CN112541174A

CN112541174A - Service data verification method, device, equipment and storage medium

Info

Publication number: CN112541174A
Application number: CN202011472837.8A
Authority: CN
Inventors: 史其选
Original assignee: Ping An Technology Shenzhen Co Ltd
Current assignee: Ping An Technology Shenzhen Co Ltd
Priority date: 2020-12-15
Filing date: 2020-12-15
Publication date: 2021-03-23
Also published as: WO2022126964A1

Abstract

The invention relates to the field of data processing, and discloses a service data verification method, a device, equipment and a storage medium. The method comprises the following steps: acquiring service data to be checked; acquiring first biological characteristic information of a current user, inquiring corresponding second biological characteristic information from a preset biological characteristic database according to user identity information in service data, comparing and identifying, and if the identification results are consistent, extracting data content to be confirmed in the service data; calling a preset AI voice conversion model, converting data contents into broadcast voice for broadcasting, and acquiring reply voice of a user; and matching the reply voice based on preset answer data, and outputting service data if the matching result is that the reply voice is consistent with the answer data. The method ensures the safety and the compliance of the business data verification process and improves the accuracy and the reliability of the verification result. In addition, the invention also relates to a block chain technology, and the service data can be stored in the block chain.

Description

Service data verification method, device, equipment and storage medium

Technical Field

The present invention relates to the field of data processing, and in particular, to a method, an apparatus, a device, and a storage medium for verifying service data.

Background

At present, when a user selects to buy a certain insurance product, a related package is selected on a purchase page, then related information is filled in, the related information is submitted to an insurance auditing server after the related information is filled in, finally, the insurance auditing server generates insurance policy information according to the related information submitted by the user, then, an applicant checks the insurance policy information, and corresponding insurance application information can be generated only after the insurance policy information passes the checking.

However, such a method for auditing and confirming information of an applicant is time-consuming and labor-consuming, and it is difficult for a user to ensure that the applicant is actually checking when checking policy information, and a related checking mechanism is also difficult to evaluate the authenticity and accuracy of a checking result, so that the accuracy of the checking result is low, and the reliability of the checking result is not strong.

Disclosure of Invention

The invention mainly aims to solve the technical problems that the accuracy of a verification result is difficult to evaluate by the conventional business data verification mode, so that the data verification accuracy is low and the reliability is not strong.

The first aspect of the present invention provides a service data verification method, where the service data verification method includes:

acquiring service data to be checked;

acquiring first biological characteristic information of a current user, and inquiring corresponding second biological characteristic information from a preset biological characteristic database according to user identity information in the service data;

comparing and identifying the first biological characteristic information with the second biological characteristic information to obtain an identification result;

if the identification result is consistent, extracting the data content to be confirmed in the service data;

calling a preset AI voice conversion model to convert the data content into broadcast voice;

broadcasting the broadcast voice, and acquiring the reply voice of the user in real time;

matching the reply voice based on preset answer data to obtain a matching result;

and if the matching result is that the reply voice is consistent with the answer data, outputting the service data.

Optionally, in a first implementation manner of the first aspect of the present invention, the comparing and identifying the first biometric information and the second biometric information to obtain an identification result includes:

extracting the face information feature based on the first biological feature information to obtain a first face feature value;

extracting the face information features based on the second biological feature information to obtain a second face feature value;

and comparing and identifying the face comparison threshold, the face first characteristic value and the face second characteristic value based on a preset face comparison threshold to obtain an identification result.

Optionally, in a second implementation manner of the first aspect of the present invention, the comparing and recognizing the face comparison threshold, the face first feature value, and the face second feature value based on a preset face comparison threshold to obtain a recognition result includes:

matching and comparing the first face characteristic value with the second face characteristic value to obtain a face similarity score;

based on a preset correction formula, carrying out correction calculation on the face similarity score to obtain a corrected face similarity score;

and comparing and identifying the corrected face similarity score with the face comparison threshold value to obtain an identification result.

Optionally, in a third implementation manner of the first aspect of the present invention, the matching the reply voice based on preset answer data to obtain a matching result includes:

performing voice recognition on the reply voice based on a voice recognition technology to obtain reply data;

and comparing and matching the reply data with preset answer data to obtain a matching result.

Optionally, in a fourth implementation manner of the first aspect of the present invention, the performing, based on the speech recognition technology, speech recognition on the reply speech to obtain reply data includes:

preprocessing the reply voice to obtain clean voice;

extracting acoustic features of the clean voice to obtain acoustic feature parameters;

and performing language processing on the acoustic characteristic parameters based on a preset language model to obtain reply data.

Optionally, in a fifth implementation manner of the first aspect of the present invention, the comparing and matching the reply data with preset answer data to obtain a matching result includes:

extracting characters of each byte in the answer data to obtain a first character;

extracting characters of each byte in the reply data to obtain second characters;

performing hash calculation on the second characters to obtain a character index corresponding to each second character;

and comparing and matching the first character and the second character based on the character index to obtain a matching result.

Optionally, in a sixth implementation manner of the first aspect of the present invention, after the obtaining the service data to be verified, the method further includes:

based on a remote communication technology, video information of a service data verification process is acquired in real time;

monitoring the verification behavior in the business data verification process based on a preset business data verification safety rule to obtain monitoring information;

performing voice prompt on the verification behavior which does not accord with the business data verification safety rule in the monitoring information;

and recording the service data verification process to generate a service data verification process video, and storing the service data verification process video in a storage unit.

