CN112929502B - Voice recognition method and system based on electric power customer service - Google Patents
Voice recognition method and system based on electric power customer service Download PDFInfo
- Publication number
- CN112929502B CN112929502B CN202110166994.4A CN202110166994A CN112929502B CN 112929502 B CN112929502 B CN 112929502B CN 202110166994 A CN202110166994 A CN 202110166994A CN 112929502 B CN112929502 B CN 112929502B
- Authority
- CN
- China
- Prior art keywords
- information
- obtaining
- power
- voice
- source address
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000012937 correction Methods 0.000 claims abstract description 58
- 238000000605 extraction Methods 0.000 claims abstract description 34
- 238000013139 quantization Methods 0.000 claims description 41
- 230000005236 sound signal Effects 0.000 claims description 23
- 238000012549 training Methods 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 11
- 241001672694 Citrus reticulata Species 0.000 claims description 10
- 238000004891 communication Methods 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 10
- 230000004048 modification Effects 0.000 claims description 7
- 238000012986 modification Methods 0.000 claims description 7
- 238000012552 review Methods 0.000 claims description 3
- 230000004044 response Effects 0.000 abstract description 14
- 230000003993 interaction Effects 0.000 abstract description 10
- 238000003062 neural network model Methods 0.000 description 15
- 238000006243 chemical reaction Methods 0.000 description 14
- 238000010586 diagram Methods 0.000 description 11
- 230000008569 process Effects 0.000 description 6
- 238000013528 artificial neural network Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000007726 management method Methods 0.000 description 3
- 238000012550 audit Methods 0.000 description 2
- 238000013178 mathematical model Methods 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 230000003925 brain function Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M3/00—Automatic or semi-automatic exchanges
- H04M3/42—Systems providing special services or facilities to subscribers
- H04M3/487—Arrangements for providing information services, e.g. recorded voice services or time announcements
- H04M3/493—Interactive information services, e.g. directory enquiries ; Arrangements therefor, e.g. interactive voice response [IVR] systems or voice portals
- H04M3/4936—Speech interaction details
-
- 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
- G10L15/26—Speech to text systems
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M3/00—Automatic or semi-automatic exchanges
- H04M3/42—Systems providing special services or facilities to subscribers
- H04M3/50—Centralised arrangements for answering calls; Centralised arrangements for recording messages for absent or busy subscribers ; Centralised arrangements for recording messages
- H04M3/51—Centralised call answering arrangements requiring operator intervention, e.g. call or contact centers for telemarketing
- H04M3/5166—Centralised call answering arrangements requiring operator intervention, e.g. call or contact centers for telemarketing in combination with interactive voice response systems or voice portals, e.g. as front-ends
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Human Computer Interaction (AREA)
- Business, Economics & Management (AREA)
- Marketing (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Telephonic Communication Services (AREA)
Abstract
The invention discloses a voice recognition method and a system based on electric power customer service.A first voice message and a first source address are obtained; judging whether the first source address has a first characteristic or not; when the first source address exists, a first correction value is obtained according to the first source address and the first voice information; inputting the first voice information and the first correction value into the keyword extraction model to obtain first power information; obtaining a first execution instruction according to the first source address and the first power information; obtaining a first region matching execution result according to the first execution instruction; when the first region matching execution result meets a first preset condition, obtaining first region power information according to the first region matching execution result; and obtaining first voice reply information according to the first region electric power information and the first correction value. The technical problem that the manual seat and the traditional self-service voice response system are limited by factors such as customer service manpower, knowledge level and consultation amount due to the adoption of a key interaction mode, and the customer experience is influenced is solved.
Description
The technical field is as follows:
the invention relates to the technical field of power customer service, in particular to a voice recognition method and system based on power customer service.
Background art:
the Power customer service Center (Power Call Center) is a necessary way for Power supply and Power utilization systems to go to the market, and is a powerful assistant for Power system innovation, concept transformation and service highlighting. The power service hotline is for 24 hours to accept business consultation, information inquiry, service complaints and power failure warranty. The third-party scheduling ten measures are committed to the working partners, namely the third-party scheduling, the third-party scheduling and the third-party scheduling are insist on legal open, fair and fair scheduling, and the safe and stable operation of the power system is guaranteed. With the wide application of traditional power customer service in recent years, the user scale is increasing, and the traffic pressure of services such as communication service hotlines, power dispatching command systems and the like of power users will increase rapidly. The manual seat and the traditional self-service voice response system are limited by factors such as customer service manpower, working time, knowledge level, telephone traffic consulting volume and the like due to the adoption of a key interaction mode, and the customer experience is seriously influenced.
However, in the process of implementing the technical solution of the invention in the embodiments of the present application, the inventors of the present application find that the above-mentioned technology has at least the following technical problems:
in the prior art, the manual seat and the traditional self-service voice response system are limited by factors such as customer service manpower, working time, knowledge level, telephone traffic consulting volume and the like due to the adoption of a key interaction mode, and the technical problem of influencing customer experience exists.
The invention content is as follows:
the embodiment of the application provides a voice recognition method and system based on electric power customer service, and solves the technical problem that in the prior art, manual agents and traditional self-service voice response systems are limited by factors such as customer service manpower, working time, knowledge level and telephone traffic consulting amount due to the fact that a key interaction mode is adopted, and customer experience is influenced. The method has the advantages that the characteristics of the incoming call voice are utilized to perform voice recognition and keyword extraction, the accuracy of power problem recognition is effectively improved by combining a neural network model, reliable problem answering content is guaranteed to be provided for users, meanwhile, in order to better fit the voice characteristics of the incoming call users, content understanding deviation is avoided, answering content is subjected to voice conversion, the quality of power customer service is effectively improved, and the user experience is improved.
In view of the foregoing problems, the embodiments of the present application provide a voice recognition method and system based on power customer service.
