CN120148519A

CN120148519A - A business abnormality early warning method and device based on voiceprint technology

Info

Publication number: CN120148519A
Application number: CN202510318313.XA
Authority: CN
Inventors: 房创辉
Original assignee: China Life Insurance Co ltd
Current assignee: China Life Insurance Co ltd
Priority date: 2025-03-18
Filing date: 2025-03-18
Publication date: 2025-06-13

Abstract

The present application discloses a method and device for warning of business anomalies based on voiceprint technology, which solves the problem that the existing technology cannot accurately identify the identity of customers, and has low recognition and processing efficiency for malicious complaints. The method for warning of business anomalies based on voiceprint technology includes the steps of: collecting voiceprint features and corresponding customer information, judging whether there are preset sensitive words in the voiceprint features according to preset rules or algorithms; in response to the presence of preset sensitive words in the first voiceprint feature, adding a classification mark to the first voiceprint feature, and saving the first voiceprint feature and the corresponding classification mark in the voiceprint database. The present application uses voiceprint technology in the insurance industry. In the complaint application, insurance claim and cancellation links, voiceprint technology can be used to more accurately identify the identity of customers and improve the recognition and processing efficiency of malicious complaints.

Description

Business abnormality early warning method and device based on voiceprint technology

Technical Field

The application relates to the technical field of identity recognition, in particular to a business abnormality early warning method and device based on voiceprint technology.

Background

Voiceprint technology is a technology that uses voice signals for personal identification. The voice print model is built by extracting the characteristics of frequency spectrum, formants and the like of personal voice and is compared with the existing voice print database, so that the aim of personal identity authentication is fulfilled. Compared with the traditional identity authentication mode, the voiceprint technology has unique advantages such as difficult impossibility, difficult theft and the like. Voiceprint recognition technology is a voiceprint recognition technology that recognizes the identity of an individual by analyzing and comparing voice signals of the individual. Compared with other voiceprint recognition technologies, the voiceprint recognition technology has the characteristics of wide applicability, no need of contact, non-invasiveness and the like. The existing voiceprint recognition technology mainly comprises three main steps of feature extraction, model training and comparison. In the feature extraction stage, the speech signal is first preprocessed, e.g., to remove noise, speech segments, etc., and features are then extracted from the speech signal, common features including mel-frequency cepstral coefficients (MFCCs), etc. In the model training stage, a large number of voiceprint samples are utilized for training and optimizing the model, and the mainly adopted model comprises a Gaussian Mixture Model (GMM), a Support Vector Machine (SVM), a deep learning model and the like. In the comparison stage, whether the identities of the individuals are matched or not is judged by carrying out similarity calculation or matching on the voiceprint features to be identified and the known voiceprint model.

At present, in links such as complaint application, insurance claim settlement and refund in the insurance industry, the problems that the identity of a client cannot be accurately identified and the identification degree and the processing efficiency of malicious complaints are low exist. There is a need for a solution that applies voiceprint technology to the insurance industry.

Disclosure of Invention

The application provides a business anomaly early warning method and device based on a voiceprint technology, which solve the problems that the identity of a customer cannot be accurately identified and the identification degree and the processing efficiency of malicious complaints are low in the prior art.

In a first aspect, an embodiment of the present application provides a method for early warning of service abnormality based on voiceprint technology, including the steps of:

collecting voiceprint characteristics and corresponding client information, and judging whether preset sensitive words exist in the voiceprint characteristics according to preset rules or algorithms;

and adding a classification mark to the first voiceprint feature in response to the existence of a preset sensitive word in the first voiceprint feature, and storing the first voiceprint feature and the corresponding classification mark in a voiceprint database.

Further, the method further comprises the steps of:

Collecting voiceprint features in complaint information and comparing the voiceprint features with voiceprint features stored in a voiceprint database;

in response to the second voice print feature in the complaint information not being present in the voice print database, storing the second voice print feature in the voice print database and establishing customer information and information associated with the second voice print feature.

