CN113645357B

CN113645357B - Call quality inspection method, device, computer equipment and computer readable storage medium

Info

Publication number: CN113645357B
Application number: CN202110900763.1A
Authority: CN
Inventors: 黎进; 杨志; 邓锐涛; 汶林丁; 刘云峰
Original assignee: Shenzhen Zhuiyi Technology Co Ltd
Current assignee: Shenzhen Zhuiyi Technology Co Ltd
Priority date: 2021-08-06
Filing date: 2021-08-06
Publication date: 2024-02-20
Anticipated expiration: 2041-08-06
Also published as: CN113645357A

Abstract

The application relates to a call quality inspection method, a call quality inspection device, computer equipment and a computer readable storage medium. The method comprises the following steps: if a trigger instruction aiming at a target quality inspection task is detected, displaying a quality inspection interface, wherein the quality inspection interface at least comprises outbound call data corresponding to the target quality inspection task and a quality inspection labeling control; acquiring quality inspection data aiming at the outbound call conversation data based on the quality inspection marking control, wherein the quality inspection data is used for representing at least one of data quality and outbound call quality of the outbound call conversation data; and storing the quality inspection data and the outbound call conversation data in a quality inspection database correspondingly. By adopting the method, the quality inspection efficiency and the quality inspection accuracy of the call quality inspection can be improved.

Description

Call quality inspection method, device, computer equipment and computer readable storage medium

Technical Field

The present invention relates to the field of artificial intelligence technology, and in particular, to a method, an apparatus, a computer device, and a computer readable storage medium for call quality inspection.

Background

With the rapid development of artificial intelligence technology, it is becoming more and more common to replace manual work with outbound robots to outbound user terminals, and currently, outbound robots play an increasingly important role in product propaganda, product sales, after-sales management and the like.

Taking the field of finance and technology as an example, a financial institution generally carries out the promotion of calling out or business promotion of calling out to a user terminal through a calling out robot, and after the calling out is finished, the calling voice which has been called out is usually manually led out to carry out the rebirth quality inspection, so as to check whether the communication process is smooth, whether the calling out robot can accurately understand and respond to the intention of the user, whether the calling conclusion output by the calling out robot is accurate, and the like, and the relevant quality inspection data are manually recorded after the quality inspection is finished by the manual work.

However, the above-described call quality inspection method has a problem of poor quality inspection efficiency and accuracy.

Disclosure of Invention

Based on the above, a call quality inspection method, a call quality inspection device, a computer device and a computer readable storage medium capable of improving the quality inspection efficiency and the quality inspection accuracy of the call quality inspection are provided.

In a first aspect, an embodiment of the present application provides a call quality inspection method, where the method includes:

if a trigger instruction aiming at a target quality inspection task is detected, displaying a quality inspection interface, wherein the quality inspection interface at least comprises outbound call data corresponding to the target quality inspection task and a quality inspection labeling control;

acquiring quality inspection data aiming at the outbound call conversation data based on the quality inspection marking control, wherein the quality inspection data is used for representing at least one of data quality and outbound call quality of the outbound call conversation data;

And storing the quality inspection data and the outbound call conversation data in a quality inspection database correspondingly.

In one embodiment, the quality inspection annotation control includes a sentence annotation control, and the displaying the quality inspection interface if a trigger instruction for the target quality inspection task is detected includes:

if a trigger instruction aiming at the target quality inspection task is detected, displaying an initial quality inspection interface, wherein the initial quality inspection interface comprises the outbound call data;

and if a selection instruction of a user for the outbound call data is detected, displaying the statement marking control in the initial quality inspection interface so as to display the quality inspection interface.

In one embodiment, the outbound call data includes an outbound call text, where the outbound call text is obtained by identifying an outbound call voice corresponding to the target quality inspection task.

In one embodiment, the sentence annotation control includes an annotation sub-control, the quality inspection data includes text recognition accurate quality inspection data, and the acquiring the quality inspection data for the outbound call data based on the quality inspection annotation control includes:

if a selection instruction of a user for the labeling sub-control is detected, a labeling sentence page is displayed;

And acquiring the text recognition accurate property check data input by the user based on the annotation statement page, wherein the text recognition accurate property check data comprises at least one of voice recognition accurate property check data and intention recognition accurate property check data.

In one embodiment, the sentence annotation control includes a performance optimization sub-control, the quality inspection data includes identification performance optimization data, and the acquiring quality inspection data for the outbound call data based on the quality inspection annotation control includes:

if a selection instruction of the user for the performance optimization sub-control is detected, displaying a similar query supplementary page;

and acquiring the recognition performance optimization data input by the user based on the similarity query supplementary page, wherein the recognition performance optimization data comprises similar question text corresponding to the outbound call text.

In one embodiment, the sentence annotation control further includes a voice playing sub-control, where the voice playing sub-control is configured to play the outbound call voice in a voice manner.

In one embodiment, the quality inspection annotation control includes a sentence quality inspection control, the sentence quality inspection control includes a plurality of different call experience label sub-controls, and the acquiring quality inspection data for the outbound call data based on the quality inspection annotation control includes:

For the outbound call data, if a selection instruction of a user for a target call experience label sub-control is detected, determining a target call experience label corresponding to the call experience label sub-control;

and acquiring the quality inspection data according to the target call experience tag.

In one embodiment, the quality inspection labeling control further includes a call result quality inspection control, and the acquiring the quality inspection data according to the target call experience tag includes:

acquiring call quality evaluation data and call result analysis data corresponding to the outbound call data based on the call result quality inspection control;

and taking the target call experience tag, the call quality evaluation data and the call result analysis data as the quality inspection data.

In one embodiment, before the storing the quality inspection data and the outbound call data in the quality inspection database, the method further includes:

generating a quality inspection auditing task according to the target quality inspection task and the quality inspection data;

correspondingly, the storing the quality inspection data and the outbound call data in the quality inspection database includes:

and if the auditing result of the quality inspection auditing task is detected to be a target auditing result, correspondingly storing the quality inspection data and the outbound call data in a quality inspection database.

In a second aspect, an embodiment of the present application provides a device for detecting call quality, where the device includes:

the display module is used for displaying a quality inspection interface if a trigger instruction aiming at a target quality inspection task is detected, wherein the quality inspection interface at least comprises outbound call data corresponding to the target quality inspection task and a quality inspection labeling control;

the quality inspection module is used for acquiring quality inspection data aiming at the outbound call data based on the quality inspection marking control, wherein the quality inspection data is used for representing at least one of data quality and outbound call quality of the outbound call data;

and the storage module is used for storing the quality inspection data and the outbound call data in a quality inspection database correspondingly.

