CN113011159A - Artificial seat monitoring method and device, electronic equipment and storage medium - Google Patents
Artificial seat monitoring method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN113011159A CN113011159A CN202110309936.2A CN202110309936A CN113011159A CN 113011159 A CN113011159 A CN 113011159A CN 202110309936 A CN202110309936 A CN 202110309936A CN 113011159 A CN113011159 A CN 113011159A
- Authority
- CN
- China
- Prior art keywords
- call
- information
- text information
- call text
- monitoring
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 228
- 238000000034 method Methods 0.000 title claims abstract description 104
- 238000012549 training Methods 0.000 claims abstract description 60
- 230000008569 process Effects 0.000 claims abstract description 45
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 27
- 238000012806 monitoring device Methods 0.000 claims abstract description 6
- 238000012545 processing Methods 0.000 claims description 25
- 238000012360 testing method Methods 0.000 claims description 13
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 12
- 238000013473 artificial intelligence Methods 0.000 abstract description 3
- 238000002372 labelling Methods 0.000 description 14
- 230000006870 function Effects 0.000 description 11
- 239000003795 chemical substances by application Substances 0.000 description 9
- 230000006854 communication Effects 0.000 description 7
- 238000013519 translation Methods 0.000 description 7
- 238000004458 analytical method Methods 0.000 description 6
- 238000003058 natural language processing Methods 0.000 description 6
- 230000003287 optical effect Effects 0.000 description 6
- 230000004044 response Effects 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 238000003672 processing method Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000003993 interaction Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 239000013307 optical fiber Substances 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 238000012552 review Methods 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- RWSOTUBLDIXVET-UHFFFAOYSA-N Dihydrogen sulfide Chemical compound S RWSOTUBLDIXVET-UHFFFAOYSA-N 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000008451 emotion Effects 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 230000002427 irreversible effect Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000036651 mood Effects 0.000 description 1
- 238000012567 pattern recognition method Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
-
- 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
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Marketing (AREA)
- Artificial Intelligence (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Finance (AREA)
- General Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Human Computer Interaction (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Telephonic Communication Services (AREA)
Abstract
The embodiment of the invention relates to the technical field of artificial intelligence, and discloses an artificial seat monitoring method, an artificial seat monitoring device, electronic equipment and a storage medium, wherein the method comprises the following steps: acquiring call voice information in a call process; converting the call voice information into call text information based on a voice recognition algorithm; inputting the call text information into a target monitoring model in real time, and determining monitoring result information of the current call service; and the target monitoring model is obtained by training the initial monitoring model according to the historical call text information and the marking information corresponding to the historical call text information. By the technical scheme of the embodiment of the invention, the service process of monitoring the manual position in real time is realized, the cost of human resources is reduced, and the technical effect of reducing the complaint rate of users is further achieved.
Description
Technical Field
The embodiment of the invention relates to the technical field of artificial intelligence, in particular to an artificial seat monitoring method and device, electronic equipment and a storage medium.
Background
In the process of providing services for customers, situations such as mood fluctuation, misunderstanding of customer intentions, unstable services and the like may occur in the manual seats of the call center, which affect customer experience, possibly cause customer complaints and irreversible influence on company images.
In the prior art, a call center usually monitors a human seat by using two monitoring methods, namely remote random monitoring and on-site overhearing. Both of the two monitoring methods require monitoring personnel to participate, and special monitoring posts need to be set, so that great waste is caused to human resources of the call center, and the monitoring efficiency is low. Moreover, monitoring personnel can only monitor the service process of a few manual seats, the monitoring of the manual seats of the whole call center cannot be guaranteed, and the service quality and the service efficiency of the call center cannot be guaranteed.
Disclosure of Invention
The embodiment of the invention provides a method and a device for monitoring an artificial seat, electronic equipment and a storage medium, which are used for realizing the technical effects of monitoring the service process of the artificial seat in real time, reducing the cost of human resources and further reducing the complaint rate of users.
In a first aspect, an embodiment of the present invention provides a method for monitoring an artificial seat, where the method includes:
acquiring call voice information in a call process;
converting the call voice information into call text information based on a voice recognition algorithm;
inputting the call text information into a target monitoring model in real time, and determining monitoring result information of the current call service; and the target monitoring model is obtained by training the initial monitoring model according to the historical call text information and the marking information corresponding to the historical call text information.
In a second aspect, an embodiment of the present invention further provides an artificial seat monitoring device, where the device includes:
the call voice information acquisition module is used for acquiring call voice information in the call process;
the call text information conversion module is used for converting the call voice information into call text information based on a voice recognition algorithm;
a monitoring result information determining module, configured to input the call text information into a target monitoring model in real time, and determine monitoring result information of the current call service; and the target monitoring model is obtained by training the initial monitoring model according to the historical call text information and the marking information corresponding to the historical call text information.
In a third aspect, an embodiment of the present invention further provides an electronic device, where the electronic device includes:
one or more processors;
a storage device for storing one or more programs,
when the one or more programs are executed by the one or more processors, the one or more processors implement the method for human seat listening according to any one of the embodiments of the present invention.
In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the artificial seat monitoring method according to any one of the embodiments of the present invention.
According to the technical scheme of the embodiment of the invention, the call voice information is acquired in the call process, the call voice information is converted into the call text information based on the voice recognition algorithm so as to acquire the call text information for display and/or analysis, the call text information is input into the target monitoring model in real time, and the monitoring result information of the current call service is determined, so that the problems that a large amount of human resources are consumed to monitor the manual seats and all the manual seats cannot be monitored in real time are solved, the service process of monitoring the manual seats in real time is realized, the cost of the human resources is reduced, and the technical effect of reducing the complaint rate of users is further achieved.
Drawings
In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, a brief description is given below of the drawings used in describing the embodiments. It should be clear that the described figures are only views of some of the embodiments of the invention to be described, not all, and that for a person skilled in the art, other figures can be derived from these figures without inventive effort.