Further preferably, the method further comprises the steps of:

and carrying out natural language processing on the complaint information, and filtering malicious complaints.

In one embodiment, the method further comprises the steps of:

and collecting voiceprint features in insurance transaction information, determining the corresponding relation between the voiceprint features and preset customer emotion according to a preset rule or algorithm, and outputting an analysis result representing the customer emotion features.

Further, the method further comprises the steps of:

And determining the corresponding relation between the analysis result of the emotion of the client and the client risk assessment value and the insurance pricing according to a preset rule or algorithm, and outputting client risk assessment and insurance pricing data.

In one embodiment, the method further comprises the steps of:

Collecting voiceprint features of a customer in complaint information, comparing the voiceprint features with the voiceprint features stored in a voiceprint database, and sending alarm information in response to the existence of a first voiceprint feature in the complaint information in the voiceprint database and the existence of a classification mark of the first voiceprint feature.

In one embodiment, the method further comprises the steps of:

collecting voiceprint features and corresponding client information in real time in any client program;

comparing the collected client information with the client information stored in the voiceprint database;

And sending alarm information in response to the fact that the customer information exists in the voiceprint database and voiceprint characteristics corresponding to the customer information are different from those stored in the voiceprint database.

Further, the method further comprises the steps of:

In a second aspect, an embodiment of the present application further provides a service abnormality early-warning device based on a voiceprint technology, which is configured to implement the service abnormality early-warning method based on the voiceprint technology according to any one of the embodiments of the first aspect, where the service abnormality early-warning device includes an acquisition module configured to acquire voiceprint features and corresponding client information. The voice print processing device comprises a voice print database, a determining module, a classifying mark adding module and a voice print processing module, wherein the voice print database is used for storing voice print characteristics and corresponding classifying marks, the determining module is used for judging whether the voice print characteristics have preset sensitive words according to preset rules or algorithms, and the classifying marks are added to the first voice print characteristics in response to the existence of the preset sensitive words of the first voice print characteristics.

In a third aspect, embodiments of the present application also provide a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a method according to any of the embodiments of the first aspect.

In a fourth aspect, an embodiment of the present application further provides an electronic device, including a memory, a processor and a computer program stored on the memory and executable by the processor, where the processor executes the computer program to implement a method according to any embodiment of the first aspect.

The above at least one technical scheme adopted by the embodiment of the application can achieve the following beneficial effects:

the application uses the voiceprint technology in insurance industry, in the links of complaint application, insurance claim settlement, refund and the like, the identity of a customer can be more accurately identified through the voiceprint technology, and the identification degree and the processing efficiency of malicious complaints are improved. Meanwhile, by combining with natural language processing technology, complaint contents can be classified and analyzed, and more refined services are provided for insurance companies. The voiceprint recognition technical scheme aims to solve the limitation of the existing complaint application processing mode and improve the processing accuracy and efficiency.

Drawings

The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute a limitation on the application. In the drawings:

FIG. 1 is a system diagram of an application scenario of the present application;

FIG. 2 is a flowchart of an abnormality pre-warning method for insurance business handling provided by an embodiment of the present application;

FIG. 3 is a flowchart of an abnormality pre-warning method for handling insurance business including judgment of sensitive words according to an embodiment of the present application;

FIG. 4 is a flowchart of an alarm method for classifying and marking complaint information provided by an embodiment of the present application;

FIG. 5 is a schematic diagram of an abnormal insurance business handling early warning process according to an embodiment of the present application;

FIG. 6 is a block diagram of an abnormality pre-warning device for insurance business handling according to an embodiment of the present application;

fig. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions of the present application will be clearly and completely described below with reference to specific embodiments of the present application and corresponding drawings. It will be apparent that the described embodiments are only some, but not all, embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.

The following describes in detail the technical solutions provided by the embodiments of the present application with reference to the accompanying drawings.

Fig. 1 is a system diagram of an application scenario of the present application, including a terminal 11/12, a server 2, and a storage device 3 (storing a voiceprint database). The server and the terminal (fixed network terminal, mobile terminal) are combined into an insurance service processing system.