In one embodiment, the quality inspection annotation control comprises a sentence annotation control, and the presentation module comprises:

the first display unit is used for displaying an initial quality inspection interface if a trigger instruction aiming at the target quality inspection task is detected, wherein the initial quality inspection interface comprises the outbound call data;

and the second display unit is used for displaying the statement marking control in the initial quality inspection interface to display the quality inspection interface if the selection instruction of the user for the outbound call data is detected.

In one embodiment, the sentence annotation control includes an annotation sub-control, the quality inspection data includes text recognition accuracy quality inspection data, and the obtaining module includes:

the third display unit is used for displaying the annotation sentence page if the selection instruction of the user for the annotation child control is detected;

the first acquisition unit is used for acquiring the text recognition accurate quality inspection data input by the user based on the annotation statement page, wherein the text recognition accurate quality inspection data comprises at least one of voice recognition accurate quality inspection data and intention recognition accurate quality inspection data.

In one embodiment, the statement annotation control comprises a performance optimization sub-control, the quality inspection data comprises identification performance optimization data, and the acquisition module comprises:

the fourth display unit is used for displaying a similar query supplementary page if a selection instruction of a user for the performance optimization sub-control is detected;

and the second acquisition unit is used for acquiring the recognition performance optimization data input by the user based on the similarity query supplementary page, wherein the recognition performance optimization data comprises similar question text corresponding to the outbound call text.

In one embodiment, the quality inspection annotation control comprises a sentence quality inspection control comprising a plurality of different conversation experience label sub-controls, and the acquisition module comprises:

the determining unit is used for determining a target call experience label corresponding to the call experience label sub-control if a selection instruction of a user for the target call experience label sub-control is detected aiming at the outbound call data;

and the third acquisition unit is used for acquiring the quality inspection data according to the target call experience tag.

In an embodiment, the quality inspection labeling control further includes a call result quality inspection control, and the third obtaining unit is specifically configured to obtain call quality evaluation data and call result analysis data corresponding to the outbound call data based on the call result quality inspection control; and taking the target call experience tag, the call quality evaluation data and the call result analysis data as the quality inspection data.

In one embodiment, the apparatus further comprises:

The generation module is used for generating a quality inspection auditing task according to the target quality inspection task and the quality inspection data;

correspondingly, the storage module is specifically configured to store the quality inspection data and the outbound call data in a quality inspection database correspondingly if it is detected that the auditing result of the quality inspection auditing task is a target auditing result.

In a third aspect, embodiments of the present application provide a computer device comprising a memory storing a computer program and a processor implementing the steps of the method of the first aspect as described above when the processor executes the computer program.

In a fourth aspect, embodiments of the present application provide a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the method of the first aspect described above.

The beneficial effects that technical scheme that this application embodiment provided include at least:

if a trigger instruction aiming at a target quality inspection task is detected, displaying a quality inspection interface, wherein the quality inspection interface at least comprises outbound call data corresponding to the target quality inspection task and a quality inspection marking control, acquiring quality inspection data aiming at the outbound call data based on the quality inspection marking control, and correspondingly storing the quality inspection data and the outbound call data in a quality inspection database, wherein the quality inspection data is used for representing at least one of data quality and outbound call quality of the outbound call data; therefore, the quality inspection personnel does not need to manually derive the call voice of the outbound call to carry out the call quality inspection, and also does not need to manually record the quality inspection data after the quality inspection is finished.

Drawings

FIG. 1-a is an application environment diagram of a call quality inspection method according to an embodiment;

FIG. 1-b is an application environment diagram of a call quality inspection method according to another embodiment;

fig. 2 is a flow chart of a call quality inspection method according to an embodiment;

FIG. 3 is a flow chart of step 201 provided in another embodiment;

FIG. 4 is a flow chart of step 202 provided in another embodiment;

FIG. 5 is a schematic diagram of an exemplary markup language page provided by another embodiment;

FIG. 6 is a flow chart of step 202 provided in another embodiment;

fig. 7 is a schematic diagram of a statement marking control corresponding to exemplary outbound call data according to another embodiment;

FIG. 8 is a schematic diagram of an exemplary similarly-to-page supplemental page provided in another embodiment;

FIG. 9 is a flow chart of step 202 provided in another embodiment;

FIG. 10 is a schematic diagram of a sentence quality control exemplary presented in a quality inspection interface, provided by another embodiment;

FIG. 11 is a flowchart illustrating a step 902 provided in another embodiment;

FIG. 12 is a schematic diagram of an exemplary call result quality control provided by another embodiment;

Fig. 13 is a flowchart illustrating a call quality inspection method according to another embodiment;

FIG. 14 is a schematic diagram of an exemplary geological inspection audit interface provided in accordance with another embodiment;

FIG. 15 is a schematic diagram of an exemplary task audit statistics interface provided by another embodiment;

fig. 16 is a block diagram of a call quality inspection device according to an embodiment;

fig. 17 is an internal structural diagram of a computer device according to an embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.

The embodiment of the application provides a call quality inspection method, a call quality inspection device, computer equipment and a computer readable storage medium, which aim to solve the technical problems of low quality inspection efficiency and poor quality inspection accuracy caused by manually guiding out call voice after an outbound call is ended, and manually recording relevant quality inspection data after the quality inspection is ended. The following will specifically describe the technical solutions of the present application and how the technical solutions of the present application solve the above technical problems by means of examples and with reference to the accompanying drawings. The following embodiments may be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

The technical solutions related to the embodiments of the present application are described below in conjunction with the scenarios applied by the embodiments of the present application.

Fig. 1-a is a schematic diagram of an implementation environment related to a call quality inspection method according to an embodiment of the present application. As shown in fig. 1-a, the implementation environment may include a computer device 101, and the computer device 101 may be a smart phone, a tablet computer, a personal computer, a notebook computer, an in-vehicle device, or the like.

In the implementation environment shown in fig. 1-a, if the computer device 101 detects a trigger instruction for a target quality inspection task, a quality inspection interface may be displayed, where the quality inspection interface includes at least outbound call data corresponding to the target quality inspection task and a quality inspection labeling control; the computer device 101 may obtain quality inspection data for the outbound call data based on the quality inspection annotation control, the quality inspection data being used to characterize at least one of a data quality and an outbound quality of the outbound call data; the computer device 101 may store the quality inspection data in the quality inspection database in correspondence with the outgoing call session data.

