CN112235469A - Method and system for quality inspection of recording of artificial intelligence call center - Google Patents
Method and system for quality inspection of recording of artificial intelligence call center Download PDFInfo
- Publication number
- CN112235469A CN112235469A CN202011116054.6A CN202011116054A CN112235469A CN 112235469 A CN112235469 A CN 112235469A CN 202011116054 A CN202011116054 A CN 202011116054A CN 112235469 A CN112235469 A CN 112235469A
- Authority
- CN
- China
- Prior art keywords
- task
- recording
- examination
- quality inspection
- initial
- 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
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M3/00—Automatic or semi-automatic exchanges
- H04M3/22—Arrangements for supervision, monitoring or testing
- H04M3/2227—Quality of service monitoring
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; 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
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Quality & Reliability (AREA)
- Stored Programmes (AREA)
Abstract
The invention discloses a method and a system for quality inspection of recording of an artificial intelligence call center, wherein the method comprises the following steps: adding an initial check task, setting a task name, a task effective time and a task invalid time, inputting a screening condition for extracting a record from the task effective time to the task invalid time to obtain a monitored seat meeting the screening condition, and generating an initial check task list; entering an initial inspection task list, selecting a monitored seat for quality inspection, and setting an initial inspection task allocation rule; and starting the primary inspection task at the effective time of the task, extracting the recording data of the monitored seat according to the distribution rule, distributing the recording data to related quality inspectors to start quality inspection and marking, and finishing the primary inspection task at the ineffective time of the task. The invention can solve the problem that the quality inspection of a large amount of call records of the existing call center is difficult, increase the number of concurrent users and improve the execution capacity of the system.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a method and a system for quality inspection of a recording of an artificial intelligence call center.
Background
The call center is used as an intangible service window for keeping close contact between enterprises and user terminals, plays an increasingly important role in product publicity, product sales, technical support, after-sales service, consultation, complaint and the like, and plays an increasingly important role in production and operation activities of the enterprises. The call center is an operation place which fully utilizes modern communication and computer technology and can automatically and flexibly process a large number of different telephone incoming and outgoing services and services. With the rapid development of computer technology and communication technology, such as the introduction of CTI (computer telephony integration), IVR (interactive voice response), ACD (automatic call distribution system) and other technologies, the development of call centers is rapid, the scale of the work is increasing, the business operation is more and more complex, and more user data are deposited.
However, in the face of the large increase of business requirements, the problems of low call, screening and service equivalence rate, increasing human cost, difficult management of huge manual seats, difficult quality inspection of huge call records, insufficient user data mining and the like are accompanied. These problems are difficult to solve by the existing call center technology.
Disclosure of Invention
The invention provides a method and a system for quality inspection of recordings in an artificial intelligent call center, aiming at solving the problem that the quality inspection of huge call recordings in the existing call center is difficult.
The invention adopts the following technical scheme:
a recording quality inspection method for an artificial intelligence call center comprises the following steps:
adding an initial check task, setting a task name, a task effective time and a task invalid time, inputting a screening condition for extracting a record from the task effective time to the task invalid time to obtain a monitored seat meeting the screening condition, and generating an initial check task list;
entering the initial inspection task list, selecting a monitored seat for quality inspection, and setting an initial inspection task allocation rule;
and starting the preliminary examination task at the task effective time, extracting the recording data of the monitored seat according to the distribution rule, distributing the recording data to related quality inspectors to start quality inspection and marking, and finishing the preliminary examination task at the task invalid time.
As a preferred embodiment, the screening conditions include: quality inspector, organization to which quality inspector belongs, recording start time, recording end time and monitored number.
As a preferred embodiment, the method further comprises the following steps of task state checking:
checking whether the task needs to be changed into a starting state or an ending state according to task effective time and task ineffective time before entering the initial inspection task list: the task effective time is the current system time, and the task is in an un-started state; after the task effective time is the current system time, the task is in a creation state; the task is in an end state before the effective time of the task is the current system time;
checking the task state in the first step at regular time;
and thirdly, the quality inspection quantity is equal to the monitored quantity, and the task is automatically ended.
The method further comprises the steps of executing an editing task on the initial inspection task in the non-started state or the suspended state, modifying the initial inspection task list, and executing a subsequent initial inspection task by using the edited new initial inspection task list.