The terminal comprises a first terminal 11 and a second terminal 12, wherein the first terminal initiates business processes such as complaints, claims, insurance application, refund and the like, and sends insurance business information and customer voice information to the server.

The storage device stores a voiceprint database, preferably a classification mark database, and can be used for training an intelligent voiceprint recognition system to determine malicious complaints and non-malicious complaints.

The voiceprint database comprises first voiceprint features, the first voiceprint features are collected customer voiceprint historical data, the first voiceprint features comprise classification marks, and the classification marks comprise presence-sensitive words and non-presence-sensitive words.

And the second voiceprint characteristics are not existed in the voiceprint database, are not acquired voiceprint characteristics, and are supplemented into the voiceprint database along with the acquisition of the second voiceprint characteristics. The second voice characteristic is supplemented into the voice characteristic database along with client information corresponding to the second voice characteristic and corresponding classification marks of whether the second voice characteristic contains sensitive words.

The second terminal is a customer service terminal, receives malicious complaint alarm information, processes real complaints and the like.

The server realizes voiceprint processing, malicious complaint recognition, insurance service pricing processing and insurance service risk assessment processing;

The server and terminal system of the present application load the computer program product of the method of the present application. The computer program product comprises a computer program or instructions which, when executed by a processor, implement a method according to any of the embodiments of the present application.

Fig. 2 is a flowchart of a business anomaly early warning method based on a voiceprint technology, which includes steps 110 to 120.

Step 110, collecting voiceprint features and corresponding client information, and judging whether preset sensitive words exist in the voiceprint features according to preset rules or algorithms.

For example, collecting first voiceprint features and corresponding client information, and judging whether preset sensitive words exist in the first voiceprint features according to preset rules or algorithms;

And voice sampling and voiceprint feature extraction are carried out, and voice samples of customers are obtained by carrying out voice sampling on the customers when complaint applications are processed. Then, based on the sound samples, voiceprint features of the customer, such as spectrum, formants, etc., are extracted.

Because the identity of the customer is queried during the complaint application or other forms of communication with the customer, the corresponding customer information can be obtained while the first voiceprint feature is collected.

And 120, adding a classification mark to the first voiceprint feature in response to the existence of a preset sensitive word in the first voiceprint feature.

Acquiring a first voiceprint feature, and judging the existence of preset sensitive words in the first voiceprint feature. And judging the first voiceprint feature with the preset sensitive word and adding a classification mark.

The preset sensitive words can be words related to complaints, refunds and claims.

For example, by machine learning, a sensitive word recognition model is constructed, and by inputting a sample of the customer's speech into the sensitive word recognition model, it can be determined whether the corresponding sensitive word exists in the customer's speech.

And (3) an updated voiceprint database, wherein voiceprint features of which the existence of the sensitive words is determined are marked.

Accurately and stably extracting the first voiceprint feature in the customer's voice is a key to achieving accurate recognition. This technique requires that the customer's voice be sampled and voiceprint features with distinguishing features be extracted from it.

For example, the related vocabulary of the malicious complaints is preset as sensitive words, the first voiceprint feature of the malicious complaints is determined, and the first voiceprint feature is added into a blacklist.

It should be noted that, steps 110 to 120 are not only applicable to customer voice sampling historical data, but also applicable to customer voice sampling real-time data, and the processing of the historical data generates a blacklist in the voiceprint database.

It should be further noted that the voiceprint database and the blacklist are 2 concepts, and customer information and corresponding voiceprint features are provided in the voiceprint database, wherein classification marks are provided on some voiceprint features, and some marks represent malicious complaints. And the voiceprint features with the malicious complaint marks form a blacklist.

In one embodiment, the method further comprises the steps of:

And sending alarm information in response to the existence of the classification mark of the first voiceprint feature.

For example, when the complaint information is received, the voiceprint feature of the complaint information is found to be the first voiceprint feature added to the blacklist, and alarm information is sent.