In a first aspect, an embodiment of the present application provides a speech recognition method based on power customer service, where the method includes: obtaining first voice information; obtaining a first source address according to the first voice information; judging whether the first source address has a first characteristic or not; when the first source address has the first characteristic, obtaining a first correction value according to the first source address and the first voice information; inputting the first voice information and the first correction value into a keyword extraction model to obtain first power information; obtaining a first execution instruction according to the first source address and the first power information, wherein the first execution instruction is used for matching the first source address and the first power information in a regional power event record library; obtaining a first region matching execution result according to the first execution instruction; when the first region matching execution result meets a first preset condition, obtaining first region power information according to the first region matching execution result; and obtaining first voice reply information according to the first region electric power information and the first correction value.
In another aspect, the present application further provides a speech recognition system based on power customer service, the system includes:
a first obtaining unit configured to obtain first voice information;
a second obtaining unit, configured to obtain a first source address according to the first voice information;
the first judging unit is used for judging whether the first source address has a first characteristic or not;
a third obtaining unit, configured to, when the first source address has the first characteristic, obtain a first correction value according to the first source address and the first voice information;
the first execution unit is used for inputting the first voice information and the first correction value into a keyword extraction model to obtain first power information;
a fourth obtaining unit, configured to obtain a first execution instruction according to the first source address and the first power information, where the first execution instruction is used to match the first source address and the first power information in a regional power event record base;
a fifth obtaining unit, configured to obtain a first region matching execution result according to the first execution instruction;
a sixth obtaining unit, configured to obtain first region power information according to the first region matching execution result;
a seventh obtaining unit, configured to obtain first voice reply information according to the first region power information and the first correction value.
In a third aspect, the present invention provides a speech recognition system based on power customer service, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method of the first aspect when executing the program.
One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:
the embodiment of the application provides a voice recognition method and system based on electric power customer service, wherein first voice information is obtained; obtaining a first source address according to the first voice information; judging whether the first source address has a first characteristic or not; when the first source address has the first characteristic, obtaining a first correction value according to the first source address and the first voice information; inputting the first voice information and the first correction value into a keyword extraction model to obtain first power information; obtaining a first execution instruction according to the first source address and the first power information, wherein the first execution instruction is used for matching the first source address and the first power information in a regional power event record library; obtaining a first region matching execution result according to the first execution instruction; when the first region matching execution result meets a first preset condition, obtaining first region power information according to the first region matching execution result; and obtaining first voice reply information according to the first region electric power information and the first correction value. According to the matched region electric power information, namely the electric power question reply content corresponding to the electric power information, in order to meet the requirements of client personnel in different regions, the first correction value is generated according to the dialect characteristics of the first voice information, the first region electric power information is corrected by the first correction value and is converted into the local dialect to reply, so that the voice characteristics of the user are met, the user can be ensured to obtain accurate and effective reply content, especially the old people at home can not have Mandarin, the information content of electric power reply can be mastered more quickly by the old people through the conversion of the dialect, and the electric power problem is solved more effectively. Therefore, the technical problem that in the prior art, due to the fact that a key interaction mode is adopted in manual seats and traditional self-service voice response systems, the manual seats and the traditional self-service voice response systems are limited by factors such as customer service manpower, working time, knowledge level and telephone traffic consulting volume, and customer experience is influenced is solved. The method has the advantages that the characteristics of the incoming call voice are utilized to perform voice recognition and keyword extraction, the accuracy of power problem recognition is effectively improved by combining a neural network model, reliable problem answering content is guaranteed to be provided for users, meanwhile, in order to better fit the voice characteristics of the incoming call users, content understanding deviation is avoided, answering content is subjected to voice conversion, the quality of power customer service is effectively improved, and the user experience is improved.
The foregoing description is only an overview of the technical solutions of the present application, and the present application can be implemented according to the content of the description in order to make the technical means of the present application more clearly understood, and the following detailed description of the present application is given in order to make the above and other objects, features, and advantages of the present application more clearly understandable.
Drawings
Fig. 1 is a schematic flow chart of a speech recognition method based on power customer service according to an embodiment of the present application.
Fig. 2 is a schematic structural diagram of a speech recognition system based on power customer service according to an embodiment of the present application.
Fig. 3 is a schematic structural diagram of an exemplary electronic device according to an embodiment of the present application.
Description of the reference numerals: a first obtaining unit 11, a second obtaining unit 12, a first judging unit 13, a third obtaining unit 14, a first executing unit 15, a fourth obtaining unit 16, a fifth obtaining unit 17, a sixth obtaining unit 18, a seventh obtaining unit 19, a bus 300, a receiver 301, a processor 302, a transmitter 303, a memory 304, and a bus interface 306.
The specific implementation mode is as follows:
the embodiment of the application provides a voice recognition method and system based on electric power customer service, and solves the technical problem that in the prior art, manual agents and traditional self-service voice response systems are limited by factors such as customer service manpower, working time, knowledge level and telephone traffic consulting amount due to the fact that a key interaction mode is adopted, and customer experience is influenced. The technical effects of performing voice recognition and keyword extraction by using the characteristics of the incoming call voice, effectively improving the accuracy of electric power problem recognition by combining with a neural network model to ensure that reliable problem answering contents are provided for users, avoiding content understanding deviation caused by the voice characteristics of incoming call users, performing voice conversion on answering contents, effectively improving the quality of electric power customer service, and improving the user experience are achieved. Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are merely some embodiments of the present application and not all embodiments of the present application, and it should be understood that the present application is not limited to the example embodiments described herein.
