CN113630505B - AI-based call center management platform - Google Patents
AI-based call center management platform Download PDFInfo
- Publication number
- CN113630505B CN113630505B CN202111069737.5A CN202111069737A CN113630505B CN 113630505 B CN113630505 B CN 113630505B CN 202111069737 A CN202111069737 A CN 202111069737A CN 113630505 B CN113630505 B CN 113630505B
- Authority
- CN
- China
- Prior art keywords
- answering
- call
- port
- target
- client
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000605 extraction Methods 0.000 claims abstract description 14
- 238000011156 evaluation Methods 0.000 claims description 36
- 238000000034 method Methods 0.000 claims description 24
- 230000011218 segmentation Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 2
- 230000036651 mood Effects 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M3/00—Automatic or semi-automatic exchanges
- H04M3/22—Arrangements for supervision, monitoring or testing
- H04M3/2218—Call detail recording
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M3/00—Automatic or semi-automatic exchanges
- H04M3/42—Systems providing special services or facilities to subscribers
- H04M3/50—Centralised arrangements for answering calls; Centralised arrangements for recording messages for absent or busy subscribers ; Centralised arrangements for recording messages
- H04M3/527—Centralised call answering arrangements not requiring operator intervention
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Quality & Reliability (AREA)
- Telephonic Communication Services (AREA)
Abstract
The invention discloses an AI-based call center management platform, which comprises a target AI answering port determination module, an answering call record extraction module, an answering call voice matching degree acquisition module, an answering call timeliness acquisition module, an answering call relevance acquisition module, a call ending service grading acquisition module, a modeling processing module, a management database, a management server and a display terminal, wherein the answering call voice parameter matching degree, the answering call timeliness and the answering call relevance corresponding to the target AI answering port are acquired, and the answering service grading of a client to the target AI answering port is acquired after the answering call is ended, so that the comprehensive answering service quality coefficient of the target AI answering port to the current call client is comprehensively estimated, the comprehensive estimation of the AI answering service quality is realized, and the estimation basis combines the subjective answering service grading of the client and the objective AI answering call parameters, thereby effectively improving the reliability of the estimation result.
Description
Technical Field
The invention belongs to the technical field of AI call center management, and particularly relates to an AI-based call center management platform.
Background
The call center is an important component of a modern enterprise, is a bridge for the enterprise to communicate with clients as a portal of the enterprise, is a foundation for the enterprise to strengthen the loyalty of clients and manage the relationship of the clients, and is increasingly valued by the enterprise in the social environment of the modern market economy with service guidance. With the comprehensive arrival of the AI age, the disadvantages of too high management cost and uneven call service quality, which are exhibited by the traditional call center management mode of manually calling, are more and more obvious. In this case, the AI call center gradually replaces the traditional manual call by virtue of its low cost and high efficiency, and becomes the most popular call management mode for enterprises at present.
For the counsel business, the counsel call received every day is counted, and the AI call center is required to answer in real time. However, AI answering is not a manual answering after all, and because of the limitation of the answering service control program, the flexibility of the answering service is poor, and when a call consultation is encountered or the answering service control program is disordered in the specific answering process, poor answering service quality is caused, so that the answering service quality of the AI call center needs to be evaluated.
However, at present, the evaluation of the answering service quality of the AI call center is mostly evaluated only according to the grading of the customer to the AI answering service, and the evaluation adopted by the evaluation mode is too strong in subjectivity, ignores the answering and answering objective parameters of the AI answering port in the whole answering and talking process, such as answering and answering voice matching degree, answering and answering timeliness, answering and answering relevance, and the like, and can truly and objectively reflect the answering and answering service quality of the AI answering port in the whole answering and talking process, so that the evaluation of the answering service quality is simply performed according to the grading of the customer to the AI answering service, and the reliability of the evaluation result is easy to be low.
