Nothing Special   »   [go: up one dir, main page]

CN116705003A - Voice work order quality inspection method, device, equipment and medium based on artificial intelligence - Google Patents

Voice work order quality inspection method, device, equipment and medium based on artificial intelligence Download PDF

Info

Publication number
CN116705003A
CN116705003A CN202310653099.4A CN202310653099A CN116705003A CN 116705003 A CN116705003 A CN 116705003A CN 202310653099 A CN202310653099 A CN 202310653099A CN 116705003 A CN116705003 A CN 116705003A
Authority
CN
China
Prior art keywords
work order
voice
quality inspection
text
voice work
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
Application number
CN202310653099.4A
Other languages
Chinese (zh)
Inventor
张博文
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ping An Bank Co Ltd
Original Assignee
Ping An Bank Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Ping An Bank Co Ltd filed Critical Ping An Bank Co Ltd
Priority to CN202310653099.4A priority Critical patent/CN116705003A/en
Publication of CN116705003A publication Critical patent/CN116705003A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06395—Quality analysis or management
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/01—Customer relationship services
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/03—Credit; Loans; Processing thereof
    • 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
    • G10L15/005—Language recognition
    • 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
    • G10L15/01—Assessment or evaluation of speech recognition systems
    • 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
    • G10L15/06—Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice
    • G10L15/063—Training
    • 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
    • G10L15/08—Speech classification or search
    • 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
    • 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
    • G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
    • G10L25/27—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique
    • G10L25/30—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique using neural networks
    • 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
    • G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
    • G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
    • G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
    • 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
    • G10L15/08—Speech classification or search
    • G10L2015/088—Word spotting
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • Human Computer Interaction (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Development Economics (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Educational Administration (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Tourism & Hospitality (AREA)
  • Artificial Intelligence (AREA)
  • Signal Processing (AREA)
  • Technology Law (AREA)
  • Evolutionary Computation (AREA)
  • Telephonic Communication Services (AREA)

Abstract

The application relates to the technical field of artificial intelligence and financial science and technology, and discloses a voice work order quality inspection method, device, equipment and medium based on artificial intelligence, which comprises the following steps: acquiring a voice work order file to be inspected; identifying a target language corresponding to the voice work order file; inputting the voice work order file into a voice conversion model corresponding to the target language according to the target language, and outputting a text file; scoring the text file according to a preset quality inspection scoring dimension and a preset scoring rule to obtain a score corresponding to each quality inspection scoring dimension; and obtaining the final score of the voice work order file to be inspected according to the score corresponding to each quality inspection score dimension. The application can analyze whether the service of the customer service personnel can meet the requirements of the customer more accurately according to the voice of the customer, and improve the accuracy of the detection of the service quality of the customer service personnel.