Summary of the application
The Power customer service Center (Power Call Center) is a necessary way for Power supply and Power utilization systems to go to the market, and is a powerful assistant for Power system innovation, concept transformation and service highlighting. The power service hotline is for 24 hours to accept business consultation, information inquiry, service complaints and power failure warranty. The third-party scheduling ten measures are committed to the working partners, namely the third-party scheduling, the third-party scheduling and the third-party scheduling are insist on legal open, fair and fair scheduling, and the safe and stable operation of the power system is guaranteed. With the wide application of traditional power customer service in recent years, the scale of users is increasing, and the traffic pressure of services such as communication service hotlines, power dispatching command systems and the like of power users is increasing rapidly. The manual seat and the traditional self-service voice response system are limited by factors such as customer service manpower, working time, knowledge level, telephone traffic consulting volume and the like due to the adoption of a key interaction mode, and the customer experience is seriously influenced. However, in the prior art, the manual seat and the traditional self-service voice response system are limited by factors such as customer service manpower, working time, knowledge level, telephone traffic consulting volume and the like due to the adoption of a key interaction mode, and the technical problem of influencing customer experience exists.
In view of the above technical problems, the technical solution provided by the present application has the following general idea:
obtaining first voice information; acquiring a first source address according to the first voice information; judging whether the first source address has a first characteristic or not; when the first source address has the first characteristic, obtaining a first correction value according to the first source address and the first voice information; inputting the first voice information and the first correction value into a keyword extraction model to obtain first power information; obtaining a first execution instruction according to the first source address and the first power information, wherein the first execution instruction is used for matching the first source address and the first power information in a regional power event record library; obtaining a first region matching execution result according to the first execution instruction; when the first region matching execution result meets a first preset condition, obtaining first region power information according to the first region matching execution result; and obtaining first voice reply information according to the first region electric power information and the first correction value. The method has the advantages that the characteristics of the incoming call voice are utilized to perform voice recognition and keyword extraction, the accuracy of power problem recognition is effectively improved by combining a neural network model, reliable problem answering content is guaranteed to be provided for users, meanwhile, in order to better fit the voice characteristics of the incoming call users, content understanding deviation is avoided, answering content is subjected to voice conversion, the quality of power customer service is effectively improved, and the user experience is improved.
Having thus described the general principles of the present application, various non-limiting embodiments thereof will now be described in detail with reference to the accompanying drawings.
Example 1:
as shown in fig. 1, an embodiment of the present application provides a speech recognition method based on power customer service, where the method includes:
step S100: obtaining first voice information;
specifically, the first voice information is obtained by call voice collection for an incoming call.
Step S200: obtaining a first source address according to the first voice information;
specifically, the address information of the incoming call is obtained according to the incoming route of the first voice message, and the first voice message can be determined from the position of the telephone signal through positioning and recognition of the telephone number.
Step S300: judging whether the first source address has a first characteristic or not;
specifically, the local language is judged by recognizing the first voice message and determining the first source address, the first characteristic is that whether the first source address has a dialect is judged, if the voice content of the incoming call has the dialect difficult to identify, corresponding processing is needed, and because the power customer service faces the whole country, and faces different customer groups, such as the old at home, the groups which can not speak the common speech can exist, but the dialect of the people is difficult to identify sometimes, and in order to better perform voice identification and improve the working efficiency of the power customer service, the embodiment of the application performs feature extraction and identification processing on the incoming call.
Step S400: when the first source address has the first characteristic, obtaining a first correction value according to the first source address and the first voice information;
further, when the first source address has the first characteristic, obtaining a first correction value according to the first source address and the first voice message, in step S400 of this embodiment of the present application, the first correction value includes:
step S410: extracting the first voice information to obtain a first audio signal;
step S420: obtaining a first audio quantization requirement according to the first audio signal;
step S430: obtaining first audio quantization information according to the first audio signal and the first audio quantization requirement;
step S440: obtaining second audio quantization information, wherein the second audio quantization information is quantization information of standard reply voice;
step S450: obtaining first quantization correction information according to the first audio quantization information and the second audio quantization information;
step S460: and obtaining the first correction value according to the first quantized correction information and the first audio signal.
Specifically, when the source address of the first voice information of the incoming call has a first characteristic, namely a dense dialect, audio signal extraction is performed according to the first voice information, digital conversion is performed on the audio signal, the requirement of digital conversion is determined according to the characteristic of the audio signal, if multi-bit digital conversion is required for an audio with high dialect audio complexity, some digital conversion only needs to be performed simply, if 16bit or 4bit is required, the corresponding audio quantization requirement is determined according to different audio characteristics, digital quantization processing is performed on the first audio signal corresponding to the first voice information according to the audio quantization requirement to obtain first audio quantization information, second audio quantization information is voice extraction of the mandarin, the information obtained by digital quantization processing is compared between the dialect corresponding to the first audio information and the mandarin, the first audio information is corrected according to a quantization difference value, and the first correction value is an audio information part which is required to be adjusted for converting the first audio signal into the mandarin.
Step S500: inputting the first voice information and the first correction value into a keyword extraction model to obtain first power information;
further, the step S500 of inputting the first speech information and the first correction value into a keyword extraction model to obtain first power information includes:
step S510: obtaining a first corrected voice according to the first voice information and the first correction value;
step S520: taking the first corrected voice as first input information;
step S530: obtaining a word library in the power industry;
step S540: taking the electric power industry word library as second input information;
step S550: inputting the first input information and the second input information into a keyword extraction model, wherein the keyword extraction model is obtained by training multiple groups of training data, and each group of the multiple groups of training data comprises: the first input information, the second input information and the identification input information comprise identification information of electric power industry words;
step S560: and obtaining a first output result of the keyword extraction model, wherein the first output result comprises first electric power information, and the first electric power information is used for representing electric power industry keywords contained in the first voice information.