Disclosure of Invention
The technical task of the invention is to provide an AI-based call center management platform which integrates subjective answering service scores and AI answering call objective parameters of clients as evaluation basis aiming at the problems in the background art, and can effectively solve the problems in the background art.
The invention is realized by the following technical scheme:
The call center management platform based on the AI comprises a target AI answering port determining module, an answering call record extracting module, an answering call voice matching degree collecting module, an answering call timeliness collecting module, an answering call relevance collecting module, a call ending service scoring collecting module, a modeling processing module, a management database, a management server and a display terminal;
the target AI answering port determining module is used for determining a target AI answering port corresponding to a current calling client when receiving a client call;
The answering call record extraction module is used for extracting the whole answering call record of the current call client from the target AI answering port when the current call client is connected;
The answering call voice matching degree acquisition module is used for processing the extracted answering call records and acquiring the call voice parameter matching degree corresponding to the target AI answering port, wherein the answering call voice matching degree acquisition module comprises a voice parameter extraction unit and a call voice parameter matching degree analysis unit of both parties;
The answering call timeliness acquisition module is used for processing the extracted answering call records and acquiring the answering call timeliness of the target AI answering port for replying the questions of the customer;
The answering call relevance acquisition module is used for processing the extracted answering call records and acquiring the answering call relevance of the target AI answering port reply customer problem;
The call ending service score acquisition module is used for acquiring answering service scores of clients to the target AI answering ports after answering the call;
The modeling processing module is used for evaluating the comprehensive answering service quality coefficient of the target AI answering port to the current calling client by combining the answering voice parameter matching degree corresponding to the target AI answering port, the answering time of the target AI answering port for replying to the client to raise the problem, the answering association degree of the target AI answering port for replying to the client to raise the problem and the answering service score of the client to the target AI answering port;
The management server is used for numbering all the AI answering ports in the AI call center, counting the total number of answering customer calls corresponding to all the AI answering ports in a set time period, and further evaluating the comprehensive answering service quality coefficient corresponding to all the answering customer calls of all the AI answering ports in the set time period according to the evaluation method of the comprehensive answering service quality coefficient of the current calling customer by the target AI answering ports, so that the AI answering ports with poor answering service quality in the set time period are identified according to the evaluation result;
the display terminal is used for displaying the number corresponding to the AI answering port with poor answering service quality.
More optimally, the specific method for determining the target AI answering port corresponding to the current calling client is to acquire the current answering state corresponding to each AI answering port in the AI call center, further screen the AI answering port in the unreceived state from the current answering state, determine the AI answering port as the AI answering port corresponding to the current calling client, and record the AI answering port as the target AI answering port.
More preferably, the speech parameters include speech speed and pitch.
More optimally, the two-party voice parameter extraction unit is used for respectively extracting the speaking voice of the client and the return voice of the target AI receiving port from the extracted receiving call records, further extracting the speaking voice parameters of the client from the speaking voice of the client and extracting the return voice parameters from the return voice of the target AI receiving port.
More optimally, the matching degree analysis unit of the return voice parameters is used for carrying out matching degree analysis on the return voice parameters of the target AI answering port, and the specific analysis process comprises the following steps:
s1, comparing speaking voice parameters of a client with AI answering port suitable call voice parameters corresponding to various speaking voice parameters of the client preset in a management database to obtain AI answering port suitable call voice parameters corresponding to the speaking voice parameters of the client;
S2, matching the return voice parameter of the target AI answering port with the appropriate return voice parameter of the AI answering port corresponding to the client speaking voice parameter, and counting the matching degree of the return voice parameter corresponding to the target AI answering port Sigma is expressed as the matching degree of the return voice parameters corresponding to the target AI receiving port, v and u are respectively expressed as the return voice speed and the return voice tone of the target AI receiving port, and v 0、u0 is respectively expressed as the suitable return voice speed and the suitable return voice tone of the AI receiving port corresponding to the client speaking voice parameters.
