CN110738524A

CN110738524A - service data management method, system, equipment and medium

Info

Publication number: CN110738524A
Application number: CN201910977537.6A
Authority: CN
Inventors: 周曦; 姚志强; 李继伟; 张锦宇
Original assignee: Shanghai Cloud From Enterprise Development Co Ltd
Current assignee: Shanghai Cloud From Enterprise Development Co Ltd
Priority date: 2019-10-15
Filing date: 2019-10-15
Publication date: 2020-01-31

Abstract

The invention provides service data management methods, systems, equipment and media, which comprise the steps of obtaining corresponding voice features through voice input, obtaining key features matched with the voice features through each cluster center in an interactive service file, obtaining voice output response through the key features, wherein each cluster center is obtained by clustering the key features of each category corresponding to service categories, and the interactive service file comprises at least of historical face information, historical voice information, key features and service information.

Description

service data management method, system, equipment and medium

Technical Field

The invention relates to the field of finance, in particular to business data management methods, systems, equipment and media.

Background

The financial business is usually handled by a company according to the production and operation condition of the company, the fund reserve condition and the future operation and development requirements of the company, funds are financed for investors and creditors of the company through scientific prediction and decision, and fund supply is organized so as to ensure the normal production and operation activity requirements of the company. The financial institution also needs to fully consider the actual requirements of the customers when conducting related business handling, and makes reasonable decisions according to the historical behaviors of the customers. However, most financial institutions lack a targeted decision for customer requirements during business handling at present, and cannot respond to the customer requirements in advance, so that not only is a lot of time and cost consumed for temporarily preparing business-related data, but also customer experience is greatly reduced, and customer efficiency is affected.

Disclosure of Invention

In view of the problems existing in the prior art, the invention provides service data management methods, systems, devices and media, and mainly solves the problems of lack of pertinence and low efficiency in service handling.

In order to achieve the above and other objects, the present invention adopts the following technical solutions.

A method for managing service data, comprising:

acquiring corresponding voice characteristics through voice input;

acquiring key features matched with the voice features through each clustering center in the interactive service file; each clustering center is obtained by clustering key features of each category corresponding to the service category;

obtaining a voice output response through the key features;

wherein the interactive service file comprises at least of historical face information, historical voice information, key features and service information.

Optionally, performing similarity comparison on the voice feature and a centroid key feature in the plurality of clustering centers to obtain a target clustering center corresponding to the voice feature;

and obtaining the similarity between the voice feature and the key feature of the target clustering center, and obtaining the service information corresponding to the key feature of which the similarity reaches a set similarity threshold when the similarity reaches the set similarity threshold.

Optionally, similarity comparison is performed between the voice feature and a centroid key feature in the plurality of clustering centers, and a clustering center corresponding to the centroid key feature with the highest similarity is determined as a target clustering center.

Optionally, the category information of the key feature is obtained according to different service categories of the interactive service profile.

Optionally, clustering is performed on the key features to obtain a plurality of clustering centers.

Optionally, the interactive service profile is created according to historical visiting data, wherein the historical visiting data includes: historical face information, historical voice information, key features and business information.

Optionally, the key features include a keyword feature and a key phrase feature.

Optionally, a feature corpus of the business information is obtained according to the business category, and key features matched with the voice features are obtained according to the feature corpus.

Optionally, a voice response library is set, and corresponding voice output response is obtained from the voice response library according to the corresponding service information obtained by the matched key features.

Optionally, feature frequency information in the speech input is counted, and the feature corpus is updated according to the feature frequency information.

Optionally, a feature frequency threshold is set, and when the feature frequency information exceeds the feature frequency threshold, the corresponding speech feature is used to update the feature corpus.

Optionally, a filtering library is set, and when the feature frequency information exceeds the feature frequency threshold, the voice features corresponding to the filtering library are screened to obtain voice features related to the service, and the feature corpus is updated.

Optionally, a service feature is obtained according to the service information, and the filter library is set according to the service feature.

A business data management system, comprising:

the characteristic acquisition module is used for acquiring corresponding voice characteristics through voice input;

the feature matching module is used for acquiring key features matched with the voice features through each clustering center in the interactive service file; each clustering center is obtained by clustering key features of each category corresponding to the service category;

the response module is used for obtaining a voice output response through the key features;

Optionally, the corpus creating module is configured to obtain a feature corpus of the business information according to the business category, and obtain a key feature matched with the speech feature according to the feature corpus.

Optionally, the system includes a classification module, configured to perform classification processing on key features corresponding to different service categories according to the interactive service profile, and acquire category information of the key features.

