CN112837684A - Service processing method and system, service processing device and readable storage medium - Google Patents
Service processing method and system, service processing device and readable storage medium Download PDFInfo
- Publication number
- CN112837684A CN112837684A CN202110024568.7A CN202110024568A CN112837684A CN 112837684 A CN112837684 A CN 112837684A CN 202110024568 A CN202110024568 A CN 202110024568A CN 112837684 A CN112837684 A CN 112837684A
- Authority
- CN
- China
- Prior art keywords
- module
- user
- questionnaire
- voice
- intention
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012545 processing Methods 0.000 title claims abstract description 58
- 238000003672 processing method Methods 0.000 title claims abstract description 30
- 238000004458 analytical method Methods 0.000 claims abstract description 55
- 230000003993 interaction Effects 0.000 claims abstract description 19
- 230000015572 biosynthetic process Effects 0.000 claims abstract description 18
- 238000003786 synthesis reaction Methods 0.000 claims abstract description 18
- 238000012360 testing method Methods 0.000 claims description 10
- 238000000034 method Methods 0.000 claims description 8
- 230000004044 response Effects 0.000 claims description 7
- 238000012549 training Methods 0.000 description 9
- 230000006870 function Effects 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 238000009434 installation Methods 0.000 description 5
- 238000007781 pre-processing Methods 0.000 description 5
- 230000008439 repair process Effects 0.000 description 5
- 241001672694 Citrus reticulata Species 0.000 description 4
- 238000004891 communication Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 239000013589 supplement Substances 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 230000010485 coping Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
- G06F40/35—Discourse or dialogue representation
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
- G10L2015/225—Feedback of the input speech
Landscapes
- Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Machine Translation (AREA)
Abstract
The invention provides a service processing method and system, a service processing device and a readable storage medium. The service processing method comprises the following steps: receiving voice of a user, carrying out voice recognition on the voice, converting the voice into a text, and carrying out semantic analysis on the text to obtain semantics; analyzing to obtain the user intention according to the semantics; acquiring scenes and questionnaires according to the user intention; performing voice synthesis on the contents of the questionnaire to obtain a voice stream, and outputting the voice stream to a user; and performing questionnaire answering according to the questionnaire and the user to complete interaction of the questionnaire. According to the technical scheme, a large amount of labor cost can be saved, the working efficiency is improved, and the user experience is improved.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a service processing method and system, a service processing device and a readable storage medium.
Background
With the increasing popularization of the internet, the squareness and broadband enter the golden period of development, and the network covers thousands of households in the south and north of the great river. Such a huge user group needs to receive tens of thousands of work orders such as installation, renewal, repair and the like every day, and each work order needs to be visited again with corresponding satisfaction, so that a company needs to operate a huge call center, a large number of telephone operators are needed to receive and track each work order, and the company cost is gradually increased. Limited manpower can cause untimely response speed to the customer due to the gradually increased service demands, a long-time waiting phenomenon often occurs, user experience is greatly influenced, more than 70% of the problems are the same or similar problems, and manpower cost is greatly wasted; during return visits, telephone operators are required to communicate with each other one by one, so that the labor cost is high, the efficiency is very low, and the number of return visits in one day is limited. Moreover, the accents of users in different regions are different, which causes great obstacles for operators to understand the services; after the waiters are received, the waiters also need to manually enter the corresponding systems, and the efficiency is very low. Therefore, how to respond to the customer demand efficiently becomes a first problem to be solved urgently.
At present, no complete intelligent solution aiming at broadband services is available, conventional broadband problems, installation, repair, renewal, return visit and the like are integrated, and the whole life cycle of a work order and the return visit thereof is completed through a robot.
Disclosure of Invention
The present invention is directed to solving at least one of the above problems.
Therefore, a first object of the present invention is to provide a service processing method.
A second object of the present invention is to provide a service processing system.
A third object of the present invention is to provide a service processing apparatus.
A fourth object of the present invention is to provide a readable storage medium.
In order to achieve the first object of the present invention, a technical solution of the present invention provides a service processing method, including: receiving voice of a user, carrying out voice recognition on the voice, converting the voice into a text, and carrying out semantic analysis on the text to obtain semantics; analyzing to obtain the user intention according to the semantics; acquiring scenes and questionnaires according to the user intention; performing voice synthesis on the contents of the questionnaire to obtain a voice stream, and outputting the voice stream to a user; and performing questionnaire answering according to the questionnaire and the user to complete interaction of the questionnaire.
This technical scheme can use manpower sparingly the cost in a large number, improves work efficiency, promotes user experience.
In addition, the technical scheme provided by the invention can also have the following additional technical characteristics:
in the above technical solution, the service processing method further includes: generating a work order based on the interaction of the completed questionnaire; and performing voice return visit on the user corresponding to the work order based on the completion of the work order.
In the technical scheme, the function of automatically generating the work order is realized, the manual workload can be effectively reduced, the labor cost is further reduced, the work efficiency is improved, and the user experience is improved.
In any of the above technical solutions, the service processing method further includes: confirming with the user by adopting the intention similar to the semantics based on the user intention which is not obtained to obtain the user intention; and adding the semantics and the user intention into a semantic library.
The technical scheme continuously optimizes the semantic library, thereby improving the recognition rate.