Specifically, the keyword is extracted according to the first voice information, the keyword is a term for describing the power problem, the power-related problem requested by the incoming call user can be known through the description of the keyword, and for the voice information with dialect problem, the first voice information needs to be corrected through a first correction value, and then the first voice information is converted into corresponding mandarin and then the keyword is extracted, so that the accuracy of extracting the keyword is ensured, and the keyword recognition and extraction are not deteriorated due to the influence of the dialect. Further, in the embodiment of the present application, in order to improve the accuracy of keyword extraction analysis, a Neural network model is added, the keyword extraction model is a Neural network model in machine learning, and a Neural Network (NN) is a complex Neural network system formed by widely interconnecting a large number of simple processing units (called neurons), reflects many basic features of human brain functions, and is a highly complex nonlinear dynamical learning system. Neural network models are described based on mathematical models of neurons. Artificial Neural Networks (Artificial Neural Networks) are a description of the first-order properties of the human brain system. Briefly, it is a mathematical model. And through training of a large amount of training data, inputting the first input information and the second input information into a neural network model, and outputting first power information.
More specifically, the training process is essentially a supervised learning process, each group of supervised data includes the first input information, the second input information and identification information including electric power industry words, the first input information and the second input information are input into a neural network model, the neural network model performs continuous self-correction and adjustment according to the identification information for identifying the input information including the electric power industry words, and the group of supervised learning is ended until the obtained output result is consistent with the identification information, and the next group of supervised learning is performed; and when the output information of the neural network model reaches the preset accuracy rate/reaches the convergence state, finishing the supervised learning process. Through the supervised learning of the neural network model, the neural network model can process the input information more accurately, so that more accurate and suitable first electric power information can be obtained, accurate electric power problem recognition can be carried out on the voice information received by the electric power customer service, then electric power problem keywords and electric power proper noun keywords can be extracted by using voice, the electric power problem of voice description can be accurately recognized, and the technical effect of reliable electric power service is provided for users.
Step S600: obtaining a first execution instruction according to the first source address and the first power information, wherein the first execution instruction is used for matching the first source address and the first power information in a regional power event record library;
specifically, the power problem reflected by the user is determined according to the first power information, and whether a power event exists at the first source address is determined by combining the first source address, such as periodic maintenance or regional problem of the power line, and the like, which are power events scheduled by the power system.
Step S700: obtaining a first region matching execution result according to the first execution instruction;
further, after obtaining the first region matching execution result according to the first execution instruction, the embodiment of the application further includes:
step S710: when the first region matching execution result does not meet the first preset condition, acquiring a power customer service question-answer database;
step S720: obtaining a first matching instruction according to the first power information, wherein the first matching instruction is used for matching in the power customer service question and answer database according to keyword information in the first power information;
step S730: obtaining a first matching result according to the first matching instruction;
step S740: and obtaining second voice reply information according to the first matching result.
Step S800: when the first region matching execution result meets a first preset condition, obtaining first region power information according to the first region matching execution result;
further, in step S800 of obtaining second voice reply information according to the first matching result, in this embodiment of the present application, the method includes:
step S810: when the first matching result is a first result, obtaining the second voice reply information according to the first matching result;
step S820: when the first matching result is a second result, acquiring first station number information according to the first electric power information;
step S830: and acquiring first switching information according to the first station number information.
Specifically, the first region matching execution result is a result of matching the first source address and the first power information in the region power event record library, and the first region matching execution result may have two cases, one is that the first region matching execution result is a corresponding power record event if the matching is successful, and the other is that the first region matching execution result is empty, which may be represented by 0, and when the first region matching execution result is empty, it indicates that the power problem is not a general problem of reported power of the region, and is an individual problem. The first predetermined condition is used for judging whether the first region matching execution result is empty, the empty condition does not meet the first predetermined condition, the problem that the current incoming call user response is related to the regional power reporting event is shown for meeting the first predetermined condition, the power condition of the event is matched at the moment, and the first region power information is corresponding power event description information generated according to the reported power event. If the first regional matching execution result is empty, the regional power provision event is not matched, at the moment, the power customer service question-answer database is searched and matched according to the power problems described in the first power information, the power customer service question-answer database is a corresponding power answer scheme generated according to different power problems, different power answer contents are made in advance according to different power problems because the frequently encountered problems of the power problems and the corresponding solutions are unified, corresponding information matching is performed according to the identified power information to obtain corresponding answer information, and the matched result is used as answer information for the incoming call user. In addition, because the set questions and answers in the power customer service question and answer database are set according to common general questions, the situation that the questions are not in the power customer service question and answer database is inevitable, the first result is that the answer content is matched, the second result is that the answer result is not matched, the situation shows that the current power questions are rare, and the corresponding professional personnel are required to carry out specific answers, at the moment, the station numbers of the professional personnel are matched according to the keywords of the power information, and manual switching is carried out, so that different services are provided according to different questions of the user, more professional services are provided for the user, on the other hand, the labor can be saved, the labor is used for the professional questions, and the technical effect of not wasting limited manpower on simple and easy-to-solve problems is achieved.
Step S900: and obtaining first voice reply information according to the first region electric power information and the first correction value.
Specifically, according to the matched electric power question reply content corresponding to the electric power information which is the regional electric power information, in order to meet the requirements of customers in different regions, the first correction value is generated according to the dialect characteristics of the first voice information, the first regional electric power information is corrected by using the first correction value and is converted into the local dialect to reply, so that the voice characteristics of the user are met, the user can be ensured to obtain accurate and effective reply content, especially the old people at home can not have Mandarin, the old people at home can be helped to master the information content of electric power reply more quickly through the conversion of the dialect, and the electric power question is solved more effectively. Therefore, the technical problem that in the prior art, due to the fact that a key interaction mode is adopted in manual seats and traditional self-service voice response systems, the manual seats and the traditional self-service voice response systems are limited by factors such as customer service manpower, working time, knowledge level and telephone traffic consulting volume, and customer experience is influenced is solved. The method has the advantages that the characteristics of the incoming call voice are utilized to perform voice recognition and keyword extraction, the accuracy of power problem recognition is effectively improved by combining a neural network model, reliable problem answering content is guaranteed to be provided for users, meanwhile, in order to better fit the voice characteristics of the incoming call users, content understanding deviation is avoided, answering content is subjected to voice conversion, the quality of power customer service is effectively improved, and the user experience is improved.