More optimally, the specific collection process corresponding to the answer time of the answer call of the target AI answer port reply customer to the question by the answer call time collection module is as follows:
the method comprises the steps of H1, counting the number of questions presented by a client from an extracted answering call record, numbering the questions presented by the client according to the sequence of the presentation time, and marking the questions as 1, 2;
H2, acquiring a proposal ending time point corresponding to each question from the extracted answering call record and a call-back time point for replying each question by the target AI answering port;
H3, subtracting the call-back time point of each question which is presented by the client from the call-back time point of each question which is replied by the target AI answering port to obtain the call-back time length of each question which is replied by the target AI answering port, and recording as t k;
H4, comparing the call-back time length of each question replied by the target AI answering port with the predefined call-back time length, and counting the call-back time length of the target AI answering port replying the questions presented by the customer Η represents the time of return call for the target AI listening port to answer the question from the customer, and t 0 represents a predefined time period for return call.
More optimally, the specific collection process corresponding to the answer call relevance of the answer call client of the target AI answer port is collected by the answer call relevance collection module as follows:
d1, intercepting problem voice information of each problem proposed by a client from the extracted answering call records;
and D2, extracting a problem keyword from the intercepted problem voice information, wherein the specific extraction method comprises the following steps of:
d21, carrying out problem text recognition on the problem voice information of each problem;
d22, extracting question keywords from the question texts corresponding to the questions;
d3, intercepting the call-back voice information of each problem presented by the target AI answering port to the client from the extracted answering voice call record;
D4, extracting the return key words of the intercepted return voice information to obtain return key words corresponding to the problems extracted by the customer replied by the target AI answering port;
D5, matching the return key words corresponding to the problems presented by the target AI answering port replying clients with the problem key words corresponding to the problems, if the return key words corresponding to the problems presented by the target AI answering port replying clients are successfully matched with the problem key words corresponding to the problems, marking the return associated index corresponding to the problems as epsilon, otherwise marking the return associated index corresponding to the problems as epsilon';
d6, counting the call relevance of the questions presented by the target AI answering port replying clients according to the call relevance index corresponding to the questions presented by the target AI answering port replying clients Xi represents the relevance of the return call of the problem presented by the target AI answering port reply client, lambda k represents the relevance index of the return call of the k problem presented by the target AI answering port reply client, and lambda k can be epsilon or epsilon'.
More optimally, the evaluation and calculation formula of the target AI answering port to the comprehensive answering service quality coefficient of the current calling client is as follows The comprehensive answering service quality coefficient of the target AI answering port to the current call client is expressed, q is expressed as an answering service grading value of the client to the target AI answering port, q 0 is expressed as a highest grading value of the call service, and alpha 1, alpha 2, alpha 3 and alpha 4 are respectively expressed as an evaluation weight proportionality coefficient of answering call voice parameter matching degree, answering call timeliness degree, answering call association degree and answering service grading to the comprehensive answering service quality coefficient, and alpha 1+ alpha 2+ alpha 3+ alpha 4 = 1.
More preferably, the magnitude relation of the alpha 1, the alpha 2, the alpha 3 and the alpha 4 is alpha 3> alpha 2> alpha 4> alpha 1.
More preferably, the specific identification process for identifying the AI answering port with poor received service quality in the set time period according to the evaluation result includes the following steps:
F1, comparing the comprehensive answering service quality coefficient corresponding to each time of answering the client call by each AI answering port in a set time period with the standard comprehensive answering service quality coefficient set in the management database, and counting the number of times of answering the client call by each AI answering port in the set time period, which is smaller than the standard comprehensive answering service quality coefficient;
F2, dividing the total number of client calls received by each AI receiving port in a set time period by the number of received client calls, in which the AI receiving port is smaller than the set standard comprehensive receiving service quality coefficient in the set time period, to obtain a corresponding poor receiving service duty ratio coefficient of each AI receiving port in the set time period;
and F3, comparing the corresponding answer service bad duty ratio coefficient of each AI answer port in the set time period with a set value, and if the corresponding answer service bad duty ratio coefficient of one AI answer port in the set time period is larger than the set value, marking the AI answer port as the answer service quality bad AI answer port.