Description

Voice work order quality inspection method, device, equipment and medium based on artificial intelligence
Technical Field
The application relates to the technical field of artificial intelligence and financial science and technology, in particular to a voice work order quality inspection method, device, equipment and medium based on artificial intelligence.
Background
With the development of socioeconomic performance, the consumer awareness and the right-of-way awareness of financial consumers are improved to some extent. In particular, in the commercial banking area, banking consumer complaints are increasing, with credit card business accounting for a significant proportion. In the face of consumer complaints, consumer rights and interests protection main responsibility must be properly dealt with, the source management is paid attention to on the basis of reinforcing investigation analysis, and the protection work of consumer rights and interests is actually done.
Currently, in complaints directed to financial consumers, major complaint channels are mail complaints, online complaints, and phone complaints. Mail complaints are the customer sending a complaint to a financial service institution via mail; online complaints refer to text communication complaints by consumers through financial application programs (such as pocket bank APP) or applets (such as Jin Guangu) and customer service personnel or artificial intelligence customer service of a financial service institution; the telephone complaint refers to the voice communication complaint of the consumer to the personnel customer service staff by dialing a telephone. Because the telephone complaints are communicated with the manual customer service personnel through real-time voice, the telephone complaints are communicated in real time, delay waiting is not needed, and the complaint demands of the consumers can be responded quickly; thus, telephone complaints are the primary complaint channels.
When the customer complaints are processed by the manual customer service, the customer records are required to be listened to and the complaints are actually processed by combining the work order description. The traditional method of voice quality inspection is to set a manual quality inspection post, and examine and grade through the later sampling of a quality inspector and listening to the call record of customer service personnel so as to ensure the service quality of the customer service personnel. However, the whole quality inspection process is operated in a manual mode, and the defects of poor timeliness, low quality inspection efficiency and the like exist; and some consumers have heavy accents and slang; often, quality inspection personnel cannot hear the voice content of the consumer, and further cannot accurately and effectively inspect the voice work order.
Disclosure of Invention
The invention provides a voice work order quality inspection method, a device, computer equipment and a storage medium based on artificial intelligence, which are used for solving the technical problems that the timeliness is poor, the quality inspection efficiency is low, the voice content of a consumer cannot be heard and the voice work order cannot be effectively inspected in the background technology.
In a first aspect, a voice industry simple substance detection method based on artificial intelligence is provided, including: acquiring a voice work order file to be inspected; identifying a target language corresponding to the voice work order file; inputting the voice work order file into a voice conversion model corresponding to the target language according to the target language, and outputting a text file; scoring the text file according to preset quality inspection scoring dimensions and preset scoring rules to obtain scores corresponding to each quality inspection scoring dimension, wherein the quality inspection scoring dimensions comprise keyword and sensitive word detection scores and quality inspection rule matching scores; and obtaining the final score of the voice work order file to be inspected according to the score corresponding to each quality inspection score dimension.
In a second aspect, an artificial intelligence based voice industry simple detection device is provided, including: the acquisition module is used for acquiring the voice work order file to be inspected; the recognition module is used for recognizing the target language corresponding to the voice work order file; the conversion module is used for inputting the voice work order file into a voice conversion model corresponding to the target language according to the target language and outputting a text file; the scoring module is used for scoring the text file according to a preset quality inspection scoring dimension and a preset scoring rule to obtain a score corresponding to each quality inspection scoring dimension, wherein the quality inspection scoring dimension comprises a keyword and sensitive word detection score and a quality inspection rule matching score; and the evaluation module is used for acquiring the final score of the voice work order file to be inspected according to the score corresponding to each quality inspection score dimension.
In a third aspect, a computer device is provided, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the artificial intelligence based voice work order quality inspection method described above when the computer program is executed by the processor.
In a fourth aspect, a computer readable storage medium is provided, where a computer program is stored, and the computer program when executed by a processor implements the steps of the artificial intelligence based voice work order quality inspection method described above.
In the scheme realized by the artificial intelligence-based voice work order quality inspection method, the device, the computer equipment and the medium, the voice work order file to be inspected is obtained; identifying a target language corresponding to the voice work order file; inputting the voice work order file into a voice conversion model corresponding to the target language according to the target language, and outputting a text file; scoring the text file according to preset quality inspection scoring dimensions and preset scoring rules to obtain scores corresponding to each quality inspection scoring dimension, wherein the quality inspection scoring dimensions comprise keyword and sensitive word detection scores and quality inspection rule matching scores; and obtaining the final score of the voice work order file to be inspected according to the score corresponding to each quality inspection score dimension. According to the voice work order quality inspection method, the language types of the utterances of the consumers are firstly obtained, then the voice work order file is converted into the text based on the voice conversion model of the language types, so that the converted text is more accurate, conversion errors caused by different language types are avoided, the voice work order is inspected according to the text file with the accurate conversion, whether the service of customer service personnel can meet the requirements of the consumers can be analyzed according to the voices of the consumers, and the accuracy of the detection of the service quality of the customer service personnel is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments of the present invention will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of an application environment of an artificial intelligence-based voice work order quality inspection method according to an embodiment of the present invention;
FIG. 2 is a flow chart of an artificial intelligence based voice work order quality inspection method according to an embodiment of the present invention;
FIG. 3 is a flowchart illustrating the step S20 in FIG. 2;
FIG. 4 is a flowchart illustrating a step S20 in FIG. 2;
FIG. 5 is a flowchart illustrating a step S40 in FIG. 2;
FIG. 6 is a schematic diagram of a voice worker simple substance detection device based on artificial intelligence according to an embodiment of the invention;
FIG. 7 is a schematic diagram of a computer device according to an embodiment of the invention;
fig. 8 is a schematic diagram of another configuration of a computer device according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be understood that the terms "comprises" and "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It is also to be understood that the terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in this specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
The voice simple substance detection method based on artificial intelligence provided by the specific embodiment of the application of the invention can be applied to an application environment as shown in figure 1, wherein the front end, the service end and the user end are respectively in communication connection through a network. Customer service personnel use a user side to communicate with a consumer (not shown in the figure) through telephone, and the user side stores audio of voice communication between the customer service personnel and the consumer to form a voice work order file; after the customer service personnel and the customer end the call, the user side sends the voice work order file to the server side, so that the server side stores the voice work order file. The quality inspection personnel uses the front end to carry out quality inspection on the voice work order, specifically, the quality inspection personnel operates the front end to call the voice work order file from the service end according to the requirement so as to carry out quality inspection on the voice work order file, and the inspection result is uploaded to the service end. The server side is used for storing the voice work order file sent by the user side; meanwhile, the server side also stores a plurality of tools for assisting quality inspection personnel in performing quality inspection on the voice work order file, such as a voice conversion model for converting audio information of the voice work order file into text information; the model can be called by the front end, so that quality inspection personnel can efficiently and accurately conduct quality inspection on the voice work order file.