To achieve the second object of the present invention, the technical solution of the present invention provides a service processing system, including: the intelligent robot comprises an intelligent robot, a core engine, a questionnaire module and a first processing module; the intelligent robot comprises a real-time voice recognition module, a semantic analysis module, a voice synthesis module and an outbound module, wherein the real-time voice recognition module receives voice of a user and performs voice recognition on the voice, the semantic analysis module converts the voice into a text and performs semantic analysis on the text to obtain semantics, the voice synthesis module performs voice synthesis on the content of a questionnaire to obtain a voice stream, and the outbound module outputs the voice stream to the user; the core engine comprises an intention analysis module and a test paper question-answering module, the intention analysis module calls the first processing module to obtain the user intention based on the semantics, the test paper question-answering module calls the questionnaire module based on the user intention to dynamically obtain scenes and questionnaires, and the questionnaire question-answering is carried out on the users according to the questionnaires to complete the interaction of the questionnaires; the questionnaire module acquires a scene and a questionnaire according to the user intention; and the first processing module analyzes to obtain the user intention according to the semantics.
The service processing system of the technical scheme can greatly save labor cost, improve working efficiency and improve user experience.
In addition, the technical scheme provided by the invention can also have the following additional technical characteristics:
in the above technical solution, the core engine further includes: the system comprises a work order generation module, an intelligent return visit module and a question and answer storage module; the work order generation module generates a work order based on the interaction of the completed questionnaire; the intelligent return visit module performs voice return visit on the user corresponding to the work order based on the condition that the work order is finished; the question-answer storage module stores each question-answer of the questionnaire question-answers with the user.
In the technical scheme, the work order module is generated to automatically generate the work order, the automatic ordering function is realized, the manual workload can be effectively reduced, the labor cost is reduced, the work efficiency is improved, and the user experience is improved.
In any of the above technical solutions, the service processing system further includes: and the analysis module is provided with at least one factor, sets the proportion of the factor and obtains the score of the work order.
In the technical scheme, the analysis module can comprehensively score factors such as response timeliness, problem solving degree and return visit satisfaction degree, the proportion of each factor can be configured, and finally a weighted average value is given as the score of the work order.
In any of the above technical solutions, the service processing system further includes: the configuration background module comprises a questionnaire configuration module, an intention configuration module, a scene configuration module and a robot configuration module; the robot system comprises a questionnaire configuration module, an intention configuration module, a scene configuration module and a robot configuration module, wherein the questionnaire configuration module configures questionnaires, questions and answers interacted with a user, the intention configuration module configures the intention of the user according to a specific scene, the scene configuration module configures and uses different scenes for coping according to different answers of the user, and the robot configuration module dynamically maintains a robot pool according to the number of the users and is used for meeting access requests of the user.
In the technical scheme, the background module is configured to input for each module, so that the service processing system can effectively operate and realize different service functions.
In any of the above technical solutions, based on that the first processing module does not obtain the user intention, the intention close to the semantics is adopted to confirm with the user, so as to obtain the user intention, and the semantics and the user intention are added into the semantic library.
In the technical scheme, the semantic library is continuously optimized, so that the recognition rate is improved.
To achieve the third object of the present invention, a technical solution of the present invention provides a service processing apparatus, including: the device comprises a memory and a processor, wherein the memory stores programs or instructions, and the processor executes the programs or instructions; wherein, the processor implements the steps of the service processing method according to any technical scheme of the present invention when executing the program or the instructions.
The service processing apparatus provided in the present technical solution implements the steps of the service processing method according to any technical solution of the present invention, and thus has all the beneficial effects of the service processing method according to any technical solution of the present invention, which are not described herein again.
In order to achieve the fourth object of the present invention, the technical solution of the present invention provides a readable storage medium, where a program or an instruction is stored, and when the program or the instruction is executed, the steps of the service processing method according to any one of the above technical solutions are implemented.
The readable storage medium provided in this technical solution implements the steps of the service processing method according to any technical solution of the present invention, and thus has all the beneficial effects of the service processing method according to any technical solution of the present invention, which are not described herein again.
Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
The above and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
fig. 1 is a schematic flow chart of a service processing method according to an embodiment of the present invention;
FIG. 2 is a second schematic flow chart of a business processing method according to an embodiment of the present invention;
fig. 3 is a third schematic flow chart of a service processing method according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a service processing system according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of an intelligent robot according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of the core engine components of one embodiment of the present invention;
FIG. 7 is a schematic diagram of a service processing system according to an embodiment of the present invention;
FIG. 8 is a block diagram of a configuration backend module according to an embodiment of the present invention;
FIG. 9 is a schematic diagram of a service processing apparatus according to an embodiment of the present invention;
FIG. 10 is a diagram of an intelligent customer service system architecture according to an embodiment of the present invention.
Wherein, the correspondence between the reference numbers and the part names in fig. 4 to 10 is:
100: business processing system, 102: intelligent robot, 1022: real-time speech recognition module, 1024: semantic analysis module, 1026: speech synthesis module, 1028: outbound module, 104: core engine, 1042: intent analysis module, 1044: test paper question-answer module, 1046: generate work order module, 1048: intelligent return visit module, 1050: question-answer saving module, 106: questionnaire module, 108: first processing module, 110: analysis module, 112: configuration background module, 1122: questionnaire configuration module, 1124: intent configuration module, 1126: scene configuration module, 1128: robot configuration module, 114: storage module, 116: work order system, 1162: installation work order, 1164: repair work order, 1166: renewal bill, 1168: complaint worksheets, 200: service processing apparatus, 210: memory, 220: a processor.