The beneficial effects of the invention are as follows:
(1) According to the invention, when a client call is received, the target AI answering port corresponding to the current call client is determined, the whole answering call record of the current call client is answered by the target AI answering port, the extracted answering call record is processed, the matching degree of the return voice parameters corresponding to the target AI answering port, the return timeliness of the problems posed by the target AI answering port and the return relevance of the problems posed by the target AI answering port are acquired, meanwhile, the answering service score of the target AI answering port is acquired after the answering call is ended, so that the comprehensive answering service quality coefficient of the target AI answering port to the current call client is comprehensively evaluated, the comprehensive evaluation of the AI answering service quality is realized, the evaluation is realized according to the combination of the subjective answering service score and the objective AI answering call parameters, and the evaluation of the current evaluation mode of the AI call center is avoided by evaluating the subjective answering service score of the client alone, so that the reliability of the evaluation result is improved.
(2) According to the invention, the comprehensive answering service quality coefficient corresponding to each AI answering port in the AI call center answering the client call in the set time period is obtained by adopting the evaluation method of the target AI answering port to the comprehensive answering service quality coefficient of the current call client, so that the AI answering port with poor answering service quality in the set time period is identified according to the evaluation result, the identification mode of the AI answering port with poor answering service quality is more practical and has higher rationality, the false identification phenomenon caused by the fact that the AI answering port is simply answering the client call once is avoided, and the accuracy of the identification of the AI answering port with poor answering service quality is improved.
Drawings
The invention will be further described with reference to the accompanying drawings, in which embodiments do not constitute any limitation of the invention, and other drawings can be obtained by one of ordinary skill in the art without inventive effort from the following drawings.
FIG. 1 is a schematic diagram of a modular connection according to the present invention;
fig. 2 is a connection schematic diagram of the voice matching degree collection module for answering the call in the invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, an AI-based call center management platform includes a target AI answer port determining module, an answer call record extracting module, an answer call voice matching degree collecting module, an answer call timeliness collecting module, an answer call relevance collecting module, a call ending service scoring collecting module, a modeling processing module, a management database, a management server and a display terminal.
The target AI answering port determining module is connected with the answering call record extracting module, the answering call record extracting module is respectively connected with the answering call voice matching degree collecting module, the answering call timeliness collecting module and the answering call relevance collecting module, the answering call voice matching degree collecting module, the answering call timeliness collecting module, the answering call relevance collecting module and the call ending service scoring collecting module are all connected with the modeling processing module, the modeling processing module is connected with the management server, and the management server is connected with the display terminal.
The target AI answering port determining module is used for determining a target AI answering port corresponding to a current calling client when receiving a client call, and the specific determining method is to acquire the current answering state corresponding to each AI answering port in the AI call center, screen the AI answering port in the unreceived state from the current answering state, determine the AI answering port as the AI answering port corresponding to the current calling client, and mark the AI answering port as the target AI answering port.
The answering call record extraction module is used for extracting the whole answering call record of the current call client from the target AI answering port when the current client call is completed.
Referring to fig. 2, the answer-back voice matching degree collection module is configured to process the extracted answer call record, collect therefrom the matching degree of the answer-back voice parameter corresponding to the target AI answer port, where the voice parameter includes a speech speed and a tone, and the answer-back voice matching degree collection module includes a two-party voice parameter extraction unit and an answer-back voice parameter matching degree analysis unit.
The double-party voice parameter extraction unit is used for respectively extracting the speaking voice of the client and the return voice of the target AI receiving port from the extracted receiving call records, further extracting the speaking voice parameters of the client from the speaking voice of the client and extracting the return voice parameters from the return voice of the target AI receiving port.