In a preferred embodiment, when the task expiration time is reached but the quality inspection task is not completed, the recorded data of the incomplete quality inspection is released.
The method further comprises a recording query, wherein the recording data of the monitored seat are checked according to the identity authority of the inquirer, and the recording is scored.
In a preferred embodiment, the step of extracting the sound recording data further includes extracting according to task priority, and prioritizing the sound recording data according to service types in advance before extracting the sound recording data.
As a preferred embodiment, the method further comprises executing a renegotiation operation on the quality inspection sound recording according to a set renegotiation process according to a request of the monitored agent for applying for an initial inspection renegotiation, wherein the renegotiation process comprises:
adding flow management, inputting a flow name, an auditor, an organization to which the auditor belongs, a flow type and a renegotiation validity period, and generating a flow management list;
and entering the process management list, and distributing the sound records applying for the initial examination and the review within the review validity period to related auditors according to the organization types to enter an approval flow.
As a preferred embodiment, the method further comprises recording the state transition situation of the sound recording data according to the following rules:
the recording which is extracted by the primary examination task is not locked and not scored, and the recording is in a system scoring state;
the record locked by the quality inspector is in a state to be primarily inspected;
recording marked by the initial examination is in an initial examination state;
recording the application of the agent for the preliminary examination and the reexamination is in an applied state of the preliminary examination and the reexamination;
after the sound recording of the primary examination and reexamination is approved, the sound recording does not enter the examination and approval flow which is the final examination and reexamination examination and approval;
after the sound records of the primary examination and reexamination pass through all the examination and approval, the sound records of the primary examination and reexamination pass through;
the recording of the primary examination repeat is rejected as the rejected of the primary examination repeat.
The invention relates to a recording quality inspection system of an artificial intelligence call center, which is stored in a storage medium in the form of machine readable program instructions, and executes the recording quality inspection method of the artificial intelligence call center when the machine readable program instructions are operated.
The method and the system for the artificial intelligence call center recording quality inspection utilize the timing task in the decoupling service system, various simple or complex scheduling works are completed by task scheduling, the concurrency problem in the conventional project is avoided, deep customized development is carried out on the basis of an open source task scheduling framework Quartz, the method and the system have a visual management interface, a task list, a task state, an execution record and an execution result can be checked, and functions of pause, recovery, immediate execution and the like are supported.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a functional module schematic diagram of a recording quality inspection system of an artificial intelligence call center according to an embodiment of the present invention.
Fig. 2 is a schematic operation flow diagram of a recording quality inspection method for an artificial intelligence call center according to an embodiment of the present invention.
Detailed Description
It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict.
In the description of the present invention, it should be noted that, as the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. appear, their indicated orientations or positional relationships are based on those shown in the drawings, and are only for convenience of description and simplicity of description, but do not indicate or imply that the referred device or element must have a specific orientation, be constructed in a specific orientation, and be operated, and thus, should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and "third" as appearing herein are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
In the description of the present invention, it should be noted that, unless otherwise explicitly stated or limited, the terms "mounted," "connected," and "connected" should be interpreted broadly, e.g., as being fixed or detachable or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.
The technical solution of the present invention will be described in detail with reference to the accompanying drawings in conjunction with embodiments.
The invention provides a recording quality inspection system of an artificial intelligence call center, which adopts JAVA as a development language and IDEA as a development tool to carry out development work of the whole project. The IDEA of IDEA is called IntelliJ IDEA, which is an integrated environment for java language development (also can be used for other languages), and IntelliJ is recognized as one of the best java development tools in the industry, and particularly, the functions of intelligent code assistant, code automatic prompting, reconfiguration, J2EE support, Ant, JUnit, CVS integration, code review, innovative GUI design and the like can be said to be supernormal. IDEA is a product of jetbrans corporation, headquartered in the czech republic of capital bragg, and developers are mainly well-known eastern european programmers. System programming language: java + Jquery; a database: mysql; the operating system is a win 1064-bit operating system; development tool idea + maven3.3.9.