A second aspect of the present invention provides a service data verification apparatus, including:

the acquisition module is used for acquiring the service data to be checked;

the query module is used for acquiring first biological characteristic information of a current user and querying corresponding second biological characteristic information from a preset biological characteristic database according to user identity information in the service data;

the identification module is used for comparing and identifying the first biological characteristic information and the second biological characteristic information to obtain an identification result;

the extraction module is used for extracting the data content to be confirmed in the service data if the identification result is consistent;

the voice conversion module is used for calling a preset AI voice conversion model and converting the data content into broadcast voice;

the broadcasting module is used for broadcasting the broadcasting voice and acquiring the reply voice of the user in real time;

the matching module is used for matching the reply voice based on preset answer data to obtain a matching result;

and the information output module is used for outputting the service data if the matching result is that the reply voice is consistent with the answer data.

Optionally, in a first implementation manner of the second aspect of the present invention, the identification module includes:

the first feature extraction unit is used for extracting the face information feature based on the first biological feature information to obtain a first face feature value;

the second feature extraction unit is used for extracting the features of the face information based on the second biological feature information to obtain a second feature value of the face;

and the characteristic value comparison unit is used for comparing and identifying the face comparison threshold value, the face first characteristic value and the face second characteristic value based on a preset face comparison threshold value to obtain an identification result.

Optionally, in a second implementation manner of the second aspect of the present invention, the feature value comparing unit is specifically configured to:

Optionally, in a third implementation manner of the second aspect of the present invention, the matching module includes:

the voice recognition unit is used for carrying out voice recognition on the reply voice based on a voice recognition technology to obtain reply data;

and the data matching unit is used for comparing and matching the reply data with preset answer data to obtain a matching result.

Optionally, in a fourth implementation manner of the second aspect of the present invention, the speech recognition unit is specifically configured to:

preprocessing the reply voice to obtain clean voice;

Optionally, in a fifth implementation manner of the second aspect of the present invention, the data matching unit is specifically configured to:

Optionally, in a sixth implementation manner of the second aspect of the present invention, the service data verification apparatus further includes a monitoring module, which is specifically configured to:

A third aspect of the present invention provides a service data verification device, where the service data verification device includes: a memory having instructions stored therein and at least one processor, the memory and the at least one processor interconnected by a line; the at least one processor calls the instructions in the memory to cause the business data verification device to perform the steps of the business data verification method described above.

A fourth aspect of the present invention provides a computer-readable storage medium having stored therein instructions, which, when run on a computer, cause the computer to perform the steps of the above-mentioned business data verification method.

In the technical scheme provided by the invention, the service data to be verified and the first biological characteristic information of the current user are acquired, and the corresponding second biological characteristic information is inquired from a preset biological characteristic database according to the user identity information in the service data; comparing and identifying the first biological characteristic information with the second biological characteristic information to obtain an identification result; if the identification result is consistent, extracting the data content to be confirmed in the policy information; calling a preset AI voice conversion model to convert the data content into broadcast voice; broadcasting the broadcast voice, and acquiring the reply voice of the current user in real time; matching the reply voice based on preset answer data to obtain a matching result; and if the matching result is that the reply voice is consistent with the answer data, outputting the service data. In the embodiment of the invention, by adopting the technical scheme, the authenticity and the accuracy of the verification result obtained after the business data is verified are ensured, and the accuracy and the reliability of the verification result are improved.

Drawings

Fig. 1 is a schematic diagram of a first embodiment of a business data verification method according to an embodiment of the present invention;

fig. 2 is a schematic diagram of a second embodiment of a business data verification method according to an embodiment of the present invention;

fig. 3 is a schematic diagram of a third embodiment of a business data verification method in the embodiment of the present invention;

fig. 4 is a schematic diagram of a fourth embodiment of a business data verification method in the embodiment of the present invention;

fig. 5 is a schematic diagram of a fifth embodiment of a business data verification method in the embodiment of the present invention;

FIG. 6 is a schematic diagram of an embodiment of a business data verification apparatus according to an embodiment of the present invention;

fig. 7 is a schematic diagram of another embodiment of a business data verification apparatus according to an embodiment of the present invention;

fig. 8 is a schematic diagram of an embodiment of a service data verification device in the embodiment of the present invention.

Detailed Description

The embodiment of the invention provides a business data verification method, a business data verification device, business data verification equipment and a storage medium, wherein the business data with verification and biological characteristic information of a user are obtained, the biological characteristic information corresponding to the business data is inquired from a biological characteristic database and is compared to ensure that the current user identity is consistent with the user identity in the business data, a preset AI voice conversion model is called to convert the data content to be confirmed in the business data into broadcast voice, the broadcast voice is played for the user, the reply voice of the user is obtained, the reply voice is matched based on answer data, and if the matching is consistent, the business data is output. The method ensures the authenticity of the whole service data verification process and improves the accuracy and reliability of the verification result obtained after verification.

The terms "first," "second," "third," "fourth," and the like in the description and in the claims, as well as in the drawings, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It will be appreciated that the data so used may be interchanged under appropriate circumstances such that the embodiments described herein may be practiced otherwise than as specifically illustrated or described herein. Furthermore, the terms "comprises," "comprising," or "having," and any variations thereof, are intended to cover non-exclusive inclusions, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

For convenience of understanding, a specific flow of the embodiment of the present invention is described below, and referring to fig. 1, a first embodiment of a service data verification method in the embodiment of the present invention includes:

101, acquiring service data to be checked;

it is to be understood that the executing subject of the present invention may be a service data verification apparatus, and may also be a terminal or a server, which is not limited herein. The embodiment of the present invention is described by taking a server as an execution subject.

If the business data to be verified is the insurance policy information applied by the applicant needing verification, the current business data verification process is the user's insurance verification process, and obtaining the business data to be verified is obtaining the insurance policy information to be verified. The insurance policy management method comprises the steps that a insurance application user inputs related personal identity information on an insurance application system, after the insurance application system receives the related personal identity information, insurance policy information corresponding to the user personal identity information is called from an insurance policy information database on the system, wherein the user personal identity information comprises basic personal identity information such as names, identification numbers and the like, and the insurance policy information comprises data information such as insurance policy numbers, identity information of the insurance application, insurance types and insurance contents.