Further, the embodiment of the present application further includes:
step S1010: acquiring first communication identification information according to the first voice information;
step S1020: obtaining first historical voice according to the first communication identification information;
step S1030: obtaining a first historical power information base according to the first historical voice;
step S1040: judging whether the first historical power information base contains the first power information or not;
step S1050: if yes, obtaining a first problem solving rate according to the first power information and the first historical power information base;
step S1060: when the first problem solving rate does not meet a second preset condition, acquiring second station number information according to the first electric power information;
step S1070: and obtaining second switching information according to the second station number information.
Specifically, communication identification information of the incoming call user, namely the incoming call personal identity information of the user, is determined according to the first voice information. The historical incoming call information of the user can be searched through the communication identification information, the power problem content of the historical incoming call is determined according to the historical incoming call information, if the power problem of the incoming call is the same as the previous problem, analysis of the solving efficiency of the same problem is carried out through the first historical power information base, whether the problem is solved or not can be determined according to the reaction interval time of the same problem of the incoming call user, if the incoming call interval time is short, the current problem is not solved according to the previous method, if the interval time is long, the previous problem is solved, the solution rate is determined according to the feedback interval time of the problem, if the interval time is long enough, problem matching is carried out according to a normal flow to determine the answering content, if the interval time is short, if the problem is just called last day, the inquiry is continued, at the moment, it is determined that the problem is not solved according to the previous answering content, at the problem is subjected to professional manual matching, a professional station number is determined, the answering is carried out according to the station number for switching, professional service is provided, the user, the problem that the user is not subjected to manual switching, the problem that the problem is found because the manual switching information is not found, the problem is not found, the manual switching information is found is not satisfactory, the problem is found, the characteristic that the manual switching information is directly found is not found, the manual switching of the user, and the manual switching is not found is directly solved, and the user, the user is not found out of the manual switching information is not found by the manual switching information is easily, and the characteristic that the manual switching is not found by the manual switching is directly, and the manual switching is improved, and the manual switching is provided.
Further, the embodiment of the present application further includes:
step S1110: acquiring preset time information;
step S1120: acquiring a first power information connection library according to the preset time information, the first source address and the first power information;
step S1130: and when the information quantity in the first power information connection base meets a third preset condition, first feedback information is obtained.
Further, after obtaining the first feedback information, step S1130 in this embodiment of the present application includes:
step S1131: acquiring first review information according to the first feedback information;
step S1132: obtaining a first auditing result according to the first auditing information;
step S1133: when the first examination result meets a fourth preset condition, judging whether the first feedback information is contained in the region power event record library;
step S1134: and when the first feedback information is not included, obtaining a first storage instruction according to the first feedback information, wherein the first storage instruction is used for storing the first feedback information in the region power event record library.
Specifically, if a large number of power events in the same area reflecting the same power problem are received in the same time period, the possibility of area power events exists, feedback is performed according to the number of received calls, the first feedback information is that aggregated power events in the same area are subjected to system feedback, the system receives the first feedback information and then sends the first feedback information to a server background management department for reminding, the background management carries out auditing on the content of the first feedback information, whether the power problem has the characteristics of the power event or not is judged, and the requirement of relevant regulations is met, if the power problem meets the requirements, the auditing is passed, if the power event meets the requirements and is not recorded in a regional power event record library, the first feedback information is used as the power event for identification, the content in the first feedback information is stored in the regional power event record library, the regional power event record library is updated, corresponding reply content is generated according to the event and stored together, and the reply content is used for automatically replying to subsequent calls, so that the service efficiency is improved, the incoming calls are avoided, or artificial conditions are caused, and the power feedback content is utilized to carry out the power analysis, so that the power analysis and the busy state of the power network is enriched.
Example 2:
based on the same inventive concept as the speech recognition method based on the power customer service in the foregoing embodiment, the present invention further provides a speech recognition system based on the power customer service, as shown in fig. 2, the system includes:
a first obtaining unit 11, where the first obtaining unit 11 is configured to obtain first voice information;
a second obtaining unit 12, where the second obtaining unit 12 is configured to obtain a first source address according to the first voice information;
a first determining unit 13, where the first determining unit 13 is configured to determine whether the first source address has a first characteristic;
a third obtaining unit 14, where the third obtaining unit 14 is configured to obtain a first correction value according to the first source address and the first voice information when the first source address has the first characteristic;
the first execution unit 15 is configured to input the first voice information and the first correction value into a keyword extraction model, so as to obtain first power information;
a fourth obtaining unit 16, where the fourth obtaining unit 16 is configured to obtain a first execution instruction according to the first source address and the first power information, where the first execution instruction is used to match the first source address and the first power information in a regional power event record base;
a fifth obtaining unit 17, where the fifth obtaining unit 17 is configured to obtain a first region matching execution result according to the first execution instruction;
a sixth obtaining unit 18, where the sixth obtaining unit 18 is configured to obtain first region power information according to the first region matching execution result;
a seventh obtaining unit 19, where the seventh obtaining unit 19 is configured to obtain the first voice reply information according to the first region power information and the first correction value.
Further, the system further comprises:
an eighth obtaining unit, configured to extract and obtain a first audio signal from the first speech information;
a ninth obtaining unit configured to obtain a first audio quantization requirement from the first audio signal;
a tenth obtaining unit, configured to obtain first audio quantization information according to the first audio signal and the first audio quantization requirement;
an eleventh obtaining unit configured to obtain second audio quantization information, which is quantization information of standard reply speech;
a twelfth obtaining unit, configured to obtain first quantization modification information according to the first audio quantization information and the second audio quantization information;
a thirteenth obtaining unit configured to obtain the first correction value from the first quantized correction information and the first audio signal.