The return voice parameter matching degree analysis unit is used for carrying out matching degree analysis on the return voice parameters of the target AI answering port, and the specific analysis process comprises the following steps:
s1, comparing speaking voice parameters of a client with AI answering port suitable call voice parameters corresponding to various speaking voice parameters of the client preset in a management database to obtain AI answering port suitable call voice parameters corresponding to the speaking voice parameters of the client;
S2, matching the return voice parameter of the target AI answering port with the appropriate return voice parameter of the AI answering port corresponding to the client speaking voice parameter, and counting the matching degree of the return voice parameter corresponding to the target AI answering port Sigma is expressed as the matching degree of the return voice parameters corresponding to the target AI receiving port, v and u are respectively expressed as the return voice speed and the return voice tone of the target AI receiving port, v 0、u0 is respectively expressed as the suitable return voice speed and the suitable return voice tone of the AI receiving port corresponding to the client speaking voice parameters, wherein the closer the return voice parameters of the target AI receiving port are to the suitable return voice parameters of the AI receiving port corresponding to the client speaking voice parameters, the greater the matching degree of the return voice parameters is.
In the embodiment, the collection of the matching degree of the return voice parameters corresponding to the target AI answering port intuitively reflects the matching degree of the return voice parameters and the client speaking voice parameters in the answering call record, and provides the correlation coefficient of the return voice parameter matching for the subsequent evaluation of the comprehensive answering service quality coefficient.
The answering call time acquisition module is used for processing the extracted answering call records, and acquiring the answering call time of the target AI answering port for replying the problem of the customer, wherein the specific acquisition process is as follows:
the method comprises the steps of H1, counting the number of questions presented by a client from an extracted answering call record, numbering the questions presented by the client according to the sequence of the presentation time, and marking the questions as 1, 2;
H2, acquiring a proposal ending time point corresponding to each question from the extracted answering call record and a call-back time point for replying each question by the target AI answering port;
H3, subtracting the call-back time point of each question which is presented by the client from the call-back time point of each question which is replied by the target AI answering port to obtain the call-back time length of each question which is replied by the target AI answering port, and recording as t k;
H4, comparing the call-back time length of each question replied by the target AI answering port with the predefined call-back time length, and counting the call-back time length of the target AI answering port replying the questions presented by the customer Η represents the time of the target AI answering port to answer the call of the question posed by the client, t 0 represents the predefined time of the call, wherein the time of the target AI answering port to answer the call of each question is closer to the predefined time of the call, and the time of the call is larger.
In this embodiment, the collection of the timeliness of the answering of the questions of the target AI answering port corresponding to the answering client intuitively reflects the timeliness of the answering of the questions of the target AI answering port in the answering call record, and provides a correlation coefficient of the timeliness of the answers for the subsequent evaluation of the comprehensive answering service quality coefficient.
The answering call relevance acquisition module is used for processing the extracted answering call records, and acquiring the answering call relevance of the target AI answering port reply customer problem, and the specific acquisition process is as follows:
d1, intercepting problem voice information of each problem proposed by a client from the extracted answering call records;
and D2, extracting a problem keyword from the intercepted problem voice information, wherein the specific extraction method comprises the following steps of:
d21, carrying out problem text recognition on the problem voice information of each problem;
D22, performing stop word removal and word segmentation processing on the problem text corresponding to each problem, and extracting problem keywords from the obtained word segmentation;
d3, intercepting the call-back voice information of each problem presented by the target AI answering port to the client from the extracted answering voice call record;
and D4, extracting the call key words of the intercepted call voice information, wherein the specific extraction method is as follows:
D41, performing return text recognition on return voice information of each problem;
D42, performing stop word removal and word segmentation processing on the call text corresponding to each problem, and further extracting call keywords from the obtained word segmentation to obtain call keywords corresponding to each problem extracted by a target AI answering port reply client;
D5, matching the return key words corresponding to the problems presented by the target AI answering port replying clients with the problem key words corresponding to the problems, if the target AI answering port replying clients present that the return key words corresponding to certain problems are successfully matched with the problem key words corresponding to the problems, marking the return associated index corresponding to the problems as epsilon, otherwise marking the return associated index corresponding to the problems as epsilon ', wherein epsilon > epsilon';
d6, counting the call relevance of the questions presented by the target AI answering port replying clients according to the call relevance index corresponding to the questions presented by the target AI answering port replying clients Xi represents the relevance of the return call of the problem presented by the target AI answering port reply client, lambda k represents the relevance index of the return call of the k problem presented by the target AI answering port reply client, and lambda k can be epsilon or epsilon'.