102, acquiring first biological characteristic information of a current user, and inquiring corresponding second biological characteristic information from a preset biological characteristic database according to user identity information in service data;

after a user enters the insurance application system, the user inputs personal identity information to acquire corresponding insurance policy information, before the insurance policy information is used for starting an insurance verification process, the system prompts the user to start a mobile terminal camera, and collects the biological characteristic information of the current operation user in real time to be used as first biological characteristic information to perform identity verification.

And the identity information database of the applicant in the insurance application system is in butt joint with the biological characteristic information database of the public security network, and corresponding biological characteristic information is extracted from the biological characteristic information database as second biological characteristic information according to the identity information of the applicant in the policy information. The first biometric information of the user is information including physiological characteristics (fingerprint, iris, facial phase, DNA, etc.) or behavior characteristics (gait, keystroke habit, etc.) inherent to the human body.

103, comparing and identifying the first biological characteristic information with the second biological characteristic information to obtain an identification result;

after the first biological characteristic information and the second biological characteristic information are acquired, based on a biological characteristic identification technology, the personal identification verification is carried out by utilizing the inherent physiological characteristics (fingerprint, iris, facial phase, DNA and the like) or the behavior characteristics (gait, keystroke habit and the like) of the human body, wherein the identification verification operation mainly carries out information identification and comparison by utilizing the physiological characteristics, specifically, the acquired facial image can be selected as an identification basis, and the corresponding face information in the public security network is called for comparison and identification.

After the first biological characteristic information and the second biological characteristic information are identified, comparing each piece of characteristic information, and then obtaining an identification result and feeding the identification result back to the server to be used as a basis for whether the next step is executed or not.

104, if the identification result is consistent, extracting the data content to be confirmed in the service data;

if the first biological characteristic information is consistent with the second biological characteristic information in identification and comparison, the operation user in the current underwriting process is proved to be consistent with the actual applicant, namely the current underwriting process is performed by the actual applicant, insurance application data needing to be confirmed by the applicant in the corresponding policy information is extracted from the insurance application system, wherein the insurance application data needing to be confirmed by the applicant in the policy information comprises: identity information of the applicant, insurance type, settlement information of insurance and the like.

105, calling a preset AI voice conversion model to convert the data content into broadcast voice;

when a user starts to prepare for the confirmation process of insurance policy information, the insurance application system extracts insurance application data to be confirmed, calls an AI voice conversion model arranged on the system, inputs the insurance application data into the AI voice conversion model, performs voice conversion by using an AI voice conversion technology in the model, adjusts the sound effect and generates broadcast voice. The technology of converting data content into voice by using AI voice conversion technology belongs to the prior art, and is not described herein again.

106, broadcasting the broadcast voice and acquiring the reply voice of the user in real time;

after receiving prompt information of starting a policy information confirmation process by a user, the insurance application system extracts corresponding broadcast voice according to the policy information and broadcasts the insurance application. After each item of content needing to be confirmed in the policy information is broadcasted, the user needs to perform voice confirmation and reply, and in the process, the system can collect reply voice of the user in real time and store the reply voice in the storage unit.

107, matching the reply voice based on preset answer data to obtain a matching result;

for insurance data to be confirmed in insurance policy information, the insurance system can perform a reply setting in advance, namely answer data, for checking and matching the reply of the user, after the reply voice of the user is obtained, the reply voice of the user is preprocessed first, clean voice is extracted, then voice recognition technology is utilized to perform voice recognition processing on the reply voice, the reply data is obtained after the voice recognition processing, then the answer data is utilized to perform matching on the reply voice, a matching result is obtained, and only when the reply of the user is consistent with the answer data, the insurance information can be generated.

And 108, if the matching result is that the reply voice is consistent with the answer data, outputting the service data.

And acquiring a matching result of the user reply data and the preset answer data, when the matching result is consistent, indicating that the user has finished checking the policy information, and confirming all contents of the policy information, outputting the policy information after the user confirms the policy information by the system, and ending the whole checking and protecting process. And when the matching results are inconsistent, the user is indicated that the policy information is in doubt, or the policy information has wrong content and needs to be changed, the user can apply for interrupting the underwriting process and feed the condition back to the insurance company.

In the embodiment of the invention, the service data with the verification and the biological characteristic information of the current user are obtained, the corresponding biological characteristic information is inquired from the biological characteristic database and is compared to ensure that the identity of the current user is consistent with the identity of the user in the service data, then the AI voice conversion technology in a preset AI voice conversion model is utilized to convert the data content to be confirmed in the service data into the broadcast voice, the broadcast is carried out on the user, the reply voice of the current user is obtained, the reply voice is matched based on the answer data, and if the match is consistent, the service data is output. By the technical scheme of the embodiment of the invention, the authenticity and the accuracy of the verification result obtained after the verification are ensured, and the accuracy and the reliability of the verification result are improved.

Referring to fig. 2, a second embodiment of the service data verification method according to the embodiment of the present invention includes:

201, acquiring service data to be checked;

202, acquiring first biological characteristic information of a current user, and inquiring corresponding second biological characteristic information from a preset biological characteristic database according to user identity information in service data;

after a user enters the insurance application system, the user inputs personal identity information to acquire corresponding insurance policy information, before the insurance policy information is used for starting an insurance verification process, the system prompts the user to start a mobile terminal camera, and collects the biological characteristic information of the current operation user in real time to be used as first biological characteristic information to perform identity verification. And the identity information database of the applicant in the insurance application system is in butt joint with the biological characteristic information database of the public security network, and corresponding biological characteristic information is extracted from the biological characteristic information database as second biological characteristic information according to the identity information of the applicant in the policy information.