Fig. 1-b is a schematic diagram of another implementation environment related to a call quality inspection method according to an embodiment of the present application. As shown in fig. 1-b, the implementation environment may further include a computer device 101 and a server 102, where the computer device 101 and the server 102 may communicate over a wired network or a wireless network.

The computer device 101 may be a smart phone, a tablet computer, a personal computer, a notebook computer, a vehicle-mounted device, etc.; the server 102 may be one server or may be a server cluster including a plurality of servers.

In the implementation environment shown in fig. 1-b, if the computer device 101 detects a trigger instruction for a target quality inspection task, a quality inspection interface may be requested to the server 102, and after the computer device 101 receives the quality inspection interface sent by the server 102, the quality inspection interface may be displayed, where the quality inspection interface includes at least outbound call data corresponding to the target quality inspection task and a quality inspection labeling control; the computer device 101 may obtain quality inspection data for the outbound call data based on the quality inspection annotation control, the quality inspection data being used to characterize at least one of a data quality and an outbound quality of the outbound call data; the computer device 101 may send the acquired quality inspection data to the server 102, and the server 102 may store the quality inspection data in the quality inspection database corresponding to the outbound call data.

In one embodiment, as shown in fig. 2, a call quality inspection method is provided, and the method is applied to the computer device 101 in fig. 1-a for illustration, and includes the following steps:

In step 201, if the computer device detects a trigger instruction for a target quality inspection task, a quality inspection interface is displayed.

Currently, automatic outbound of a user terminal by using an outbound robot instead of manual work is becoming more and more common, and the automatic outbound is widely applied in a plurality of industries. For example, in the field of financial science and technology, a financial institution will generally make an external call to a user terminal through an external caller, or make a business promotion external call, in the field of e-commerce, an e-commerce platform makes a satisfaction return visit to the user terminal through the external caller, and so on.

Specifically, for a service scenario, in the outbound process, an outbound robot generally needs to perform voice recognition on outbound call voices of outbound objects to obtain texts, then performs intention recognition on the texts to obtain intention, and the outbound robot matches an outbound knowledge base of the service scenario according to the intention to obtain automatic dialogue sentences of outbound nodes triggered by the intention, and feeds the automatic dialogue sentences back to terminals of the outbound objects, so that automatic dialogue in the outbound process is realized. However, in the actual application process, there are often situations that the outbound robot cannot accurately converse due to the reasons that the outbound robot cannot accurately identify the semantic intention of the outbound object, and the like, which affects the outbound quality.

In the embodiment of the application, in order to perform quality inspection on the outbound quality of the outbound robot, after the outbound is finished, the computer device may configure quality inspection tasks, each quality inspection task is associated with a corresponding quality inspection interface, and the quality inspection personnel may perform call quality inspection based on the quality inspection interface. Of course, the quality inspection task may also be configured by the quality inspection personnel themselves.

In this embodiment of the present application, the target quality inspection task may be any quality inspection task, and the target quality inspection task may be a quality inspection task configured for one complete session, or may be a quality inspection task configured for a part of sessions in one complete session, or may be a quality inspection task configured for a plurality of complete sessions, or the like.

Taking the example that the target quality inspection task is the quality inspection task configured for one complete session, one complete session may be a complete outbound session procedure for one outbound object. The computer equipment can acquire dialogue data in the outbound process, the dialogue data can comprise automatic dialogue sentences of the outbound robot and outbound dialogue voices of outbound objects, and generates a quality inspection interface corresponding to a target quality inspection task by using the dialogue data, and optionally, the computer equipment can store the generated quality inspection interface in association with a task identifier corresponding to the target quality inspection task.

In the embodiment of the application, the generated quality inspection interface at least comprises outbound call data corresponding to the target quality inspection task and a quality inspection labeling control. The quality inspection labeling control can be used for quality inspection personnel to input quality inspection data aiming at outbound call conversation data, and the outbound call conversation data can be obtained according to outbound call conversation voice of an outbound call object in an outbound call process. The following describes the way to obtain the outbound call data.

In one possible implementation manner, the outbound call data may include call data acquired based on the outbound call voice, for example, the outbound call data may be an outbound call text obtained by performing voice recognition on the outbound call voice, the outbound call data may also be the outbound call voice, and the outbound call data may also be a combination of the outbound call text and the outbound call voice, that is, the outbound call data may be in a voice form, or may be in a text form, or may be a combination of the voice form and the text form, which is not limited herein.

In another possible embodiment, the outbound call data may further include call data and an intention acquired based on the outbound call voice, that is, the outbound call data may further include an intention of the outbound call voice, for example, the outbound call data may include an outbound call text obtained by identifying the outbound call voice and an intention of the outbound call voice, the outbound call data may further include an intention of the outbound call voice and the outbound call voice, and the outbound call data may further include an intention of the outbound call text, the outbound call voice and the outbound call voice.

After the computer equipment configures the target quality inspection task, the target quality inspection task can be distributed to the quality inspection account of the quality inspection personnel, so that after the quality inspection personnel logs in the quality inspection account, the target quality inspection task can be triggered to carry out conversation quality inspection. For example, the quality inspector may click on the target quality inspection task to input a trigger instruction, of course, the quality inspector may also input a trigger instruction for the target quality inspection task through voice, the quality inspector may also input a trigger instruction for the target quality inspection task through facial expression, gesture motion, and so on.

The trigger instruction can carry a task identifier of the target quality inspection task, so that after the computer equipment detects the trigger instruction, a quality inspection interface corresponding to the task identifier is acquired and displayed.

In step 202, the computer device obtains quality inspection data for outbound call data based on the quality inspection annotation control.

After the computer equipment displays the quality inspection interface, the quality inspection personnel can conduct conversation quality inspection according to the outbound conversation data.

Taking the example that the outbound call data comprises call data acquired based on outbound call voice, the quality inspection interface can also comprise automatic dialogue sentences of the outbound robot, and the computer equipment can display the call data and the automatic dialogue sentences in a dialogue form in the quality inspection interface. In this way, the quality inspection personnel can determine whether the automatic dialogue statement of the outbound robot aiming at the outbound call data is accurate according to the dialogue logic of the context in the quality inspection interface, and can input the quality inspection data based on the quality inspection marking control after the automatic dialogue statement is determined.