In this embodiment, the collection of the relevance of the answer to the question and the return call of the reply client corresponding to the target AI answer port intuitively reflects the relevance of the answer to the question and the return call of the reply client in the answer call record, and provides the relevance coefficient of the return call relevance for the subsequent evaluation of the comprehensive answer service quality coefficient.
In the embodiment, the matching degree, the time degree and the relevance degree of the return call are collected from the whole call answering record of the current call client as the AI call answering objective parameters, because the matching degree of the return call voice parameters can reflect the experience of the client on the return call speed and the tone of the target AI call port, the better the matching degree is, the better the client receives the return call voice of the target AI call port, the time degree of the return call can reflect the time degree of the return call of the target AI call port for answering the client, the better the call mood of the client is, the more the call mood of the client is influenced, and the more the relevance degree of the return call of the target AI call port for answering the call of the client is, so that the more the client can obtain accurate response, thereby the influence on the output objectivity of the return call quality of the target AI call port is reflected by the matching degree of the return call voice parameters, the time degree of the return call and the relevance degree of the return call.
And the call ending service score acquisition module is used for acquiring the answering service score of the client to the target AI answering port after answering the call.
The management database is used for storing the suitable call-back voice parameters of the AI answering ports corresponding to the various speaking voice parameters of the clients, storing standard comprehensive answering service quality coefficients, storing the evaluation weight proportion coefficients of the answering voice parameter matching degree, the answering time degree, the answering call relevance degree and the answering service score to the comprehensive answering service quality coefficients, and storing the setting value of the answering service poor proportion coefficient.
The modeling processing module is used for evaluating the comprehensive answering service quality coefficient of the target AI answering port to the current call client by combining the matching degree of the call voice parameters corresponding to the target AI answering port, the call time of the target AI answering port for replying to the problem of the client, the call relevance of the target AI answering port for replying to the problem of the client and the answering service score of the client to the target AI answering port The comprehensive answering service quality coefficient of the target AI answering port to the current call client is expressed, q is expressed as the answering service grading value of the client to the target AI answering port, q 0 is expressed as the highest grading value of the answering service, and alpha 1, alpha 2, alpha 3 and alpha 4 are respectively expressed as the evaluation weight proportionality coefficient of the answering call voice parameter matching degree, the answering call timeliness degree, the answering call association degree and the answering service grading to the comprehensive answering service quality coefficient, and the corresponding size relationship of alpha 1+ alpha 2+ alpha 3+ alpha 4 = 1, alpha 2, alpha 3 and alpha 4 is alpha 3> alpha 2> alpha 4> alpha 1.
According to the method, when a client call is received, the target AI answering port corresponding to the current call client is determined, the whole answering call record of the current call client is answered by the target AI answering port, the extracted answering call record is processed, the matching degree of the return voice parameters corresponding to the target AI answering port, the return timeliness of the problems posed by the target AI answering port and the return relevance of the problems posed by the target AI answering port are collected, meanwhile, the answering service score of the target AI answering port is collected after the answering call is ended, and therefore the comprehensive answering service quality coefficient of the target AI answering port to the current call client is comprehensively estimated, comprehensive evaluation of AI answering service quality is achieved, the evaluation is conducted according to the subjective answering service score and the objective AI answering call parameters of the client, and the fact that the current evaluation mode of the quality of answering service of the AI call center is simply evaluated according to the subjective answering service score of the client is avoided is achieved, and reliability of an evaluation result is improved.