Detailed Description
In order that the above objects, features and advantages of the present invention can be more clearly understood, a more particular description of the invention will be rendered by reference to the appended drawings. It should be noted that the embodiments and features of the embodiments of the present application may be combined with each other without conflict.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced in other ways than those specifically described herein, and therefore the scope of the present invention is not limited by the specific embodiments disclosed below.
A service processing method and system, a service processing apparatus, and a readable storage medium according to some embodiments of the present invention are described below with reference to fig. 1 to 10.
Example 1:
as shown in fig. 1, the present embodiment provides a service processing method, including the following steps:
step S102, receiving voice of a user, carrying out voice recognition on the voice, converting the voice into a text, and carrying out semantic analysis on the text to obtain semantics;
step S104, analyzing to obtain the user intention according to the semantics;
step S106, acquiring scenes and questionnaires according to the user intention;
step S108, carrying out voice synthesis on the questionnaire content to obtain a voice stream, and outputting the voice stream to a user;
and step S110, performing questionnaire answering according to the questionnaire and the user, and finishing interaction of the questionnaire.
In the related technology, business processes such as repair, installation and renewal in the broadband industry are various, and if manual service is adopted, the labor cost is increased, and the handling efficiency is low.
In the embodiment, firstly, the voice of the user is acquired, the voice is processed to obtain the semantics, then, the scene and the questionnaire are selected according to the semantics, the content of the questionnaire is output to the user in a voice stream mode, and finally, the questionnaire is answered with the user according to the questionnaire based on the answer of the user, so that the questionnaire interaction is completed. The cost of using manpower sparingly can be in a large number, improves work efficiency, promotes user experience.
In the embodiment, the voice library is arranged, after the voice of the user is obtained, the words of the user are automatically recognized in real time according to the voice library, and the contexts of synonymy, similarity and the like of different languages are analyzed according to semantic analysis to obtain reasonable meanings and know real user intentions.
In the embodiment, based on voice recognition, the Chinese and English with mandarin and slight mouth are intelligently recognized; intelligently identifying the real intention of the user based on semantic analysis; when the method is applied to the broadband industry, the manual work can be replaced, the related services are handled, the user operation is simple, the feedback is timely, the labor cost is reduced, the manual workload is reduced, the working efficiency is improved, various accents can be recognized, and the user experience is improved.
Example 2:
as shown in fig. 2, the present embodiment provides a service processing method, and in addition to the technical features of the foregoing embodiment, the present embodiment further includes the following technical features:
the service processing method also comprises the following steps:
step S202, based on the interaction of the completed questionnaire, generating a work order;
and step S204, based on the completion of the work order, performing voice return visit on the user corresponding to the work order.
In this embodiment, after the interaction of completion questionnaire, the work order is generated automatically, realizes the automatic function of placing an order, can effectual reduction artificial work load, and then reduces the cost of labor, promotes work efficiency, promotes user experience.
In the embodiment, after the work order is completed, the intelligent return visit is carried out, the return visit is carried out on the corresponding user in a voice mode, the user can feed back the information in time without operation, the user experience is improved, and the return visit efficiency is improved through the communication satisfaction degree of the intelligent robot and the user.
Example 3:
as shown in fig. 3, the present embodiment provides a service processing method, and in addition to the technical features of the foregoing embodiment, the present embodiment further includes the following technical features:
the service processing method also comprises the following steps:
step S302, based on the user intention which is not obtained, adopting the intention which is similar to the semanteme to confirm with the user, and obtaining the user intention;
step S304, adding the semantics and the user intention into a semantic library.
In this embodiment, when the user intention cannot be analyzed according to the user semantics, similar dialogues or the user intention needs to be confirmed with the user to obtain the user intention, and then the semantics and the user intention corresponding to the semantics are added into the semantic library, so that the same intention of the user corresponding to the semantics can be analyzed later, and the intelligent customer service system is more and more intelligent by slowly enriching the semantic library.
In the embodiment, the semantic library is a model, the user intention corresponding to the unrecognized semantics is obtained, the model is trained, the third-party unrecognized semantics can be continuously trained, the third-party unrecognized semantics can be used as powerful supplement of the third party, and the semantic library is continuously optimized through model training, so that the recognition rate is improved.
Example 4:
as shown in fig. 4 to fig. 6, the present embodiment provides a service processing system 100, including: intelligent robot 102, core engine 104, questionnaire module 106, and first processing module 108; the intelligent robot 102 comprises a real-time voice recognition module 1022, a semantic analysis module 1024, a voice synthesis module 1026 and an outbound module 1028, wherein the real-time voice recognition module 1022 receives voice of a user and performs voice recognition on the voice, the semantic analysis module 1024 converts the voice into a text and performs semantic analysis on the text to obtain semantics, the voice synthesis module 1026 performs voice synthesis on the content of a questionnaire to obtain a voice stream, and the outbound module 1028 outputs the voice stream to the user; the core engine 104 includes an intention analysis module 1042 and a test paper question-answering module 1044, based on semantics, the intention analysis module 1042 calls the first processing module 108 to obtain the user intention, based on the user intention, the test paper question-answering module 1044 calls the questionnaire module 106 to dynamically obtain scenes and questionnaires, and performs questionnaire question-answering with the user according to the questionnaires to complete interaction of the questionnaires; the questionnaire module 106 acquires scenes and questionnaires according to the user intention; the first processing module 108 parses the semantic meaning to obtain the user intention.