Further, the system further comprises:
a fourteenth obtaining unit, configured to obtain a first corrected voice according to the first voice information and the first correction value;
a second execution unit configured to take the first corrected voice as first input information;
a fifteenth obtaining unit, configured to obtain an electric power industry term library;
the third execution unit is used for taking the electric power industry word bank as second input information;
a first input unit, configured to input the first input information and the second input information into a keyword extraction model, where the keyword extraction model is obtained through training of multiple sets of training data, and each set of the multiple sets of training data includes: the first input information, the second input information and the identification input information comprise identification information of electric power industry words;
a sixteenth obtaining unit, configured to obtain a first output result of the keyword extraction model, where the first output result includes first power information, and the first power information is used to represent a power industry keyword included in the first voice information.
Further, the system further comprises:
a seventeenth obtaining unit, configured to obtain a power customer service question-and-answer database when the first region matching execution result does not satisfy the first predetermined condition;
an eighteenth obtaining unit, configured to obtain a first matching instruction according to the first power information, where the first matching instruction is used to perform matching in the power customer service question-answer database according to keyword information in the first power information;
a nineteenth obtaining unit, configured to obtain a first matching result according to the first matching instruction;
a twentieth obtaining unit configured to obtain second voice reply information based on the first matching result.
Further, the system further comprises:
a twenty-first obtaining unit, configured to, when the first matching result is a first result, obtain the second voice reply information according to the first matching result;
a twenty-second obtaining unit, configured to obtain, when the first matching result is a second result, first station number information according to the first power information;
and the twenty-third obtaining unit is used for obtaining first switching information according to the first station number information.
Further, the system further comprises:
a twenty-fourth obtaining unit, configured to obtain first communication identification information according to the first voice information;
a twenty-fifth obtaining unit, configured to obtain a first history voice according to the first communication identification information;
a twenty-sixth obtaining unit, configured to obtain a first historical power information base according to the first historical voice;
a second determination unit configured to determine whether the first historical power information base includes the first power information;
a twenty-seventh obtaining unit, configured to obtain a first problem solving rate according to the first power information and the first historical power information base if the first problem solving rate is included in the first problem solving rate;
a twenty-eighth obtaining unit, configured to obtain second station number information according to the first power information when the first problem resolution does not satisfy a second predetermined condition;
and the twenty-ninth obtaining unit is used for obtaining second switching information according to the second station number information.
Further, the system further comprises:
a thirtieth obtaining unit, configured to obtain preset time information;
a thirty-first obtaining unit, configured to obtain a first power information connection library according to the preset time information, the first source address, and the first power information;
a thirty-second obtaining unit, configured to obtain the first feedback information when the number of information in the first power information connection library satisfies a third predetermined condition.
Further, the system further comprises:
a thirty-third obtaining unit, configured to obtain first review information according to the first feedback information;
a thirty-fourth obtaining unit, configured to obtain a first audit result according to the first audit information;
a third determining unit, configured to determine whether the first feedback information is included in the regional power event record base when the first reviewing result satisfies a fourth predetermined condition;
a thirty-fifth obtaining unit, configured to, when the first feedback information is not included, obtain a first storage instruction according to the first feedback information, where the first storage instruction is used to store the first feedback information in the region power event record library.
Various changes and specific examples of the speech recognition method based on power customer service in the first embodiment of fig. 1 are also applicable to the speech recognition system based on power customer service in the present embodiment, and through the foregoing detailed description of the speech recognition method based on power customer service, those skilled in the art can clearly know the implementation method of the speech recognition system based on power customer service in the present embodiment, so for the brevity of the description, detailed descriptions are omitted here.
Exemplary electronic device
An electronic apparatus of an embodiment of the present application is described below with reference to fig. 3.
Fig. 3 illustrates a schematic structural diagram of an electronic device according to an embodiment of the present application.
Based on the inventive concept of a speech recognition method based on power customer service in the foregoing embodiments, the present invention further provides a speech recognition system based on power customer service, on which a computer program is stored, which when executed by a processor implements the steps of any one of the foregoing speech recognition methods based on power customer service.
Where in fig. 3 a bus architecture, represented by bus 300, may include any number of interconnected buses and bridges, the bus 300 linking together various circuits including one or more processors, represented by processor 302, and memory, represented by memory 304. The bus 300 may also link together various other circuits such as peripherals, voltage regulators, power management circuits, and the like, which are well known in the art, and therefore, will not be described any further herein. A bus interface 306 provides an interface between the bus 300 and the receiver 301 and transmitter 303. The receiver 301 and the transmitter 303 may be the same element, i.e., a transceiver, providing a means for communicating with various other systems over a transmission medium.
The processor 302 is responsible for managing the bus 300 and general processing, and the memory 304 may be used for storing data used by the processor 302 in performing operations.