The management server is used for numbering all the AI answering ports in the AI call center, counting the total number of answering customer calls corresponding to all the AI answering ports in a set time period, further evaluating the comprehensive answering service quality coefficient corresponding to all the answering customer calls of all the AI answering ports in the set time period according to the evaluation method of the comprehensive answering service quality coefficient of the current calling customer by the target AI answering ports, and further identifying the AI answering ports with poor answering service quality in the set time period according to the evaluation result, wherein the specific identification process comprises the following steps:
F1, comparing the comprehensive answering service quality coefficient corresponding to each answering client call in the set time period with the set standard comprehensive answering service quality coefficient, and counting the answering client call times of each AI answering port smaller than the set standard comprehensive answering service quality coefficient in the set time period;
F2, dividing the total number of client calls received by each AI receiving port in a set time period by the number of received client calls, in which the AI receiving port is smaller than the set standard comprehensive receiving service quality coefficient in the set time period, to obtain a corresponding poor receiving service duty ratio coefficient of each AI receiving port in the set time period;
and F3, comparing the corresponding answer service bad duty ratio coefficient of each AI answer port in the set time period with a set value, and if the corresponding answer service bad duty ratio coefficient of one AI answer port in the set time period is larger than the set value, marking the AI answer port as the answer service quality bad AI answer port.
According to the method, the comprehensive answering service quality coefficient corresponding to each AI answering port in the AI call center answering the client call in the set time period is obtained through the evaluation method of the target AI answering port to the comprehensive answering service quality coefficient of the current call client, so that the AI answering port with poor answering service quality in the set time period is identified according to the evaluation result, the identification mode of the AI answering port with poor answering service quality is more practical and has higher rationality, the false identification phenomenon caused by the fact that whether the AI answering port is the AI answering port with poor answering service quality is avoided by simply answering the comprehensive answering service quality coefficient of the client call once by the AI answering port is avoided, and therefore the accuracy of the identification of the AI answering port with poor answering service quality is improved.
The display terminal is used for displaying the number corresponding to the AI answering port with poor answering service quality.
The foregoing is merely illustrative of the structures of this invention and various modifications, additions and substitutions for those skilled in the art can be made to the described embodiments without departing from the scope of the invention or from the scope of the invention as defined in the accompanying claims.
Claims (10)
1. An AI-based call center management platform, characterized by: the system comprises a target AI answering port determining module, an answering call record extracting module, an answering call voice matching degree collecting module, an answering call timeliness collecting module, an answering call relevance collecting module, a call ending service scoring collecting module, a modeling processing module, a management database, a management server and a display terminal;
the target AI answering port determining module is used for determining a target AI answering port corresponding to a current calling client when receiving a client call;
The answering call record extraction module is used for extracting the whole answering call record of the current call client from the target AI answering port when the current call client is connected;
The answering call voice matching degree acquisition module is used for processing the extracted answering call records and acquiring the call voice parameter matching degree corresponding to the target AI answering port, wherein the answering call voice matching degree acquisition module comprises a voice parameter extraction unit and a call voice parameter matching degree analysis unit of both parties;
The answering call timeliness acquisition module is used for processing the extracted answering call records and acquiring the answering call timeliness of the target AI answering port for replying the questions of the customer;
The answering call relevance acquisition module is used for processing the extracted answering call records and acquiring the answering call relevance of the target AI answering port reply customer problem;
The call ending service score acquisition module is used for acquiring answering service scores of clients to the target AI answering ports after answering the call;
The modeling processing module is used for evaluating the comprehensive answering service quality coefficient of the target AI answering port to the current calling client by combining the answering voice parameter matching degree corresponding to the target AI answering port, the answering time of the target AI answering port for replying to the client to raise the problem, the answering association degree of the target AI answering port for replying to the client to raise the problem and the answering service score of the client to the target AI answering port;
The management server is used for numbering all the AI answering ports in the AI call center, counting the total number of answering customer calls corresponding to all the AI answering ports in a set time period, and further evaluating the comprehensive answering service quality coefficient corresponding to all the answering customer calls of all the AI answering ports in the set time period according to the evaluation method of the comprehensive answering service quality coefficient of the current calling customer by the target AI answering ports, so that the AI answering ports with poor answering service quality in the set time period are identified according to the evaluation result;
the display terminal is used for displaying the number corresponding to the AI answering port with poor answering service quality.