203, extracting the face information feature based on the first biological feature information to obtain a first face feature value;

the face image acquisition technology is utilized to acquire face information of the applicant on site, wherein the face information comprises information of a front face, a left face and a right face, and the front face image, the left face image and the right face image are acquired in real time. The method comprises the steps of performing image standard formatting processing on a front face image, a left face image and a right face image, and then performing pixel normalization processing on the front face image, the left face image and the right face image respectively by taking pixels of the front face photo, the left face photo and the right face photo as standards, so that the difference between face information in a face information database and face characteristic information of an acquired face image is reduced or even disappears, and the accuracy is improved.

And extracting the characteristic values of the processed front face image, the processed left face image and the processed right face image to be used as a first front face characteristic value, a first left face characteristic value and a first right face characteristic value. Here, the feature value refers to five sense organs of a human face, including eyes, eyebrows, nose, mouth, and ears, and may also include face shape, hair, or other facial features.

204, extracting the face information features based on the second biological feature information to obtain a second face feature value;

the face information data corresponding to the applicant is extracted from a face information database of a butt joint public security system, the face information data comprise a front face photo, a left face photo and a right face photo, characteristic values of face information are extracted according to the face information data contained in the photos, the characteristic values comprise a front face second characteristic value, a left face second characteristic value and a right face second characteristic value, and a front face comparison threshold value M2, a first side face comparison threshold value M2 and a second side face comparison threshold value M3 are set.

205, matching and comparing the first characteristic value of the face with the second characteristic value of the face to obtain a face similarity score;

respectively extracting the shortest distances from eyes to the lower edges of the front face photo, the left face photo and the right face photo in the front face photo, the left face photo and the right face photo, and respectively recording the shortest distances as a front face original size alpha 1, a left face original size alpha 2 and a right face original size alpha 3; and respectively extracting the shortest distances from the eyes in the front face image, the left face image and the right face image to the lower edges of the front face image, the left face image and the right face image, and respectively recording the shortest distances as a front face correction size beta 1, a left face correction size beta 2 and a right face correction size beta 3.

Calculating a front face alignment correction coefficient, a left face alignment correction coefficient and a right face alignment correction coefficient, matching and comparing a front face first characteristic value and a front face second characteristic value, and recording a corresponding front face similarity score R1; matching and comparing the first characteristic value of the left face with the second characteristic value of the left face, and recording a corresponding left face similarity score R2; and matching and comparing the first right face characteristic value and the second right face characteristic value to record a corresponding right face similarity score R3.

206, based on a preset correction formula, carrying out correction calculation on the face similarity score to obtain a corrected face similarity score;

calculating a corrected positive face similarity score R1 'and a corrected left face similarity score R2' according to a preset correction formula; calculating a correlation correction left face similarity score R2' according to a correlation influence formula; calculating a corrected right face similarity score R3 'according to a correction formula, and calculating an association corrected right face similarity score R3' according to an association influence formula; because the front face image, the left face image and the right face image are obtained simultaneously, when the comparison result of the front face image has a deviation from the threshold value, because the angles of the human faces are not aligned or other reasons, there may exist human face images with left faces and right faces not aligned at the same time, which may cause the deviation between the comparison result of the side face image and the threshold value to increase, that is, cause an error to increase, so that the error influence of the front face on the left face and the error influence of the left face on the right face need to be eliminated, and the correction can be performed by using an associated influence formula.

And (3) carrying out normalization processing on R1 'and R2' by integrating the comparison result of the front face and the left face to obtain a first face similarity score R, wherein the calculation formula of R is as follows:

normalizing R1 ', R2 ' and R3 ' to obtain a second face similarity score R ', R ' having the formula:

wherein, the correction formula is: r_i'＝δ_iR_i

In the correction formula, i is a positive number, i is more than or equal to 1 and less than or equal to 3, and delta i represents a Ri alignment correction coefficient; δ 1 represents an alignment correction coefficient of R1, i.e., a front face alignment correction coefficient; δ 2 denotes an alignment correction coefficient of R2, i.e., a left-face alignment correction coefficient; δ 3 denotes an alignment correction coefficient of R3, i.e., a right-face alignment correction coefficient.

When the pixel normalization processing is carried out on the face image acquired on site, the face ratio in the face image and the face photo in the face information database is not completely aligned, namely, facial features, facial shapes and the like of the face are not completely aligned, so that the error of comparison can be caused.

In addition, the formula for calculating the alignment correction coefficient is:

wherein i is a positive number, i is more than or equal to 1 and less than or equal to 3, and δ i represents a Ri alignment correction coefficient; δ 1 represents an alignment correction coefficient of R1, i.e., a front face alignment correction coefficient; δ 2 denotes an alignment correction coefficient of R2, i.e., a left-face alignment correction coefficient; δ 3 denotes an alignment correction coefficient of R3, i.e., a right-face alignment correction coefficient, α 1 is a front-face original size, α 2 is a left-face original size, α 3 is a right-face original size, β 1 is a front-face corrected size, β 2 is a left-face corrected size, and β 3 is a right-face corrected size.

In addition, the correlation impact formula is: r_i”＝(1+Q_i-1 ²)R_i'

Wherein i is a positive number, i is more than or equal to 2 and less than or equal to 3, and Q1 is used for correcting the deviation rate of the similarity of the positive face; q2 is the corrected side face similarity deviation ratio, R2 'is the corrected left face similarity score, R3' is the corrected right face similarity score, R2 'is the correlation corrected left face similarity score, and R3' is the correlation corrected right face similarity score.

207, comparing and identifying the corrected face similarity score with a face comparison threshold value to obtain an identification result;

and comparing the R with a first side face comparison threshold M2, if the R is greater than a first side face comparison threshold M2, calculating a correction side face similarity deviation rate Q2 by using a threshold deviation formula, and if the R is less than or equal to the first side face comparison threshold M2, failing the face consistency check.