Taking the example that the outbound call data comprises call data and intention acquired based on outbound call voice, as an implementation mode, quality inspection personnel can directly check whether the intention is matched with the call data, so that whether an automatic dialogue statement replied by an outbound call robot aiming at the outbound call data is accurate or not can be determined, and quality inspection data can be input based on a quality inspection labeling control after the automatic dialogue statement replied by the outbound call robot is determined; as another implementation mode, the quality inspection interface can further comprise an automatic dialogue statement of the outbound robot, the computer device displays the outbound call data and the automatic dialogue statement in a dialogue form in the quality inspection interface, so that a quality inspection person can judge whether the automatic dialogue statement replied by the outbound robot for the outbound call data is accurate or not by combining the dialogue logic and the intention of the context, and quality inspection data is obtained.

In the embodiment of the application, the quality inspection data is used for representing at least one of data quality and outbound quality of outbound call data. The quality inspection data may include a voice recognition result and an intention recognition result corresponding to the outbound call voice, so that if the voice recognition result has a deviation, the voice recognition quality is poor, that is, the quality inspection data represents that the data quality of the outbound call data is poor; if the intention recognition result has deviation, the intention recognition quality is poor, namely the quality inspection data also represents the poor data quality of the outbound call data, and the like.

The quality inspection data can also comprise results of whether the outbound is connected, whether the conversation process is too noisy, whether the outbound robot and the outbound target have the condition of mutually robbing, and the like, and if the outbound is not connected, the noise is too loud or the problem of mutually robbing exists, the quality inspection data represents that the outbound quality of the outbound conversation data is poor.

In step 203, the computer device stores the quality inspection data and the outbound call data in the quality inspection database.

The computer equipment correspondingly stores the quality inspection data input by the quality inspection personnel based on the outbound call data and the outbound call data in the quality inspection database, so that if the data quality of the outbound call data represented by the quality inspection data and/or the outbound call quality are poor, the computer equipment can perform performance optimization on a machine learning model such as a voice recognition model, an intention recognition model and the like used by the outbound robot in the outbound call process so as to improve the outbound call quality of the outbound robot.

In the above embodiment, if the computer device detects a trigger instruction for a target quality inspection task, displaying a quality inspection interface, where the quality inspection interface includes at least outbound call data corresponding to the target quality inspection task and a quality inspection labeling control, and then acquiring quality inspection data for the outbound call data based on the quality inspection labeling control, where the quality inspection data corresponds to the outbound call data and is stored in a quality inspection database, and the quality inspection data is used to characterize at least one of data quality and outbound call quality of the outbound call data; therefore, the quality inspection personnel does not need to manually derive the call voice of the outbound call, and also does not need to manually record the quality inspection data after the quality inspection is finished, and the quality inspection personnel can input the quality inspection data of the outbound call data based on the quality inspection marking control through the quality inspection interface.

In one embodiment, based on the embodiment shown in fig. 2, referring to fig. 3, this embodiment relates to a process how a computer device presents a quality inspection interface. The quality inspection annotation control in this embodiment includes a sentence annotation control, as shown in fig. 3, step 201 includes steps 301 and 302:

in step 301, if the computer device detects a trigger instruction for a target quality inspection task, an initial quality inspection interface is displayed.

The quality inspection personnel can click the target quality inspection task to input a trigger instruction, and can input the trigger instruction for the target quality inspection task through voice or input the trigger instruction for the target quality inspection task through facial expression and gesture actions.

After the computer equipment detects the trigger instruction aiming at the target quality inspection task, the computer equipment displays an initial quality inspection interface, wherein the initial quality inspection interface comprises outbound call data, namely the computer equipment firstly displays the outbound call data in the initial quality inspection interface.

It can be understood that the number of outbound call data may be plural in the outbound session, that is, the outbound robot and the outbound object perform multiple dialogue interactions, so that after detecting a trigger instruction for the target quality inspection task, the computer device displays all outbound call data corresponding to the target quality inspection task in the initial quality inspection interface, and the quality inspection personnel may select one outbound call data or plural outbound call data for quality inspection as required.

The outbound call data may include outbound call text, which is obtained by identifying outbound call voices corresponding to the target quality inspection tasks. As an implementation manner, after the computer device obtains the outbound call voice of the outbound call object, the outbound call voice may be input to the voice recognition model, so as to obtain the outbound call text output by the voice recognition model.

Of course, the outbound call data may also include an intention of an outbound call object corresponding to the outbound call text and the outbound call voice. The computer device may input the text of the outbound call obtained by the voice recognition to the intention recognition model to obtain the intention.

In step 302, if the computer device detects a selection instruction of the user for outbound call data, a sentence annotation control is displayed in the initial quality inspection interface to display the quality inspection interface.

Based on the initial quality inspection interface, a user (i.e. a quality inspection personnel) can select outbound call data needing quality inspection, and input a selection instruction for the outbound call data, and if the computer equipment detects the selection instruction for the outbound call data, the sentence annotation control for the outbound call data is displayed in the initial quality inspection interface, so that the initial quality inspection interface with the sentence annotation control and the outbound call data is the quality inspection interface.

In the following, a procedure for detecting a selection instruction of a user for outbound call data by the computer device will be described.

In one possible implementation manner, a selection box corresponding to the outbound call data can be displayed in the quality inspection interface, and if the quality inspection personnel selects the selection box, the computer equipment detects a selection instruction of a user for the outbound call data, so that a display statement marking control is triggered.

In another possible implementation manner, if the computer device detects that the mouse pointer is suspended within a preset range around the outbound call data in the quality inspection interface, the computer device determines that the selection instruction of the user for the outbound call data is detected, that is, if the quality inspection personnel moves the mouse pointer, if the mouse pointer is suspended above or around the outbound call data, the computer device detects the selection instruction of the user for the outbound call data.

Of course, the quality inspector may also press the outbound call data for a long time to input a selection instruction for the outbound call data, etc., without being particularly limited herein.

In this way, if the computer equipment detects a trigger instruction aiming at a target quality inspection task, the computer equipment firstly displays an initial quality inspection interface, and if the computer equipment detects a selection instruction aiming at outbound call data input by a user in the initial quality inspection interface, then displays a sentence annotation control in the initial quality inspection interface so as to display the quality inspection interface, so that the sentence annotation control can be displayed according to the actual quality inspection requirements of quality inspection personnel, and the quality inspection flexibility is improved.

In one embodiment, based on the embodiment shown in fig. 3, referring to fig. 4, this embodiment relates to a process of how a computer device obtains quality inspection data for outgoing call data based on a quality inspection annotation control. In this embodiment, the sentence annotation control includes an annotation sub-control, and the quality inspection data includes text recognition accurate quality inspection data, as shown in fig. 4, step 202 includes step 401 and step 402:

in step 401, if the computer device detects a selection instruction of the user for the labeling sub-control, the labeling sentence page is displayed.

Illustratively, after the quality inspector selects the outbound call data (as described above, the quality inspector may select the outbound call data by selecting the selection frame, suspending the mouse pointer, long-pressing, etc.), the computer device displays a sentence annotation control corresponding to the outbound call data, where the sentence annotation control includes an annotation sub-control.

And if the computer equipment detects a selection instruction of the user for the labeling sub-control, displaying a labeling sentence page. Illustratively, the quality inspector may click on the labeling sub-control to enter a selection instruction for the labeling sub-control.

In step 402, the computer device obtains text recognition accurate quality inspection data input by a user based on the annotation statement page.

In the embodiment of the application, the labeling sentence page is used for inputting text recognition accuracy quality inspection data by a quality inspection personnel, and in a possible implementation manner, the labeling sentence page may include a voice recognition result, an intention correction control and a voice recognition result correction control.

The voice recognition result included in the annotation statement page is an outbound call text obtained by recognizing outbound call voice of an outbound call object; the intention recognition result is an intention obtained by carrying out intention recognition on the outbound call voice or the outbound call text.

Illustratively, referring to FIG. 5, FIG. 5 is a page schematic of a markup language page. The voice recognition result correction control is shown as a "correct voice recognition result" control in fig. 5 and a correct voice recognition result input box associated with the control, and the computer device may default that the correct voice recognition result input box is empty or the same as the voice recognition result displayed in the markup sentence page, that is, the default voice recognition result is correct. Thus, if the quality inspection personnel modifies the default result in the correct voice recognition result input box, namely, the quality inspection personnel inputs the correct voice recognition result in the voice recognition result correction control, the computer equipment determines that the voice recognition result has deviation and counts the voice recognition accuracy.

Optionally, the sentence annotation control corresponding to the outbound call data displayed by the computer device further includes a voice playing sub-control, where the voice playing sub-control is used for voice playing of the outbound call voice. If the quality inspection personnel inputs a trigger instruction aiming at the voice playing sub-control, the computer equipment plays the voice of the outbound call corresponding to the outbound call data, so that the quality inspection personnel can determine whether the voice recognition result is correct or not by listening to the voice of the outbound call and comparing the listened content with the voice recognition result displayed in the annotation statement page, and if the voice recognition result is incorrect, the quality inspection personnel can input the correct voice recognition result in the voice recognition result correction control.

With continued reference to FIG. 5, the intent correction control may be a "correct intent" control as shown in FIG. 5 and a drop-down option control associated with the control, which the computer device may default to be null or the same as the intent recognition result, i.e., the default intent recognition result is correct. Thus, if the quality inspector modifies the default result in the drop-down option control, i.e., the quality inspector inputs a correct intention in the intention correction control, the computer device determines that the intention recognition result has a deviation and counts the statistics of the intention understanding accuracy. Optionally, the drop down option control may provide all intents that the corresponding dialog node in the current business scenario can trigger.

Of course, the drop down option control may also be replaced with an input box, so that the quality inspector can directly enter the correct intent in the input box.

Optionally, the annotation sentence page may further include an intent recognition error cause analysis control, with continued reference to fig. 5, where the intent recognition error cause analysis control may include a plurality of intent recognition error cause selection sub-controls, and illustratively, the plurality of intent recognition error cause selection sub-controls may include a "lack intent" sub-control, a "lack educational data" sub-control, a "speech recognition error" sub-control, and a "other cause" sub-control, so that if an intent recognition error occurs, a quality inspector may select a corresponding sub-control from the plurality of selection sub-controls, and the computer device may record the cause of the intent recognition error to facilitate a subsequent intent recognition model optimization operation.

Optionally, the annotation sentence page may further include a voice recognition error cause analysis control, with continued reference to fig. 5, the voice recognition error cause analysis control may include a plurality of voice recognition error cause selection sub-controls, the plurality of voice recognition error cause selection sub-controls may include a "normal voice but text error" sub-control, a "missed recognition part voice" sub-control, a "background voice recognized" sub-control, a "noise recognized" sub-control, a "other voice recognized" sub-control, and the like, and if the voice recognition result includes a voice recognition error cause represented by the above-mentioned certain sub-control, a quality inspector may select a corresponding sub-control from the plurality of selection sub-controls, and the computer device may record the cause of the voice recognition error, so as to facilitate the subsequent voice recognition model optimization operation.

The following briefly describes the meaning of each voice recognition error cause, and normal voice but with wrong words refers to voice recognition results outside the service recognition range in the process of outbound conversation, and the voice recognition results are counted into statistics of voice recognition error numbers; missing part of voice refers to the situation that voice is not recognized; recognizing background voice, namely voice recognition results of people except main dialog people in the process of outbound dialog, wherein the voice recognition results are counted into statistics of voice recognition error numbers; noise is identified, namely, the noise in the voice of the outbound call is identified as words, and the voice recognition result is counted into statistics of voice recognition errors; other voice is identified, which means that the voice is ambiguous, can not be distinguished, or the outbound call data is of an unsupported dialect type, etc., and the voice identification result is counted into statistics of voice identification error number. Therefore, the quality inspector can select the corresponding selection sub-control according to the reason of the actual voice recognition error, and the optimization of the related model algorithm used in the outbound process by the computer equipment is facilitated.

Optionally, the labeling sentence page may further include a semantic recognition result confirmation control, please continue to refer to fig. 5, where the semantic recognition result confirmation control is "the same as the original meaning" and "different from the original meaning" as "the semantic recognition case" and "the two associated confirmation sub-controls in fig. 5, and the quality inspector determines the semantic recognition case according to the voice recognition result and selects in the labeling sentence page. The text semantic meaning representing the voice recognition result is identical to the semantic meaning actually intended to be expressed by the outbound object, the text semantic meaning representing the voice recognition result is different from the semantic meaning, the semantic meaning actually intended to be expressed by the outbound object is different from the semantic meaning, and the voice recognition sentence meaning error rate statistics can be counted by a quality inspector aiming at the selection result of the two confirmation sub-controls.

In summary, in the embodiment of the present application, the text recognition accuracy quality inspection data input by the quality inspection personnel based on the labeling sentence page includes at least one of speech recognition accuracy quality inspection data and intention recognition accuracy quality inspection data. The voice recognition accurate quality control data may include, for example, the correct voice recognition result, the cause of the voice recognition error, and the intention recognition accurate quality control data may include, for example, the correct intention, the cause of the intention recognition error, and the like.

According to the embodiment, the quality inspection personnel can input the text identification accurate quality inspection data based on the labeling sentence page displayed by the computer equipment, and the computer equipment can automatically store the text identification accurate quality inspection data, so that manual recording of the quality inspection personnel is not needed, and the quality inspection efficiency is improved. In addition, the computer equipment can optimize the voice recognition model, the intention recognition model and other models used in the outbound process based on the stored text recognition accurate quality detection data, thereby being beneficial to improving the outbound quality.

In one embodiment, based on the embodiment shown in fig. 3, referring to fig. 6, this embodiment relates to another exemplary process of how a computer device obtains quality control data for outgoing call data based on a quality control annotation control. The sentence annotation control in this embodiment includes a performance optimization sub-control, and the quality inspection data includes identification performance optimization data, as shown in fig. 6, and step 202 includes steps 601 and 602:

In step 601, if the computer device detects a selection instruction of the user for the performance optimization sub-control, a similar query supplementary page is displayed.

Illustratively, after the quality inspector selects the outbound call data (as described above, the quality inspector may select the outbound call data by selecting the selection frame, suspending the mouse pointer, long-pressing, etc.), the computer device displays a sentence annotation control corresponding to the outbound call data, where the sentence annotation control includes a performance optimization sub-control.

Referring to fig. 7, fig. 7 is a schematic diagram illustrating a sentence annotation control corresponding to outbound call data. As shown in fig. 7, the sentence annotation control may include a performance optimization sub-control, and may further include an annotation sub-control, a voice playing sub-control, and the like related to the foregoing embodiment.

Illustratively, the quality inspector may click on the performance optimization sub-control to input a selection instruction for the performance optimization sub-control, so that the computer device presents the similar query supplementary page if it detects a selection instruction for the performance optimization sub-control by the user.

In step 602, the computer device obtains identification performance optimization data entered by the user based on the similarity query supplemental page.

The recognition performance optimization data comprises similar question text corresponding to the outbound call text.

Referring to fig. 8, for an exemplary illustration of a page of an exemplary supplemental page of similar questions, a quality inspector can quickly enter similar question text based on the supplemental page of similar questions.

In one possible implementation, the supplementary page may further include a voice recognition result, that is, an outbound call text, and the quality inspector may input similar question text matching the outbound call text in the supplementary page. For example, if the outbound call text of the outbound call object expresses a certain intention in the mode a, the one or more similar question texts input by the quality inspector may express the intention corresponding to the outbound call text in the modes B, C, etc.

Therefore, the computer equipment stores the identification performance optimization data input by the user based on the similarity query supplementary page as quality inspection data, so that manual recording of quality inspection personnel is not needed, and quality inspection efficiency is improved. In addition, the computer equipment can optimize semantic understanding models such as an intention recognition model and the like used in the outbound process based on the recognition performance optimization data, so that the outbound quality is improved.

In one embodiment, based on the embodiment shown in fig. 3, referring to fig. 9, this embodiment relates to another exemplary process of how a computer device obtains quality control data for outgoing call data based on a quality control annotation control. In this embodiment, the quality inspection labeling control includes a sentence quality inspection control, where the sentence quality inspection control includes a plurality of different call experience label sub-controls, and step 202 includes steps 901 and 902 shown in fig. 9:

In step 901, for outbound call data, if a selection instruction of a user for a target call experience label sub-control is detected, determining a target call experience label corresponding to the call experience label sub-control.

In this embodiment of the present application, a selection item is configured in the quality inspection interface for each outbound call data, for example, may be a selection box, and if a quality inspection person selects the selection item of the outbound call data, the computer device displays a sentence quality inspection control corresponding to the outbound call data in the quality inspection interface.

Referring to fig. 10, fig. 10 is an interface schematic diagram illustrating sentence quality control in a quality inspection interface by an exemplary computer device. As shown in fig. 10, the sentence quality control corresponding to the outbound call data may include a plurality of different call experience label sub-controls, including, for example, a "loud" call experience label sub-control, a "poor" call experience label sub-control, and a "mutually robbed" call experience label sub-control.

It should be noted that, the conversation experience label sub-control may be configured by the quality inspector, and the conversation experience label sub-control is used for marking the relevant data of the customer experience in the current dialogue by the quality inspector.

In step 902, the computer device obtains quality inspection data according to the target call experience tag.

Thus, the quality inspector can select the conversation experience label of the current conversation, for example, if the current conversation is too noisy, the quality inspector clicks the conversation experience label sub-control with excessive noise, and the computer device determines the target conversation experience label clicked by the quality inspector, so as to obtain quality inspection data, wherein the quality inspection data is the conversation experience label represented by the target conversation experience label.

Based on the embodiment shown in fig. 9, in a possible implementation manner, the quality inspection labeling control further includes a call result quality inspection control, referring to fig. 11, and step 902 includes step 9021 and step 9022 shown in fig. 11.

In step 9021, the computer device obtains call quality evaluation data and call result analysis data corresponding to the outbound call data based on the call result quality control.

Referring to fig. 12, fig. 12 is an interface diagram of an exemplary call result quality control. As shown in fig. 12, the call result quality inspection control may include a call quality evaluation sub-control and a call analysis result sub-control, where the call quality evaluation sub-control may include a "run" option, a "no run" option and an "no call" option, and the quality inspection personnel may select a corresponding option according to the actual situation of the current outbound call, and the computer device may obtain call quality evaluation data, and similarly, the quality inspection personnel may input call result analysis data based on the call analysis result sub-control, where the call result analysis data may include, for example, an outgoing call state (such as completed) of the current outbound call, and so on.

In step 9022, the computer device uses the target call experience tag, the call quality assessment data, and the call result analysis data as quality inspection data.

The computer equipment stores the target call experience label, the call quality evaluation data and the call result analysis data as quality inspection data, so that manual recording of quality inspection personnel is not needed, and the quality inspection efficiency is improved. In addition, the computer equipment can optimize the models such as a voice recognition model, an intention recognition model and the like used in the outbound process based on the stored quality inspection data, so that the outbound quality is improved.

In one embodiment, referring to fig. 13, based on the embodiment shown in fig. 2, this embodiment is directed to a quality inspection auditing process. As shown in fig. 13, step 203 further includes step 204:

in step 204, the computer device generates a quality inspection audit task according to the target quality inspection task and the quality inspection data.

In the embodiment of the application, the computer equipment can also be configured with a quality inspection auditing task for auditing the quality inspection data of the auditing personnel, and after the auditing is passed, the computer equipment can correspondingly store the quality inspection data and the outbound call data in the quality inspection database.

As one implementation mode, the computer equipment generates a quality inspection auditing task according to the target quality inspection task and the quality inspection data, and distributes the quality inspection auditing task to the quality inspection auditing account of the auditing personnel, so that the auditing personnel logs in the quality inspection auditing account of the auditing personnel, and the quality inspection auditing task can be triggered to conduct quality inspection auditing.

Illustratively, after the auditor triggers a quality inspection audit task (e.g., clicks on the quality inspection audit task), the computer device presents a quality inspection audit interface that includes at least outbound call data and corresponding quality inspection data. Thus, the auditor can check whether the quality inspection data aiming at the outbound call data is accurate or not.

Illustratively, referring to FIG. 14, FIG. 14 is a schematic diagram of an exemplary geological inspection audit interface. As shown in fig. 14, the outbound call data, the quality inspection data and the audit result input control may include, for example, a "reject and next pass" sub-control and a "pass and next pass" sub-control, and if the quality inspection data for the outbound call data displayed on the current quality inspection audit interface is accurate, an auditor may click on the "pass and next pass" sub-control, so that the computer device determines that the audit result of the quality inspection audit task is passed.

Correspondingly, step 203 includes:

in step 2031, if the computer device detects that the auditing result of the quality inspection auditing task is the target auditing result, the quality inspection data and the outbound call data are correspondingly stored in the quality inspection database.

The target audit result characterizes that the audit result of the quality inspection audit task is passing audit, namely, if the audit result of the quality inspection audit task is passing audit, the computer equipment correspondingly stores the quality inspection data and the outbound call data in the quality inspection database.

In one possible implementation, the audit result input control may further include an "end audit" sub-control, and after the auditor clicks the "end audit" sub-control, the computer device may count that relevant information of the quality inspection audit task within the preset time period is displayed on the task audit statistics interface, and the relevant information includes a total amount, throughput, reject amount and non-audit amount of the quality inspection task within the preset time period.

Referring to fig. 15, fig. 15 is a schematic diagram of an exemplary task audit statistics interface, on which an auditor can end audit in advance if part of data is audited and the whole data is known, so as to save audit time.

In one possible implementation, after the auditing is finished, if the amount of the overruling is equal to a preset threshold, for example, equal to zero, the computer device may mark the task state of the quality inspection auditing task as completed; if the reject amount is greater than a preset threshold, for example, greater than zero, the computer device marks the task state of the quality inspection auditing task as incomplete, and the target quality inspection task corresponding to the quality inspection auditing task is issued to the quality inspection account of the quality inspection personnel again for quality inspection again, so that the accuracy of quality inspection data can be improved.

In one embodiment, a call quality inspection method is provided, and the method is applied to the computer device 101 in fig. 1-a for illustration, and includes the following steps:

and a step a, if the computer equipment detects a trigger instruction aiming at a target quality inspection task, displaying an initial quality inspection interface, wherein the initial quality inspection interface comprises outbound call data.

The outbound call data comprise outbound call text, and the outbound call text is obtained by identifying outbound call voice corresponding to the target quality inspection task.

And b, if the computer equipment detects a selection instruction of the user for the outbound call data, displaying a statement marking control in the initial quality inspection interface so as to display the quality inspection interface.

The quality inspection interface at least comprises outbound call data corresponding to the target quality inspection task and a quality inspection annotation control, and the quality inspection annotation control comprises a statement annotation control.

And c, the computer equipment acquires quality inspection data aiming at the outbound call data based on the quality inspection marking control, wherein the quality inspection data is used for representing at least one of data quality and outbound call quality of the outbound call data.

The sentence annotation control comprises an annotation sub-control, the quality inspection data comprises text recognition accuracy quality inspection data, the computer equipment obtains the quality inspection data aiming at outbound call data based on the quality inspection annotation control, and the sentence annotation control comprises: if the computer equipment detects a selection instruction of a user for the labeling sub-control, a labeling sentence page is displayed; the computer device obtains text recognition accurate quality inspection data input by a user based on the annotation sentence page, wherein the text recognition accurate quality inspection data comprises at least one of voice recognition accurate quality inspection data and intention recognition accurate quality inspection data.

The sentence annotation control comprises a performance optimization sub-control, the quality inspection data comprises identification performance optimization data, the computer equipment obtains quality inspection data aiming at outbound call data based on the quality inspection annotation control, and the method comprises the following steps: if the computer equipment detects a selection instruction of a user aiming at the performance optimization sub-control, displaying a similar query supplementary page; the computer device obtains recognition performance optimization data input by the user based on the similarity query supplementary page, wherein the recognition performance optimization data comprises similar question text corresponding to the outbound call text.

The sentence annotation control further comprises a voice playing sub-control, and the voice playing sub-control is used for playing voice of the outbound call conversation.

The quality inspection annotation control comprises a sentence quality inspection control, the sentence quality inspection control comprises a plurality of different call experience label sub-controls, the computer equipment obtains quality inspection data aiming at outbound call data based on the quality inspection annotation control, and the quality inspection annotation control comprises: aiming at outbound call data, if a selection instruction of a user aiming at a target call experience label sub-control is detected, determining a target call experience label corresponding to the call experience label sub-control; and the computer equipment acquires quality inspection data according to the target call experience tag.

The quality inspection labeling control further comprises a call result quality inspection control, and the computer equipment acquires quality inspection data according to the target call experience label, and comprises: the computer equipment acquires call quality evaluation data corresponding to the outbound call data and call result analysis data based on the call result quality inspection control; the computer device takes the target call experience label, the call quality evaluation data and the call result analysis data as quality inspection data.

And d, the computer equipment generates a quality inspection auditing task according to the target quality inspection task and the quality inspection data.

And e, if the computer equipment detects that the auditing result of the quality inspection auditing task is the target auditing result, the quality inspection data and the outbound call data are correspondingly stored in a quality inspection database.

It should be understood that, although the steps in the above-described flowcharts are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described above may include a plurality of sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, and the order of execution of the sub-steps or stages is not necessarily sequential, but may be performed alternately or alternately with at least a part of the sub-steps or stages of other steps or other steps.