The embodiment of the invention can be used for the quality inspection service of complaints of credit card service departments of customer service centers of certain commercial banks; the credit card business department of the customer service center is provided with twenty user ends, a front end and a service end, and the twenty user ends are respectively in communication connection with the front end and the service end. Of course, the server may also be used in other departments, such as a savings card department, a financial department, etc., and when one server corresponds to multiple departments, the server may divide different storage areas according to each department, so that each storage area corresponds to one department. The user terminal is in communication connection with the cellular network or the telecommunication network so as to receive the telephone of the consumer at any time, and customer service personnel communicate with the consumer in real time through voice. The user side can store voice communication as audio files at regular time, for example, when the user side performs voice communication with the terminal of the consumer, the user side stores the voice in the communication process as audio files every other minute, and when the user side finishes voice communication with the terminal of the consumer, all the audio files in the voice communication are combined and stored; in addition, the user side can store the audio file of the voice communication after finishing the voice communication with the consumer. The quality inspector can directly call the stored audio file in real-time communication from the user terminal when the user terminal is in voice communication with the consumer terminal, can also directly connect the interface of the user terminal, and can answer the real-time voice communication between the user terminal and the consumer terminal so as to conduct real-time supervision and quality detection on the user terminal; after the user terminal and the consumer terminal end the voice communication, the quality inspector can call the voice work order file uploaded to the service terminal by the user terminal to the front end so as to detect the voice quality of the user terminal.
The user side and the front end can be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers and portable wearable devices. The server may be implemented by a stand-alone server or a server cluster formed by a plurality of servers. The present invention will be described in detail with reference to specific examples.
Referring to fig. 2, fig. 2 is a schematic flow chart of a voice work order quality inspection method based on artificial intelligence according to an embodiment of the present invention, which includes the following steps:
s10: and acquiring a voice work order file to be inspected.
In the above steps, the voice work order file is an audio file, and is a voice record of communication between customer service personnel and consumers, for example, when a complaint department of a credit card service center of a commercial bank answers a complaint about repayment stage, the consumer calls the complaint department by dialing a phone, one of the customer service personnel of the complaint department receives a phone call of the consumer, and a voice of telephone communication between the two people is the voice work order file. The execution main body for acquiring the voice work order file to be inspected is the front end, and the front end is controlled by the inspection personnel, or the front end can acquire the voice work order file according to the inspection rules set by the inspection personnel. The front end can be the voice work order file of the voice communication which is directly called from the user end, or can be called from the server end after the user end uploads the voice work order file to the server end. It should be noted that, because the user end does not completely correspond to the customer service personnel one by one, the user end generally does not store the voice work order file, so that the customer service personnel are prevented from listening to the voice work order files of other customer service personnel after replacing the user terminal; therefore, after the user end finishes the current voice communication with the consumer terminal, the corresponding voice work order file is immediately uploaded to the server end, and then if the calling request of the front end is not received within a certain preset time, the voice work order file in the user end is deleted; and if the call request of the front end is received within a certain preset time, the voice work order file is sent to the front end and then the voice work order file in the user end is deleted.
S20: and identifying the target language corresponding to the voice work order file.
In the step, after the front end reads the voice work order file, the target language corresponding to the voice work order file is identified. With the development of social economy, the popularization of networks is promoted, a great number of services tend to be networked, obvious regional removal phenomenon exists, the service objects of customer service staff are consumers from all places of the country, the corresponding consumers in all places also contain local accents, especially some consumers with older ages are difficult to speak standard mandarin, the content of voice communication brings heavy local accents, and even some consumers can directly communicate with the customer service staff by using local dialects. For example, a business bank opens hundreds of branches and thousands of business institutions in the country, and is distributed in various provinces, cities and autonomous regions in the country, the service objects of the business bank are people in various places in the country, the credit card business is related to each branch or business institution, when credit card consumers in different areas meet the problem of requiring complaints, the business bank dials a call to a complaint department, complaints are performed by adopting a mandarin or local dialect, the telephone dialed by the consumer is a consumer terminal, the telephone dialed by the consumer is a user terminal, and the user terminal records the telephone communication voice of customer service personnel and the consumer about the complaint content to form a voice work order file. The front end inputs the voice work order file into a trained language identification model and outputs the language of the consumer in the voice work order file. The language is not the type of language understood by the ordinary person, but the corresponding dialect. Common languages include northeast officials, beijing officials, ji Lu officials, jianliao officials, chinese officials, lanyingzhen officials, southwest officials, jianghuai officials, jin dialects, hui dialects, min dialects, yue dialects, hakka dialects, ganges dialects, ping dialects, etc., and each of the above languages may be subdivided according to actual needs. The language identification model is obtained by a professional technician through a series of training, and the model is put into a server after the training is finished so as to be conveniently called by the front end. In addition, the front end can also send the voice work order file to the service end, after the service end inputs the voice work order file into the language identification model, the service end outputs the target language, and then the service end sends the target language to the front end. The service end can also send the language identification model to the front end, and the front end outputs the target language after inputting the voice work order file into the language identification model. The training process of the special technician on the language training model is as follows: firstly, collecting a corpus, namely collecting a large amount of corpus data comprising slang and small languages, including text and voice data; then extracting features related to slang and small languages, such as pronunciation features, structural features, language habits and the like, from the collected text and voice data; then, establishing a dictionary, and establishing a dictionary corresponding to slang and small languages based on the feature extraction result, wherein in the process, related knowledge such as linguistics, phonology and the like is required to be used for analysis and processing; then, building a model, training the model by adopting a statistical model or a machine learning algorithm according to knowledge such as linguistics and phonology, analyzing and identifying rules of slang and small languages, building the model, and obtaining the trained model as a language identification model; the technician then uploads the trained language identification model to the server. Preferably, before the technician uploads the language identification model to the server, model evaluation can be performed, the trained language identification model is evaluated, the performance and the accuracy of different algorithms are compared, and finally, the language type of each voice message can be accurately identified by the trained language identification model.
S30: and inputting the voice work order file into a voice conversion model corresponding to the target language according to the target language, and outputting a text file.
In the above steps, each language corresponds to a voice conversion model, and the text corresponding to the voice can be more accurately recognized by using the corresponding voice conversion model. The server stores a plurality of voice conversion models, and each voice conversion model corresponds to one language. After the front end recognizes the target language corresponding to the voice work order file, the corresponding voice conversion model is searched at the server according to the target language, the voice work order file is input into the corresponding voice conversion model, and then information output by the voice conversion model is received, wherein the information is the text file of the voice work order file. Similarly, the voice conversion model stored in the server is obtained after the technician is trained, specifically, the technician inputs a plurality of audio files in the same language into a neural network model for training, and other training processes are conventional technical means of the technician except that the input audio files are limited to the same language.
S40: and scoring the text file according to a preset quality inspection scoring dimension and a preset scoring rule to obtain a score corresponding to each quality inspection scoring dimension, wherein the quality inspection scoring dimension comprises a keyword and sensitive word detection score and a quality inspection rule matching score.
In the step, quality detection of two dimensions, namely a keyword and sensitive word detection score and a quality inspection rule matching score, is preset, corresponding scoring rules are respectively set for the two dimensions, and the text file is subjected to assessment scoring according to each quality inspection scoring dimension and the scoring rule corresponding to the quality inspection scoring dimension, so that the score corresponding to each scoring dimension is correspondingly obtained. Performing keyword and sensitive word detection scoring on the text file according to a preset scoring rule to obtain scores corresponding to the keyword and sensitive word detection scoring; performing quality inspection rule matching scoring on the text file according to a preset scoring rule to obtain a score corresponding to the quality inspection rule matching scoring; the quality inspection rules comprise a text matching rule set preset according to quality inspection content; the text matching rule set includes word rules, phrase rules, and script rules. In general, when a consumer complains, the existing service is not necessarily satisfactory, and a professional customer service person must first express an apology to the consumer, so that keywords include words related to "disfigurement", "sorry", "bad meaning", "troublesome matter addition" and the like that must be said. Sensitive words refer to terms that are not suitable for presentation, such as related statements that a customer attendant says "must …", "cannot satisfy …", etc. to a customer that indicate a strong processing attitude to the customer or a reluctance to process a customer's appeal. Scoring is carried out according to the keywords and the keyword scoring rules to obtain keyword scores, and the sensitive words are scored according to the sensitive words and the scoring rules to obtain sensitive word scores. The keyword scoring rule comprises the number and type of occurrence of keywords, the keywords are not occurring or the number of occurrence is excessive, and the score of the dimension is relatively low; the sensitive word scoring rules include the number and type of occurrences of the sensitive word, the more the sensitive word occurs, the lower the score for that dimension.
S50: and obtaining the final score of the voice work order file to be inspected according to the score corresponding to each quality inspection scoring dimension.
In the above step, the scores of the two dimensions may be added to obtain the final score of the voice work order file. The higher the final score, the more specialized the customer service personnel can be.
According to the scheme, when the quality of the voice work order is detected, the target languages of the languages in the voice work order are identified according to the files in the voice work order, so that when the audio files are converted into the text files, the information of the languages can be combined and converted into the text files corresponding to the target languages, and the audio in the voice work order files can be accurately converted into the text files, so that the expression content of consumers can be more accurately identified, and whether the service of the customer service personnel can meet the requirements of the consumers or not can be more accurately analyzed according to the voices of the customer service personnel, and the accuracy of detecting the service quality of the customer service personnel is improved.
Referring to fig. 3, in some embodiments, the step S20 includes:
s21: and extracting the voice information of the consumer from the voice work order file.
In the above steps, the voice work order file is generated at the user side, specifically, when the customer service personnel and the consumer carry out voice communication with the consumer terminal through the user side, the user side is provided with two receiving channels, namely an audio signal from a microphone of the user side and an audio signal from a network interface, wherein the microphone collects the voice of the customer service personnel to form the audio signal; the network interface captures audio signals that the consumer speaks through the consumer terminal. When the user side combines the audio signals collected by the microphone and the audio signals collected by the network interface into a voice work order file, each section of audio signals are added with a corresponding interface label, the audio signals collected by the microphone are added with labels of customer service personnel, and the audio signals collected by the network interface are added with labels of consumers. When the front end extracts the sound information of the consumer from the voice work order file, the audio signal of the consumer is extracted according to the label of the consumer, and the sound information of the consumer can be extracted.
S22: and inputting the sound information into a preset language identification model, and outputting a target language corresponding to the sound information.
In this step, the voice information inputted into the predetermined language identification model is only the voice of the consumer. Since the customer service personnel are professional service personnel, the service standard includes speaking in the designated Mandarin language, so that the language of the service personnel is not required to be identified. Here, an audio signal including only the voice information of the consumer is input into the language recognition model, and the speaking accent of the consumer is output to obtain the target language. In the subsequent process of inputting the voice conversion model, the audio signal of the consumer can be input into the voice conversion model corresponding to the target language, the audio signal of the customer service personnel can be input into the voice conversion model corresponding to the Mandarin, and then the texts converted by the two voice conversion models are respectively combined to form an output text file.
Referring to fig. 4, in some embodiments, the step S20 further includes:
s23: and reading the consumer information of the consumer in the voice work order file.
S24: and identifying the target language based on the personal information in the consumer information.
In the above step, when the consumer terminal calls to the user terminal, the user terminal may obtain the terminal ID of the consumer terminal, then obtain consumer information of the consumer according to the terminal ID, and store the consumer information in the voice work order file, so that the front end may read the consumer information of the consumer from the voice work order file. The front end obtains the residence address or the home address of the consumer based on the personal information in the consumer information, and identifies the language corresponding to the residence address or the home address according to the residence address or the home address, so that the target language of the consumer in the voice work order file can be identified. In a specific embodiment, the terminal ID of the consumer terminal may be a mobile phone number, and the user terminal may query, according to the location of the mobile phone number, a region corresponding to the location, and further obtain a language corresponding to the region, so as to identify the target language. In some other embodiments, the consumer information includes an identification card number, and the user side determines a native address of the consumer according to the identification card number, so as to obtain a language corresponding to the native address, and then the target language can be identified. In addition, the relevant address information of the consumer can be obtained according to other effective address information in the personal information, and the corresponding language is determined according to the address information, so that the target language is obtained.
Referring to fig. 5, in some embodiments, the step S40 includes:
s41: acquiring a first text of an customer service person in the text file;
s42: extracting a first half text from the first text;
s43: and extracting keywords from the first half text, and grading the first text based on the keywords and a preset keyword grading rule.
In the above steps, the keywords are used for scoring customer service personnel, mainly based on forward addition. In the communication with consumers, customer service personnel have a state of sorry in the early stage of communication due to the demand of the complaints of the consumers, and take the emotion of the consumer as a main part, and communicate specific complaint contents after the emotion of the consumer is stable so as to solve the complaints of the consumers. Therefore, in the communication process of the first half of customer service staff, the expressed word eyes have related word eyes such as sorry, disfigurement, inconvenience, trouble and the like. After a user side obtains a voice work order file and a text file corresponding to the voice work order file, first, extracting a first text of a customer service person according to a label of the customer service person in the text file; the front part of the field in the first text is then extracted as the first half of the text. And extracting keywords from the front half text, wherein a keyword word stock is stored in the front end, a plurality of keywords are arranged in the keyword word stock, and the front end matches each keyword in the front half text and the keyword word stock to obtain successfully matched keywords. And scoring the first half text according to the keywords and the keyword scoring rule, namely scoring the first text of the customer service staff to obtain the keyword score. In some embodiments, the specific method for extracting the first half text from the first text is: acquiring the length of a first text to obtain the number of characters; obtaining the extraction length corresponding to the character number of the first text according to the mapping relation between the preset character number and the extraction length; and extracting the first half text from the first text according to the extraction length. For example, there are one thousand characters in the first text, i.e., the number of characters is 1000; in a preset mapping relation between the number of characters and the extraction length, the extraction length corresponding to the number of characters 1000 is 300, and the first 300 characters in the first text are extracted to be used as the first half text. In some embodiments, the mapping relationship between the preset number of characters and the extraction length may be a proportional relationship, for example, the first text has one thousand characters, that is, the number of characters is 1000, the corresponding mapping relationship is 20%, that is, the extraction length refers to 20% of the number of characters, and the extracted first half text is the first 200 characters of the first text. Through the steps, whether the customer service personnel can timely calm the emotion of the customer or not when dealing with the complaint of the customer can be accurately obtained, and the service quality of the customer service personnel can be rapidly and efficiently accurately evaluated.
In some embodiments, after the step of step S41, the method further includes:
s44: and extracting a second half text from the first text.
S45: and extracting sensitive words from the second half text, and scoring the first text based on the sensitive words and a preset sensitive word scoring rule.
In the above steps, the sensitive words are used for scoring customer service personnel, mainly based on negative deduction. In the latter half of consumer complaints, customer service personnel need to reply positively to the consumer's complaint content, indicating that the complaint is to be handled, and if withholding or indicating that the complaint cannot be handled, the image of banking business in the mind of the consumer is reduced. Therefore, the complaint content is not allowed to appear in the second half, and related words such as unresolved and unresolved are not solved. In this embodiment, a field in the latter part of the first text of the customer service person is extracted as the latter half text. And extracting sensitive words from the second half text, wherein a sensitive word stock is stored in the front end, a plurality of sensitive words are arranged in the sensitive word stock, and the front end matches each sensitive word in the second half text and the sensitive word stock to obtain the successfully matched sensitive word. And scoring the second text according to the sensitive words and the scoring rule of the sensitive words, namely scoring the first text of the customer service personnel to obtain the score of the sensitive words.
In some embodiments, the specific method for extracting the second half text from the first text is as follows: acquiring the length of a first text to obtain the number of characters; obtaining the extraction length corresponding to the character number of the first text according to the mapping relation between the preset character number and the extraction length; and extracting the second half text from the first text according to the extraction length. For example, there are two thousand characters in the first text, i.e., the number of characters is 2000; in the mapping relation between the preset character number and the extraction length, the extraction length corresponding to the character number 2000 is 500, and the last 500 characters in the first text are extracted to be used as the second half text. In some embodiments, the mapping relationship between the preset number of characters and the extraction length may be a proportional relationship, for example, the first text has two thousand characters, i.e. the number of characters is 2000, the corresponding mapping relationship is 25%, i.e. the extraction length refers to 25% of the number of characters, and the extracted second half text is the last 500 characters of the first text. Through the steps, whether a processing method for solving the complaint content of the consumer can be provided when customer service personnel deal with the complaint of the consumer can be accurately obtained, and the service quality of the customer service personnel can be rapidly and efficiently evaluated accurately.
In some embodiments, the step of step S20 further includes:
s25: and preprocessing the voice work order file, wherein the preprocessing comprises noise reduction processing and filtering processing.
S26: and identifying the target language corresponding to the voice work order file based on the preprocessed voice work order file.
In the above steps, the signal preprocessing is performed on the voice work order file, and mainly the processing such as denoising, noise reduction, filtering, voice segmentation and the like is performed on the voice work order file. For example, noise reduction processing refers to the ability to detect and reduce noise in real time by techniques such as algorithms and digital signal processing. For example, ANC (active noise reduction technique) may collect ambient noise through a microphone, analyze and process it, and output a waveform opposite to the noise, thereby canceling or reducing noise interference. After the voice work order file is preprocessed, noise, environmental noise and the like when customer service personnel communicate with consumers can be removed, the signal power of effective sound is improved, the follow-up voice feature extraction is convenient to achieve more accuracy, and the identified languages of the consumers are more accurate.
In some embodiments, after the step of obtaining the final score of the voice work order file to be inspected according to the score corresponding to each quality inspection score dimension, the method further includes:
S60: and analyzing the emotion value of the voice work order file.
S70: and grading the customer service personnel corresponding to the voice work order file according to the emotion value and the final score.
In the above steps, the customer service personnel often complain about the customer or the customer is excited in the process of communicating with the customer, and the customer service personnel must be calm. Therefore, in standard service, customer service personnel must not communicate with the customer using very fast frequency-changing fundamental frequencies, relatively sharp tones, very fast speech rates; the fundamental frequency can be obtained by calculating the period and amplitude information of the extracted voice signal, the tone can be extracted by an automatic voice recognition system based on acoustic analysis, the voice speed can be estimated by using a voice synthesis engine, extracting fundamental frequency contours, accent positions and the like, the voice speed is usually estimated by calculating short-time energy and zero crossing rate change in the voice signal, and the voice speed can be controlled by resampling the voice signal. The front end can input the voice work order file into an emotion value model, and the emotion value model finally outputs emotion values. And then, the customer service personnel are rated according to the emotion value and the final score, and the higher the emotion value is towards the intermediate value, the higher the final score is, and the higher the corresponding customer service personnel are rated. In some embodiments, the voice work order file is preprocessed to extract the fundamental frequency, the tone and the speech speed, and then the three characteristics are input into the emotion value model to obtain the emotion value.
In some embodiments, the execution body of the method may also be a user side; when customer service personnel use the user terminal to communicate with the consumer terminal of the consumer, the user terminal converts the voice acquired in real time into a voice work order file, and executes the steps of S20-S50, and then the final score is loaded on the user terminal, so that the customer service personnel can know the service quality of the customer service personnel in real time.
Therefore, in the scheme, the language types of the speaking of the consumers are firstly obtained, and then the voice work order file is converted into the text based on the voice conversion model of the language types, so that the converted text is more accurate, and conversion errors caused by different language types are avoided, and therefore, the voice work order is inspected according to the text file with the accurate conversion, whether the service of the customer service personnel can meet the requirements of the consumers can be analyzed according to the voices of the consumers, and the accuracy of the detection of the service quality of the customer service personnel is improved.
It should be understood that the sequence number of each step in the foregoing embodiment does not mean that the execution sequence of each process should be determined by the function and the internal logic, and should not limit the implementation process of the embodiment of the present invention.
In some embodiments, an artificial intelligence-based voice simple detection device is provided, and the artificial intelligence-based voice simple detection device corresponds to the artificial intelligence-based voice simple detection method in the above embodiments one by one. As shown in fig. 6, the voice worker simple inspection device based on artificial intelligence includes an acquisition module 101, an identification module 102, a conversion module 103, a scoring module 104 and an evaluation module 105. The functional modules are described in detail as follows: an acquisition module 101, configured to acquire a voice work order file to be inspected;
the recognition module 102 is configured to recognize a target language corresponding to the voice work order file;
the conversion module 103 is configured to input the voice work order file into a voice conversion model corresponding to the target language according to the target language, and output a text file;
the scoring module 104 is configured to score the text file according to a preset quality inspection scoring dimension and a preset scoring rule, so as to obtain a score corresponding to each quality inspection scoring dimension, where the quality inspection scoring dimension includes a keyword and sensitive word detection score and a quality inspection rule matching score;
and the evaluation module 105 is used for acquiring the final score of the voice work order file to be inspected according to the score corresponding to each quality inspection score dimension.
In some embodiments, the identification module 102 is specifically configured to:
extracting the sound information of the consumer from the voice work order file;
and inputting the sound information into a preset language identification model, and outputting a target language corresponding to the sound information.
In some embodiments, the identification module 102 is specifically configured to:
reading consumer information of consumers in the voice work order file;
and identifying the target language based on the personal information in the consumer information.
In some embodiments, the scoring module 104 is specifically configured to:
acquiring a first text of an customer service person in the text file;
extracting a first half text from the first text;
and extracting keywords from the first half text, and grading the first text based on the keywords and a preset keyword grading rule.
In some embodiments, scoring module 104 is further configured to:
extracting a second half text from the first text;
and extracting sensitive words from the second half text, and scoring the first text based on the sensitive words and a preset sensitive word scoring rule.
In some embodiments, the identification module 102 is further configured to:
Preprocessing the voice work order file, wherein the preprocessing comprises noise reduction processing and filtering processing;
and identifying the target language corresponding to the voice work order file based on the preprocessed voice work order file.
In some embodiments, the artificial intelligence based voice industry simple inspection device is further configured to:
analyzing emotion values of the voice work order file;
and grading the customer service personnel corresponding to the voice work order file according to the emotion value and the final score.
The invention provides a voice work simple substance detection device based on artificial intelligence, which firstly acquires the language types of the speaking of a consumer, and then converts a voice work sheet file into a text based on a voice conversion model of the language types, so that the converted text is more accurate, and conversion errors caused by different language types are avoided, thereby carrying out quality detection on the voice work sheet according to the text file with accurate conversion, analyzing whether the service of customer service personnel can meet the requirements of the consumer according to the voice of the consumer more accurately, and improving the accuracy of detecting the service quality of the customer service personnel.
For specific limitation of the voice work simple substance detection device based on artificial intelligence, reference may be made to the limitation of the voice work simple substance detection method based on artificial intelligence hereinabove, and the description thereof will not be repeated here. All or part of the modules in the voice engineering simple detection device based on the artificial intelligence can be realized by software, hardware and a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules. The modules in all embodiments of the present invention may be implemented by a general-purpose integrated circuit, such as a CPU (central processing unit), or by an ASIC (application specific integrated circuit).
In one embodiment, a computer device is provided, which may be a server, and the internal structure of which may be as shown in fig. 7. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes non-volatile and/or volatile storage media and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface of the computer device is for communicating with an external client via a network connection. When the computer program is executed by a processor, the function or the step of a voice work simple substance detection method service end side based on artificial intelligence is realized, firstly, the language type of a customer speaking is obtained, then, a voice work sheet file is converted into a text based on a voice conversion model of the language type, so that the converted text is more accurate, the conversion error caused by different language types is avoided, the voice work sheet is subjected to quality inspection according to the text file with the accurate conversion, whether the service of customer service personnel can meet the requirements of the customer can be analyzed according to the voice of the customer more accurately, and the accuracy of the detection of the service quality of the customer service personnel is improved.
In one embodiment, a computer device is provided, which may be a client, the internal structure of which may be as shown in fig. 8. The computer device includes a processor, a memory, a network interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface of the computer device is for communicating with an external server via a network connection. When the computer program is executed by a processor, the function or the step of a voice work simple substance detection method on the user side based on artificial intelligence is realized, the language type of a customer speaking is firstly obtained, then the voice work sheet file is converted into a text based on a voice conversion model of the language type, so that the converted text is more accurate, the conversion error caused by different language types is avoided, the voice work sheet is subjected to quality inspection according to the text file with accurate conversion, whether the service of customer service personnel can meet the requirements of the customer can be analyzed according to the voice of the customer more accurately, and the accuracy of the detection of the service quality of the customer service personnel is improved.
In one embodiment, a computer device is provided comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the steps of when executing the computer program:
acquiring a voice work order file to be inspected;
identifying a target language corresponding to the voice work order file;
inputting the voice work order file into a voice conversion model corresponding to the target language according to the target language, and outputting a text file;
scoring the text file according to preset quality inspection scoring dimensions and preset scoring rules to obtain scores corresponding to each quality inspection scoring dimension, wherein the quality inspection scoring dimensions comprise keyword and sensitive word detection scores and quality inspection rule matching scores;
and obtaining the final score of the voice work order file to be inspected according to the score corresponding to each quality inspection score dimension.
Firstly, the language types of the speaking of the consumers are obtained, then, the voice work order file is converted into the text based on the voice conversion model of the language types, so that the converted text is more accurate, and conversion errors caused by different language types are avoided, and therefore, the quality of the voice work order is checked according to the text file with accurate conversion, whether the service of customer service staff can meet the requirements of the consumers can be analyzed according to the voices of the consumers, and the accuracy of the detection of the service quality of the customer service staff is improved.
In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of:
acquiring a voice work order file to be inspected;
identifying a target language corresponding to the voice work order file;
inputting the voice work order file into a voice conversion model corresponding to the target language according to the target language, and outputting a text file;
scoring the text file according to preset quality inspection scoring dimensions and preset scoring rules to obtain scores corresponding to each quality inspection scoring dimension, wherein the quality inspection scoring dimensions comprise keyword and sensitive word detection scores and quality inspection rule matching scores;
and obtaining the final score of the voice work order file to be inspected according to the score corresponding to each quality inspection score dimension.
Firstly, the language types of the speaking of the consumers are obtained, then, the voice work order file is converted into the text based on the voice conversion model of the language types, so that the converted text is more accurate, and conversion errors caused by different language types are avoided, and therefore, the quality of the voice work order is checked according to the text file with accurate conversion, whether the service of customer service staff can meet the requirements of the consumers can be analyzed according to the voices of the consumers, and the accuracy of the detection of the service quality of the customer service staff is improved.
It should be noted that, the functions or steps implemented by the computer readable storage medium or the computer device may correspond to the relevant descriptions of the server side and the client side in the foregoing method embodiments, and are not described herein for avoiding repetition.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, i.e. the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-described functions.
The above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention, and are intended to be included in the scope of the present invention.

