Disclosure of Invention
The embodiment of the invention provides a method and a device for verifying identity, which are used for improving the accuracy of user identity verification under the condition that a user does not sense the identity.
The method for verifying the identity provided by the embodiment of the invention comprises the following steps:
acquiring verification information; the authentication information is used for authenticating the identity of the user;
sequentially executing a plurality of preset verification processes, and determining that the user passes the identity verification after the user passes each verification process; wherein, executing any verification process comprises:
determining a first verification problem from a problem library of the verification process, and displaying the first verification problem to the user;
acquiring voice data of the user for answering the first verification question by voice;
extracting a first voiceprint and a first answer of the user from the speech data;
determining whether the user passes the verification process or not according to the first voiceprint and the first answer of the user; if yes, entering the next verification process.
Optionally, the determining whether the user passes the verification process according to the first voiceprint and the first answer of the user includes:
generating a voiceprint score corresponding to the first voiceprint and a question score corresponding to the first answer;
if the voiceprint score exceeds a first threshold and the question score exceeds a second threshold, determining that the user passes the verification process;
if the voiceprint score does not exceed the first threshold value or the question score does not exceed the second threshold value, determining a second verification question in a question bank of the verification process so as to execute the verification process again for the user, and if the number of times of repeatedly executing the verification process reaches a preset number of times, determining that the user does not pass the verification process.
Optionally, the generating a voiceprint score corresponding to the first voiceprint includes:
determining whether a second voice print corresponding to the identity exists in a voice print database or not according to the identity in the verification information; the standard voiceprint of each user is recorded in the voiceprint database;
and if so, generating the voiceprint score according to the first voiceprint and the second voiceprint.
Optionally, the method further includes:
if it is determined that the second voiceprint corresponding to the identity does not exist in the voiceprint database, pre-registering the first voiceprint into the voiceprint database;
and after the user is determined to pass the identity authentication, taking the first voiceprint pre-registered in the voiceprint database as a standard voiceprint of the user.
Optionally, before determining that the user passes the verification process according to the first voiceprint and the first answer of the user, the method further includes:
and matching the first voiceprint with each voiceprint in a voiceprint blacklist, and if determining that a third voiceprint matched with the first voiceprint exists in the voiceprint blacklist, determining that the user does not pass the identity authentication.
Optionally, after determining that the user passes the authentication, the method further includes:
and inputting the first voiceprint into a voiceprint training model so as to update the standard voiceprint of the user.
In the technical scheme, when the user is authenticated, a plurality of authentication processes need to be performed on the user, voiceprints and problem answers of the user for answering questions are extracted in each authentication process, and the voiceprints and the problem answers of the user are scored respectively to determine whether the user passes through the current authentication process. In addition, the comprehensiveness of the problem of user identity verification can be realized through multiple verification processes, and the accuracy of the whole user identity verification is further improved.
Correspondingly, the embodiment of the invention also provides a device for verifying the identity, which comprises:
an acquisition unit and a processing unit;
the acquisition unit is used for acquiring verification information; the authentication information is used for authenticating the identity of the user;
the processing unit is used for sequentially executing a plurality of preset verification processes, and after the user passes each verification process, determining that the user passes the identity verification; wherein, executing any verification process comprises:
determining a first verification problem from a problem library of the verification process, and displaying the first verification problem to the user;
acquiring voice data of the user for answering the first verification question by voice;
extracting a first voiceprint and a first answer of the user from the speech data;
determining whether the user passes the verification process or not according to the first voiceprint and the first answer of the user; if yes, entering the next verification process.
Optionally, the processing unit is specifically configured to:
generating a voiceprint score corresponding to the first voiceprint and a question score corresponding to the first answer;
if the voiceprint score exceeds a first threshold and the question score exceeds a second threshold, determining that the user passes the verification process;
if the voiceprint score does not exceed the first threshold value or the question score does not exceed the second threshold value, determining a second verification question in a question bank of the verification process so as to execute the verification process again for the user, and if the number of times of repeatedly executing the verification process reaches a preset number of times, determining that the user does not pass the verification process.