One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:
the embodiment of the application provides a voice recognition method and a voice recognition system based on electric power customer service, wherein first voice information is obtained; obtaining a first source address according to the first voice information; judging whether the first source address has a first characteristic or not; when the first source address has the first characteristic, obtaining a first correction value according to the first source address and the first voice information; inputting the first voice information and the first correction value into a keyword extraction model to obtain first power information; obtaining a first execution instruction according to the first source address and the first power information, wherein the first execution instruction is used for matching the first source address and the first power information in a regional power event record library; obtaining a first region matching execution result according to the first execution instruction; when the first region matching execution result meets a first preset condition, obtaining first region power information according to the first region matching execution result; and obtaining first voice reply information according to the first region electric power information and the first correction value. According to the matched region electric power information, namely the electric power question reply content corresponding to the electric power information, in order to meet the requirements of client personnel in different regions, the first correction value is generated according to the dialect characteristics of the first voice information, the first region electric power information is corrected by the first correction value and is converted into the local dialect to reply, so that the voice characteristics of the user are met, the user can be ensured to obtain accurate and effective reply content, especially the old people at home can not have Mandarin, the information content of electric power reply can be mastered more quickly by the old people through the conversion of the dialect, and the electric power problem is solved more effectively. Therefore, the technical problem that in the prior art, due to the fact that a key interaction mode is adopted in manual seats and traditional self-service voice response systems, the manual seats and the traditional self-service voice response systems are limited by factors such as customer service manpower, working time, knowledge level and telephone traffic consulting volume, and customer experience is influenced is solved. The method has the advantages that the characteristics of the incoming call voice are utilized to perform voice recognition and keyword extraction, the accuracy of power problem recognition is effectively improved by combining a neural network model, reliable problem answering content is guaranteed to be provided for users, meanwhile, in order to better fit the voice characteristics of the incoming call users, content understanding deviation is avoided, answering content is subjected to voice conversion, the quality of power customer service is effectively improved, and the user experience is improved.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a system for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction system which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks. While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (9)
1. A voice recognition method based on power customer service, wherein the method comprises:
collecting call voice to obtain first voice information;
obtaining a first source address according to the incoming call route of the first voice message;
judging whether the first source address has a first characteristic according to the identification of the first source address and the first voice information, wherein the first characteristic is used for judging whether a dialect exists in the first source address;
when the first source address has the first characteristic, obtaining a first correction value according to the first source address and the first voice information, where the first correction value is an audio information portion that needs to be adjusted to convert a first audio signal into mandarin chinese, and the method includes:
extracting the first voice information to obtain a first audio signal, wherein the source address of the first voice information has a first characteristic, namely a dense dialect;
obtaining a first audio quantization requirement according to the first audio signal;
obtaining first audio quantization information according to the first audio signal and the first audio quantization requirement;
obtaining second audio quantization information, the second audio quantization information being a pairPutonghui (Mandarin Chinese character)Line audio extraction and digital quantization processing are carried out to obtain information;
obtaining first quantization correction information according to the first audio quantization information and the second audio quantization information;
obtaining the first correction value according to the first quantized correction information and the first audio signal;
inputting the first voice information and the first correction value into a keyword extraction model to obtain first power information;
obtaining a first execution instruction according to the first source address and the first power information, wherein the first execution instruction is used for matching the first source address and the first power information in a regional power event record library;
obtaining a first region matching execution result according to the first execution instruction;
matching the first region matching execution result with a corresponding power record event, judging whether the first region matching execution result is empty or not, and if the first region matching execution result is empty, not meeting a first preset condition;
when the first region matching execution result meets a first preset condition, obtaining first region power information according to the first region matching execution result, wherein the first region power information is corresponding power event description information generated according to a reported power event;
and obtaining first voice reply information according to the first region electric power information and the first correction value.
2. The method of claim 1, wherein the inputting the first voice information and the first correction value into a keyword extraction model to obtain first power information comprises:
obtaining a first corrected voice according to the first voice information and the first correction value;
taking the first corrected voice as first input information;
obtaining a word library in the power industry;
taking the electric power industry word library as second input information;
inputting the first input information and the second input information into a keyword extraction model, wherein the keyword extraction model is obtained by training multiple groups of training data, and each group of the multiple groups of training data comprises: the first input information, the second input information and the identification input information comprise identification information of electric power industry words;
and obtaining a first output result of the keyword extraction model, wherein the first output result comprises first electric power information, and the first electric power information is used for representing electric power industry keywords contained in the first voice information.
3. The method of claim 1, wherein the obtaining a first zone matching execution result according to the first execution instruction comprises:
when the first region matching execution result does not meet the first preset condition, acquiring a power customer service question-answer database;
obtaining a first matching instruction according to the first power information, wherein the first matching instruction is used for matching in the power customer service question and answer database according to keyword information in the first power information;
obtaining a first matching result according to the first matching instruction;
and obtaining second voice reply information according to the first matching result.
4. The method as claimed in claim 3, wherein obtaining first region power information according to the first region matching execution result after obtaining second voice reply information according to the first matching result when the first region matching execution result satisfies a first predetermined condition comprises:
when the first matching result is a first result, obtaining the second voice reply information according to the first matching result;
when the first matching result is a second result, acquiring first station number information according to the first electric power information;
and acquiring first switching information according to the first station number information.
5. The method of claim 1, wherein the method comprises:
acquiring first communication identification information according to the first voice information;
obtaining first historical voice according to the first communication identification information;
obtaining a first historical power information base according to the first historical voice;
judging whether the first historical power information base contains the first power information or not;
if yes, obtaining a first problem solving rate according to the first power information and the first historical power information base;
when the first problem solving rate does not meet a second preset condition, acquiring second station number information according to the first electric power information;
and obtaining second switching information according to the second station number information.
6. The method of claim 1, wherein the method comprises:
acquiring preset time information;
acquiring a first power information connection library according to the preset time information, the first source address and the first power information;
and when the information quantity in the first power information connection library meets a third preset condition, obtaining first feedback information.
7. The method of claim 6, wherein the obtaining the first feedback information comprises:
acquiring first review information according to the first feedback information;
obtaining a first auditing result according to the first auditing information;
when the first checking result meets a fourth preset condition, judging whether the first feedback information is contained in the region power event record library;
and when the first feedback information is not included, obtaining a first storage instruction according to the first feedback information, wherein the first storage instruction is used for storing the first feedback information in the region power event record library.