Fig. 1 is a schematic flow chart of a method for monitoring an artificial seat according to a first embodiment of the present invention;
fig. 2 is a schematic flow chart of a manual seat monitoring method according to a second embodiment of the present invention;
fig. 3 is a schematic flow chart of a method for monitoring an artificial seat according to a third embodiment of the present invention;
fig. 4 is a schematic flow chart of a manual seat monitoring method according to a fourth embodiment of the present invention;
fig. 5 is a schematic structural diagram of an artificial seat monitoring device according to a fifth embodiment of the present invention;
fig. 6 is a schematic structural diagram of an electronic device according to a sixth embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.
Example one
Fig. 1 is a schematic flow chart of a manual seat monitoring method according to an embodiment of the present invention, where the present embodiment is applicable to a situation where a call between a manual seat and a user is monitored, the method may be executed by a manual seat monitoring apparatus, the apparatus may be implemented in a form of software and/or hardware, and the hardware may be an electronic device, and optionally, the electronic device may be a PC terminal, and the like.
As shown in fig. 1, the method of this embodiment specifically includes the following steps:
and S110, acquiring call voice information in the call process.
The communication process may be a process of a manual seat communicating with a user, and the communication voice information may be audio information in the communication process.
Specifically, the call between the human seat and the user may be initiated through any one of a mobile phone, a landline, voice software and a voice web page, and when a call request is received, a human seat is selected from the idle human seats and allocated to the user, so as to solve a problem or transact a service for the user. When the conversation between the manual seat and the user starts, the conversation voice information in the conversation process is obtained. The call voice information in the call process can be continuously acquired for monitoring and processing.
And S120, converting the call voice information into call text information based on a voice recognition algorithm.
The speech recognition algorithm is a pattern recognition method based on speech characteristic parameters, and the speech recognition algorithm can enable a computer to convert speech information into text information through recognition and understanding. The call text information may be text information corresponding to the call voice information.
Specifically, after the call voice information is acquired, the call voice information may be processed based on a voice recognition algorithm, and the audio information is converted into text information, so as to obtain call text information. This may be when a longer pause in speech is detected, for example: and 1s, converting the call voice information between two pauses into call text information, or converting the acquired call voice information into the call text information in real time.
It should be noted that the speech recognition algorithm is mainly divided into three major categories, the first category is a model matching method, including Vector Quantization (VQ), Dynamic Time Warping (DTW), and the like; the second category is probabilistic methods, including Gaussian Mixed Models (GMMs), Hidden Markov Models (HMMs), etc.; the third category is a discriminator classification method, such as a Support Vector Machine (SVM), an Artificial Neural Network (ANN), a Deep Neural Network (DNN), and the like, and various combination methods. A suitable speech recognition algorithm may be selected based on the monitoring requirement, and the specific speech recognition algorithm is not specifically limited in this embodiment.
S130, inputting the call text information into the target monitoring model in real time, and determining the monitoring result information of the current call service.
And the target monitoring model is obtained by training the initial monitoring model according to the historical call text information and the marking information corresponding to the historical call text information. The interception result information may be an output result of the target interception model, and may be, for example, a quality of service level, question answering assistance information, or the like.
Specifically, the call text information is input into the target monitoring model in real time, and the call text information can be analyzed and the monitoring result information of the current call service can be generated.
It should be noted that the target listening model may be a Natural Language Processing (NLP) model, and the NLP model may be used to study Language problems in human-to-human interaction and human-to-computer interaction, so that a machine can understand and process Language like a human, which is a branch of artificial intelligence. The NLP model may specifically include word segmentation, grammar labeling, syntax analysis, information classification, information retrieval, and the like.
It should be further noted that, if the interception result information includes a service quality level, the service quality of the current call service may be determined by the target interception model, for example: the quality of service level can be determined for the whole call service process, or the quality of service level can be determined for each reply of the manual position, so that the quality of service of each manual position can be monitored. If the monitoring result information comprises question answering auxiliary information, the information of the data and the like required by the reply user can be determined through the target monitoring model, so that the manual position can accurately and quickly answer the user, and a large amount of time is saved.
According to the technical scheme of the embodiment of the invention, the call voice information is acquired in the call process, the call voice information is converted into the call text information based on the voice recognition algorithm so as to acquire the call text information for display and/or analysis, the call text information is input into the target monitoring model in real time, and the monitoring result information of the current call service is determined, so that the problems that a large amount of human resources are consumed to monitor the manual seats and all the manual seats cannot be monitored in real time are solved, the service process of monitoring the manual seats in real time is realized, the cost of the human resources is reduced, and the technical effect of reducing the complaint rate of users is further achieved.
Example two
Fig. 2 is a schematic flow chart of a manual seat monitoring method according to a second embodiment of the present invention, and in this embodiment, reference may be made to the technical solution of this embodiment for a conversion manner of call voice information and call text information and a processing manner of call text information and monitoring result information based on the foregoing embodiments. The same or corresponding terms as those in the above embodiments are not explained in detail herein.
As shown in fig. 2, the method of this embodiment specifically includes the following steps:
s210, acquiring call voice information in the call process.
S220, initializing the voice recognition resource pool, and dividing the voice recognition resource pool into at least one sub-resource pool.
The speech recognition resource pool can be a resource set for speech recognition divided from resources of a computer or a server. A resource pool is a configuration mechanism for partitioning resources of a computer or server, etc. The resources may include computing resources, storage resources, and/or network resources, among others.
Specifically, basic resources required for speech recognition may be configured for the speech recognition resource pool to initialize the speech recognition resource pool. Further, in order to improve the voice recognition response speed, the voice recognition resource pool may be divided into a plurality of sub-resource pools, for example, 500 sub-resource pools. The division mode of the voice recognition resource pool can be equally divided into multiple parts, and the specific division mode can be set according to actual requirements.
And S230, distributing a sub-resource pool for the call voice information, and converting the call voice information into call text information based on a voice recognition algorithm and the sub-resource pool.
Specifically, when a call between the human seat and the user starts, one sub-resource pool may be selected from the free sub-resource pools and allocated to the current call, that is, allocated to the call voice message. And calling resources in the sub-resource pool to process the call voice information according to a voice recognition algorithm to obtain call text information.
And S240, displaying the call text information on the terminal equipment of the current manual position in real time.
Specifically, the call text information translated by the interaction between the manual seat and the user can be displayed on the terminal equipment of the manual seat in real time. The advantage of displaying the call text information on the terminal equipment of the current manual position in real time is that: the method and the system have the advantages that the manual seat can conveniently check and review the call process, the problem that the user is forgotten by the manual seat is avoided, and the intention of the user can be understood according to call text information when the content expressed by the user is not understood.
And S250, inputting the call text information into the target monitoring model in real time, and determining the monitoring result information of the current call service.
And S260, if the monitoring result of the current call service is the warning information, highlighting the call text information corresponding to the warning information.
The warning message may be a message output based on the target monitoring model, and may be caused by a forbidden language, a contraindication, and the like in a reply of the manual seat during a call, or may be caused by a reply content not corresponding to a user requirement, or may be caused by dissatisfaction of an expression of the user on the manual seat in a language of the user.
Specifically, if the monitoring result of the current call service is the warning information, in order to remind the manual seat to pay attention to the service attitude and ensure the service quality, the call text information triggering the warning information may be highlighted, so that the manual seat can be pertinently improved and the service quality is improved.
On the basis of the foregoing embodiments, if the monitoring result information includes the warning information, the monitoring result information may respond to the warning information, and the responding manner may include at least one of the following:
and responding to the warning information, and sending warning prompt information to the terminal equipment of the current manual position.
The warning prompt information may be information describing a reason for warning and performing reminding, for example: "detect the forbidden language in the current call and please note the service quality. "and the like.
Specifically, in response to the warning information, warning prompt information may be generated according to the warning information, and the warning prompt information may be pushed to the terminal device of the current human seat to remind the human seat of paying attention.
Optionally, the short prompt message may also be sent to the terminal device for managing the agent corresponding to the current manual agent.
The management seat may be a person who manages a manual seat, and may be a seat group leader, for example.
Specifically, the warning prompt information may be generated based on the warning information, for example: "warning is received by the manual agent XX number. "and the like. And the prompt information can be pushed to the terminal equipment for managing the seat.
And secondly, responding to the warning information, and sending warning prompt information to the terminal equipment of the current manual seat and the terminal equipment of the management seat corresponding to the current manual seat.
Specifically, in response to the warning information, warning prompt information may be generated according to the warning information, and the warning prompt information may be pushed to the terminal device of the current human seat to remind the human seat of paying attention. In addition, the warning prompt information can be sent to the management seat, so that the management seat can pay attention to the current call in time. Meanwhile, the management seat can check the current call voice text information and/or monitor the current call voice information in real time, and the management seat can provide help for the current manual seat when appropriate so as to promote the current call process.
And thirdly, responding to the warning information, cutting off the call of the current manual seat, and switching the current call service to the management seat corresponding to the current manual seat.
Specifically, the communication between the current manual seat and the user can be cut off in response to the warning information, so that the user is prevented from generating discontent emotion. And the current call service is switched to the management seat corresponding to the current manual seat so that the management seat completes subsequent services. At this time, the call text information can be displayed on the terminal device of the management agent, so that the management agent can review the previous call content, understand the user requirements, and provide services for the user better.
On the basis of the foregoing embodiments, if the warning information includes at least two warning levels, a target processing manner corresponding to the warning level may be determined according to the warning level, and the current call service may be processed according to the target processing manner.
The alert level may be a level classified according to the alert information, and the target processing mode may be an operation of processing the current call.
Specifically, the warning level may be determined based on the warning information. Further, a processing method corresponding to the current warning level may be determined from the correspondence between warning levels and processing methods established in advance, and may be set as the target processing method. And the call service can be processed according to the target processing mode so as to ensure good service quality.
For example, the processing mode corresponding to the primary warning may be: displaying conversation text information triggering warning information on terminal equipment of the manual seat in a highlighting mode in real time so as to warn the manual seat, and receiving a first-level warning by terminal equipment of a management seat corresponding to the current manual seat without displaying in a highlighting mode in real time; the corresponding processing mode of the secondary warning may be: displaying the call text information triggering the warning information on the terminal equipment of the manual seat in a highlighting mode in real time, sending a secondary warning to the manual seat, displaying the current call text information on the terminal equipment of the management seat corresponding to the current manual seat in a highlighting mode in real time, and reminding the management seat to pay attention to the call service process immediately; the processing mode corresponding to the three-level warning may be: displaying the call text information triggering the warning information on the terminal equipment of the manual seat in a highlighting mode in real time, sending a three-level warning to the manual seat, displaying the current call text information on the terminal equipment of the management seat corresponding to the current manual seat in a highlighting mode in real time, taking over the current service process by the management seat, and completing the subsequent user service.
Note that the warning levels may be a primary warning, a secondary warning, or the like, or may be a low-level warning, a middle-level warning, a high-level warning, or the like, the number and naming method of the levels may be set according to actual requirements, and the design of the processing method corresponding to each warning level may be set according to actual requirements, which is not particularly limited in this embodiment.
The technical scheme of the embodiment of the invention comprises the steps of acquiring call voice information in the call process, initializing a voice recognition resource pool, dividing the call voice information into at least one sub-resource pool, distributing the sub-resource pool for the call voice information to improve the response speed of voice recognition, converting the call voice information into call text information based on a voice recognition algorithm and the sub-resource pool, displaying the call text information on the terminal equipment of the current manual position in real time to acquire the call text information for display and/or analysis, inputting the call text information into a target monitoring model in real time, determining the monitoring result information of the current call service, highlighting the call text information corresponding to the warning information to enable the manual position to pay attention and correct, solving the problem that the manual position cannot be timely provided with help and suggestion when a large amount of human resources are consumed to monitor the manual position, the service process of monitoring the manual seat in real time and reminding the manual seat are achieved, the service level of the manual seat is improved, and the complaint rate of users is reduced.
EXAMPLE III
Fig. 3 is a schematic flow chart of a method for monitoring an artificial seat according to a third embodiment of the present invention, and in this embodiment, reference may be made to the technical solution of this embodiment for a determination method and an update method of a target monitoring model based on the foregoing embodiments. The same or corresponding terms as those in the above embodiments are not explained in detail herein.
As shown in fig. 3, the method of this embodiment specifically includes the following steps:
and S310, determining a target monitoring model.
In this embodiment, in order to make the monitoring effect of the target monitoring model more expected, an initial monitoring model may be established in advance, and the initial monitoring model may be trained using the historical call text information and the tagging information.
Specifically, the target listening model may be determined by:
step one, historical call text information and marking information corresponding to the historical call text information are obtained.
The historical call text information may be pre-recorded call text information, and may be symbolic text information, for example: excellent call template, alert call text message, etc. The labeling information may be information for labeling some sentences in the historical call text information, for example: card number, identification card number, etc., and the label information may also be keyword information in the history communication text information, for example: complaints, regulatory agencies, leaders, supervisors, etc.
Specifically, symbolic text information may be selected from the historical call text database as the historical call text information, or all call text information in the historical call text database may be selected as the historical call text information. After the history call text information is acquired, information to be labeled in the history call text information may be labeled, for example: and hiding the privacy information, labeling the keyword information machine and the like. And then, the monitoring result information of the call service can be determined according to the marking information.
It should be noted that, the process of labeling the historical call text information may be that the computer performs labeling based on a preset labeling rule, or may be manually labeled. The computer labeling can improve the speed and save labor, but sentences which are difficult to understand by the computer can exist. The information in the historical call text information can be better understood by using manual labeling, but the labeling speed is low, and a large amount of human resources are needed. Of course, the labeling process may also be a process combining computer labeling and manual labeling, and the specific labeling process may be selected according to actual requirements, which is not specifically limited in this embodiment.
And secondly, training the initial monitoring model based on the historical call text information and the marking information corresponding to the historical call text information to obtain a target monitoring model.
Specifically, the pre-established initial monitoring model can be trained according to the historical call text information and the label information corresponding to the historical call text information, when the monitoring model obtained through training meets the use requirement, the monitoring model obtained through training can be used as the target monitoring model, if the monitoring model obtained through training does not meet the use requirement, the training can be continued until the model meets the requirement, and the model is used as the target monitoring model.
The specific training process may include:
and dividing the historical call text information and the label information corresponding to the historical call text information into a training sample set and a test sample set according to a preset rule.
The preset rule may be a rule that the historical call text information is divided according to a preset proportion. The training sample set is used for training the monitoring model, and the testing sample set is used for evaluating the effect of the monitoring model.
Specifically, the historical call text information and the label information corresponding to the historical call text information are divided into a training sample set and a test sample set according to a preset rule, wherein 80% of the historical call text information and the label information corresponding to the historical call text information are generally used as the training sample set, and 20% of the historical call text information and the label information corresponding to the historical call text information are used as the test sample set. The specific preset rule may be set according to actual conditions, and is not specifically limited in this embodiment.
And training the initial monitoring model based on the training sample set to obtain the monitoring model to be used.
The process of training the initial listening model based on the training sample set may include: inputting the current training text data into a pre-constructed initial monitoring model aiming at each training text data in the training sample set to obtain actual monitoring result information corresponding to the current training text data; and correcting a loss function in the initial monitoring model based on the actual monitoring result information and theoretical monitoring result information corresponding to the current training text data, and training to obtain the monitoring model to be used by taking the convergence of the loss function as a training target.
Specifically, each training text data in the training sample set may be input into the initial monitoring model to obtain an output value corresponding to the training text data, and the output value is used as actual monitoring result information corresponding to the training text data. Based on the actual monitoring result information and the theoretical monitoring result information corresponding to the training text data, a loss value between the actual monitoring result information and the theoretical monitoring result information can be calculated, and model parameters in the current monitoring model are adjusted based on the loss value. The training error of the loss function, i.e., the loss parameter, may be used as a condition for detecting whether the loss function reaches convergence currently, for example, whether the training error is smaller than a preset error or whether the error variation trend tends to be stable, or whether the current iteration number is equal to the preset number. If the detection reaches the convergence condition, for example, the training error of the loss function is smaller than the preset error or the error change tends to be stable, indicating that the monitoring model training is finished, the iterative training may be stopped at this time. If the current condition is not reached, training text data can be further obtained to train the current monitoring model until the training error of the loss function is within the preset range. When the training error of the loss function reaches convergence, the current listening model can be used as the listening model to be used.
And inputting the test sample set into the monitoring model to be used to obtain an output result, and when the accuracy of the output result meets a preset accuracy condition, taking the monitoring model to be used as a target monitoring model.
Specifically, the model effect of the monitoring model to be used can be measured according to model evaluation methods such as model accuracy and the like. The process of determining the listening model may include: inputting the test text information in the test sample set into the monitoring model to be used, and determining the accuracy of the monitoring model to be used based on the monitoring result information of the monitoring model to be used and the monitoring result information corresponding to the test text information in the test sample set; when the accuracy reaches a preset accuracy threshold, taking the monitoring model to be used as a target monitoring model; and if the accuracy rate does not reach the accuracy rate threshold value, obtaining training text information in the training sample set, and continuing training the monitoring model to be used until the accuracy rate of the monitoring model to be used reaches a preset accuracy rate threshold value.
It should be noted that the indexes for evaluating the interception model to be used may further include Precision (Precision), Recall (Recall), and comprehensive evaluation index (F1-Measure).
And S320, acquiring call voice information in the call process.
S330, converting the call voice information into call text information based on a voice recognition algorithm.
S340, inputting the call text information into the target monitoring model in real time, and determining the monitoring result information of the current call service.
And S350, performing incremental training on the target monitoring model according to the call text information and the real monitoring result information corresponding to the call text information to update the target monitoring model.
Since a large amount of call text information is generated every day in the call service center, the text information can be stored in a historical call text database and can be used for updating and maintaining a target monitoring model. Meanwhile, as the business is continuously updated, the content of the conversation service carried out by the manual position is also amplified, so that the model can better accord with the business scene by updating the target monitoring model.
Specifically, the real monitoring result information of each call text message can be determined, and the target monitoring model is subjected to incremental training based on each call text message and the real monitoring result information corresponding to each call text message, so that the model effect of the target monitoring model is improved, and the target monitoring model is more suitable for the current call scene.
On the basis of the above embodiments, the monitoring result information may further include the current service level; in order to facilitate visual display of the service quality of each artificial seat, each call text message of each artificial seat and the service grade corresponding to each call text message can be obtained; and generating a monitoring result table according to each call text message and each service level.
The service level may be excellent, good, general, primary warning, secondary warning, tertiary warning, etc., or may be other types of levels, such as a color-based rating: red grade, yellow grade, green grade, etc. The monitoring result table may be a table for recording service levels of each call service of each human agent, and the monitoring result table may further include a result of performing statistical analysis on the service levels of each human agent, a result of performing overall statistical analysis on the call center, and the like.
Specifically, the service levels of the manual seats and the call services may be recorded in the monitoring result table, and further, the information recorded in the monitoring result table may be counted, for example, the ratio of each service level in each manual seat, the ratio of each service level of all manual seats corresponding to each management seat, the ratio of each service level in the call center, and the like may be counted. According to the monitoring result table, the working condition of each artificial seat in the call center can be known, so that targeted training can be performed on the artificial seat with poor service, and further, the service quality of the call center is improved.
Optionally, the service level may be determined in the following manner: inputting the call text information into a target monitoring model in real time, and determining probability values corresponding to all service levels; and determining the service level according to the probability value.
Specifically, the call text information may be input into the target monitoring model in real time, and a probability value corresponding to each service level of the current call text information may be obtained. Furthermore, the service level corresponding to the maximum probability value can be used as the service level corresponding to the current call text information. The service level may change as the call text message continues to increase until the call is ended, in which case the original service level may be updated based on the newly determined service level.
Exemplarily, the call text information is input into the target monitoring model in real time, and the probability values corresponding to the call text information and the service levels are determined as follows: excellent 25%, good 38%, general 20%, primary alert 8%, secondary alert 5%, tertiary alert 4%, at which time the service level of the call text message can be determined to be good. If the call text information is increased after a period of time, the probability value of the call text information corresponding to each service level becomes: excellent 45%, good 23%, general 20%, primary warning 8%, secondary warning 2%, tertiary warning 2%, at which point the service level at which the call text information can be updated is excellent.
On the basis of the embodiments, the call text information to be recommended is determined based on the service level in the monitoring result table, and the call text information to be recommended is sent to the target terminal device, wherein the target terminal device comprises a terminal device of an artificial seat and/or a terminal device of a management seat.
Specifically, in order to improve the service level of the human agent, a series of excellent call text messages may be searched according to the service level in the monitoring result table, and a series of call text messages with typical errors may also be searched according to the service level in the monitoring result table. And the searched call text information is used as the call text information to be recommended and is sent to the target terminal equipment, so that the manual seat and/or the management seat can learn, and the service level is improved.
Illustratively, if the service level settings are: excellent, good, general, primary warning, secondary warning and tertiary warning, the typical call text information suitable for the front side can be searched from the call text information with excellent service level, the typical call text information suitable for the back side can be searched from the call text information with three-level warning service level, and then the call text information to be recommended is generated according to the typical call text information on the front side and the typical call text information on the back side and is pushed to the target terminal equipment for the human seat and/or the management seat to check.
According to the technical scheme of the embodiment of the invention, the target monitoring model is determined to improve the monitoring accuracy of the monitoring model, the call voice information is obtained in the call process, the call voice information is converted into the call text information based on the voice recognition algorithm to obtain the call text information for display and/or analysis, the call text information is input into the target monitoring model in real time to determine the monitoring result information of the current call service, the problems that a large amount of human resources are consumed to monitor the manual seats and all the manual seats cannot be monitored in real time are solved, the service process of accurately monitoring the manual seats in real time is realized, the cost of the human resources is reduced, and the technical effect of reducing the complaint rate of users is achieved.
Example four
As an alternative implementation of the foregoing embodiments, fig. 4 is a schematic flow chart of an artificial seat monitoring method according to a fourth embodiment of the present invention. The same or corresponding terms as those in the above embodiments are not explained in detail herein.
As shown in fig. 4, the method of this embodiment specifically includes the following steps:
1. when a user dials a hotline, if all the manual seats are in a working state, the user needs to queue for waiting; and if the manual seats in the idle state exist, allocating the user to any manual seat in the idle state so that the manual seat can answer the call.
2. When the human agent answers the call, resources are allocated for the current call by using the voice recognition resource pool.
The voice recognition resource pool is a voice recognition resource service pool initialized when the system is started, and the voice recognition resource pool is initialized in advance to improve the response speed of the voice recognition due to the fact that the voice recognition consumes large network and computing resources.
3. The method comprises the steps of obtaining a real-time translation text (call text information) based on a voice recognition technology, displaying the real-time translation text at the front end of an artificial seat, inputting the real-time translation text into an intelligent monitoring model (target monitoring model), and determining a monitoring result (monitoring result information).
When the manual seat answers the call and the user starts to talk, the talk voice data is collected in real time for voice recognition, and the recognition result, namely the talk text information, is pushed to the front end of the manual seat and the intelligent monitoring model.
The manual seat front end refers to a front end system designed and developed for the manual seat by the call center, and is used for displaying user identity information and service information and performing service inquiry and service handling. The front end of the artificial seat can receive the voice recognition result and display the real-time translation text of the artificial seat and the user in real time.
The intelligent monitoring model is an NLP model trained by using a large number of marked call center conversation texts, the NLP model receives a real-time translation text as an input and outputs an artificial seat service level for judging the service level of the current artificial seat service process.
4. Feeding back the monitoring result to the front end of the manual seat, and if the monitoring result contains warning information, highlighting in real time; and if the monitoring result does not contain the warning information, feeding back the monitoring result without highlighting.
If abnormal conditions occur, namely the monitoring result contains warning information, the warning information is highlighted and displayed at the front end of the manual seat in time so as to remind the manual seat to pay attention to the speech termination by taking the user as the center, and the service quality is improved.
5. Judging whether the monitoring result contains warning information, if so, sending the warning information to the front end of the seat group leader, and executing 6; if not, the manual seat continues to finish the subsequent communication.
The seat group leader is a manual seat which is responsible for arranging the working content of the manual seat and monitoring the working quality of the group, and is also responsible for managing the service quality and the service efficiency of the group of users besides answering the incoming calls of the users. The seat group leader can check the monitoring results of each manual seat in the group in real time.
6. And judging whether the warning information is primary warning or not, if so, receiving the warning information only by the seat group leader, and if not, highlighting the front end of the seat group leader in real time and executing step 7.
It should be noted that the primary warning is the weakest, the tertiary warning is the strongest, and the secondary warning is between the primary warning and the secondary warning. When a first-level warning occurs, the seat group leader only receives warning information. When a secondary warning or a tertiary warning occurs, the seat group leader receives the warning information, and the warning information is highlighted at the front end of the seat group leader so that the seat group leader can find and track the service process of the non-compliant users in time and the service quality and the service efficiency of the users of the group are ensured.
7. If the warning information is a three-level warning, the seat group leader takes over the telephone.
If three-level warning occurs, the seat group leader can take over the service process immediately, so that the service can not cause adverse effect on the user, the service quality of the user of the group is improved, and the complaint rate of the user is reduced.
According to the technical scheme, when the manual seat answers a call, resources are allocated for the current call through the voice recognition resource pool, the real-time translation text is obtained based on the voice recognition technology, the real-time translation text is displayed at the front end of the manual seat and is input into the intelligent monitoring model, the current call is processed according to the monitoring result output by the intelligent monitoring model, the problems that a large amount of human resources are consumed to monitor the manual seat and all the manual seats cannot be monitored in real time are solved, the service process of monitoring the manual seat in real time is achieved, the cost of the human resources is reduced, and the technical effect of reducing the complaint rate of users is achieved.
EXAMPLE five
Fig. 5 is a schematic structural diagram of an artificial seat monitoring device according to a fifth embodiment of the present invention, where the device includes: a call voice information acquisition module 510, a call text information conversion module 520 and a monitoring result information determination module 530.
The call voice information obtaining module 510 is configured to obtain call voice information in a call process; a call text information conversion module 520, configured to convert the call voice information into call text information based on a voice recognition algorithm; a monitoring result information determining module 530, configured to input the call text information into a target monitoring model in real time, and determine monitoring result information of the current call service; and the target monitoring model is obtained by training the initial monitoring model according to the historical call text information and the marking information corresponding to the historical call text information.
Optionally, the monitoring result information includes warning information.
Optionally, the apparatus further comprises: and the first processing module is used for responding to the warning information and sending warning prompt information to the terminal equipment of the current manual position.
Optionally, the apparatus further comprises: and the second processing module is used for responding to the warning information and sending warning prompt information to the terminal equipment of the current manual position and the terminal equipment of the management position corresponding to the current manual position.
Optionally, the apparatus further comprises: and the third processing module is used for responding to the warning information, cutting off the conversation of the current manual seat and switching the current conversation service to the management seat corresponding to the current manual seat.
Optionally, the warning information includes at least two warning levels; the device further comprises: and the service processing module is used for determining a target processing mode corresponding to the warning grade according to the warning grade and processing the current call service according to the target processing mode.
Optionally, the apparatus further comprises: the sub-resource pool dividing module is used for initializing a voice recognition resource pool and dividing the voice recognition resource pool into at least one sub-resource pool; correspondingly, the call text information conversion module 520 is further configured to allocate a sub-resource pool for the call voice information, and convert the call voice information into call text information based on a voice recognition algorithm and the sub-resource pool.
Optionally, the apparatus further comprises: and the text display module is used for displaying the call text information on the terminal equipment of the current manual position in real time.
Optionally, the apparatus further comprises: and the highlight text display module is used for highlighting the call text information corresponding to the warning information if the monitoring result of the current call service is the warning information.
Optionally, the apparatus further comprises: the target monitoring model determining module is used for determining a target monitoring model; the target monitoring model determining module is specifically used for acquiring historical call text information and marking information corresponding to the historical call text information; and training an initial monitoring model based on the historical call text information and the marking information corresponding to the historical call text information to obtain a target monitoring model.
Optionally, the target monitoring model determining module is further configured to divide the historical call text information and the label information corresponding to the historical call text information into a training sample set and a test sample set according to a preset rule; training the initial monitoring model based on the training sample set to obtain a monitoring model to be used; and inputting the test sample set into the monitoring model to be used to obtain an output result, and when the accuracy of the output result meets a preset accuracy condition, taking the monitoring model to be used as a target monitoring model.
Optionally, the apparatus further comprises: and the target monitoring model updating module is used for carrying out incremental training on the target monitoring model according to the call text information and the real monitoring result information corresponding to the call text information so as to update the target monitoring model.
Optionally, the monitoring result information includes a current service level; the device further comprises: the monitoring result table generating module is used for acquiring each call text message of each artificial seat and the service level corresponding to each call text message; and generating a monitoring result table according to the call text information and the service grades.
Optionally, the monitoring result information determining module 530 is further configured to input the call text information into a target monitoring model in real time, and determine probability values corresponding to the service levels; and determining the service level according to the probability value.
Optionally, the apparatus further comprises: and the text recommendation module is used for determining the call text information to be recommended based on the service level in the monitoring result table and sending the call text information to be recommended to target terminal equipment, wherein the target terminal equipment comprises terminal equipment of an artificial seat and/or terminal equipment of a management seat.
According to the technical scheme of the embodiment of the invention, the call voice information is acquired in the call process, the call voice information is converted into the call text information based on the voice recognition algorithm so as to acquire the call text information for display and/or analysis, the call text information is input into the target monitoring model in real time, and the monitoring result information of the current call service is determined, so that the problems that a large amount of human resources are consumed to monitor the manual seats and all the manual seats cannot be monitored in real time are solved, the service process of monitoring the manual seats in real time is realized, the cost of the human resources is reduced, and the technical effect of reducing the complaint rate of users is further achieved.
The artificial seat monitoring device provided by the embodiment of the invention can execute the artificial seat monitoring method provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.
It should be noted that, each unit and module included in the above-mentioned artificial seat monitoring apparatus are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding function can be realized; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the embodiment of the invention.
EXAMPLE six
Fig. 6 is a schematic structural diagram of an electronic device according to a sixth embodiment of the present invention. FIG. 6 illustrates a block diagram of an exemplary electronic device 60 suitable for use in implementing embodiments of the present invention. The electronic device 60 shown in fig. 6 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiment of the present invention.
As shown in fig. 6, the electronic device 60 is in the form of a general purpose computing device. The components of the electronic device 60 may include, but are not limited to: one or more processors or processing units 601, a system memory 602, and a bus 603 that couples various system components including the system memory 602 and the processing unit 601.
The system memory 602 may include computer system readable media in the form of volatile memory, such as Random Access Memory (RAM)604 and/or cache memory 605. The electronic device 60 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage system 606 may be used to read from and write to non-removable, nonvolatile magnetic media (not shown in FIG. 6, commonly referred to as a "hard drive"). Although not shown in FIG. 6, a magnetic disk drive for reading from and writing to a removable, nonvolatile magnetic disk (e.g., a "floppy disk") and an optical disk drive for reading from or writing to a removable, nonvolatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 603 by one or more data media interfaces. Memory 602 may include at least one program product having a set (e.g., at least one) of program modules that are configured to carry out the functions of embodiments of the invention.
A program/utility 608 having a set (at least one) of program modules 607 may be stored, for example, in memory 602, such program modules 607 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which examples or some combination thereof may comprise an implementation of a network environment. The program modules 607 generally perform the functions and/or methods of the described embodiments of the invention.
The processing unit 601 executes various functional applications and data processing by running programs stored in the system memory 602, for example, implementing the manual seat listening method provided by the embodiment of the present invention.
EXAMPLE seven
An embodiment of the present invention further provides a storage medium containing computer-executable instructions, where the computer-executable instructions are executed by a computer processor to perform a method for artificial seat listening, and the method includes:
acquiring call voice information in a call process;
converting the call voice information into call text information based on a voice recognition algorithm;
inputting the call text information into a target monitoring model in real time, and determining monitoring result information of the current call service; and the target monitoring model is obtained by training the initial monitoring model according to the historical call text information and the marking information corresponding to the historical call text information.
Computer storage media for embodiments of the invention may employ any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Computer program code for carrying out operations for embodiments of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (18)
1. A method for monitoring artificial seats is characterized by comprising the following steps:
acquiring call voice information in a call process;
converting the call voice information into call text information based on a voice recognition algorithm;
inputting the call text information into a target monitoring model in real time, and determining monitoring result information of the current call service; and the target monitoring model is obtained by training the initial monitoring model according to the historical call text information and the marking information corresponding to the historical call text information.
2. The method of claim 1, wherein the listening result information comprises warning information.
3. The method of claim 2, further comprising:
and responding to the warning information, and sending warning prompt information to the terminal equipment of the current manual position.
4. The method of claim 2, further comprising:
and responding to the warning information, and sending warning prompt information to the terminal equipment of the current manual position and the terminal equipment of the management position corresponding to the current manual position.
5. The method of claim 2, further comprising:
and responding to the warning information, cutting off the call of the current manual seat, and switching the current call service to the management seat corresponding to the current manual seat.
6. The method of claim 2, wherein the warning information includes at least two warning levels;
the method further comprises the following steps:
and determining a target processing mode corresponding to the warning grade according to the warning grade, and processing the current call service according to the target processing mode.
7. The method of claim 1, further comprising, prior to the converting the call speech information into call text information based on a speech recognition algorithm:
initializing a voice recognition resource pool, and dividing the voice recognition resource pool into at least one sub-resource pool;
correspondingly, the converting the call voice information into call text information based on the voice recognition algorithm includes:
and allocating a sub-resource pool for the call voice information, and converting the call voice information into call text information based on a voice recognition algorithm and the sub-resource pool.
8. The method of claim 1, after the converting the call voice information into call text information based on the voice recognition algorithm, further comprising:
and displaying the call text information on the terminal equipment of the current manual position in real time.
9. The method of claim 8, further comprising:
and if the monitoring result of the current call service is warning information, highlighting the call text information corresponding to the warning information.
10. The method of claim 1, further comprising:
determining a target monitoring model;
the determining of the target monitoring model comprises:
acquiring historical call text information and marking information corresponding to the historical call text information;
and training an initial monitoring model based on the historical call text information and the marking information corresponding to the historical call text information to obtain a target monitoring model.
11. The method of claim 10, wherein training an initial listening model based on the historical call text information and the label information corresponding to the historical call text information comprises:
dividing the historical call text information and the label information corresponding to the historical call text information into a training sample set and a test sample set according to a preset rule;
training the initial monitoring model based on the training sample set to obtain a monitoring model to be used;
and inputting the test sample set into the monitoring model to be used to obtain an output result, and when the accuracy of the output result meets a preset accuracy condition, taking the monitoring model to be used as a target monitoring model.
12. The method of claim 1, further comprising:
and performing incremental training on the target monitoring model according to the call text information and the real monitoring result information corresponding to the call text information so as to update the target monitoring model.
13. The method of claim 1, wherein the interception result information includes a current service level; the method further comprises the following steps:
acquiring each call text message of each artificial seat and a service level corresponding to each call text message;
and generating a monitoring result table according to the call text information and the service grades.
14. The method of claim 13, wherein the inputting the call text information into the target monitoring model in real time and determining the monitoring result information of the current call service comprises:
inputting the call text information into a target monitoring model in real time, and determining probability values corresponding to all service levels;
and determining the service level according to the probability value.
15. The method of claim 13, further comprising:
and determining the call text information to be recommended based on the service level in the monitoring result table, and sending the call text information to be recommended to target terminal equipment, wherein the target terminal equipment comprises terminal equipment of an artificial seat and/or terminal equipment of a management seat.
16. An artificial seat monitoring device, comprising:
the call voice information acquisition module is used for acquiring call voice information in the call process;
the call text information conversion module is used for converting the call voice information into call text information based on a voice recognition algorithm;
a monitoring result information determining module, configured to input the call text information into a target monitoring model in real time, and determine monitoring result information of the current call service; and the target monitoring model is obtained by training the initial monitoring model according to the historical call text information and the marking information corresponding to the historical call text information.
17. An electronic device, characterized in that the electronic device comprises:
one or more processors;
a storage device for storing one or more programs,
when executed by the one or more processors, cause the one or more processors to implement the method of human agent listening as recited in any of claims 1-15.
18. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the method of artificial seat listening according to any one of claims 1-15.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110309936.2A CN113011159A (en) | 2021-03-23 | 2021-03-23 | Artificial seat monitoring method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110309936.2A CN113011159A (en) | 2021-03-23 | 2021-03-23 | Artificial seat monitoring method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113011159A true CN113011159A (en) | 2021-06-22 |
Family
ID=76405643
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110309936.2A Pending CN113011159A (en) | 2021-03-23 | 2021-03-23 | Artificial seat monitoring method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113011159A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113946673A (en) * | 2021-12-20 | 2022-01-18 | 深圳市一号互联科技有限公司 | Semantic-based intelligent customer service routing processing method and device |
| CN114422647A (en) * | 2021-12-24 | 2022-04-29 | 上海浦东发展银行股份有限公司 | Digital person-based agent service method, apparatus, device, medium, and product |
| CN115695656A (en) * | 2022-12-29 | 2023-02-03 | 北京青牛技术股份有限公司 | Method and system for evaluating and managing call content quality |
-
2021
- 2021-03-23 CN CN202110309936.2A patent/CN113011159A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113946673A (en) * | 2021-12-20 | 2022-01-18 | 深圳市一号互联科技有限公司 | Semantic-based intelligent customer service routing processing method and device |
| CN113946673B (en) * | 2021-12-20 | 2022-04-08 | 深圳市一号互联科技有限公司 | Semantic-based intelligent customer service routing processing method and device |
| CN114422647A (en) * | 2021-12-24 | 2022-04-29 | 上海浦东发展银行股份有限公司 | Digital person-based agent service method, apparatus, device, medium, and product |
| CN115695656A (en) * | 2022-12-29 | 2023-02-03 | 北京青牛技术股份有限公司 | Method and system for evaluating and managing call content quality |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108962282B (en) | Voice detection analysis method and device, computer equipment and storage medium | |
| CN113011159A (en) | Artificial seat monitoring method and device, electronic equipment and storage medium | |
| CN107273531B (en) | Telephone number classification identification method, device, equipment and storage medium | |
| CN105261362A (en) | Conversation voice monitoring method and system | |
| CN111212190A (en) | Conversation management method, device and system based on conversation strategy management | |
| EP3633536A1 (en) | Partial automation of text chat conversations | |
| CN113240436A (en) | Method and device for online customer service call technical quality inspection | |
| US11095601B1 (en) | Connection tier structure defining for control of multi-tier propagation of social network content | |
| CN110955770A (en) | Intelligent dialogue system | |
| US11870937B2 (en) | Methods for smart gas call center feedback management and Internet of things (IoT) systems thereof | |
| CN108965613A (en) | Method, storage medium and application server based on the monitoring of collection system quality | |
| US20220392434A1 (en) | Reducing biases of generative language models | |
| CN116828109A (en) | Intelligent evaluation method and system for telephone customer service quality | |
| US10924611B2 (en) | Voice recognition system and call evaluation setting method | |
| US11783840B2 (en) | Video conference verbal junction identification via NLP | |
| US20110197206A1 (en) | System, Method And Program Product For Analyses Based On Agent-Customer Interactions And Concurrent System Activity By Agents | |
| CN111970311B (en) | Session segmentation method, electronic device and computer readable medium | |
| CN115221892A (en) | Work order data processing method and device, storage medium and electronic equipment | |
| CN111949776B (en) | User tag evaluation method and device and electronic equipment | |
| US11468367B2 (en) | Method and apparatus for cognitive system training with incorporation of probability and uncertainty to achieve deterministic results | |
| CN114155845A (en) | Service determination method and device, electronic equipment and storage medium | |
| CN112632241A (en) | Method, device, equipment and computer readable medium for intelligent conversation | |
| US9116980B1 (en) | System, method, and computer program for determining a set of categories based on textual input | |
| CN110334341B (en) | Corpus recommendation method, apparatus, medium and electronic device based on data analysis | |
| CN111080482B (en) | Quantitative evaluation method for power consumption client emotion indexes |
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 |