The recording quality inspection system of the artificial intelligent call center can be stored in a storage medium in a machine readable program instruction form, and when the machine readable program instruction runs, the recording quality inspection method of the artificial intelligent call center is executed. The storage medium may be, for example, ROM, RAM, a removable hard disk, a magnetic disk, an optical disk, a usb disk, etc. The machine readable program instruction is obtained by programming in a JAVA language under an IDEA development environment, meanwhile, a timing task in a decoupling service system is utilized, various simple or complex scheduling works are completed by task scheduling, the concurrency problem in a conventional project is avoided, deep customized development is carried out on the basis of an open source task scheduling framework Quartz, a visual management interface is provided, a task list, a task state, an execution record and an execution result can be viewed, and functions of pausing, recovering, immediate execution and the like are supported.
Referring to fig. 1, as an embodiment of the recording quality inspection system of the artificial intelligence call center, the functional modules thereof may include: the system comprises a task creating module 11, a task distributing module 12, a quality inspection executing module 13, a recording inquiring module 14 and a repeating module 15, and specifically:
and the task creating module 11 is used for adding an initial inspection task, setting a task name, a task effective time and a task invalid time, inputting a screening condition for extracting the recording from the task effective time to the task invalid time, obtaining a monitored seat meeting the screening condition, and generating an initial inspection task list.
And the task allocation module 12 is used for entering an initial inspection task list, selecting a monitored seat for quality inspection, and setting an initial inspection task allocation rule.
And the quality inspection execution module 13 is used for starting the initial inspection task at the task effective time, extracting the recording data of the monitored seat according to the distribution rule, distributing the recording data to the related quality inspector to start quality inspection and marking, and finishing the initial inspection task at the task ineffective time.
And the recording query module 14 is used for checking the recording data of the monitored seat according to the identity authority of the inquirer and scoring the recording.
And the rechecking module 15 is used for executing the rechecking operation on the quality inspection record according to the set rechecking flow according to the request of the monitored agent for applying the primary inspection rechecking.
The specific application of the above functional modules in the system of the present invention is described as follows:
1) Description of the function
The recording query can check the recording condition of the agent and can score the recording.
2) Business rules
(1) Recording and checking conditions of the role:
1. the administrator can view all the recordings;
2. the quality inspection manager can check all the recordings in the group;
3. the quality inspector can check all the recordings in the large group;
4. the seat can check the recording of the seat.
(2) Circulation condition of recording state:
1. recording scored by the machine is in a system scoring state, and recording sampled by the primary detection task is in the system scoring state when not locked and not scored;
2. the machine locked by the quality inspector marks and records the state to be initially inspected;
3. the recording marked by the initial examination is in an initial examination state, and the recording which is selected by the retest task and is not locked and marked is in the initial examination state;
4. recording the application of the agent for the preliminary examination and the reexamination is in an applied state of the preliminary examination and the reexamination;
5. after the sound recording of the primary examination and reexamination is approved, the sound recording does not enter the examination and approval flow which is the final examination and reexamination examination and approval;
6. after the sound records of the primary examination and reexamination pass through all the examination and approval, the sound records of the primary examination and reexamination pass through;
7. the recording of the primary examination repeat is rejected as the rejected of the primary examination repeat.
(3) When the system of the invention is executed on a computer, a recording inquiry page is generated to execute the operation of recording inquiry, and the operation steps are as follows:
1. clicking a recording query sub-menu and displaying a recording query page;
2. inputting query conditions, clicking for query, and displaying query recording results;
3. clicking a reset button to empty the query condition content;
4. clicking a retraction (expansion) button, and retracting (expanding) the query function;
5. clicking a tune and score button, and entering a score page to score.
1) Description of the function:
setting an initial inspection task allocation rule, automatically extracting seat recording data according to the rule by the system, allocating the seat recording data to relevant quality inspectors for quality inspection, and having the functions of adding, editing, pausing, closing and deleting;
the administrator, the call center manager, the quality cultivation supervisor and the quality inspector have the authority of the initial inspection task.
2) And (4) business rules:
firstly, recording can be extracted after starting, and recording can not be extracted after pausing and ending;
when adding the task, the name of the task, the effective time and the expiration time of the task, the organization of the quality inspector, the starting and ending time of the recording, the number of the monitored persons and the quality inspector need to be filled;
selecting 2016/8/1 task effective time and 2016/8/31 task invalid time, namely extracting the recording with the recording time within 2016/8/1-2016/8/31;
fourthly, the extraction task can be performed within the time of the quality inspection task taking effect and failure;
selecting the organization to which the quality inspector belongs, and displaying the quality inspector under the organization to do a quality inspection task;
one sound recording screening condition corresponds to one service type;
and seventhly, customizing a return visit task: and (3) revisiting the tasks, and increasing the screening conditions for extracting the record: return visit item, return visit list number, return visit status;
clicking the number of the sound records in the initial inspection task list to obtain the number of the sound records which can be quality inspected and correspond to the task;
ninthly, the screening conditions of the tasks cannot be overlapped so as to avoid inaccurate acquisition quantity, such as task conditions: 10 strips are extracted in 0-50 minutes, and the overlapping condition that 10 strips are extracted in 30-60 minutes can cause the data quantity to be repeatedly calculated;
and the sound record corresponding to the task can be selected from the initial inspection task list for release at the frequency of red (R), and the released sound record can be used for reestablishing the task quality inspection.
1. before entering an initial inspection task list, checking whether the task needs to be changed into started or finished according to the task effective time and the task invalid time;
2. the timed task checks the task in1 every 10 minutes;
3. the quality inspection quantity is equal to the total number of the monitored objects, and the task is automatically ended.
3) The method comprises the following operation steps:
(1) newly-built preliminary examination task
Clicking an adding button, and displaying an adding initial inspection task page;
inputting an adding condition, clicking to submit, and jumping to a page of the initial inspection task list after the submission is completed.
The click movement or the whole movement of the selected organization can select the monitored person. Inputting the number of the agent in the input box of the monitored person, clicking an adding button, adding the agent to the monitoring list, clicking a searching button, inquiring whether the agent exists in the monitoring list, selecting the number of the agent in the monitoring list, clicking a deleting button, and then rejecting the agent from the monitoring list. And when the monitored person does not input the agent number in the input box, clicking an adding button to prompt the monitored person to fill in.
The failure time of the quality inspection task is the current system time, the task is in an un-started state, and the task is automatically started after one minute. And after the failure time of the quality inspection task is the current system time, the task is the creation state. And before the failure time of the quality inspection task is the current system time, the task is in an end state.
And clicking a task name link on the initial inspection task list page to display a detailed list.
Clicking the total number of the monitored persons on the initial inspection task list page, and jumping out of a pop-up box to display the detail of the monitored condition.
Clicking the total number of quality testers in the initial inspection task list, and jumping out of a pop-up box to display the detail of the quality inspection monitoring condition.
(2) Editing the task of first inspection
And after the initial inspection task is successfully established, clicking an editing button on an initial inspection task list page, and entering a modification task page. Only the not-started state can be edited.
After editing the task, clicking a submit button, and re-distributing the task after pressing the modification.
(3) Starting the task of initial inspection
When the initial inspection task is successfully established, the initial inspection task can be started by clicking a starting button on a page;
and when the initial inspection task is suspended, clicking a starting button corresponding to the initial inspection task to start the initial inspection task.
(4) Suspending the task of initial examination
And when the initial inspection task is started successfully, clicking a pause button on the page to pause the initial inspection task.
(5) Close the task of initial inspection
And after the initial inspection task is started successfully, clicking a closing button on the page to close the initial inspection task.
(6) Deleting the task of initial inspection
When the initial inspection task is successfully established, the initial inspection task can be deleted before being started;
when the initial inspection task is finished, the initial inspection task can be deleted;
and when a delete button on the page is clicked, a delete popup window appears, and the initial inspection task can be deleted after the click determination is carried out.
(7) Detailed query of preliminary examination task
When the initial inspection task exists, the query button on the page can be clicked, and the detailed description of the initial inspection task can be checked;
and clicking the record number in the initial examination task specification to check the quality examination details of the record.
And when the quality inspection task is finished and the quality inspection task is not completely finished, clicking the release button can release the record which is not subjected to quality inspection. One or more corresponding records of the tasks can be selected or all the records can be selected to be released, and the released records can be subjected to task quality inspection again.
(8) Task of inquiry and initial inspection
Manually inputting the initial inspection task name, the task effective time and the task failure time, selecting the task state, selecting the name and the job number of a quality inspector, selecting the recording start time and the recording failure time, and clicking a query button to query the initial inspection task meeting the conditions;
clicking a reset button to empty the query condition content;
clicking the retract (expand) button, the query function is retracted (expanded).
(III) quality inspection execution module 13
1) Description of the function
The quality inspector carries out initial inspection grading on the allocated sound recording;
the administrator, the call center manager, the quality cultivation supervisor and the quality inspector have initial inspection permission.
2) Business rules
Only one piece of recording data can be extracted for quality inspection at one time, and the next piece can be extracted after the quality inspection is finished;
the exclusive task can be used for quality inspection of all the records, and the exclusive task is only allocated to one quality inspector to be subjected to quality inspection;
the extraction tasks need to be extracted according to task priority, and according to the sequence of A, B, C and the like, the B level can be extracted after the A level is extracted; prioritized extraction according to organization type, such as extracting complaint types preferentially;
if the intelligent scoring shows the intelligent scoring result, the quality inspector can modify the scoring on the basis of the intelligent scoring, and if the scoring is deducted, root cause analysis and user problem classification corresponding to the quality inspection item must be filled;
clicking the primary check remark, popping up an input box, inputting a mark deduction reason and inputting 1500 characters;
the quality inspector finishes all quality inspections, and the task is finished;
the state of the record which is locked by the quality inspector but is not subjected to quality inspection is changed into the state to be subjected to initial inspection;
after the initial examination grading is finished, changing the recording state into the original examination;
the quality inspection item hits the fatal item, and the quality inspection is totally 0.
3) Description of the operation
(1) Entering an initial examination task list, clicking a query button, and extracting recording data;
(2) when the initial inspection task is not inspected, clicking the query button to prompt that an item which is not inspected exists and please inspect;
(3) clicking a primary examination button corresponding to the recording number in the list, and displaying a scoring page;
(4) grading and inputting an initial examination remark, if a deduction item exists, filling in root cause analysis and user problem classification, clicking to submit, and completing the initial examination;
(5) when no initial inspection task exists at present, the extraction of the initial inspection task can prompt that no initial inspection task exists.
(IV) repeating module 15
1) Description of the function
And executing the repeating operation according to the set repeating flow, and having an adding and editing function. The double-conference flow comprises two operations of auditing and double-conference, a plurality of roles can be set for auditing, and double-conference is the final operation.
The administrator, the call center manager and the quality inspection supervisor have the process management authority.
2) Business rules
Each large group (organization to which the quality inspector belongs) can be configured with at most one repeating process;
the name, organization and type of the process are bound filling, and the period of validity of the renegotiation is bound filling;
the application review validity period can be left unfilled and is defaulted to three days;
the application for the reply validity period is input as days, the agent can apply for the reply in the validity period, and the agent cannot apply for the reply when the validity period exceeds the validity period;
the double-conference flow comprises two operations of auditing and double-conference, and the selected job talent has the authority to do the operation;
there must be one review in the flow, and the review must be the last operation;
after the flow is edited, the method is only effective for the subsequent flow, the sound recording and the order which are circulated and completed before are not influenced, and if a business uses the flow, the flow cannot be edited;
when the agent is in a double-meeting state, only the approval flow of the organization in which the agent is positioned is displayed.
3) Description of the operation
(1) Add flow management
1. And clicking an adding button to display an adding process management page.
Each large group can be configured with at most one repeating process;
the name, organization and type of the process are bound filling, and the period of validity of the renegotiation is bound filling;
the application review validity period can be left unfilled and is defaulted to three days;
the application for the reply validity period is input as days, the agent can apply for the reply in the validity period, and the agent cannot apply for the reply when the validity period exceeds the validity period;
the double-conference flow has two operations of auditing and double-conference, and the selected job talent has the right to do the operation
2. After the page mandatory item is input, an adding button is clicked, and an operation flow can be added;
inputting the name of the process, applying for the validity period of the review, selecting the type of the process, the product line and the organization.
Selecting and auditing the operation name, selecting a quality inspection manager by the role, and auditing the agent by the quality inspection manager after applying for a reexamination;
the job name is selected for reemergence, the role is selected for a quality inspection manager, the quality inspection manager performs auditing after applying for reemergence on behalf of the agent, and the quality inspection manager can score again;
clicking a deletion button corresponding to the operation flow in the list to delete the corresponding operation flow;
after the operation flow is filled, a new flow record can be added by clicking the save button.
(2) Editing process management
Clicking a corresponding editing button of the process management to enter a process management editing page;
changing the organization into a quality inspector, reselecting the organization, clicking a storage button, successfully editing, returning to a flow management list page, and simultaneously displaying modification time and a modifier;
(3) query flow management
Inputting a process name, selecting an organization, clicking a query button, querying a recording result meeting the condition, clicking a reset button, and emptying the task name content;
clicking the retract (expand) button, the query function is retracted (expanded).
The technical scheme of the invention is further explained in detail by combining the specific operation process of the recording quality inspection method of the artificial intelligent call center.
Referring to fig. 2, the recording quality inspection method of the artificial intelligence call center mainly comprises the following steps:
step S1: adding an initial check task, setting a task name, a task effective time and a task invalid time, inputting a screening condition for extracting a record from the task effective time to the task invalid time to obtain a monitored seat meeting the screening condition, and generating an initial check task list;
step S2: entering an initial inspection task list, selecting a monitored seat for quality inspection, and setting an initial inspection task allocation rule;
step S3: and starting the primary inspection task at the effective time of the task, extracting the recording data of the monitored seat according to the distribution rule, distributing the recording data to related quality inspectors to start quality inspection and marking, and finishing the primary inspection task at the ineffective time of the task.
In step S1, the screening conditions include: quality inspector, organization to which quality inspector belongs, recording start time, recording end time and monitored number.
Preferably, the method between sound recordings of the present invention may further comprise: and (3) checking the task state, wherein the specific checking method comprises the following steps:
before entering an initial inspection task list, checking whether the task needs to be changed into a starting state or an ending state according to task effective time and task invalid time: the task effective time is the current system time, and the task is in an un-started state; after the task effective time is the current system time, the task is in a creation state; the task is in an end state before the effective time of the task is the current system time;
checking the task state in the first step at regular time;
and thirdly, the quality inspection quantity is equal to the monitored quantity, and the task is automatically ended.
Preferably, the method for quality inspection of recorded sound further comprises: and executing an editing task on the initial inspection task in the non-started state or the suspended state, and modifying the initial inspection task list so as to execute a subsequent initial inspection task by the edited new initial inspection task list.
Preferably, the method for quality inspection of recorded sound further comprises: and when the task failure time is reached but the initial inspection task is not finished, releasing the recorded data of which the quality inspection is not finished.
Preferably, the method for quality inspection of recorded sound further comprises: and recording inquiry, namely checking the recording data of the monitored seat according to the identity authority of an inquirer and performing scoring operation on the recording.
Preferably, the method for quality inspection of recorded sound further comprises: the step of extracting the recording data comprises the steps of extracting according to task priority, and prioritizing the recording data according to service types in advance before extracting the recording data.
Preferably, the method for quality inspection of recorded sound further comprises: according to the request of the monitored agent for applying the primary examination and the reexamination, the reexamination operation is executed to the quality-examined sound recording according to the set reexamination process, and the reexamination process comprises the following steps:
adding flow management, inputting a flow name, an auditor, an organization to which the auditor belongs, a flow type and a renegotiation validity period, and generating a flow management list;
and entering a flow management list, and distributing the sound records applying for the primary examination and the reexamination within the reexamination validity period to related auditors according to the organization types to enter an approval flow.
Preferably, the method for quality inspection of recorded sound further comprises: recording the state circulation condition of the recording data according to the following rules:
the recording which is extracted by the primary examination task is not locked and not scored, and the recording is in a system scoring state;
the record locked by the quality inspector is in a state to be primarily inspected;
recording marked by the initial examination is in an initial examination state;
recording the application of the agent for the preliminary examination and the reexamination is in an applied state of the preliminary examination and the reexamination;
after the sound recording of the primary examination and reexamination is approved, the sound recording does not enter the examination and approval flow which is the final examination and reexamination examination and approval;
after the sound records of the primary examination and reexamination pass through all the examination and approval, the sound records of the primary examination and reexamination pass through;
the recording of the primary examination repeat is rejected as the rejected of the primary examination repeat.
The artificial intelligence call center recording quality inspection method utilizes the timing task in the decoupling service system, various simple or complex scheduling works are completed by task scheduling, the concurrency problem in the conventional project is avoided, deep customized development is carried out on the basis of an open source task scheduling framework Quartz, the artificial intelligence call center recording quality inspection method has a visual management interface, a task list, a task state, an execution record and an execution result can be checked, and functions of pause, recovery, immediate execution and the like are supported.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; the modifications or substitutions do not cause the essence of the corresponding technical solutions to depart from the scope of the technical solutions of the embodiments of the present invention, and are all included in the scope of the present invention.
The techniques, shapes, and configurations not described in detail in the present invention are all known techniques.
Claims (10)
1. A recording quality inspection method for an artificial intelligence call center is characterized by comprising the following steps:
adding an initial check task, setting a task name, a task effective time and a task invalid time, inputting a screening condition for extracting a record from the task effective time to the task invalid time to obtain a monitored seat meeting the screening condition, and generating an initial check task list;
entering the initial inspection task list, selecting a monitored seat for quality inspection, and setting an initial inspection task allocation rule;
and starting the preliminary examination task at the task effective time, extracting the recording data of the monitored seat according to the distribution rule, distributing the recording data to related quality inspectors to start quality inspection and marking, and finishing the preliminary examination task at the task invalid time.
2. The method of claim 1, wherein the screening condition comprises: quality inspector, organization to which quality inspector belongs, recording start time, recording end time and monitored number.
3. The artificial intelligence call center recording quality inspection method of claim 2, further comprising task status verification:
checking whether the task needs to be changed into a starting state or an ending state according to task effective time and task ineffective time before entering the initial inspection task list: the task effective time is the current system time, and the task is in an un-started state; after the task effective time is the current system time, the task is in a creation state; the task is in an end state before the effective time of the task is the current system time;
checking the task state in the first step at regular time;
and thirdly, the quality inspection quantity is equal to the monitored quantity, and the task is automatically ended.
4. The method as claimed in claim 1, further comprising editing the initial test task in an un-started state or a suspended state, and modifying the initial test task list to edit a new initial test task list to perform a subsequent initial test task.
5. The method as claimed in claim 1, wherein the recorded data of unfinished quality inspection is released when the task expiration time has been reached but the quality inspection of the preliminary inspection task is unfinished.
6. The method as claimed in claim 1, further comprising a recording inquiry for checking the recording data of the monitored agents and scoring the recording according to the identity rights of the inquirer.
7. The artificial intelligence call center recording quality inspection method of claim 1, wherein the step of extracting the recording data further comprises extracting according to task priority, and prioritizing the recording data according to business type in advance before extracting the recording data.
8. The method for quality inspection of audio recordings in an artificial intelligence call center as claimed in claim 1, further comprising performing a review operation on the quality-inspected audio recordings according to a set review process based on a request from the agent to be monitored for an initial review, the review process comprising:
adding flow management, inputting a flow name, an auditor, an organization to which the auditor belongs, a flow type and a renegotiation validity period, and generating a flow management list;
and entering the process management list, and distributing the sound records applying for the initial examination and the review within the review validity period to related auditors according to the organization types to enter an approval flow.
9. The artificial intelligence call center recording quality inspection method of claim 8, further comprising recording the status flow condition of the recording data according to the following rules:
the recording which is extracted by the primary examination task is not locked and not scored, and the recording is in a system scoring state;
the record locked by the quality inspector is in a state to be primarily inspected;
recording marked by the initial examination is in an initial examination state;
recording the application of the agent for the preliminary examination and the reexamination is in an applied state of the preliminary examination and the reexamination;
after the sound recording of the primary examination and reexamination is approved, the sound recording does not enter the examination and approval flow which is the final examination and reexamination examination and approval;
after the sound records of the primary examination and reexamination pass through all the examination and approval, the sound records of the primary examination and reexamination pass through;
the recording of the primary examination repeat is rejected as the rejected of the primary examination repeat.