8. A speech recognition system based on power customer service, wherein the system is applied to the method of any one of claims 1-7, and the system comprises:
the first obtaining unit is used for acquiring call voice to obtain first voice information;
a second obtaining unit, configured to obtain a first source address according to an incoming call route of the first voice message;
the first judging unit is used for judging whether the first source address has a first characteristic, and the first characteristic is used for judging whether a dialect exists in the first source address;
a third obtaining unit, configured to, when the first source address has the first characteristic, obtain a first correction value according to the first source address and the first voice information, where the first correction value is an audio information portion that is required to be adjusted to convert the first audio signal into mandarin chinese;
the first execution unit is used for inputting the first voice information and the first correction value into a keyword extraction model to obtain first power information;
a fourth obtaining unit, configured to obtain a first execution instruction according to the first source address and the first power information, where the first execution instruction is used to match the first source address and the first power information in a regional power event record base;
a fifth obtaining unit, configured to obtain a first region matching execution result according to the first execution instruction;
a sixth obtaining unit, configured to obtain first region power information according to the first region matching execution result;
a seventh obtaining unit, configured to obtain first voice reply information according to the first region power information and the first correction value;
an eighth obtaining unit, configured to extract and obtain a first audio signal from the first speech information;
a ninth obtaining unit, configured to obtain a first audio quantization requirement according to the first audio signal;
a tenth obtaining unit, configured to obtain first audio quantization information according to the first audio signal and the first audio quantization requirement;
an eleventh obtaining unit configured to obtain second audio quantization information, which is quantization information of standard reply speech;
a twelfth obtaining unit, configured to obtain first quantization modification information according to the first audio quantization information and the second audio quantization information;
a thirteenth obtaining unit configured to obtain the first correction value from the first quantized correction information and the first audio signal.
9. A power customer service based speech recognition system comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the method of any one of claims 1-7 when executing the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110166994.4A CN112929502B (en) | 2021-02-05 | 2021-02-05 | Voice recognition method and system based on electric power customer service |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110166994.4A CN112929502B (en) | 2021-02-05 | 2021-02-05 | Voice recognition method and system based on electric power customer service |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112929502A CN112929502A (en) | 2021-06-08 |
| CN112929502B true CN112929502B (en) | 2023-03-28 |
Family
ID=76171016
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110166994.4A Active CN112929502B (en) | 2021-02-05 | 2021-02-05 | Voice recognition method and system based on electric power customer service |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112929502B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113345429B (en) * | 2021-06-18 | 2022-03-29 | 图观(天津)数字科技有限公司 | Semantic analysis method and system based on complex scene |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106201424B (en) * | 2016-07-08 | 2019-10-01 | 北京甘为乐博科技有限公司 | A kind of information interacting method, device and electronic equipment |
| CN107665710B (en) * | 2016-07-27 | 2021-02-09 | 上海博泰悦臻网络技术服务有限公司 | Mobile terminal voice data processing method and device |
| CN106340308A (en) * | 2016-08-25 | 2017-01-18 | 北京云知声信息技术有限公司 | Speech reply method and device |
| CN108161933A (en) * | 2017-12-07 | 2018-06-15 | 北京康力优蓝机器人科技有限公司 | Interactive mode selection method, system and reception robot |
| CN110427472A (en) * | 2019-08-02 | 2019-11-08 | 深圳追一科技有限公司 | The matched method, apparatus of intelligent customer service, terminal device and storage medium |
| CN111193834B (en) * | 2019-12-16 | 2022-04-15 | 北京淇瑀信息科技有限公司 | Man-machine interaction method and device based on user sound characteristic analysis and electronic equipment |
| CN111161708A (en) * | 2019-12-31 | 2020-05-15 | 中国银行股份有限公司 | Voice information processing method and device |
| CN111326136B (en) * | 2020-02-13 | 2022-10-14 | 腾讯科技（深圳）有限公司 | Voice processing method and device, electronic equipment and storage medium |
| CN112199498A (en) * | 2020-09-27 | 2021-01-08 | 中国建设银行股份有限公司 | Man-machine conversation method, device, medium and electronic equipment for endowment service |
-
2021
- 2021-02-05 CN CN202110166994.4A patent/CN112929502B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN112929502A (en) | 2021-06-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11004013B2 (en) | Training of chatbots from corpus of human-to-human chats | |
| US11632463B2 (en) | Automated systems and methods for natural language processing with speaker intention inference | |
| US6898277B1 (en) | System and method for annotating recorded information from contacts to contact center | |
| US7003079B1 (en) | Apparatus and method for monitoring performance of an automated response system | |
| CN107886949A (en) | A kind of content recommendation method and device | |
| CN109977202A (en) | A kind of intelligent customer service system and its control method | |
| CN106230985B (en) | One kind is based on Internet of Things big data processing method, system and service processing end | |
| CN111739519A (en) | Dialogue management processing method, device, equipment and medium based on voice recognition | |
| CN110942326A (en) | User satisfaction prediction method and device and server | |
| CN110516057B (en) | Petition question answering method and device | |
| CN110955770A (en) | Intelligent dialogue system | |
| CN113064992A (en) | Complaint work order structured processing method, device, equipment and storage medium | |
| CN112929502B (en) | Voice recognition method and system based on electric power customer service | |
| CN117688165B (en) | Multi-edge collaborative customer service method, device, equipment and readable storage medium | |
| CN115017289A (en) | Method and device for serving customers based on knowledge base | |
| CN112860873A (en) | Intelligent response method, device and storage medium | |
| US20240162743A1 (en) | Power Demand Side Speech Interaction Method and System | |
| CN114418366B (en) | Data processing method and device for intelligent cloud interview | |
| CN113111157B (en) | Question-answer processing method, device, computer equipment and storage medium | |
| CN113705199A (en) | Work order priority confirmation method, work order priority confirmation device, work order priority confirmation electronic equipment, work order priority confirmation medium and work order priority confirmation product | |
| CN113645364A (en) | Intelligent voice outbound method facing power dispatching | |
| CN114401348A (en) | Method for monitoring and reminding operator customer service seat before quality inspection | |
| TWM617221U (en) | Automated customer service platform | |
| CN111192008A (en) | Self-help guide system and method | |
| CN113345429B (en) | Semantic analysis method and system based on complex scene |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |