recording module 105 is used to display the text information preset by the video recording function on the video recording interface, record the video of the user reading the text information, and send the video to the server so as to extract the user's voiceprint information from the video.

[0181] In some embodiments, Figure 6 As shown, Figure 6 1 is a schematic diagram of the structure of a voiceprint recording device provided in an embodiment of the present application, wherein the voiceprint recording device is used to execute the aforementioned voiceprint recording method. The voiceprint recording device can be configured in a terminal or a server.

[0182] like Figure 6 As shown, the voiceprint recording device 200 includes a video receiving module 201, a dual-factor extraction module 202, a failure retry module 203, a registration failure module 204 and a registration success module 205.

[0183] The video receiving module 201 is used to obtain the video sent by the terminal device, where the video is obtained by recording the user reading the text information;

[0184] The dual-factor extraction module 202 is used to start the first thread and the second thread using the asynchronous thread pool; wherein the first thread extracts frames from the video and sends the frames to the identity verification module for identity verification; the second thread extracts the sound from the video recorded by the user and sends it to the voiceprint registration module for voiceprint registration;

[0185] Failure retry module 203, used for repeatedly extracting frames from the video and sending the frames to the human evidence verification module for human evidence verification if the result of the human evidence verification is failure;

[0186] The registration failure module 204 is used to terminate the first thread and the second thread if the number of repetitions of the first thread is greater than or equal to a preset threshold, or the result of the voiceprint registration of the second thread is a failure, and to clear the registration, and send the result of the voiceprint recording failure to the terminal device;

[0187] The registration success module 205 is used to send the result of successful voiceprint registration to the terminal device if the result of the identity verification is passed and the result of the voiceprint registration is successful.

[0188] It should be noted that those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described device and each module can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0189] The above-mentioned device can be implemented in the form of a computer program. Figure 7 Runs on the computer device shown.

[0190] See also Figure 7 , Figure 7 is a schematic block diagram of a computer device provided in an embodiment of the present application. The computer device may be a server or a terminal device. Figure 7 The computer device includes a processor, a memory and a network interface connected through a system bus, wherein the memory may include a non-volatile storage medium and an internal memory.

[0191] The non-volatile storage medium can store an operating system and a computer program. The computer program includes program instructions, and when the program instructions are executed, the processor can execute any voiceprint recording method.

[0192] The processor is used to provide computing and control capabilities and support the operation of the entire computer equipment.

[0193] The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor, the processor can execute any voiceprint recording method.

[0194] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0195] It should be understood that the processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0196] In one embodiment, the processor is used to run a computer program stored in a memory to implement the following steps: when a trigger operation for opening the application is detected, obtaining user identity information of the application; determining, based on the user identity information, a user category to which the user corresponding to the user identity information belongs, the user category including a first category of users and a second category of users, wherein the first category of users is a user who has not recorded voiceprint information, and the second category of users is a user who has recorded voiceprint information; if the user category to which the user corresponding to the user identity information belongs is the first category of users, loading a preset guide page into a display interface of the application, responding to the trigger operation and opening a display interface containing the guide page, the guide page being used to guide the user to record voiceprint; if it is detected that the user triggers the guide page, displaying a video recording interface; displaying text information preset by the video recording function on the video recording interface, recording a video of the user reading the text information, and sending the video to a server so as to extract the user's voiceprint information from the video.

[0197] In some embodiments, the processor is also used to implement the following steps: obtaining a guidance strategy corresponding to the user, the guidance strategy being determined based on user characteristic information of the user; generating at least one alternative guidance page based on the guidance strategy; selecting one from at least one of the alternative guidance pages as the guidance page based on a preset period, the preset period including a preset duration or number of logins; loading the preset guidance page into the display interface of the application, the preset guidance page including a guidance page obtained by switching the alternative guidance page based on a period of time or number of logins.

[0198] In some embodiments, the processor is further configured to implement: sending an acquisition instruction for acquiring the guidance strategy to a server, so that the server feeds back the guidance strategy corresponding to the user according to the acquisition instruction.

[0199] The server determines at least one corresponding alternative guidance item based on the user's user characteristic information, and determines a corresponding guidance strategy for the user based on at least one of the alternative guidance items, wherein the user characteristic information includes login frequency, activity level, service level, transaction frequency, geographic location, service items and / or risk preference, and the alternative guidance items include gift certificates, points, discounts, lottery opportunities, cash red envelopes, memberships, event tickets, data permissions and / or virtual costumes.

[0200] In some embodiments, the processor is used to implement: if it is detected that the user triggers the guide page, the liveness detection function is also turned on; before the text information preset for the video recording function is displayed, the processor is also used to implement the following steps: based on the liveness detection function, the video recording interface is controlled to change color, and a facial image corresponding to the face of the user illuminated by the color change is collected; the facial image is sent to the server for liveness detection and the detection result fed back by the server is obtained; if the detection result fed back by the server is liveness, the step of displaying the text information preset for the video recording function is executed.

[0201] In some embodiments, after determining, based on the user identity information, the user category to which the user corresponding to the user identity information belongs, the processor is further used to implement: if the user category to which the user corresponding to the user identity information belongs is the second category of users, opening the display interface of the application in response to the trigger operation.

[0202] In some embodiments, the processor is used to implement the following steps: receiving a video recording instruction sent by a terminal device, and starting a video recording function in response to the video recording instruction, wherein the video recording instruction is generated when it is determined that the user category is the first category of user and the user triggers a guide page displayed on the display interface of the application.

[0203] A video sent by the terminal device is obtained, where the video is obtained by recording the user reading the text information aloud.

[0204] An asynchronous thread pool is used to start a first thread and a second thread; wherein the first thread extracts frames from the video and sends the frames to a person identification verification module for person identification verification; the second thread extracts sound from the video recorded by the user and sends it to a voiceprint registration module for voiceprint registration.

[0205] If the result of the person-witness verification is failure, repeat the step of extracting frames from the video and sending the frames to the person-witness verification module for person-witness verification.

[0206] If the number of repetitions of the first thread is greater than or equal to a preset threshold, or the result of the voiceprint registration of the second thread is a failure, the first thread and the second thread are terminated, the registration is cleared, and the result of the voiceprint recording failure is sent to the terminal device.

[0207] If the result of the identity verification is passed and the result of the voiceprint registration is successful, the result of the successful voiceprint registration is sent to the terminal device.

[0208] A computer-readable storage medium is also provided in an embodiment of the present application. The computer-readable storage medium stores a computer program. The computer program includes program instructions. The processor executes the program instructions to implement any voiceprint recording method provided in the embodiments of the present application.

[0209] The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiment, such as a hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc., equipped on the computer device.

[0210] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A voiceprint recording method, characterized in that: The voiceprint recording is applied to a terminal device, and the terminal device includes an application program, including: When a trigger operation for starting the application is detected, obtaining user identity information of the application; Determine, according to the user identity information, a user category to which the user corresponding to the user identity information belongs, the user category including a first category of users and a second category of users, wherein the first category of users is a user whose voiceprint information has not been recorded, and the second category of users is a user whose voiceprint information has been recorded; If the user category to which the user corresponding to the user identity information belongs is the first category of user, loading a preset guide page into the display interface of the application, responding to the trigger operation and opening the display interface including the guide page, the guide page being used to guide the user to perform voiceprint recording; If it is detected that the user triggers the guide page, display the video recording interface; The text information preset by the video recording function is displayed on the video recording interface, a video of the user reading the text information is recorded, and the video is sent to a server so as to extract the voiceprint information of the user from the video.

2. The method according to claim 1, characterized in that The step of loading a preset guide page into a display interface of the application program includes: Acquire a guidance strategy corresponding to the user, where the guidance strategy is determined according to user feature information of the user; generating at least one alternative guide page according to the guide strategy; Selecting one from at least one of the candidate guide pages as the guide page according to a preset period, wherein the preset period includes a preset duration or a number of logins; The preset guide page is loaded into the display interface of the application, wherein the preset guide page includes a guide page obtained by switching the candidate guide page based on time or number of logins.

3. The method according to claim 1, characterized in that The obtaining of the guidance strategy corresponding to the user includes: Sending an acquisition instruction for acquiring the guidance strategy to a server, so that the server feeds back the guidance strategy corresponding to the user according to the acquisition instruction; The server determines at least one corresponding alternative guidance item based on the user's user characteristic information, and determines a corresponding guidance strategy for the user based on at least one of the alternative guidance items, wherein the user characteristic information includes login frequency, activity level, service level, transaction frequency, geographic location, service items and / or risk preference, and the alternative guidance items include gift certificates, points, discounts, lottery opportunities, cash red envelopes, memberships, event tickets, data permissions and / or virtual costumes.

4. The method according to claim 1, characterized in that If it is detected that the user triggers the guide page, the liveness detection function is also turned on; before displaying the text information preset by the video recording function, the method also includes: Controlling the video recording interface to change color based on the liveness detection function, and collecting a face image corresponding to the face of the user illuminated by the color change; Sending the face image to the server for liveness detection and obtaining the detection result fed back by the server; If the detection result fed back by the server is a living body, the step of displaying the text information preset by the video recording function is executed.

5. The method according to claim 1, characterized in that After determining, based on the user identity information, the user category to which the user corresponding to the user identity information belongs, the method further includes: If the user category to which the user corresponding to the user identity information belongs is the second category of user, the display interface of the application is opened in response to the trigger operation.

6. A voiceprint recording method, characterized in that: The voiceprint recording is applied to the server, including: Acquire a video sent by the terminal device, where the video is obtained by recording the user reading the text information aloud; Using an asynchronous thread pool to start a first thread and a second thread; wherein the first thread extracts frames from the video and sends the frames to a human identity verification module for human identity verification; the second thread extracts sound from the video recorded by the user and sends the sound to a voiceprint registration module for voiceprint registration; If the result of the human-evidence verification is failure, repeating the step of extracting frames from the video and sending the frames to the human-evidence verification module for human-evidence verification; If the number of repetitions in the first thread is greater than or equal to a preset threshold, or the result of the voiceprint registration in the second thread is a failure, the first thread and the second thread are terminated, the registration is cleared, and the result of the voiceprint recording failure is sent to the terminal device; If the result of the identity verification is passed and the result of the voiceprint registration is successful, the result of the successful voiceprint registration is sent to the terminal device.

7. A voiceprint recording device, configured in a terminal device, characterized in that: include: An identity acquisition module, used to acquire user identity information of the application when a trigger operation for opening the application is detected; an identity classification module, configured to determine, based on the user identity information, a user category to which the user corresponding to the user identity information belongs, the user category comprising a first category of users and a second category of users, wherein the first category of users are users whose voiceprint information has not been recorded, and the second category of users are users whose voiceprint information has been recorded; a guide loading module, configured to load a preset guide page to the display interface of the application program if the user category to which the user corresponding to the user identity information belongs is the first category of user, and respond to the trigger operation and open the display interface including the guide page, wherein the guide page is used to guide the user to perform voiceprint recording; An interface display module, used to display a video recording interface if it is detected that a user triggers the guide page; The video recording module is used to display the text information preset by the video recording function on the video recording interface, record the video of the user reading the text information, and send the video to the server so as to extract the voiceprint information of the user from the video.

8. A voiceprint recording device, configured on a server, characterized in that: include: A video receiving module, used for acquiring a video sent by the terminal device, wherein the video is obtained by recording the user reading the text information aloud; A dual-factor extraction module, configured to start a first thread and a second thread using an asynchronous thread pool; wherein the first thread extracts frames from the video and sends the frames to a person identification verification module for person identification verification; and the second thread extracts sound from the video recorded by the user and sends the sound to a voiceprint registration module for voiceprint registration; A failure retry module, used for repeating the step of extracting frames from the video and sending the frames to the human identification verification module for human identification verification if the result of the human identification verification is failure; A registration failure module, configured to terminate the first thread and the second thread if the number of repetitions of the first thread is greater than or equal to a preset threshold, or the result of the voiceprint registration of the second thread is a failure, and to clear the registration, and send the result of the voiceprint recording failure to the terminal device; The registration success module sends the result of successful voiceprint registration to the terminal device if the result of the identity verification is passed and the result of the voiceprint registration is successful.

9. A computer device, characterized in that: The computer device includes a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein the memory stores a policy model, wherein when the computer program is executed by the processor, the voiceprint recording method according to any one of claims 1 to 6 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements the voiceprint recording method according to any one of claims 1 to 6.