In one embodiment, as shown in fig. 16, there is provided a call quality inspection apparatus, including:

the display module 100 is configured to display a quality inspection interface if a trigger instruction for a target quality inspection task is detected, where the quality inspection interface includes at least outbound call data corresponding to the target quality inspection task and a quality inspection labeling control;

An obtaining module 200, configured to obtain quality inspection data for the outbound call data based on the quality inspection labeling control, where the quality inspection data is used to characterize at least one of data quality and outbound call quality of the outbound call data;

and the storage module 300 is configured to store the quality inspection data and the outbound call data in a quality inspection database correspondingly.

In one embodiment, the quality inspection annotation control comprises a sentence annotation control, and the presentation module 100 comprises:

In one embodiment, the sentence annotation control includes an annotation sub-control, the quality inspection data includes text recognition accuracy quality inspection data, and the obtaining module 200 includes:

In one embodiment, the statement labeling control includes a performance optimization sub-control, the quality inspection data includes identification performance optimization data, and the obtaining module 200 includes:

In one embodiment, the quality inspection annotation control includes a sentence quality inspection control, the sentence quality inspection control includes a plurality of different call experience label sub-controls, and the obtaining module 200 includes:

In one embodiment, the apparatus further comprises:

correspondingly, the storage module 300 is specifically configured to store the quality inspection data and the outbound call data in a quality inspection database correspondingly if it is detected that the auditing result of the quality inspection auditing task is a target auditing result.

The call quality inspection device provided in this embodiment may execute the above call quality inspection method embodiment, and its implementation principle and technical effects are similar, and will not be described herein again. The specific limitation of the call quality inspection device can be referred to the limitation of the call quality inspection method, and is not repeated herein. All or part of the modules in the call quality inspection device can be realized by software, hardware and a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.

In one embodiment, there is also provided a computer device as shown in fig. 17, which may be a terminal, and an internal structural diagram thereof may be as shown in fig. 17. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless mode can be realized through WIFI, an operator network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement a call quality inspection method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, can also be keys, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.

It will be appreciated by those skilled in the art that the structure shown in fig. 17 is merely a block diagram of a portion of the structure associated with the present application and is not limiting of the computer device to which the present application applies, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is provided comprising a memory and a processor, the memory having stored therein a computer program, the processor when executing the computer program performing the steps of:

In one embodiment, the quality inspection annotation control comprises a statement annotation control, and the processor when executing the computer program further performs the steps of:

In one embodiment, the sentence annotation control comprises an annotation sub-control, the quality inspection data comprises text recognition accuracy quality inspection data, and the processor when executing the computer program further performs the steps of:

In one embodiment, the statement annotation control comprises a performance optimization sub-control, the quality inspection data comprises identification performance optimization data, and the processor when executing the computer program further performs the steps of:

In one embodiment, the quality inspection annotation control comprises a sentence quality inspection control, the sentence quality inspection control comprises a plurality of different conversation experience label sub-controls, and the processor when executing the computer program further implements the steps of:

In one embodiment, the quality inspection labeling control further comprises a call result quality inspection control, and the processor further realizes the following steps when executing the computer program:

In one embodiment, the processor when executing the computer program further performs the steps of:

Those skilled in the art will appreciate that implementing all or part of the above-described methods may be accomplished by way of a computer program, which may be stored on a non-transitory computer readable storage medium and which, when executed, may comprise the steps of the above-described embodiments of the methods. Any reference to memory, storage, database, or other medium used in the various embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus (Ramb microsecond) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.

In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of:

In one embodiment, the quality inspection annotation control comprises a statement annotation control, which when executed by the processor further performs the steps of:

In one embodiment, the sentence annotation control comprises an annotation sub-control, the quality inspection data comprises text recognition accurate quality inspection data, and the computer program when executed by the processor further performs the steps of:

In one embodiment, the statement annotation control comprises a performance optimization sub-control, the quality inspection data comprises identification performance optimization data, and the computer program when executed by the processor further performs the steps of:

In one embodiment, the quality inspection annotation control comprises a sentence quality inspection control comprising a plurality of different conversation experience label sub-controls, the computer program when executed by the processor further implementing the steps of:

In one embodiment, the quality control further comprises a call result quality control, and the computer program when executed by the processor further implements the steps of:

In one embodiment, the computer program when executed by the processor further performs the steps of:

correspondingly, the storing the quality inspection data and the outbound call data in the quality inspection database includes: and if the auditing result of the quality inspection auditing task is detected to be a target auditing result, correspondingly storing the quality inspection data and the outbound call data in a quality inspection database.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The foregoing examples illustrate only a few embodiments of the invention, which are described in detail and are not to be construed as limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention. Accordingly, the scope of protection of the present invention is to be determined by the appended claims.

Claims

1. A method for quality testing of a call, the method comprising:

the quality inspection data and the outbound call data are correspondingly stored in a quality inspection database;

the outbound call data comprises outbound call text, the outbound call text is obtained by identifying outbound call voice corresponding to the target quality inspection task, the quality inspection annotation control comprises a statement annotation control, the statement annotation control comprises a performance optimization sub-control, the quality inspection data comprises identification performance optimization data, and the quality inspection data aiming at the outbound call data is obtained based on the quality inspection annotation control and comprises the following steps:

2. The method according to claim 1, wherein the presenting the quality inspection interface if the trigger instruction for the target quality inspection task is detected comprises:

3. The method of claim 1, wherein the sentence annotation control comprises an annotation sub-control, the quality inspection data comprises text recognition accurate quality inspection data, the acquiring quality inspection data for the outgoing call data based on the quality inspection annotation control comprises:

4. The method of claim 1 or 3, wherein the sentence annotation control further comprises a voice play sub-control for voice playing the outbound call voice.

5. The method of claim 1, wherein the quality control labeling control comprises a sentence quality control comprising a plurality of different call experience label sub-controls, wherein the obtaining quality control data for the outgoing call data based on the quality control labeling control comprises:

6. The method of claim 5, wherein the quality control labeling control further comprises a call result quality control, the obtaining the quality control data according to the target call experience tag comprising:

7. The method of claim 1, wherein before storing the quality inspection data in the quality inspection database in correspondence with the outgoing call session data, further comprises:

8. A call quality inspection device, the device comprising:

The storage module is used for storing the quality inspection data and the outbound call data in a quality inspection database correspondingly;

the outbound call data comprise outbound call text, the outbound call text is obtained by identifying outbound call voice corresponding to the target quality inspection task, the quality inspection annotation control comprises a statement annotation control, the statement annotation control comprises a performance optimization sub-control, the quality inspection data comprise identification performance optimization data, and the acquisition module comprises:

9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 7 when the computer program is executed.

10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 7.