Comparing R1 ' with a positive face comparison threshold, if R1 ' is larger than the positive face comparison threshold M1, calculating a corrected positive face similarity deviation rate Q1 by using a threshold deviation formula, and if R1 ' is smaller than or equal to the positive face comparison threshold M1, failing to pass the face consistency check;

comparing the R 'with a second side face comparison threshold, and if the R' is larger than the second side face comparison threshold, the human face consistency comparison is passed; and if R' is less than or equal to the first side face comparison threshold, the human face consistency check is not passed.

The comparison results of the front face, the left face and the right face are integrated, and the normalization processing is carried out on the R1 ', the R2 ' and the R3 ', so that the situation that the insurance applicant is falsely authenticated by using a personal photo to start the insurance application process is avoided. And setting a front face comparison threshold M1, a first side face comparison threshold M2 and a second side face comparison threshold M3 to sequentially compare whether the front face, the left face and the right face are consistent, and stopping subsequent operation steps once the corrected similarity score is less than or equal to the corresponding threshold, and directly judging that the face consistency check does not pass.

Wherein, the threshold deviation formula is:

wherein i is a positive number, i is more than or equal to 1 and less than or equal to 2, and Q1 is used for correcting the deviation rate of the similarity of the positive face; q2 is a correction side face similarity deviation ratio.

208, if the identification result is consistent, extracting the data content to be confirmed in the service data;

209, calling a preset AI voice conversion model to convert the data content into broadcast voice;

210, broadcasting the broadcast voice and acquiring the reply voice of the user in real time;

211, matching the reply voice based on preset answer data to obtain a matching result;

212, if the matching result is that the reply voice is consistent with the answer data, outputting the service data.

And acquiring a matching result of the user reply data and the preset answer data, when the matching result is consistent, indicating that the user has finished checking the policy information, and confirming all contents of the policy information, and outputting the user policy information after the user confirms the policy information by the system, wherein the whole checking and protecting process is finished. And when the matching results are inconsistent, the user is indicated that the policy information is in doubt, or the policy information has wrong content and needs to be changed, the user can apply for interrupting the underwriting process and feed the condition back to the insurance company.

In the embodiment of the invention, the identity of the current user is identified by a face identification technology, the identity of the current user is ensured to be consistent with the identity of the user in the service data to be verified, then an AI voice conversion technology is utilized, the data content required to be confirmed by the user is voice broadcast and reply voice is obtained in the verification process, and the identity of the current user is verified by the method according to the confirmation result of the reply voice verification on the data content to be confirmed, so that the authenticity of the verification process is ensured, and the reliability of the verification result is improved.

Referring to fig. 3, a third embodiment of the service data verification method according to the embodiment of the present invention includes:

301, acquiring service data to be checked;

if the business data to be verified is the insurance policy information applied by the applicant needing verification, the current business data verification process is the user's insurance verification process, and obtaining the business data to be verified is obtaining the insurance policy information to be verified. The insurance policy information database is used for storing the insurance policy information corresponding to the personal information of the user.

302, acquiring first biological characteristic information of a current user, and inquiring corresponding second biological characteristic information from a preset biological characteristic database according to user identity information in service data;

after a user enters the insurance application system, the user inputs personal identity information to acquire corresponding insurance policy information, before the insurance policy information is used for starting an insurance verification process, the system prompts the user to start a mobile terminal camera, collects the biological characteristic information of the current operation user in real time and uses the biological characteristic information as first biological characteristic information to verify the identity of the insurance application person.

The identity information database of the applicant in the insurance application system is in butt joint with the biological characteristic information database of the public security network, corresponding biological characteristic information is extracted from the biological characteristic information database as second biological characteristic information according to the identity information of the applicant in the policy information, wherein the first biological characteristic information of the user is information including inherent physiological characteristics (fingerprints, irises, facial features, DNA and the like) or behavior characteristics (gait, keystroke habits and the like) of the human body.

303, comparing and identifying the first biological characteristic information with the second biological characteristic information to obtain an identification result;

After the first biological characteristic information and the second biological characteristic information are identified, comparing each characteristic information, and then obtaining an identification result and feeding back the identification result to the application information to be used as a basis for whether the next step is executed or not.

304, if the identification result is consistent, extracting the data content to be confirmed in the service data;

305, calling a preset AI voice conversion model to convert the data content into broadcast voice;

306, broadcasting the broadcast voice and acquiring the reply voice of the user in real time;

307, preprocessing the reply voice to obtain clean voice;

because the surrounding environment is uncontrollable and the relevant reply voice acquired by the system may be doped with environmental noise in the voice reply process of the user, the reply voice needs to be preprocessed. Among them, the preprocessing is mainly silence removal, noise processing and speech enhancement.

Firstly, a voice signal of the reply voice is extracted, voice and non-voice signal time periods are distinguished in the voice signal, a starting point of the voice signal is accurately determined, and then effective voice segments are detected from a continuous voice stream. The method comprises two aspects, namely detecting a starting point of effective voice, namely a front end point, and detecting an end point of the effective voice, namely a rear end point.

Then noise suppression is carried out, the spectrum characteristic of background noise is stabilized, the amplitude is very stable at a certain spectrum or a plurality of spectrums, a small section of background noise is assumed to be the background noise at the beginning, grouping and Fourier transformation are carried out from the initial background noise, and the grouping is averaged to obtain the spectrum of the noise. The noise reduction process is to obtain noise-reduced voice after carrying out reverse compensation on the noise-containing voice, and then eliminate the influence of environmental noise on the voice by utilizing spectral subtraction based on a short-time spectrum estimation enhancement algorithm and an improved form thereof.

308, extracting acoustic features of the clean voice to obtain acoustic feature parameters;

the received reply voice is preprocessed to obtain effective clean voice, voice signals and voice waveforms are extracted from the clean voice, and acoustic feature extraction is carried out on each frame of waveform, so that a multi-dimensional vector, namely acoustic feature parameters, can be obtained.