Optionally, a clustering module is included, configured to cluster the key features to obtain a plurality of clustering centers.

Optionally, a feature recognition module is included, configured to compare the voice feature with a centroid key feature in a plurality of the clusters to obtain the target cluster center corresponding to the voice feature; and obtaining the similarity between the voice feature and the key feature of the target clustering center, and obtaining the service information corresponding to the key feature of which the similarity reaches a set similarity threshold when the similarity reaches the set similarity threshold.

Optionally, a statistics module is included, configured to count feature frequency information in the speech input, and update the feature corpus according to the feature frequency information.

Optionally, the corpus updating module is configured to set a feature frequency threshold, and when the feature frequency information exceeds the feature frequency threshold, use the corresponding speech feature for updating the feature corpus.

Optionally, the filtering module is configured to set a filtering library, and when the feature frequency information exceeds the feature frequency threshold, filter the corresponding voice feature through the filtering library to obtain a voice feature related to a service, and update the feature corpus.

an apparatus, comprising:

or more processors, and

or more machine readable media having instructions stored thereon that, when executed by the or more processors, cause the device to perform the business data management method.

one or more machine readable media having instructions stored thereon that, when executed by the one or more processors, cause an apparatus to perform the business data management methods described herein.

As described above, the business data management methods, systems, devices and media according to the invention have the following advantages.

Through creating an interactive service file, the requirement can be analyzed according to historical data, and a targeted decision can be made; the voice response information of the service to be handled can be acquired by combining voice interaction with the data file of the service to be handled, so that the voice interaction efficiency is improved, and the user experience is enhanced.

Drawings

Fig. 1 is a flowchart of a business data management method in an embodiment of the invention.

Fig. 2 is a block diagram of a business data management system in an embodiment of the invention.

Fig. 3 is a schematic structural diagram of a terminal device in an embodiment of the present invention.

Fig. 4 is a schematic structural diagram of a terminal device in another embodiment of the present invention.

Detailed Description

The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict.

It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention, and the components related to the present invention are only shown in the drawings rather than drawn according to the number, shape and size of the components in actual implementation, and the type, quantity and proportion of each component in actual implementation can be random changes, and the layout of the components may be more complicated.

Referring to fig. 1, the present invention provides business data management methods, which include steps S01-S03.

In step S01, the corresponding speech feature is acquired by the speech input:

the voice input can comprise voice record information of historical visits of the client, and voice information of the client in service handling can be collected in real time through the audio collection equipment and sent to the server side to extract voice characteristics.

In step S02, obtaining key features matching with the voice features through each clustering center in the interactive service profile; each clustering center is obtained by clustering key features of each category corresponding to the service category:

taking banking business as an example, when a client enters banks to transact financial business, the banks can collect facial image information of the client through a camera device arranged in a business transaction area and record voice communication information in the business transaction process of the client.

The interactive service archive is provided with a feature corpus, and all key features of corresponding clients are acquired according to historical visiting data of each client and are used for updating the feature corpus.

In , all historical key features corresponding to a client can be clustered according to different classes of services handled by the client, a clustering algorithm such as K-means can be adopted to cluster all key feature information of the client, and a plurality of clustering centers are obtained.

When a client transacts business, the client face information is collected, the collected face image is compared with the face image in the face database, the client information is identified according to the face image, and an interactive business file corresponding to the collected face image is obtained. The method can be used for processing business information according to history in the interactive business file, predicting the client behavior, outputting corresponding voice information aiming at the predicted behavior and carrying out targeted voice interaction with the client. The interactive service file can be displayed to corresponding staff through a display interface, so that the staff can adjust the strategy according to the historical data of the client and carry out voice communication according to the client requirement; and voice interaction can also be carried out with the customer through the self-service terminal.

The method comprises the steps of collecting voice input of a client through voice collecting equipment such as a microphone arranged in a service handling area, obtaining voice input information and voice features corresponding to the voice input information, comparing the voice features with key feature information in a feature corpus of an interactive service file, comparing the voice features with centroid key features of a plurality of clustering centers to obtain target clustering centers corresponding to the voice features when performing feature comparison according to the key feature information in the feature corpus, comparing the key features in the corresponding target clustering centers one by one at to calculate the similarity between the key features and the voice features, and adopting a cosine similarity calculation method or an Euclidean distance calculation method to obtain the key features with the highest similarity to the voice features so as to obtain the service information of the corresponding key features.