10. An artificial intelligence call center recording quality inspection system, wherein the artificial intelligence call center recording quality inspection system is stored in a storage medium in the form of machine-readable program instructions, and when the machine-readable program instructions are executed, the artificial intelligence call center recording quality inspection method according to any one of claims 1 to 9 is executed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011116054.6A CN112235469A (en) | 2020-10-19 | 2020-10-19 | Method and system for quality inspection of recording of artificial intelligence call center |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011116054.6A CN112235469A (en) | 2020-10-19 | 2020-10-19 | Method and system for quality inspection of recording of artificial intelligence call center |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112235469A true CN112235469A (en) | 2021-01-15 |
Family
ID=74118844
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011116054.6A Pending CN112235469A (en) | 2020-10-19 | 2020-10-19 | Method and system for quality inspection of recording of artificial intelligence call center |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112235469A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110251845A1 (en) * | 2008-12-17 | 2011-10-13 | Nec Corporation | Voice activity detector, voice activity detection program, and parameter adjusting method |
| CN104168394A (en) * | 2014-06-27 | 2014-11-26 | 国家电网公司 | Sampling quality inspection method and system in calling center |
| CN105975514A (en) * | 2016-04-28 | 2016-09-28 | 朱宇光 | Automatic quality testing method and system |
| CN111787168A (en) * | 2020-06-28 | 2020-10-16 | 上海安录宇高数据技术有限公司 | Voice recognition quality inspection allocation method based on artificial intelligence technology |
-
2020
- 2020-10-19 CN CN202011116054.6A patent/CN112235469A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110251845A1 (en) * | 2008-12-17 | 2011-10-13 | Nec Corporation | Voice activity detector, voice activity detection program, and parameter adjusting method |
| CN104168394A (en) * | 2014-06-27 | 2014-11-26 | 国家电网公司 | Sampling quality inspection method and system in calling center |
| CN105975514A (en) * | 2016-04-28 | 2016-09-28 | 朱宇光 | Automatic quality testing method and system |
| CN111787168A (en) * | 2020-06-28 | 2020-10-16 | 上海安录宇高数据技术有限公司 | Voice recognition quality inspection allocation method based on artificial intelligence technology |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9176729B2 (en) | System and method for prioritizing and remediating defect risk in source code | |
| US7506312B1 (en) | Method and system for automatically determining risk areas to retest | |
| US8010584B1 (en) | Relational database method for technology program management | |
| CN103631713B (en) | ERP software automatization test system and method | |
| CN108108297A (en) | The method and apparatus of automatic test | |
| CN106951369B (en) | Management method and device for joint debugging test | |
| US20050144150A1 (en) | Remote process capture, identification, cataloging and modeling | |
| US20020165742A1 (en) | Feature centric release manager method and system | |
| CN107943694B (en) | Test data generation method and device | |
| CN112486557A (en) | Development-based complete life cycle management platform and method | |
| KR102232859B1 (en) | Method for automatically generating a worker pool based on the functional elements and difficulty of crowdsourcing based projects for artificial intelligence training data generation | |
| CN107368428A (en) | A kind of method for testing software and software testing system | |
| US20140282354A1 (en) | Automated team assembly system and method | |
| CN109344053A (en) | Interface coverage test method, system, computer equipment and storage medium | |
| Abdellatif et al. | Software analytics to software practice: a systematic literature review | |
| CN111258881B (en) | Intelligent test system for workflow test | |
| CN109656615A (en) | A method of permission early warning is carried out based on code method significance level | |
| Izquierdo-Cortazar et al. | Towards automated quality models for software development communities: The QualOSS and FLOSSMetrics case | |
| CN112235469A (en) | Method and system for quality inspection of recording of artificial intelligence call center | |
| US20040267814A1 (en) | Master test plan/system test plan database tool | |
| JP2011257947A (en) | Method, apparatus and program for preparing test plan | |
| CN115511317A (en) | Wind control strategy configuration method and device | |
| Bukhari et al. | Determining the factors affecting the accuracy of effort estimates for different application and task types | |
| US20070118531A1 (en) | Issues database system and method | |
| Khraiwesh | Configuration management measures in CMMI |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210115 |