The alarm information can be displayed on a display of the customer service personnel through a GUI, and can also be sent out in a luminous mode through an indicator lamp to remind the customer service personnel that the current business has risks.

In one embodiment, the method further comprises the steps of:

comparing the collected client information with the client information stored in the voiceprint database to determine whether the client information exists in the voiceprint database;

It should be noted that, the voice print database stores historical samples of the client voice, so that the voice print database includes both voice print features and corresponding user information.

By comparing the customer information collected by the current sampling with the customer information in the voiceprint database, whether the customer of the current sampling exists in the voiceprint database or not can be known, and after the customer information collected by the current sampling exists in the voiceprint database, whether the customer of the current sampling is the owner or not is determined by comparing the historical sampling of the customer voice and the voiceprint characteristics of the current sampling.

And sending alarm information in response to the existence of the client information in the voiceprint database, wherein the first voiceprint characteristics are different from voiceprint characteristics corresponding to the client information stored in the voiceprint database.

Fig. 3 is a flowchart of an early warning method for handling abnormality of insurance business including judgment of sensitive words according to an embodiment of the present application.

Further, before judging whether the preset sensitive word exists, the method further comprises the steps of:

step 210, comparing the voiceprint features in the collected complaint information with the voiceprint features stored in the voiceprint database.

For example, collecting a second voiceprint feature and comparing the second voiceprint feature to the first voiceprint feature, determining whether the second voiceprint feature is present in the voiceprint database;

Wherein the first voiceprint feature is stored in a voiceprint feature database, the method specifically comprises the following steps:

Acquiring characteristic data, inputting the characteristic data into a first processing module, and outputting a voiceprint database with first voiceprint characteristics;

the feature data comprises a customer voice history sample (i.e., voiceprint features stored in a voiceprint database) including a first voiceprint feature;

The first processing module is used for determining the corresponding first voiceprint characteristics of the client voice history samples according to a preset rule or algorithm and storing the first voiceprint characteristics in a voiceprint database.

For example, by integrating historical data of the client voice of the historical samples of the client voice, voiceprint data collection is performed on voice samples of different clients, so as to construct a voiceprint database.

For example, customer voiceprint data may be collected by telephone recording or an online complaint system and stored in association with complaint information.

And in the subsequent processing process of the insurance business, determining the identity of the client by comparing the voiceprint characteristics of the client with data in the voiceprint database through the association of the voiceprint database and the client information.

For example, a voiceprint recognition engine is integrated in the system and is used for recognizing and verifying voiceprint data of clients, so that accuracy and authenticity of complaint information are ensured.

Acquiring a first data set, inputting a voiceprint database with a first voiceprint characteristic and the first data set into a second processing module, and outputting a second data set;

the first data set contains a current sample of the collected customer speech (i.e., voiceprint features in the complaint information), which contains second voiceprint features.

It should be noted that, in the present application, the first voiceprint feature refers to a voiceprint feature collected in the history, and the second voiceprint feature refers to a voiceprint feature collected currently.

For example, a customer contacts an insurance company via a telephone to discuss about business related matters such as complaint application, insurance claim or refund, and extracts voiceprint information in the current sample of the customer voice obtained by answering the telephone.

The second processing module is used for comparing the second voiceprint characteristics of the current sample of the client voice with the first voiceprint characteristics in the voiceprint database and searching for the first voiceprint characteristics consistent with the second voiceprint characteristics.

And comparing the second voiceprint feature in the current sample of the extracted client voice with the first voiceprint feature in the voiceprint database, and judging whether the first voiceprint feature in the voiceprint database is consistent with the second voiceprint feature in the current sample of the client voice. If the first voiceprint feature is consistent with the second voiceprint feature, the second voiceprint feature is considered a client present in the voiceprint database.

The second data set includes a comparison of whether there is a voiceprint in the voiceprint database that is currently sampled by the client voice.

The comparison result comprises the second voiceprint feature in the voiceprint database and also comprises the second voiceprint feature in the voiceprint database.