And comparing the similarity of the voice features with the centroid key features in the plurality of clustering centers, and determining the clustering center corresponding to the centroid key feature with the highest similarity as a target clustering center.

In step S03, with the key feature, a speech output response is obtained:

in embodiment, a voice response library can be set to output corresponding voice response information according to different service information, comprehensive analysis can be performed according to historical access data of all customers to obtain key features in the customer voice information and specific information of the transacted service, and corresponding voice response information is recorded for each key feature and corresponding service information.

In the embodiment, the occurrence frequency of the voice features in the voice input information of the client may be counted, a feature frequency threshold may be set, and when the counted voice feature frequency exceeds the set feature frequency threshold and the feature corpus does not have corresponding key features, the corresponding voice features may be input into the feature corpus to update the feature corpus, and at the same time, the business information that the client needs to handle is recorded, and the corresponding voice features are associated with the business information.

In , a corresponding service feature can be obtained according to the service information, such as a customer needs to handle credit service, the service feature can include vocabularies related to credit such as credit degree and amount of stay, and the service feature is input into a database to create a filter library.

In the embodiment, when the face information of the client is not matched with the information in the face database, and it is determined that the client is visited for the first time, the corresponding voice feature information may be obtained according to the voice input information of the client, the similarity calculation may be directly performed on the voice feature information and the service feature information in the filtering library, when the similarity reaches a set threshold, the service information corresponding to the voice information is obtained, and the voice response information is obtained according to the service information.

Referring to fig. 2, the present invention provides service data management systems for executing the service data management method in the foregoing method embodiments.

In , the business data management system includes a feature obtaining module 10, a feature matching module 11, and a response module 12, where the feature obtaining module 10 is configured to assist in executing step S01 described in the foregoing method embodiment, the feature matching module 11 is configured to assist in executing step S02 described in the foregoing method embodiment, and the response module 12 is configured to assist in executing step S03 described in the foregoing method embodiment.

In , the system includes a corpus creating module for obtaining a feature corpus of business information according to business categories, and obtaining key features matching with the voice features according to the feature corpus.

In an embodiment, the system includes a classifying module configured to classify key features corresponding to different service classes according to the interaction service profiles to obtain class information of the key features, and in another embodiment, the system further includes a clustering module configured to cluster the key features to obtain a plurality of clustering centers.

In the embodiment, the system includes a feature recognition module configured to compare a similarity between a voice feature and a centroid key feature in a plurality of clusters to obtain a target cluster center corresponding to the voice feature, obtain a similarity between the voice feature and the key feature of the target cluster center, and obtain service information corresponding to the key feature whose similarity with the voice feature reaches a set similarity threshold when the similarity reaches the set similarity threshold.

In , the system includes a statistics module for counting feature frequency information in the speech input and updating a feature corpus according to the feature frequency information.

In an embodiment , the system includes an expectation update module that sets a feature frequency threshold and uses a corresponding speech feature to update the feature corpus when the feature frequency information exceeds the feature frequency threshold.

In , the system includes a filtering module configured to set a filtering library, and when the feature frequency information exceeds a feature frequency threshold, filter a corresponding speech feature through the filtering library to obtain a service-related speech feature, and update the feature corpus.

devices are also provided, which may include or more processors and or more machine-readable media having instructions stored thereon, which when executed by the or more processors cause the device to perform the method of fig. 1. in practical applications, the device may act as a terminal device, which may also act as a server, examples of the terminal device may include a smart phone, a tablet computer, an e-book reader, an MP3 (Moving Picture Experts Group Audio Layer III) player, an MP4 (Moving Picture Experts Group Audio Layer IV) player, a laptop, a car computer, a desktop computer, a set-top box, a smart television, a wearable device, etc., and the present embodiment is not limited to a specific device.

The present embodiment also provides nonvolatile readable storage media, where or multiple modules (programs) are stored in the storage media, and when the or multiple modules are applied to a device, the device may execute instructions (instructions) included in the service data management method in fig. 1 according to the present embodiment.

Fig. 3 is a schematic diagram of a hardware structure of a terminal device according to an embodiment of the present application , as shown in the figure, the terminal device may include an input device 1100, a -th processor 1101, an output device 1102, a -th memory 1103, and at least communication buses 1104, where the communication buses 1104 are used to implement communication connections between elements, the -th memory 1103 may include a high-speed RAM memory, and may also include a non-volatile storage NVM, such as at least disk memories, and the -th memory 1103 may store various programs for performing various processing functions and implementing method steps of the embodiment.

Alternatively, the -th processor 1101 may be, for example, a Central Processing Unit (CPU), an Application Specific Integrated Circuit (ASIC), a Digital Signal Processor (DSP), a Digital Signal Processing Device (DSPD), a Programmable Logic Device (PLD), a field programmable array (FPGA), a controller, a microcontroller, a microprocessor, or other electronic components, and the processor 1101 is coupled to the input device 1100 and the output device 1102 through a wired or wireless connection.

The input device 1100 may optionally include a variety of input devices, for example, at least types of input devices may include a user-oriented interface, a device-oriented interface, a programmable interface of software, a camera, and a sensor, optionally, the device-oriented interface may be a wired interface for data transmission between devices, or may also be a hardware-inserted interface (e.g., USB interface, serial port, etc.) for data transmission between devices, optionally, the user-oriented interface may be, for example, user-oriented control keys, a voice input device for receiving voice input, and a touch-sensitive device (e.g., a touch screen, a touch pad, etc.) for receiving user touch input, optionally, the programmable interface of software may be, for example, an entry for a user to edit or modify a program, such as an input pin interface or an input interface of a chip, etc., and the output device 1102 may include an output device such as a display, a sound, etc.

In this embodiment, the processor of the terminal device includes a function for executing each module of the speech recognition apparatus in each device, and specific functions and technical effects may refer to the above embodiments, which are not described herein again.

Fig. 4 is a schematic diagram of a hardware structure of a terminal device according to another embodiments of the present application, fig. 4 is specific embodiments of the implementation process of fig. 3.

The second processor 1201 executes the computer program code stored in the second memory 1202 to implement the method described in fig. 1 in the above embodiment.

The secondary memory 1202 may comprise Random Access Memory (RAM), and may also include non-volatile memory, such as at least disk memories.

Optionally, the -th processor 1201 is disposed in the processing component 1200, the terminal device may further include a communication component 1203, a power supply component 1204, a multimedia component 1205, a voice component 1206, an input/output interface 1207, and/or a sensor component 1208.

The processing component 1200 may include or more second processors 1201 to execute instructions to perform all or part of the steps of the method illustrated in fig. 1 described above, furthermore, the processing component 1200 may include or more modules to facilitate interaction between the processing component 1200 and other components, for example, the processing component 1200 may include a multimedia module to facilitate interaction between the multimedia component 1205 and the processing component 1200.

The power component 1204 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the terminal device.

The multimedia components 1205 include a display screen that provides output interfaces between the terminal device and the user in embodiments the display screen may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). if the display screen includes a touch panel, the display screen may be implemented as a touch screen to receive input signals from the user.

The voice component 1206 comprises Microphones (MICs) configured to receive external voice signals when the terminal device is in an operational mode, such as a voice recognition mode, for example, the received voice signals may be stored in the second memory 1202 or transmitted via the communication component 1203, in embodiments the voice component 1206 further comprises speakers for outputting voice signals.

The input/output interface 1207 provides an interface between the processing component 1200 and peripheral interface modules, which may be click wheels, buttons, etc. These buttons may include, but are not limited to: a volume button, a start button, and a lock button.

The sensor assembly 1208 can include or more sensors to provide various aspects of status assessment for the terminal device, for example, the sensor assembly 1208 can detect the open/closed status of the terminal device, the relative positioning of the assemblies, the presence or absence of user contact with the terminal device.

The communication component 1203 is configured to facilitate wired or wireless communication between the terminal device and other devices the terminal device may have access to a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof in embodiments, the terminal device may include a SIM card slot therein for inserting a SIM card such that the terminal device may log onto a GPRS network to establish communication with a server via the internet.

As can be seen from the above, the communication component 1203, the voice component 1206, the input/output interface 1207 and the sensor component 1208 referred to in the embodiment of fig. 4 can be implemented as the input device in the embodiment of fig. 3.

In summary, the service data management methods, systems, devices and media of the present invention can predict customer behavior according to customer historical needs through the interactive data files, make decisions according to customer needs, improve work efficiency, classify the key feature information, reduce the range of speech feature recognition, improve recognition efficiency, and obtain the speech response corresponding to the service information according to the interactive service files and customer speech input, thereby effectively enhancing user interactive experience.

It will be appreciated by those skilled in the art that modifications and variations can be made to the disclosed embodiments without departing from the spirit and scope of the invention, and therefore, is equivalent to modifications and variations that would be apparent to those skilled in the art without departing from the spirit and scope of the invention as disclosed in the appended claims.

Claims

The service data management method of types 1 and is characterized by comprising the following steps:

acquiring corresponding voice characteristics through voice input;

acquiring key features matched with the voice features through each clustering center in the interactive service file; each clustering center is obtained by clustering key features of each category corresponding to the service category;

obtaining a voice output response through the key features;

wherein the interactive service file comprises at least of historical face information, historical voice information, key features and service information.
2. The traffic data management method according to claim 1,

comparing the similarity of the voice features with mass center key features in a plurality of clustering centers to obtain a target clustering center corresponding to the voice features;

and obtaining the similarity between the voice feature and the key feature of the target clustering center, and obtaining the service information corresponding to the key feature of which the similarity reaches a set similarity threshold when the similarity reaches the set similarity threshold.
3. The traffic data management method according to claim 2,

and comparing the similarity of the voice features with the centroid key features in the plurality of clustering centers, and determining the clustering center corresponding to the centroid key feature with the highest similarity as a target clustering center.
4. The business data management method of claim 1, wherein the category information of key features is obtained according to different business categories of the interactive business profile.
5. The traffic data management method according to claim 4,

and clustering the key features to obtain a plurality of clustering centers.
6. The business data management method of claim 1, wherein the interactive business profile is created according to historical visiting data, wherein the historical visiting data comprises: historical face information, historical voice information, key features and business information.
7. The business data management method of claim 6, wherein the key features comprise keyword features and key phrase features.
8. The method according to claim 1, wherein a feature corpus of the service information is obtained according to a service category, and a key feature matching the speech feature is obtained according to the feature corpus.
9. The service data management method according to claim 1, wherein a voice response library is provided, and corresponding voice output responses are obtained from the voice response library according to the corresponding service information obtained from the matched key features.
10. The method according to claim 8, wherein the feature frequency information in the speech input is counted, and the feature corpus is updated according to the feature frequency information.
11. The traffic data management method according to claim 10, wherein a feature frequency threshold is set, and when the feature frequency information exceeds the feature frequency threshold, the corresponding speech feature is used to update the feature corpus.
12. The method according to claim 11, wherein a filter library is provided, and when the characteristic frequency information exceeds the characteristic frequency threshold, the corresponding voice features are filtered through the filter library to obtain voice features related to the service, and the feature corpus is updated.
13. The method according to claim 9, wherein a service characteristic is obtained according to the service information, and the filter library is set according to the service characteristic.
14, A service data management system, comprising:

the characteristic acquisition module is used for acquiring corresponding voice characteristics through voice input;

the feature matching module is used for acquiring key features matched with the voice features through each clustering center in the interactive service file; each clustering center is obtained by clustering key features of each category corresponding to the service category;

the response module is used for obtaining a voice output response through the key features;

wherein the interactive service file comprises at least of historical face information, historical voice information, key features and service information.
15. The business data management system according to claim 14, comprising a corpus creation module configured to obtain a feature corpus of the business information according to a business category, and obtain a key feature matching the speech feature according to the feature corpus.
16. The business data management system of claim 15, wherein the key features comprise keyword features, key phrase features.
17. The service data management system according to claim 14, comprising a classification module, configured to perform classification processing on key features corresponding to different service categories according to the interactive service profile, so as to obtain category information of the key features.
18. The business data management system of claim 14, comprising a clustering module configured to cluster the key features to obtain a plurality of cluster centers.
19. The system according to claim 18, comprising a feature recognition module, configured to compare the voice feature with a centroid key feature in a plurality of the clusters to obtain the target cluster center corresponding to the voice feature; and obtaining the similarity between the voice feature and the key feature of the target clustering center, and obtaining the service information corresponding to the key feature of which the similarity reaches a set similarity threshold when the similarity reaches the set similarity threshold.
20. The system according to claim 15, comprising a statistics module for counting feature frequency information in the speech input and updating the feature corpus according to the feature frequency information.
21. The system according to claim 15, wherein the corpus updating module is configured to set a feature frequency threshold, and when the feature frequency information exceeds the feature frequency threshold, the corresponding speech feature is used to update the feature corpus.
22. The system according to claim 21, wherein the filtering module is configured to set a filtering library, and when the feature frequency information exceeds the feature frequency threshold, filter the corresponding speech features through the filtering library to obtain speech features related to the service, and update the feature corpus.
An apparatus of the type , comprising:

or more processors, and

or more machine readable media having instructions stored thereon that, when executed by the or more processors, cause the device to perform the method of or more of claims 1-13.
One or more machine-readable media 24, , having instructions stored thereon, which when executed by the one or more processors , cause an apparatus to perform the method of of claims 1-13.