In the process of extracting acoustic characteristic parameters, pre-filtering is firstly carried out, then A/D conversion is carried out, pre-emphasis is carried out through a first-order finite excitation response high-pass filter, according to the short-time stable characteristic of voice, the voice can be subjected to frame division processing by taking a frame as a unit, a Hamming window is adopted to carry out windowing on a frame of voice so as to reduce the influence of Gibbs effect, then fast Fourier transform is carried out, a time domain signal is converted into a power spectrum of the signal, a group of Mel frequency scale is used for marking linearly distributed triangular window filters (24 triangular window filters in total), the power spectrum of the signal is filtered, logarithm is obtained based on the output of the triangular window filter group, the correlation among all dimensional signals is removed, the signal is mapped to a low-dimensional space, spectral weighting is carried out, low-order and high-order parameters are inhibited, cepstrum mean value reduction is carried out, differential parameters representing the dynamic characteristic of the voice are added into the, and finally obtaining acoustic characteristic parameters.

309, performing language processing on the acoustic characteristic parameters based on a preset language model to obtain reply data;

inputting the acoustic characteristic parameters into a language model, firstly calculating the distance between the characteristic vector sequence of the voice and each pronunciation template by the language model, then judging and correcting the grammar result and the semantics by using a tool in the language model, particularly determining the meaning of some homophones only through a context structure, and finally obtaining reply data.

310, comparing and matching the reply data with preset answer data to obtain a matching result;

and for the part of the insurance information needing to remind the applicant of confirmation, correct answer data is preset, relevant reply data is obtained after voice recognition processing, matching is carried out on the answer data and the applicant reply to obtain a matching result, and the applicant can continue insurance application only after the matching result is consistent. If the matching result is inconsistent, the system outputs prompt information to indicate that the user has errors in the verification information.

And 311, if the matching result is that the reply data is consistent with the answer data, outputting the service data.

And acquiring a matching result of the user reply data and the preset answer data, when the matching result is consistent, indicating that the user has finished checking the policy information, and confirming all contents of the policy information, wherein the system outputs user insurance application information generated after the user confirms the policy information, and the whole underwriting process is finished at this moment. And when the matching results are inconsistent, the user is indicated that the policy information is in doubt, or the policy information has wrong content and needs to be changed, the user can apply for interrupting the underwriting process and feed the condition back to the insurance company.

According to the embodiment of the invention, the data information to be confirmed in the policy information is converted into the voice information for broadcasting through the AI voice conversion technology, the confirmation reply voice of the current user is obtained, the reply of the current user is identified and verified by utilizing the voice identification technology, the data content to be confirmed is ensured to be clear for the current user, and the confirmation information is correct, so that the accuracy of the verification result is improved.

Referring to fig. 4, a fourth embodiment of the service data verification method according to the embodiment of the present invention includes:

401, acquiring service data to be checked;

if the business data to be verified is the insurance policy information applied by the applicant needing verification, the current business data verification process is the user's insurance verification process, and obtaining the business data to be verified is obtaining the insurance policy information to be verified. The insurance policy information comprises the data information of insurance policy number, the identity information of the insurance applicant, insurance type, insurance content and the like.

402, acquiring first biological characteristic information of a current user, and inquiring corresponding second biological characteristic information from a preset biological characteristic database according to user identity information in the service data;

403, comparing and identifying the first biological characteristic information with the second biological characteristic information to obtain an identification result;

404, if the identification result is consistent, extracting the data content to be confirmed in the service data;

405, calling a preset AI voice conversion model to convert the data content into broadcast voice;

406, broadcasting the broadcast voice and acquiring the reply voice of the user in real time;

407, performing voice recognition on the reply voice based on a voice recognition technology to obtain reply data;

after the reply voice of the user is acquired, voice recognition processing is carried out on the reply voice by utilizing a voice recognition technology, corresponding characteristic parameters are recognized from the voice waveform according to the voice waveform corresponding to the reply voice, and then the voice is converted into reply data according to the characteristic parameters.

408, extracting characters of each byte in the answer data to obtain a first character;

in the process of matching answer data with reply voice, firstly storing the answer data in an answer data storage window, extracting characters of each byte in the answer data to obtain a first character, and storing the first character in a random access memory in the answer data storage window.

The random access memory of the answer data storage window is 16 individually addressable distributed random access memories, wherein each individually addressable distributed random access memory has two independent read ports.

The answer data storage window starts from the 1 st individually addressable distributed random access memory, the first character is sequentially placed into the 16 individually addressable distributed random access memories, if the first character is placed into the last 1 individually addressable distributed random access memory and then the first character which is not placed is still present, the first character which is not placed is sequentially placed into the 16 individually addressable distributed random access memories again, and the operation is sequentially circulated until all the first characters of the preset bytes are placed into the 16 individually addressable distributed random access memories, wherein each individually addressable distributed random access memory stores at most 1 kilobyte of the first character.

409, extracting characters of each byte in the reply data to obtain second characters;

and performing voice recognition on the reply voice by using a voice recognition technology, obtaining clean voice after preprocessing, then extracting acoustic features, performing mode matching and language processing on the extracted acoustic features through a preset acoustic model and a preset language model, and extracting each character in the reply voice to obtain a second character.

410, performing hash calculation on the second characters to obtain a character index corresponding to each second character;

calculating a hash value of the second character in a hash calculation unit, determining a character index corresponding to the second character according to the hash value, and storing the character index in a random memory of the hash calculation unit, wherein the character index is position information in the random memory of the answer data window corresponding to the second character.

411, comparing and matching the first character and the second character based on the character index to obtain a matching result;

generating and storing a read address of an answer data storage window according to the character index, reading and storing 32-byte first characters from 16 individually addressable distributed random access memories in the answer data storage window according to the read address, and selecting and storing effective 26-byte first characters from the 32-byte first characters according to the lower 4 bits of the read address; in the initiated matching search command, matching a 4-byte second character stored in the lower 24 bits of the shift register with a 4-byte first character in a 26-byte first character according to the character index; if the 4-byte second character is matched and hit, at least two character indexes corresponding to the current second character are obtained in the second period, matching operation is carried out on the current second character in the 16-23 bits of the shift register and the corresponding 2-byte first character in the 26-byte first character according to the at least two character indexes corresponding to the current second character, if the current second character is successfully matched, matching is continued to be carried out on the next current second character, and the like is carried out until the matching is finished, and if the current second character is failed to be matched, the matching is finished.

And 412, if the matching result is consistent, outputting the service data.

In the embodiment of the invention, the voice recognition technology is utilized to recognize the acquired reply voice of the current user, and the verification and comparison are carried out according to the preset answer data, so that the confirmation reply of the current user is real and accurate, the reliability of the verification is improved, and the authenticity and the accuracy of the verification result are also improved.

Referring to fig. 5, a fifth embodiment of the service data verification method according to the embodiment of the present invention includes:

501, acquiring service data to be checked;

502, acquiring first biological characteristic information of a current user, and inquiring corresponding second biological characteristic information from a preset biological characteristic database according to user identity information in service data;

503, comparing and identifying the first biological characteristic information and the second biological characteristic information to obtain an identification result;

504, if the identification result is consistent, extracting the data content to be confirmed in the service data;

505, calling a preset AI voice conversion model to convert the data content into broadcast voice;

506, broadcasting the broadcast voice and acquiring the reply voice of the user in real time;

507, matching the reply voice based on preset answer data to obtain a matching result;

508, if the matching result is that the reply voice is consistent with the answer data, outputting the service data;

509, acquiring video information of the service data verification process in real time based on a remote communication technology;

when the service data verification process is an underwriting process, the video information of the service data verification process is acquired, namely after the policy information is confirmed, the video information of the user in the underwriting process can be acquired in real time based on a remote communication technology, specifically, the remote communication technology is mainly realized by utilizing the camera function of the mobile terminal, namely when the user starts the underwriting process, the camera function of the mobile terminal is started, and the whole underwriting process is recorded.

510, monitoring the verification behavior in the business data verification process based on a preset business data verification safety rule to obtain monitoring information;

when the business data verification process is a verification process, the business data verification safety rule is a verification safety rule, and the verification behavior is a verification behavior. Monitoring the user's underwriting behavior in the process of acquiring the video information of the underwriting process in real time, monitoring the underwriting behavior which does not accord with the regulations in the underwriting process according to the regulated underwriting safety rules, and making corresponding prompts by utilizing the voice communication function of remote communication in time. The real-time monitoring process mainly comprises the steps that a camera of a mobile terminal is used for collecting real-time state videos of the underwriting process in real time and sending the videos to corresponding insurance company related personnel, and then the insurance company related personnel monitor the whole underwriting process in real time.

511, carrying out voice prompt on the verification behavior which does not accord with the business data verification safety rule in the monitoring information;

the real-time monitoring of the underwriting process mainly utilizes the mobile terminal camera shooting function of the applicant to carry out remote communication, and carries out voice prompt on underwriting behaviors which do not accord with the underwriting safety regulations in time. The insurance behavior not conforming to the insurance safety regulations mainly means that an applicant leaves a camera shooting range in the insurance application process or does interference by irrelevant personnel in the insurance application process, when the insurance behavior not conforming to the insurance safety regulations is monitored in real time, the user is reminded in real time by using a voice communication technology, specifically, if the user is found to have behavior not conforming to the safety regulations in the monitoring process or third person interferes in the insurance application process, related personnel of an insurance company can remind the user in time, and the insurance application process can be interrupted according to the condition, so that the insurance behavior of the user in the whole insurance application process is safe and effective, related video records are made in time, and the insurance application process is convenient to trace back subsequently.

And 512, recording the service data verification process, generating a service data verification process video, and storing the service data verification process video in a storage unit.

In the whole underwriting process, the whole underwriting process is recorded by utilizing the camera shooting function of the mobile terminal, an underwriting process video is generated, and the underwriting process video is stored in the storage unit. When the follow-up tracing of the underwriting process is required, the underwriting process video can be extracted from the storage unit.

In the embodiment of the invention, the identity of the current user is identified and verified by combining the face identification technology, the AI voice conversion technology and the remote communication technology, the identity of the current user is ensured to be consistent with the identity of the user in the service data to be verified, the AI voice conversion technology in the AI voice conversion model is utilized to convert the part needing attention in the service data to be verified into the broadcast voice for broadcasting, and the remote communication technology is utilized to monitor the whole service data verification process, so that the safety and the compliance of the service data verification process are ensured to a great extent, and the accuracy and the reliability of the verification result are improved.

In the above description of the service data verification method in the embodiment of the present invention, the service data verification apparatus in the embodiment of the present invention is described below with reference to fig. 6, and an embodiment of the service data verification apparatus in the embodiment of the present invention includes:

an obtaining module 601, configured to obtain service data to be verified;

the query module 602 is configured to obtain first biometric information of a current user, and query, according to user information in the service data, corresponding second biometric information from a preset biometric database;

the identification module 603 is configured to compare the first biometric characteristic information with the second biometric characteristic information and identify the first biometric characteristic information and the second biometric characteristic information to obtain an identification result;

an extracting module 604, configured to extract data content to be confirmed in the service data if the identification result is consistent;

the voice conversion module 605 is configured to call a preset AI voice conversion model, and convert the data content into broadcast voice;

the broadcasting module 606 is used for broadcasting the broadcasting voice and acquiring the reply voice of the user in real time;

a matching module 607, configured to match the reply voice based on preset answer data to obtain a matching result;

an information output module 608, configured to output the service data if the matching result is that the reply voice is consistent with the answer data.

In the embodiment of the invention, the business data verification device is used for executing the steps of the business data verification method, the business data with verification and the biological characteristic information of the user are obtained, the biological characteristic information corresponding to the business data is inquired from the biological characteristic database and is compared to ensure that the user identity is consistent with the user identity in the business data to be verified, an AI voice conversion model is used for converting the data content to be confirmed in the business data into broadcast voice, the user is played, the reply voice of the user is obtained, the reply voice is matched based on the answer data, and if the matching is consistent, the business data is output. The device runs the steps of the business data verification method of the embodiment, so that the authenticity and the reliability of the verification process are improved, and the accuracy of the verification result is also improved.

Referring to fig. 7, another embodiment of the service data verification apparatus according to the embodiment of the present invention includes:

an obtaining module 601, configured to obtain service data to be verified;

the query module 602 is configured to acquire first biometric information of a current user, and query, according to user identity information in the service data, corresponding second biometric information from a preset biometric database;

Optionally, the identifying module 603 includes:

a first feature extraction unit 6031, configured to perform face information feature extraction based on the first biological feature information to obtain a first feature value of a face;

a second feature extraction unit 6032, configured to perform face information feature extraction based on the second biological feature information to obtain a second feature value of the face;

and the feature value comparison unit 6033 is configured to compare and identify the face comparison threshold, the face first feature value, and the face second feature value based on a preset face comparison threshold, and obtain an identification result.

Optionally, the characteristic value comparing unit 6033 is specifically configured to:

Optionally, the matching module 607 includes:

a voice recognition unit 6071, configured to perform voice recognition on the reply voice based on a voice recognition technology to obtain reply data;

and a data matching unit 6072, configured to compare and match the reply data with preset answer data to obtain a matching result.

Optionally, the speech recognition unit 6071 is specifically configured to:

preprocessing the reply voice to obtain clean voice;

Optionally, the data matching unit 6072 is specifically configured to:

Optionally, the monitoring module 609 is specifically configured to:

and recording the service data verification process to generate a video of the verification process, and storing the video of the service data verification process in a storage unit.

In the embodiment of the invention, a monitoring module is added in the business data verification device and is used for acquiring the video information of the business data verification process in real time, monitoring the verification behavior and prompting the verification behavior which does not accord with the business data verification safety rule in the monitoring process in time. The device improves the authenticity of the service data verification process, realizes the monitoring of verification behaviors, and simultaneously improves the accuracy and reliability of verification results.

Fig. 6 and fig. 7 describe the service data verification apparatus in the embodiment of the present invention in detail from the perspective of the modular functional entity, and the service data verification apparatus in the embodiment of the present invention is described in detail from the perspective of hardware processing.

Fig. 8 is a schematic structural diagram of a service data verification apparatus according to an embodiment of the present invention, where the service data verification apparatus 800 may generate relatively large differences due to different configurations or performances, and may include one or more processors (CPUs) 810 (e.g., one or more processors) and a memory 820, and one or more storage media 830 (e.g., one or more mass storage devices) storing an application 833 or data 832. Memory 820 and storage medium 830 may be, among other things, transient or persistent storage. The program stored in the storage medium 830 may include one or more modules (not shown), and each module may include a series of instructions for the business data verifying apparatus 800. Further, processor 810 may be configured to communicate with storage medium 830, and execute a series of instruction operations in storage medium 830 on business data validation apparatus 800.

The business data validation apparatus 800 may also include one or more power supplies 840, one or more wired or wireless network interfaces 880, one or more input-output interfaces 860, and/or one or more operating systems 831, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, and the like. Those skilled in the art will appreciate that the business data validation device architecture illustrated in FIG. 8 does not constitute a limitation of business data validation devices, and may include more or fewer components than illustrated, or some components may be combined, or a different arrangement of components.

The block chain is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, a consensus mechanism, an encryption algorithm and the like. A block chain (Blockchain), which is essentially a decentralized database, is a series of data blocks associated by using a cryptographic method, and each data block contains information of a batch of network transactions, so as to verify the validity (anti-counterfeiting) of the information and generate a next block. The blockchain may include a blockchain underlying platform, a platform product service layer, an application service layer, and the like.

The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium, and which may also be a volatile computer-readable storage medium, having stored therein instructions, which, when run on a computer, cause the computer to perform the steps of the business data validation method.

It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.

The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a read-only memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims

1. A service data verification method is characterized in that the service data verification method comprises the following steps:

acquiring service data to be checked;

2. The business data verification method of claim 1, wherein the comparing and identifying the first biometric information and the second biometric information to obtain an identification result comprises:

3. The business data verification method according to claim 2, wherein the comparing and recognizing the face comparison threshold, the face first feature value and the face second feature value based on a preset face comparison threshold to obtain a recognition result comprises:

4. The service data verification method according to any one of claims 1 to 3, wherein the matching the reply voice based on preset answer data to obtain a matching result comprises:

5. The service data verification method according to claim 4, wherein the performing voice recognition on the reply voice based on the voice recognition technology to obtain the reply data comprises:

preprocessing the reply voice to obtain clean voice;

6. The business data verification method according to claim 4, wherein the comparing and matching the reply data with the preset answer data to obtain the matching result comprises:

7. The service data verification method according to any one of claims 1 to 3, further comprising, after the obtaining the service data to be verified:

8. A service data verification apparatus, wherein the service data verification apparatus comprises:

the acquisition module is used for acquiring the service data to be checked;

9. A service data verification device, characterized in that the service data verification device comprises:

a memory having instructions stored therein and at least one processor, the memory and the at least one processor interconnected by a line;

the at least one processor invoking the instructions in the memory to cause the business data validation device to perform the steps of the business data validation method of any of claims 1-7.

10. A computer-readable storage medium having instructions stored thereon, which when executed by a processor implement the steps of the business data verification method according to any one of claims 1-7.