Step 220, in response to the second voice print feature in the complaint information not being present in the voice print database, storing the second voice print feature in the voice print database and establishing information associated with the customer information and the second voice print feature.

The absence of the second voiceprint feature in the voiceprint database indicates that the client is not an existing client.

By establishing a malicious complaint model, known malicious complaints are marked and classified and matched with the first voiceprint features, so that the malicious complaints are distinguished from real complaints, and the processing efficiency and accuracy are improved.

For example, in response to the second voice print feature not being present in the voice print database, the second voice print feature is stored in the voice print database and customer information and information associated with the second voice print feature is established.

The second voiceprint features that do not exist in the voiceprint database create a new entry in the voiceprint database, namely the voiceprint features and corresponding customer information.

And comparing the second voiceprint characteristics of the current sample of the customer voice, finding the second voiceprint characteristics which do not exist in the voiceprint database, and recording the second voiceprint characteristics into the voiceprint database as new customer voiceprint characteristics.

If the second voice characteristic in the current sample of the client voice exists in the voice print database, the second voice characteristic is considered to be the existing client.

Further preferably, in step 230, natural language processing is performed on the complaint information to filter malicious complaints.

Step 240, step 110, collecting voiceprint features and corresponding client information, and judging whether preset sensitive words exist in the voiceprint features according to preset rules or algorithms;

step 250, step 120, in response to the presence of a preset sensitive word in the first voiceprint feature, adding a classification mark to the first voiceprint feature, and storing the first voiceprint feature and the corresponding classification mark in a voiceprint database.

In one embodiment, as shown in fig. 4, before determining whether the preset sensitive word exists, the method further includes the steps of:

Step 310, step 110, collecting voiceprint features and corresponding client information, and judging whether preset sensitive words exist in the voiceprint features according to preset rules or algorithms.

Step 320, step 120, in response to the presence of a preset sensitive word in the first voiceprint feature, adding a classification mark to the first voiceprint feature, and storing the first voiceprint feature and the corresponding classification mark in a voiceprint database.

Step 330, collecting the voiceprint characteristics of the customer in the complaint information, comparing the voiceprint characteristics with the voiceprint characteristics stored in the voiceprint database, and sending alarm information in response to the existence of the first voiceprint characteristics in the complaint information in the voiceprint database and the existence of the classification mark in the first voiceprint characteristics.

And sending alarm information in response to the first voiceprint feature in the voiceprint database carrying the classification mark.

And obtaining a customer voice sample, obtaining a second voice characteristic of the customer voice sample, comparing the voice characteristic with a voice database, judging that the second voice characteristic exists in the voice database, and directly generating classification information to remind customer service if the second voice characteristic is provided with a classification mark.

It should be noted that in one embodiment of the two embodiments, customer voices are sampled, set sensitive words are identified for the sampled second voice characteristics, if the second voice characteristics are found to have the sensitive words, a classification information prompt is sent to customer service, and meanwhile, classification marks are added for the voice characteristics and a blacklist is added.

Another embodiment is to sample the customer speech, find that the second voiceprint feature of the sample is present in the voiceprint database and has a flag, which indicates that the voiceprint feature has a history of uttering the set-sensitive word and is flagged, and then issue a classification information reminder to the customer service regardless of whether the first voiceprint feature is recognized to utter the set-sensitive word this time.

For example, by machine learning, a sensitive word recognition processing module is constructed, and by inputting the customer speech samples into the sensitive word recognition processing module, it can be determined whether the corresponding sensitive word exists in the customer speech.

And adding marks to the first voiceprint features in the voiceprint database to generate classification information in response to the existence of preset sensitive words in the first voiceprint features.

In the process of sampling customer voices and constructing a voiceprint database, identifying sensitive words in the customer voices, marking first voiceprint features of the identified sensitive words, and adding a voiceprint blacklist.

For example, a mark is added for a first voiceprint feature of a set sensitive word, a voiceprint blacklist is added, the voiceprint feature is classified and recorded to enter the blacklist, and classification information is generated.

The classification information can be to remind the customer service of carrying out voice call with the customer from the display interface or the voice prompt aspect, so as to remind the customer service of carrying out abnormal treatment.

In one embodiment, the method further comprises the steps of:

Collecting voiceprint features in insurance transaction information, determining the corresponding relation between the voiceprint features and preset customer emotion according to a preset rule or algorithm, and outputting an analysis result representing the customer emotion features;

And determining the corresponding relation between the second voice pattern characteristics and the preset customer emotion according to a preset rule or algorithm, and outputting an analysis result of the customer emotion.

Acquiring a voiceprint database with first voiceprint characteristics, inputting the voiceprint database into a fourth processing module, and outputting a fifth data set;

The fourth processing module is used for determining the corresponding relation between the voiceprint database with the first voiceprint characteristics or the first data set and the preset customer emotion.

The fifth data set contains analysis results of the emotion of the customer.

For example, voiceprint recognition technology can be combined with natural language processing and other technologies to conduct intelligent analysis and emotion recognition on customer complaint information, so that insurance companies can better understand customer requirements and solve problems.

The insurance company can monitor and evaluate the call quality of the telephone sales personnel by utilizing the voiceprint recognition technology, so that the sales efficiency and the service quality are improved. Meanwhile, emotion and demand of the clients can be analyzed through voiceprint recognition technology, so that more intelligent client communication is realized.

Further, the method further comprises the steps of:

And the fifth processing module is used for determining the corresponding relation between the analysis result of the emotion of the customer and the customer risk assessment value and the insurance pricing according to a preset rule or algorithm.

The sixth data set is customer risk assessment and insurance pricing data.

The voiceprint recognition technology can be combined with other data sources, such as voice emotion analysis, speech speed and the like, and is used for customer risk assessment and insurance pricing, so that risk management capability of insurance companies is improved.

In one embodiment, the method further comprises the step of collecting voiceprint features and corresponding customer information in real time in any one of the customer service programs.

For example, as shown in fig. 5, the customer service answers the call of the customer, and compares the voiceprint features in the answering process, if the corresponding voiceprint feature is found in the voiceprint database, the identity of the customer can be directly confirmed, if the voiceprint feature is not found, the customer is indicated to be a new customer, the current sample of the customer voice is extracted, and the new voiceprint feature is collected.

In one embodiment, the method further comprises the step of comparing the collected customer information with customer information stored in the voiceprint database.

And for the clients with confirmed identities, distinguishing whether the clients have history records for speaking the set sensitive words or not according to whether the corresponding voiceprint features in the voiceprint database are marked, and if so, directly sending classification information to customer service.

If the history record of the sensitive word is not set or a new customer is not set, identifying the voiceprint feature of the conversation, judging whether the set sensitive word exists or not, if the set sensitive word exists, marking the voiceprint feature, and sending classification information to customer service.

If the set sensitive word does not exist, no classification information is sent out.

Fig. 6 is a block diagram of a business anomaly early warning device based on a voiceprint technology according to an embodiment of the present application, which is configured to implement the business anomaly early warning method based on the voiceprint technology according to any one of the embodiments of the first aspect, where the business anomaly early warning method includes:

an acquisition module 410, configured to acquire voiceprint features and corresponding client information;

The determining module 420 is configured to determine whether a preset sensitive word exists in the voiceprint feature according to a preset rule or algorithm, and further configured to add a classification mark to the first voiceprint feature in response to the existence of the preset sensitive word in the first voiceprint feature, and store the first voiceprint feature and the corresponding classification mark in a voiceprint database.

In one embodiment, the acquiring module includes a first acquiring unit configured to acquire voiceprint features and corresponding client information.

The determining module comprises a first determining unit which is used for judging whether preset sensitive words exist in the voiceprint features according to a preset rule or algorithm. The voice print processing method further comprises a second determining unit, wherein the second determining unit is used for responding to the fact that the first voice print feature has preset sensitive words, adding classification marks to the first voice print feature, and storing the first voice print feature and the corresponding classification marks in a voice print database. .