In this embodiment, the real-time speech recognition module 1022 obtains the speech of the user, processes the speech, the semantic analysis module 1024 obtains the semantic meaning, the first processing module 108 obtains the user intention by parsing according to the semantic meaning, the questionnaire module 106 selects a scene and a questionnaire according to the semantic meaning, the intention analysis module 1042 calls the first processing module 108 to obtain the user intention, the test paper questionnaire module 1044 calls the questionnaire module 106 to dynamically obtain the scene and the questionnaire, perform questionnaire answering according to the questionnaire and the user, complete interaction of the questionnaire, the speech synthesis module 1026 performs speech synthesis on the content of the questionnaire to obtain a speech stream, the outbound module 1028 outputs the speech stream to the user, the service processing system 100 can greatly save labor cost, improve work efficiency, and improve user experience.
In this embodiment, a voice library is provided, after the voice of the user is obtained, the real-time voice recognition module 1022 automatically recognizes the words of the user in real time according to the voice library, and the first processing module 108 analyzes the synonymous, similar and other contexts of different languages according to semantic analysis to obtain a reasonable meaning and know the real user intention.
In the embodiment, based on voice recognition, the Chinese and English with mandarin and slight mouth are intelligently recognized; intelligently identifying the real intention of the user based on semantic analysis; when the method is applied to the broadband industry, the manual work can be replaced, the related services are handled, the user operation is simple, the feedback is timely, the labor cost is reduced for a company, the manual workload is reduced, the working efficiency is improved, various accents can be recognized, and the user experience is improved.
Example 5:
as shown in fig. 6, the present embodiment provides a service processing system 100, and in addition to the technical features of the above embodiment, the present embodiment further includes the following technical features:
the core engine 104 further includes: a work order generation module 1046, an intelligent return visit module 1048 and a question and answer storage module 1050; the work order generation module 1046 generates a work order based on the interaction of the completed questionnaire, and the work order generation module 1046 generates a work order; the intelligent return visit module 1048 performs voice return visit on the user corresponding to the work order based on the state of the work order being completed; the question-answer saving module 1050 saves each question and answer to the questionnaire with the user.
In this embodiment, after the interaction of the questionnaire is completed, the work order generation module 1046 automatically generates the work order, so that the automatic ordering function is realized, the manual workload can be effectively reduced, the labor cost is reduced, the work efficiency is improved, and the user experience is improved.
In this embodiment, after the work order is completed, the intelligent return visit module 1048 performs intelligent return visit, and returns visit to the corresponding user in a voice manner, so that the user can timely feed back without operation, user experience is improved, and the return visit efficiency is improved by communication satisfaction between the intelligent robot and the user.
In this embodiment, the question-answer saving module 1050 saves each question and answer of the questionnaire with the user, which is convenient for later searching and analysis.
Example 6:
as shown in fig. 7, the present embodiment provides a service processing system 100, and in addition to the technical features of the above embodiment, the present embodiment further includes the following technical features:
the business processing system 100 further includes: and the analysis module 110, wherein the analysis module 110 is provided with at least one factor, the proportion of the factor is set, and the grade of the work order is obtained.
In this embodiment, the analysis module 110 gives a real score for each work order user for the comprehensive evaluation of the satisfaction revisit. The analysis module 110 may perform comprehensive scoring on factors such as response timeliness, problem resolution, return visit satisfaction, etc., and the proportion of each factor may be configured, and finally, a weighted average value is given as the score of the work order, so as to facilitate subsequent improvement on the problems related to the work order.
Example 7:
as shown in fig. 7 and fig. 8, the present embodiment provides a service processing system 100, and in addition to the technical features of the above embodiments, the present embodiment further includes the following technical features:
the business processing system 100 further includes: a configuration backend module 112, the configuration backend module 112 including a questionnaire configuration module 1122, an intent configuration module 1124, a scene configuration module 1126, and a robot configuration module 1128; the questionnaire configuration module 1122 configures questionnaires, questions and answers interacting with a user, the intention configuration module 1124 configures user intentions according to specific scenes, the scene configuration module 1126 configures different scenes for coping according to different answers of the user, and the robot configuration module 1128 dynamically maintains a robot pool according to the number of users for meeting access requests of the user.
In this embodiment, the background module 112 is configured as a basis of the whole service processing system 100, and the background module 112 is configured to perform input for each module, so that the service processing system 100 can effectively operate to implement different service functions.
Specifically, the configuration backend module 112 mainly implements questionnaire configuration, intention configuration, scene configuration, and robot configuration.
The intent configuration module 1124 configures user intent based on a particular scenario, and may include, for example, configuring basic intent, which refers to intent that is available in most scenarios and questions, and configuring general intent, which refers to intent that is not available in a particular scenario.
The scene configuration module 1126 configures different scenes for response according to different answers of the user, for example, the scene configuration module may configure scene names, scene association questionnaires and other basic information, and the configuration policy configures current intentions, next intentions, location nodes and the like.
The robot configuration module 1128 dynamically maintains the pool of robots according to the number of users to satisfy the access request of the users, for example, initializing the robots and adding scenes to the robots, where initializing the robots means initializing a certain number of available robots to the pool of robots according to the configuration, and if the pool of robots has a large amount of waiting, expanding the number of the pool of robots during the use. Adding scenes to the robot means that the idle robots in the robot pool are distributed to each scene, so that the robot establishes connection with a user.
Example 8:
in addition to the technical features of the above embodiments, the present embodiment further includes the following technical features:
based on the fact that the first processing module 108 does not obtain the user intention, the user intention is confirmed by adopting the intention similar to the semantics, and the semantics and the user intention are added into a semantic library.
In this embodiment, when the first processing module 108 cannot analyze the user intention according to the user semantics, it is necessary to confirm the similar dialogues or the user intentions with the user to obtain the user intention, and then add the semantics and the user intention corresponding to the semantics into the semantic library, so as to analyze the same intention of the user corresponding to the semantics in the future, and by slowly enriching the semantic library, the intelligent customer service system becomes more and more intelligent.
In the embodiment, the semantic library is a model, the user intention corresponding to the unrecognized semantics is obtained, the model is trained, the third-party unrecognized semantics can be continuously trained, the third-party unrecognized semantics can be used as powerful supplement of the third party, and the semantic library is continuously optimized through model training, so that the recognition rate is improved.
Example 9:
as shown in fig. 9, the present embodiment provides a service processing apparatus 200, including: a memory 210 and a processor 220, the memory 210 storing programs or instructions, the processor 220 executing the programs or instructions; wherein the processor 220, when executing the program or instructions, implements the steps of the service processing method according to any embodiment of the present invention.
Example 10:
the embodiment provides a readable storage medium, which stores a program or instructions, and when the program or instructions are executed by the processor 220, the steps of the service processing method of any of the above embodiments are implemented.
The specific embodiment is as follows:
the embodiment provides a business processing system 100, an AI-based intelligent customer service system, and aims to solve the problem that a large amount of labor is needed to complete the work in the past, further liberate the labor cost of enterprises, improve the work order efficiency, and improve the user experience.
The embodiment overcomes the defects of business processing such as repair, installation and renewal in the existing broadband industry, and solves the core problems of user intention identification, automatic answer, automatic ordering, intelligent return visit and the like, so that the user is easy to operate and feeds back timely; the company reduces the labor cost and reduces the labor workload; the recognition strength of various accents is improved, and the user experience is improved.
The design idea of the embodiment is as follows: based on voice recognition, intelligently recognizing Chinese and English with mandarin, or with accent; intelligently identifying the real intention of the user based on semantic analysis; continuously training the semantics which cannot be identified by a third party by taking model training as a core, and taking the semantics as powerful supplement of the third party; by taking the intelligent satisfaction degree return visit as a means, the return visit efficiency is improved through the communication satisfaction degree between the intelligent robot and the user. In the whole embodiment, voice recognition is the basis, semantic analysis is the key, model training is the highlight, and automatic return visit is the means.
The technical points of the embodiment are as follows: automatically identifying the words of the user in real time according to the voice library; according to semantic analysis, synonymy, similar and other contexts of different languages are analyzed to obtain reasonable meanings and know real intentions; according to data preprocessing, summarizing and warehousing common questions for automatic answering; according to model training, a semantic library is continuously optimized, and the recognition rate is improved; automatically generating a work order of a corresponding type according to an analysis result; and configuring different questionnaires and scenes according to the configuration background, and intelligently performing return visit investigation.
The embodiment relies on voice recognition and semantic analysis, accurately and conveniently recognizes the intention of the user through voice, and particularly relates to a real-time semantic analysis algorithm and a phonetics model training algorithm.
The business processing system 100 (i.e., the intelligent customer service system) of the present embodiment is structured as shown in fig. 10, and the operation depends on powerful support of various functional modules, and the intelligent customer service system is divided into a configuration background module 112, an intelligent robot 102, a core engine 104, a preprocessing module (i.e., a first processing module 108), a questionnaire module 106, an analysis module 110, and a storage module 114.
The configuration background module 112 is responsible for questionnaire configuration, scene configuration, questionnaire and scene association, scene policy configuration, robot initialization and other detailed functions.
The intelligent robot 102 is responsible for interacting with a user, and integrates interactive functional components such as a real-time voice recognition module 1022, a semantic analysis module 1024, a voice synthesis module 1026 and an outbound module 1028.
The core engine 104 is a core of the intelligent customer service, and includes an intention analysis module 1042, a test paper question-answer module 1044, a work order generation module 1046, an intelligent revisit module 1048, and a question-answer storage module 1050, and the core engine 104 implements functions of intention analysis, questionnaire selection, answer recording, work order generation, automatic revisit, and the like.
The pre-processing module (the first processing module 108) is responsible for recognizing and analyzing the utterance of the user, so as to grasp the real intention of the user.
The questionnaire module 106 is responsible for different questionnaire surveys according to different scenes, so that different scenes are intelligentized in a questionnaire form.
The analysis module 110 is responsible for comprehensive evaluation of satisfaction revisit and gives a real score for each work order user.
Specifically, each module is explained in detail:
(1) intelligent robot 102
The intelligent robot 102 is a forerunner of the whole intelligent customer service system, and a conversation with a user is started through the intelligent robot 102; the responsibilities are mainly listening and speaking: the real-time speech recognition module 1022 receives the speech of the user, and invokes third-party speech software to convert the speech into a text; the semantic analysis module 1024 then performs semantic analysis on the received text, analyzes the user's intention, and transmits the analyzed result to the core engine 104 for processing. The said processing is to call the third-party speech software by the speech synthesis module 1026 according to the processing result of the core engine 104, convert the result into a speech stream, and transmit the speech stream to the outbound system (outbound module 1028) for speech reply.
(2) Configuring the backend module 112
The background configuration module 112 is a Web configuration background, and the background configuration module 112 is the basis of the whole intelligent customer service system and inputs the information for each module; the responsibility is mainly divided into questionnaire configuration, intention configuration, scene configuration and robot configuration.
First, the questionnaire configuration module 1122 is used for questionnaire configuration, and refers to a questionnaire interacted with by a user and questions and answers therein. The questionnaire configuration mainly comprises the following steps:
1) and (3) configuring a questionnaire: assigning basic information such as company number, company name, questionnaire number, questionnaire name and the like to which the questionnaire belongs;
2) configuration problems and admission conditions: assigning basic information such as problem serial numbers, problem types, placeholder types, problem descriptions, problem admission conditions and the like;
3) choice question configuration standard answer: the standard answer configuration is carried out on the choice questions, such as: answer choices, standard answers, associated public customs, etc.;
4) selecting subject standard answer configuration public dialect: only basic dialogs such as positive (can, good, kay, good, etc.) or negative are configured;
5) selecting subject standard answers and configuring private dialect: refers to the configuration of the dialogies for the answer to the question, e.g. (recovered, recovered last week, recovered for days, etc.).
Second, the intention configuration module 1124 configures an intention, which is a specific scenario, for considering questions or intentions that the user is likely to present in advance, so as to intelligently respond to the questions or intentions when the user presents the questions or intentions. The intended configuration is mainly divided into the following aspects:
1) configuring a basic intention: meaning an intent that can be used in most scenarios and problems, such as positive, negative, etc.;
2) the general intent of the configuration is: the intention is used only in a certain scene, such as the intention of poor network and poor service in the broadband fault return visit scene.
Third, the scene configuration module 1126 is used for scene configuration, which refers to a questionnaire that deals with different scenes according to different answers of the user. The scene configuration is mainly divided into the following aspects:
1) configuring a scene: configuring basic information such as scene names and scene association questionnaires;
2) configuring a strategy: refers to configuring the current intent, the next intent, the location node (agree, disagree), etc.
Fourth, the robot configuration module 1128 is configured for robot configuration, which means that a pool of robots is dynamically maintained to satisfy access requests of visiting users according to the number of users.
1) Initializing the smart robot 102: the robot configuration module 1128 initializes a certain number of available intelligent robots 102 to a pool of robots, according to the configuration; in the using process, if a large amount of waiting robot pools exist, the number of the robot pools is increased;
2) add scenarios to the intelligent robot 102: the intelligent robots 102 are assigned to the scenes in order to establish a connection with the user.
(3) Core engine 104
The core engine 104 is the heart of the entire intelligent customer service, and all intelligence and control is performed in the core engine 104. The responsibilities are mainly divided into questionnaire answering, automatic return visit, automatic order building, result storage and the like.
First, based on semantics, the intent analysis module 1042 invokes the first processing module 108 to obtain the user intent.
Second, the questionnaire answering (examination questionnaire answering) means that a corresponding scene is dynamically selected according to the intention of the user, and interaction of a questionnaire is completed according to the answer of the user, and is realized through the examination questionnaire answering module 1044.
Third, the automatic return visit is to automatically return visits to the completed work orders according to the status of each work order, and is realized through the intelligent return visit module 1048, and the return visit process calls the test paper question-answer module 1044.
Fourthly, automatically creating a work order means that if the user has finished the response of the configuration questionnaire, the intelligent customer service system automatically generates a work order, the work order is subsequently tracked and implemented by the system, the work order is realized by the work order generation module 1046, and the work order generated by the work order generation module 1046 is arranged in the work order system 116, and comprises a loading work order 1162, a repairing work order 1164, a renewal work order 1166 and a complaint work order 1168.
Fifthly, result saving means that each question and answer interacted with the user is recorded, so that the search and analysis can be performed later, the result is saved in the storage module 114 through the question and answer saving module 1050.
(4) Pre-processing module
The preprocessing module (i.e., the first processing module 108) is used for calling a pre-configured dialect for the semantics of the user transmitted by the intelligent robot 102, and accurately analyzing the answer or intention of the user through a certain algorithm technology; if the answer or intention of the user cannot be analyzed, similar dialogues or the intention of the user need to be confirmed by the user, then the semantics are added into a semantic library, the same answer or intention of the user can be successfully analyzed next time, and meanwhile, the semantic library can be slowly enriched, so that the intelligent voice system is more and more intelligent, wherein the semantic library is continuously optimized according to model training, and the recognition rate is improved.
(5) Questionnaire module 106
The questionnaire module 106 is a system for answering answers by the user according to the scene configured by the configuration background and the questionnaire thereof. It is the key of intelligent customer service system. Not only can a suitable scene and a questionnaire thereof be selected according to the intention, but also the next-step intention question selection is carried out according to the user answers.
(6) Analysis module 110
The analysis module 110 evaluates the entire intelligent customer service system, and comprehensively scores factors such as response timeliness, problem resolution, return visit satisfaction and the like, so that the proportion of each factor can be configured, and finally a weighted average value is given as the score of the work order.
In summary, the embodiment of the invention has the following beneficial effects:
1. in the embodiment, based on voice recognition, the Chinese and English with mandarin and slight mouth are intelligently recognized; intelligently identifying the real intention of the user based on semantic analysis; when the method is applied to the broadband industry, the manual work can be replaced, the related services are handled, the user operation is simple, the feedback is timely, the labor cost is reduced, the manual workload is reduced, the working efficiency is improved, various accents can be recognized, and the user experience is improved.
2. In the embodiment, after the work order is completed, the intelligent return visit is carried out, the return visit is carried out on the corresponding user in a voice mode, the user can feed back the information in time without operation, the user experience is improved, and the return visit efficiency is improved through the communication satisfaction degree of the intelligent robot and the user.
3. In the embodiment, the semantic library is a model, the user intention corresponding to the unrecognized semantics is obtained, the model is trained, the third-party unrecognized semantics can be continuously trained, the third-party unrecognized semantics can be used as powerful supplement of the third party, and the semantic library is continuously optimized through model training, so that the recognition rate is improved.
In the present invention, the terms "first", "second", and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance; the term "plurality" means two or more unless expressly limited otherwise. The terms "mounted," "connected," "fixed," and the like are to be construed broadly, and for example, "connected" may be a fixed connection, a removable connection, or an integral connection; "coupled" may be direct or indirect through an intermediary. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations.
In the description of the present invention, it is to be understood that the terms "upper", "lower", "left", "right", "front", "rear", and the like indicate orientations or positional relationships based on those shown in the drawings, and are only for convenience of description and simplification of description, but do not indicate or imply that the referred device or unit must have a specific direction, be constructed in a specific orientation, and be operated, and thus, should not be construed as limiting the present invention.
In the description herein, the description of the terms "one embodiment," "some embodiments," "specific embodiments," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The above is only a preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes will occur to those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A method for processing a service, comprising:
receiving voice of a user, carrying out voice recognition on the voice, converting the voice into a text, and carrying out semantic analysis on the text to obtain semantics;
analyzing to obtain the user intention according to the semantics;
acquiring scenes and questionnaires according to the user intention;
performing voice synthesis on the content of the questionnaire to obtain a voice stream, and outputting the voice stream to the user;
and performing questionnaire questioning and answering according to the questionnaire and the user to complete interaction of the questionnaire.
2. The traffic processing method according to claim 1, further comprising:
generating a work order based on the completion of the interaction of the questionnaire;
and performing voice return visit on the user corresponding to the work order based on the completion of the state of the work order.
3. The traffic processing method according to claim 2, further comprising:
confirming the user with an intention similar to the semantic meaning based on the fact that the user intention is not obtained, and obtaining the user intention;
and adding the semantics and the user intention into a semantic library.
4. A transaction system (100) comprising:
the intelligent robot (102) comprises a real-time voice recognition module (1022), a semantic analysis module (1024), a voice synthesis module (1026) and an outbound module (1028), wherein the real-time voice recognition module (1022) receives voice of a user and performs voice recognition on the voice, the semantic analysis module (1024) converts the voice into a text, and performs semantic analysis on the text to obtain semantics; the voice synthesis module (1026) carries out voice synthesis on the contents of the questionnaire to obtain a voice stream, and the outbound module (1028) outputs the voice stream to the user;
the core engine (104), the core engine (104) includes an intention analysis module (1042) and a test paper question and answer module (1044), based on the semantics, the intention analysis module (1042) calls the first processing module (108) to obtain the user intention, based on the user intention, the test paper question and answer module (1044) calls the questionnaire module (106) to dynamically obtain the scene and the questionnaire, and performs questionnaire question and answer with the user according to the questionnaire to complete the interaction of the questionnaire;
the questionnaire module (106), the questionnaire module (106) obtaining scenes and questionnaires according to the user intention;
the first processing module (108), the first processing module (108) analyzes to obtain the user intention according to the semantic meaning.
5. The transaction system (100) of claim 4, wherein the core engine (104) further comprises:
a work order generation module (1046), the work order generation module (1046) generating a work order based on the interaction of the questionnaire;
the intelligent revisiting module (1048) is used for performing voice revisiting on the user corresponding to the work order based on the completion of the state of the work order;
a question-answer saving module (1050), wherein the question-answer saving module (1050) saves each question and answer of the questionnaire question and answer with the user.
6. The transaction system (100) of claim 5, further comprising:
the analysis module (110), the analysis module (110) is equipped with at least one factor, set up the proportion of the said factor, obtain the grade of the said work order.
7. The transaction system (100) of claim 4, further comprising:
a configuration backend module (112), the configuration backend module (112) comprising a questionnaire configuration module (1122), an intent configuration module (1124), a scene configuration module (1126), and a robot configuration module (1128);
wherein the questionnaire configuration module (1122) configures questionnaires, questions and answers interacting with the user, the intention configuration module (1124) configures user intentions according to specific scenes, the scene configuration module (1126) configures responses using different scenes according to different answers of the user, and the robot configuration module (1128) dynamically maintains a pool of robots for satisfying the user's access request according to the number of the users.
8. The transaction system (100) of claim 4, wherein the user intent is determined by an intent similar to the semantic meaning based on the first processing module (108) not obtaining the user intent, and wherein the semantic meaning and the user intent are added to a semantic library.
9. A traffic processing apparatus (200), comprising:
a memory (210) storing programs or instructions;
a processor (220) that executes the program or instructions;
wherein the processor (220), when executing the program or instructions, implements the steps of the traffic processing method according to any of claims 1 to 3.
10. A readable storage medium, characterized in that it stores thereon a program or instructions which, when executed by a processor, implement the steps of the traffic processing method according to any one of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110024568.7A CN112837684A (en) | 2021-01-08 | 2021-01-08 | Service processing method and system, service processing device and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110024568.7A CN112837684A (en) | 2021-01-08 | 2021-01-08 | Service processing method and system, service processing device and readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112837684A true CN112837684A (en) | 2021-05-25 |
Family
ID=75928846
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110024568.7A Pending CN112837684A (en) | 2021-01-08 | 2021-01-08 | Service processing method and system, service processing device and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112837684A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104933084A (en) * | 2015-05-04 | 2015-09-23 | 上海智臻网络科技有限公司 | Method, apparatus and device for acquiring answer information |
| WO2015141700A1 (en) * | 2014-03-18 | 2015-09-24 | Kabushiki Kaisha Toshiba | Dialogue system construction support apparatus and method |
| CN106992009A (en) * | 2017-05-03 | 2017-07-28 | 深圳车盒子科技有限公司 | Vehicle-mounted voice exchange method, system and computer-readable recording medium |
| CN110677543A (en) * | 2019-09-12 | 2020-01-10 | 国网河北省电力有限公司电力科学研究院 | Interaction method of intelligent customer service system |
| CN110895940A (en) * | 2019-12-17 | 2020-03-20 | 集奥聚合(北京)人工智能科技有限公司 | Intelligent voice interaction method and device |
| CN112185383A (en) * | 2020-09-30 | 2021-01-05 | 深圳供电局有限公司 | Processing method and system for customer service return visit |
-
2021
- 2021-01-08 CN CN202110024568.7A patent/CN112837684A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015141700A1 (en) * | 2014-03-18 | 2015-09-24 | Kabushiki Kaisha Toshiba | Dialogue system construction support apparatus and method |
| CN104933084A (en) * | 2015-05-04 | 2015-09-23 | 上海智臻网络科技有限公司 | Method, apparatus and device for acquiring answer information |
| CN106992009A (en) * | 2017-05-03 | 2017-07-28 | 深圳车盒子科技有限公司 | Vehicle-mounted voice exchange method, system and computer-readable recording medium |
| CN110677543A (en) * | 2019-09-12 | 2020-01-10 | 国网河北省电力有限公司电力科学研究院 | Interaction method of intelligent customer service system |
| CN110895940A (en) * | 2019-12-17 | 2020-03-20 | 集奥聚合(北京)人工智能科技有限公司 | Intelligent voice interaction method and device |
| CN112185383A (en) * | 2020-09-30 | 2021-01-05 | 深圳供电局有限公司 | Processing method and system for customer service return visit |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110769124B (en) | Electric power marketing customer communication system | |
| US7907705B1 (en) | Speech to text for assisted form completion | |
| CN105592237B (en) | A kind of method, apparatus and intelligent customer service robot of session switching | |
| CN110263144A (en) | A kind of answer acquisition methods and device | |
| US7103553B2 (en) | Assistive call center interface | |
| CN111192060B (en) | Full-channel self-service response implementation method based on power IT service | |
| US7606714B2 (en) | Natural language classification within an automated response system | |
| EP1976255B1 (en) | Call center with distributed speech recognition | |
| US20050002502A1 (en) | Apparatus and method for processing service interactions | |
| US20120221331A1 (en) | Method and Apparatus for Automatically Building Conversational Systems | |
| CN110825858A (en) | Intelligent interaction robot system applied to customer service center | |
| KR102163081B1 (en) | Interactive voice bot server and unmanned counsel system | |
| CN112885348A (en) | AI-combined intelligent voice electric marketing method | |
| CN111739519A (en) | Dialogue management processing method, device, equipment and medium based on voice recognition | |
| CN111462726B (en) | Method, device, equipment and medium for answering out call | |
| US20040042591A1 (en) | Method and system for the processing of voice information | |
| CN115858744A (en) | Outbound method, device and storage medium based on AI | |
| CN109977243A (en) | A kind of unidirectional interview method and system based on video record | |
| CN112150694A (en) | Intelligent voice electric charge payment urging system and method | |
| CN111858874A (en) | Conversation service processing method, device, equipment and computer readable storage medium | |
| CN112837684A (en) | Service processing method and system, service processing device and readable storage medium | |
| CN116340482A (en) | Multi-skill customer service auxiliary product based on enterprise WeChat combined with NLP engine | |
| CN116485409A (en) | Intelligent voice technology-based power service scene application method | |
| CN114040056B (en) | Method, device and storage medium for automatically checking user information | |
| US7343288B2 (en) | Method and system for the processing and storing of voice information and corresponding timeline information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210525 |
|
| RJ01 | Rejection of invention patent application after publication |