Claims (10)

1. The voice work order quality inspection method based on artificial intelligence is characterized by comprising the following steps:
acquiring a voice work order file to be inspected;
identifying a target language corresponding to the voice work order file;
Inputting the voice work order file into a voice conversion model corresponding to the target language according to the target language, and outputting a text file;
scoring the text file according to preset quality inspection scoring dimensions and preset scoring rules to obtain scores corresponding to each quality inspection scoring dimension, wherein the quality inspection scoring dimensions comprise keyword and sensitive word detection scores and quality inspection rule matching scores;
and obtaining the final score of the voice work order file to be inspected according to the score corresponding to each quality inspection score dimension.
2. The method for quality inspection of voice work orders based on artificial intelligence according to claim 1, wherein the step of identifying the target language corresponding to the voice work order file comprises:
extracting the sound information of the consumer from the voice work order file;
and inputting the sound information into a preset language identification model, and outputting a target language corresponding to the sound information.
3. The method for quality inspection of voice work orders based on artificial intelligence according to claim 1, wherein the step of identifying the target language corresponding to the voice work order file comprises:
Reading consumer information of consumers in the voice work order file;
and identifying the target language based on the personal information in the consumer information.
4. The method for quality inspection of voice work orders based on artificial intelligence according to claim 1, wherein the step of scoring the text file according to a preset quality inspection scoring dimension and a preset scoring rule to obtain a score corresponding to each quality inspection scoring dimension comprises:
acquiring a first text of an customer service person in the text file;
extracting a first half text from the first text;
and extracting keywords from the first half text, and grading the first text based on the keywords and a preset keyword grading rule.
5. The artificial intelligence based voice work order quality inspection method of claim 4, wherein after the step of obtaining the first text of the customer in the text file, comprising:
extracting a second half text from the first text;
and extracting sensitive words from the second half text, and scoring the first text based on the sensitive words and a preset sensitive word scoring rule.
6. The method for quality inspection of voice work orders based on artificial intelligence according to claim 1, wherein the step of identifying the target language corresponding to the voice work order file comprises:
preprocessing the voice work order file, wherein the preprocessing comprises noise reduction processing and filtering processing;
and identifying the target language corresponding to the voice work order file based on the preprocessed voice work order file.
7. The method for quality inspection of voice work orders based on artificial intelligence according to claim 1, wherein after the step of obtaining the final score of the voice work order file to be inspected according to the score corresponding to each quality inspection score dimension, the method further comprises:
analyzing emotion values of the voice work order file;
and grading the customer service personnel corresponding to the voice work order file according to the emotion value and the final score.
8. Device is examined to pronunciation worker simple substance based on artificial intelligence, a serial communication port, include:
the acquisition module is used for acquiring the voice work order file to be inspected;
the recognition module is used for recognizing the target language corresponding to the voice work order file;
the conversion module is used for inputting the voice work order file into a voice conversion model corresponding to the target language according to the target language and outputting a text file;
The scoring module is used for scoring the text file according to a preset quality inspection scoring dimension and a preset scoring rule to obtain a score corresponding to each quality inspection scoring dimension, wherein the quality inspection scoring dimension comprises a keyword and sensitive word detection score and a quality inspection rule matching score;
and the evaluation module is used for acquiring the final score of the voice work order file to be inspected according to the score corresponding to each quality inspection score dimension.
9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the artificial intelligence based voice work order quality inspection method of any of claims 1 to 7 when the computer program is executed.
10. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the steps of the artificial intelligence based voice work order quality inspection method of any of claims 1 to 7.
CN202310653099.4A 2023-06-02 2023-06-02 Voice work order quality inspection method, device, equipment and medium based on artificial intelligence Pending CN116705003A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310653099.4A CN116705003A (en) 2023-06-02 2023-06-02 Voice work order quality inspection method, device, equipment and medium based on artificial intelligence

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310653099.4A CN116705003A (en) 2023-06-02 2023-06-02 Voice work order quality inspection method, device, equipment and medium based on artificial intelligence

Publications (1)

Publication Number Publication Date
CN116705003A true CN116705003A (en) 2023-09-05

Family

ID=87840379

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310653099.4A Pending CN116705003A (en) 2023-06-02 2023-06-02 Voice work order quality inspection method, device, equipment and medium based on artificial intelligence

Country Status (1)

Country Link
CN (1) CN116705003A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117610569A (en) * 2023-11-24 2024-02-27 中国电信股份有限公司技术创新中心 Operation and maintenance work sheet quality inspection method, device, equipment and medium

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117610569A (en) * 2023-11-24 2024-02-27 中国电信股份有限公司技术创新中心 Operation and maintenance work sheet quality inspection method, device, equipment and medium

Similar Documents

Publication Publication Date Title
CN112804400B (en) Customer service call voice quality inspection method and device, electronic equipment and storage medium
CN110378562B (en) Voice quality inspection method, device, computer equipment and storage medium
WO2019210557A1 (en) Voice quality inspection method and device, computer device and storage medium
CN111311327A (en) Service evaluation method, device, equipment and storage medium based on artificial intelligence
CN110177182B (en) Sensitive data processing method and device, computer equipment and storage medium
US12033163B2 (en) Systems and methods for detecting complaint interactions
CN107886951B (en) Voice detection method, device and equipment
CN111597818B (en) Call quality inspection method, device, computer equipment and computer readable storage medium
CN110929011A (en) Conversation analysis method, device and equipment
CN116705003A (en) Voice work order quality inspection method, device, equipment and medium based on artificial intelligence
CN114818649A (en) Service consultation processing method and device based on intelligent voice interaction technology
CN113709313B (en) Intelligent quality inspection method, device, equipment and medium for customer service call data
CN115983285A (en) Questionnaire auditing method, device, electronic equipment and storage medium
CN115831125A (en) Speech recognition method, device, equipment, storage medium and product
CN114707515A (en) Method and device for judging dialect, electronic equipment and storage medium
CN110853674A (en) Text collation method, apparatus, and computer-readable storage medium
CN115022471A (en) Intelligent robot voice interaction system and method
CN114356982A (en) Marketing compliance checking method and device, computer equipment and storage medium
CN114067842A (en) Customer satisfaction degree identification method and device, storage medium and electronic equipment
CN113822029A (en) Customer service assistance method, device and system
CN111916073B (en) Robot outbound control method and device, server, and computer-readable storage medium
WO2023173966A1 (en) Speech identification method, terminal device, and computer readable storage medium
CN113179352B (en) Call quality determination method, call quality determination device, and computer-readable storage medium
CN113726962B (en) Method and device for evaluating service quality, electronic device and storage medium
CN113254621B (en) Seat call prompting method and device, computer equipment and storage medium

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