In one embodiment, the determining module further includes a third determining unit, configured to collect the first voiceprint feature and compare the first voiceprint feature with voiceprint features stored in the voiceprint database, and determine whether the first voiceprint feature exists in the voiceprint database.

The method also comprises a fourth determining unit, which is used for responding to the fact that the first voiceprint feature does not exist in the voiceprint database, storing the first voiceprint feature in the voiceprint database and establishing information of the association of the client information and the first voiceprint feature.

The above embodiment is used to implement any one of the steps 210 to 250.

In one embodiment, the output module 430 further includes a first output unit, configured to determine a correspondence between the first voiceprint feature and a preset customer emotion according to a preset rule or algorithm, and output an analysis result of the customer emotion.

The foregoing embodiment is configured to implement any one of the first voiceprint feature of the embodiment and a preset customer emotion.

In one embodiment, the system further comprises an output module, wherein the output module comprises a second output unit for sending alarm information in response to the existence of the classification mark of the first voiceprint feature.

The above embodiment is configured to implement any one of the first voiceprint feature presence classification mark embodiments in step 120.

In one embodiment, the determining module further includes a fifth determining unit, configured to compare the collected client information with client information stored in the voiceprint database, and determine whether the client information exists in the voiceprint database.

The output module comprises a third output unit and is used for responding to the existence of the client information in the voiceprint database, and the first voiceprint characteristics are different from voiceprint characteristics corresponding to the client information stored in the voiceprint database, and sending alarm information.

The above embodiment is used to implement any one of the comparison embodiments with respect to the client information stored in the voiceprint database in step 120.

In one embodiment, the system further comprises an output module, and the output module comprises a fourth output unit, and is used for sending alarm information in response to the first voiceprint feature existing in the voiceprint database being provided with a classification mark.

In one embodiment, the system further comprises an output module, and the output module comprises a fifth output unit, and the fifth output unit is used for determining the corresponding relation between the analysis result of the emotion of the client and the client risk assessment value and the insurance pricing according to a preset rule or algorithm, and outputting the client risk assessment and the insurance pricing data.

The above embodiment is any one of embodiments for realizing correspondence between analysis results of emotion of a customer and customer risk assessment values and insurance pricing.

The above embodiment is used for calculating any one of customer risk assessment and insurance pricing according to analysis results of emotion and demand of customers.

In order to realize the method of the embodiments of the present application, as shown in fig. 1, the embodiment of the present application further provides a service abnormality pre-warning system based on a voiceprint technology, which includes a computer device, configured to operate the service abnormality pre-warning method based on a voiceprint technology according to any one of the embodiments of the first aspect, and generate data of adding classification marks to first voiceprint features. For example, the computer device loads the various software operating modules of the embodiment shown in FIG. 6.

In one embodiment, the storage means 3 are also comprised.

It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The application therefore also proposes a computer-readable storage medium, on which a computer program is stored, which program, when being executed by a processor, implements a method according to any of the embodiments of the application.

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 flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations 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.

Further, the present application also proposes an electronic device (or computing device) comprising a memory, a processor and a computer program stored on the memory and executable by the processor, said processor implementing a method according to any of the embodiments of the present application when said computer program is executed.

In one typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory. The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media.

Fig. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present application. The electronic device 600 shown is merely an example and should not be construed to limit the functionality and scope of use of embodiments of the present application. The voice print technology-based business anomaly pre-warning method comprises the steps of collecting voice print characteristics and corresponding client information, judging whether preset sensitive words exist in the voice print characteristics according to preset rules or algorithms, adding classification marks to the first voice print characteristics in response to the preset sensitive words exist in the first voice print characteristics, and storing the first voice print characteristics and the corresponding classification marks in a voice print database when the one or more programs are run by the one or more processors 620. Preferably, the method further comprises any of the steps of the method of the previous embodiments of the application.