Optionally, the processing unit is specifically configured to:
determining whether a second voice print corresponding to the identity exists in a voice print database or not according to the identity in the verification information; the standard voiceprint of each user is recorded in the voiceprint database;
and if so, generating the voiceprint score according to the first voiceprint and the second voiceprint.
Optionally, the processing unit is further configured to:
if it is determined that the second voiceprint corresponding to the identity does not exist in the voiceprint database, pre-registering the first voiceprint into the voiceprint database;
and after the user is determined to pass the identity authentication, taking the first voiceprint pre-registered in the voiceprint database as a standard voiceprint of the user.
Optionally, the processing unit is further configured to:
and matching the first voiceprint with each voiceprint in a voiceprint blacklist before the user passes the verification process according to the first voiceprint and the first answer of the user, and determining that the user does not pass the identity verification if a third voiceprint matched with the first voiceprint is determined to exist in the voiceprint blacklist.
Optionally, the processing unit is further configured to:
after the user is determined to be verified, inputting the first voiceprint to a voiceprint training model to update a standard voiceprint of the user.
Correspondingly, an embodiment of the present invention further provides a computing device, including:
a memory for storing program instructions;
and the processor is used for calling the program instructions stored in the memory and executing the identity verification method according to the obtained program.
Accordingly, embodiments of the present invention also provide a computer-readable non-volatile storage medium, which includes computer-readable instructions, and when the computer-readable instructions are read and executed by a computer, the computer is caused to execute the above method for verifying an identity.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the present invention will be described in further detail with reference to the accompanying drawings, and it is apparent that the described embodiments are only a part of the embodiments of the present invention, not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 schematically shows a system architecture to which the method for authenticating identity provided by the embodiment of the present invention is applicable, and the system architecture may include an anti-fraud authentication system, a video authentication system, a telephone authentication system, a voiceprint authentication system, a decision making system, and a question bank.
The anti-fraud verification system is configured with various anti-fraud policy models for performing anti-fraud verification on the user, specifically, order information, credit investigation information and the like of the user are input into each anti-fraud policy model, whether the user is a fraudulent user is judged, if yes, a service request initiated by the user is directly rejected, and if not, the next verification can be performed.
The phone verification system verifies the identity of the user in a phone form, the video verification system verifies the identity of the user in a video form, after the user is confirmed to pass anti-fraud verification, the phone verification system or the video verification system can be selected, then questions can be selected from a question bank to the user, answer of the questions answered by the user through voice is obtained, the voiceprint verification system identifies the voiceprint of the user in the voice of the user, and the voiceprint of the user is matched with the standard voiceprint of the user, which is configured in the voiceprint verification system in advance, so that a matching result is generated.
And the decision-making system determines whether the user passes the identity authentication or not by combining the question answers of the user and the voiceprint matching result of the user.
Based on the above description, fig. 2 exemplarily shows a flow of a method for verifying an identity according to an embodiment of the present invention.
As shown in fig. 2, the process specifically includes:
step 201, obtaining verification information.
The authentication information is used for authenticating the identity of the user when the user requests banking business; the authentication information may include an identification of the user, a contact address, and the like.
Step 202, a plurality of predefined authentication procedures are executed in sequence, and after the user passes each authentication procedure, it is determined that the user passes the identity authentication.
In order to ensure the accuracy of the verification, a plurality of verification processes are preset, when the user is subjected to identity verification, the plurality of verification processes are sequentially executed, a second verification process is executed after a first verification process passes, and a third verification process is executed after a second verification process passes until all verification processes pass. Each verification process corresponds to a problem library of the verification process, and the types of the problem libraries of different verification processes are different.
When any verification process is executed, a first verification problem can be determined from a problem library of the verification process and displayed to a user, so that voice data of the user for answering the first verification problem in a voice mode is obtained, a first voiceprint and a first answer of the user are extracted from the voice data, whether the user passes the verification process or not is determined according to the first voiceprint and the first answer of the user, if yes, the next verification process is started, and if not, the user is directly determined to fail in identity verification.
When determining whether the user passes the verification process according to the first voiceprint and the first answer of the user, the first voiceprint and the first answer may be scored respectively, and then whether the user passes the verification process is determined according to the scores.
Specifically, a voiceprint score corresponding to the first voiceprint and a question score corresponding to the first answer are generated, and the voiceprint score and the question score are respectively compared with corresponding threshold values, and the following conditions can be provided:
in the first situation, if the voiceprint score exceeds a first threshold and the problem score exceeds a second threshold, determining that the user can enter the next verification process after passing the verification process;
determining a second verification problem in a problem base of the verification process if the voiceprint score does not exceed a first threshold or the problem score does not exceed a second threshold, so that the verification process is executed again for the user by adopting the second verification problem, and if the voiceprint score of the user is determined to exceed the first threshold and the problem score exceeds the second threshold within the preset times, determining that the user passes through the verification process and can enter the next verification process; if the voiceprint score of the user does not exceed the first threshold value and the problem score does not exceed the second threshold value after the number of times of repeatedly executing the verification process reaches the preset number of times, it can be determined that the user does not pass the verification process, namely, the user does not pass the whole identity verification;
and in a third case, if the voiceprint score does not exceed the first threshold and the problem score does not exceed the second threshold, it can be determined that the user does not pass the verification process, that is, the user does not pass the whole identity verification.
In the embodiment of the present invention, when generating the voiceprint score corresponding to the first voiceprint, the voiceprint database may be read according to the identity in the verification information, where the standard voiceprint of each user is recorded in the voiceprint database. And if a second voiceprint corresponding to the identity in the verification information, namely a standard voiceprint corresponding to the user identity, is read in the voiceprint database, comparing the first voiceprint with the second voiceprint to generate a voiceprint score.
In addition, the standard voiceprint of the user may not be recorded in the voiceprint database, that is, the second voiceprint corresponding to the identity is not determined according to the identity in the verification information, and the first voiceprint of the user is pre-registered in the voiceprint database; and if the user passes the identity authentication after the subsequent determination, taking the first voiceprint pre-registered in the voiceprint database as the standard voiceprint of the user.
In any one verification process, if the video verification system is adopted to verify the identity of the user, the first face image of the user can be obtained, and whether the user passes the verification process or not is determined according to the first voiceprint, the first answer and the first face image of the user. Specifically, the face score, the voiceprint score and the question score of the user are determined, and whether the user passes the verification process or not is determined by combining the three scores. In determining the facial score of the user, the first facial image of the user may be compared to facial images indicated by the identifiers in the image database to generate a facial score.
In the embodiment of the present invention, a voiceprint blacklist may also be set, a voiceprint of each fraudulent user is configured in the voiceprint blacklist, and the voiceprint blacklist is used to determine in advance whether the user is a fraudulent user, that is, before generating a voiceprint score corresponding to a first voiceprint, the first voiceprint is matched with each voiceprint in the voiceprint blacklist, and if it is determined that a third voiceprint matched with the first voiceprint exists, it is determined that the user does not pass the verification. And a voiceprint blacklist is set, and the accuracy of blacklist identification is effectively improved based on blacklist comparison of living body characteristics.
After the user is determined to pass the verification, the standard voiceprint of the user in the voiceprint database can be updated by adopting the first voiceprint in the current identity verification, specifically, the first voiceprint is input into the voiceprint training model to update the standard voiceprint of the user, and the updated standard voiceprint of the user is updated into the voiceprint database.
It should be noted that, whether a video authentication system or a telephone authentication system is adopted, the voiceprint database and the voiceprint blacklist can be shared by the embodiments of the present invention.
In practical application, the embodiment of the invention can be used in the borrowing scene of banking business, and when a user initiates a transaction, the anti-fraud verification system can make a decision to directly pass, reject or need further verification according to the wind control variable of the user transaction. And when the fact that one-step verification is needed is determined, a telephone verification system or a video verification system is selected, and therefore the next step of identity verification is conducted. For example, a user enters a video verification room through a cloud live broadcast technology, a bypass is opened for live broadcast, after a video stream starts to be pushed, the bypass pushes the video stream to a recording platform, a question bank generates verification questions and displays the verification questions to the user, the user answers the questions through voice, the recording platform converts the real-time video stream into an audio stream and pushes the audio stream to a voiceprint verification system, voiceprint features are extracted through voiceprint application in the voiceprint verification system and are compared with stored voiceprints in a voiceprint database, and if the voiceprint database is determined to have standard voiceprints of the user, the comparison is carried out according to the standard voiceprints and the extracted voiceprints; if the voiceprint database does not have the standard voiceprint of the user, the extracted voiceprint is pre-registered in the voiceprint database. And sending the question result answered by the user and the voiceprint comparison result into a decision system, carrying out weighting decision by the decision system, and determining whether the user passes the identity authentication or fails the identity authentication or enters the next authentication process. Here, if the user passes the authentication, the voiceprint pre-registration is converted into the formal registration, and the voiceprint model is re-run by combining the voiceprint feature which passes the comparison with the voiceprint feature which passes the latest, so that the voiceprint feature is optimized. In this example, the voiceprint feature passed through by the borrowing scene can also be used for identity verification in other scenes such as customer service consultation, card change, password change and the like.
In the technical scheme, when the user is authenticated, a plurality of authentication processes need to be performed on the user, voiceprints and problem answers of the user for answering questions are extracted in each authentication process, and the voiceprints and the problem answers of the user are scored respectively to determine whether the user passes through the current authentication process. In addition, the comprehensiveness of the problem of user identity verification can be realized through multiple verification processes, and the accuracy of the whole user identity verification is further improved.
Based on the same inventive concept, fig. 3 exemplarily shows a structure of an apparatus for verifying an identity, which may perform a flow of a method for verifying an identity according to an embodiment of the present invention.
An apparatus for verifying identity, comprising:
an acquisition unit 301 and a processing unit 302;
the obtaining unit 301 is configured to obtain authentication information; the authentication information is used for authenticating the identity of the user;
the processing unit 302 is configured to sequentially execute a plurality of predefined authentication procedures, and after the user passes each authentication procedure, determine that the user passes the identity authentication; wherein, executing any verification process comprises:
determining a first verification problem from a problem library of the verification process, and displaying the first verification problem to the user;
acquiring voice data of the user for answering the first verification question by voice;
extracting a first voiceprint and a first answer of the user from the speech data;
determining whether the user passes the verification process or not according to the first voiceprint and the first answer of the user; if yes, entering the next verification process.
Optionally, the processing unit 302 is specifically configured to:
generating a voiceprint score corresponding to the first voiceprint and a question score corresponding to the first answer;
if the voiceprint score exceeds a first threshold and the question score exceeds a second threshold, determining that the user passes the verification process;
if the voiceprint score does not exceed the first threshold value or the question score does not exceed the second threshold value, determining a second verification question in a question bank of the verification process so as to execute the verification process again for the user, and if the number of times of repeatedly executing the verification process reaches a preset number of times, determining that the user does not pass the verification process.
Optionally, the processing unit 302 is specifically configured to:
determining whether a second voice print corresponding to the identity exists in a voice print database or not according to the identity in the verification information; the standard voiceprint of each user is recorded in the voiceprint database;
and if so, generating the voiceprint score according to the first voiceprint and the second voiceprint.
Optionally, the processing unit 302 is further configured to:
if it is determined that the second voiceprint corresponding to the identity does not exist in the voiceprint database, pre-registering the first voiceprint into the voiceprint database;
and after the user is determined to pass the identity authentication, taking the first voiceprint pre-registered in the voiceprint database as a standard voiceprint of the user.
Optionally, the processing unit 302 is further configured to:
and matching the first voiceprint with each voiceprint in a voiceprint blacklist before the user passes the verification process according to the first voiceprint and the first answer of the user, and determining that the user does not pass the identity verification if a third voiceprint matched with the first voiceprint is determined to exist in the voiceprint blacklist.
Optionally, the processing unit 302 is further configured to:
after the user is determined to be verified, inputting the first voiceprint to a voiceprint training model to update a standard voiceprint of the user.
Based on the same inventive concept, an embodiment of the present invention further provides a computing device, including:
a memory for storing program instructions;
and the processor is used for calling the program instructions stored in the memory and executing the identity verification method according to the obtained program.
Based on the same inventive concept, the embodiment of the present invention also provides a computer-readable non-volatile storage medium, which includes computer-readable instructions, and when the computer-readable instructions are read and executed by a computer, the computer-readable instructions cause the computer to execute the above method for verifying the identity.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.