The electronic device 600 further comprises input means 630 and output means 640. The processor 620, the memory means 610, the input means 630 and the output means 640 in the electronic device may be connected by a bus or by other means, in the figure by way of example by a bus 650.

The storage device 610 is a computer readable storage medium, and may be used to store a software program, a computer executable program, and a module unit, such as program instructions corresponding to a method for determining a cloud bottom height in an embodiment of the present application. The storage device 610 may mainly include a storage program area that may store an operating system, an application program required for at least one function, and a storage data area that may store data created according to the use of a terminal, etc. In addition, the storage 610 may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid-state storage device. In some examples, the storage device 610 may further include memory remotely located with respect to the processor 620, which may be connected via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The input device 630 may be used to receive input numeric, character information, or voice information, and to generate key signal inputs related to user settings and function control of the electronic device. The output device 640 may include an electronic device such as a display screen, a speaker, etc.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises an element.

The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and variations of the present application will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. which come within the spirit and principles of the application are to be included in the scope of the claims of the present application.

Claims

1. A method for early warning of business anomalies based on voiceprint technology, characterized by comprising the steps of:

Collect voiceprint features and corresponding customer information, and determine whether there are preset sensitive words in the voiceprint features based on preset rules or algorithms;

In response to the presence of a preset sensitive word in the first voiceprint feature, a classification mark is added to the first voiceprint feature, and the first voiceprint feature and the corresponding classification mark are saved in a voiceprint database.

2. According to claim 1, the method for early warning of abnormal business operations based on voiceprint technology further comprises the steps of:

Collect the voiceprint features in the complaint information and compare them with the voiceprint features stored in the voiceprint database;

In response to the second voiceprint feature in the complaint information not existing in the voiceprint database, the second voiceprint feature is stored in the voiceprint database and information associating the customer information with the second voiceprint feature is established.

3. The service abnormality early warning method based on voiceprint technology according to claim 1 is characterized by further comprising the steps of:

Collect voiceprint features in insurance transaction information, determine the correspondence between the voiceprint features and preset customer emotions according to preset rules or algorithms, and output analysis results representing customer emotional characteristics.

4. The method for early warning of abnormal business operations based on voiceprint technology according to claim 1, characterized in that it further comprises the steps of:

The voiceprint features of the customer are collected in the complaint information and compared with the voiceprint features stored in the voiceprint database. In response to the presence of the first voiceprint feature in the complaint information in the voiceprint database and the presence of a classification mark in the first voiceprint feature, an alarm message is sent.

5. The method for early warning of abnormal business operations based on voiceprint technology according to claim 1, characterized in that it further comprises the steps of:

Collect voiceprint features and corresponding customer information in real time in any customer service program;

Compare the collected customer information with the customer information stored in the voiceprint database;

In response to the customer information existing in the voiceprint database, and the voiceprint feature corresponding to the customer information is different from that stored in the voiceprint database, an alarm message is sent.

6. The method for early warning of abnormal business operations based on voiceprint technology according to claim 2, characterized in that it further comprises the steps of:

Perform natural language processing on complaint information to filter out malicious complaints.

7. The method for early warning of abnormal business operations based on voiceprint technology according to claim 3, characterized in that it further comprises the steps of:

The correspondence between the analysis results of the customer's emotions and the customer's risk assessment value and insurance pricing is determined according to preset rules or algorithms, and the customer's risk assessment and insurance pricing data are output.

8. A service abnormality warning device based on voiceprint technology, characterized in that it is used to implement the service abnormality warning method based on voiceprint technology as described in any one of claims 1 to 7, comprising:

Acquisition module, used to collect voiceprint features and corresponding customer information;

The determination module is used to determine whether there are preset sensitive words in the voiceprint feature according to preset rules or algorithms; it is also used to add a classification mark to the first voiceprint feature in response to the presence of preset sensitive words in the first voiceprint feature, and save the first voiceprint feature and the corresponding classification mark in the voiceprint database.

9. A computer-readable storage medium having a computer program stored thereon, wherein when the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

10. An electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 7 when executing the computer program.