2. The AI-based call center management platform of claim 1, wherein: the specific method for determining the target AI answering port corresponding to the current calling client is to acquire the current answering state corresponding to each AI answering port in the AI call center, screen the AI answering port in the unreceived state, determine the AI answering port as the AI answering port corresponding to the current calling client, and record the AI answering port as the target AI answering port.
3. The AI-based call center management platform of claim 1, wherein: the speech parameters include speech speed and pitch.
4. The AI-based call center management platform of claim 1, wherein: the two-party voice parameter extraction unit is used for respectively extracting the speaking voice of the client and the return voice of the target AI receiving port from the extracted receiving call records, further extracting the speaking voice parameters of the client from the speaking voice of the client and extracting the return voice parameters from the return voice of the target AI receiving port.
5. The AI-based call center management platform of claim 1, wherein: the return voice parameter matching degree analysis unit is used for carrying out matching degree analysis on the return voice parameters of the target AI answering port, and the specific analysis process comprises the following steps:
s1, comparing speaking voice parameters of a client with AI answering port suitable call voice parameters corresponding to various speaking voice parameters of the client preset in a management database to obtain AI answering port suitable call voice parameters corresponding to the speaking voice parameters of the client;
S2, matching the return voice parameters of the target AI answering port with the appropriate return voice parameters of the AI answering port corresponding to the client speaking voice parameters, wherein the statistical return voice parameter matching degree sigma corresponding to the target AI answering port is expressed as the return voice parameter matching degree corresponding to the target AI answering port, v and u are respectively expressed as the return voice speed and the return voice tone of the target AI answering port, and v 0 and u 0 are respectively expressed as the appropriate return voice speed and the appropriate return voice tone of the AI answering port corresponding to the client speaking voice parameters.
6. The AI-based call center management platform of claim 1, wherein: the specific collection process corresponding to the answer call timeliness of the target AI answer port reply customer question is as follows:
the method comprises the steps of H1, counting the number of questions presented by a client from an extracted answering call record, numbering the questions presented by the client according to the sequence of the presentation time, and marking the questions as 1, 2;
H2, acquiring a proposal ending time point corresponding to each question from the extracted answering call record and a call-back time point for replying each question by the target AI answering port;
H3, subtracting the call-back time point of each question which is presented by the client from the call-back time point of each question which is replied by the target AI answering port to obtain the call-back time length of each question which is replied by the target AI answering port, and marking as t k;
and H4, comparing the call-back time length of the target AI answering port for replying each problem with the predefined call-back time length, wherein the call-back time eta for counting the problems presented by the target AI answering port replying clients is represented as the call-back time length of the target AI answering port replying the problems presented by the clients, and t0 is represented as the predefined call-back time length.
7. The AI-based call center management platform of claim 1, wherein: the specific collection process corresponding to the answer call relevance of the answer call relevance collection module for collecting the answer questions of the target AI answer port reply clients is as follows:
d1, intercepting problem voice information of each problem proposed by a client from the extracted answering call records;
and D2, extracting a problem keyword from the intercepted problem voice information, wherein the specific extraction method comprises the following steps of:
d21, carrying out problem text recognition on the problem voice information of each problem;
d22, extracting question keywords from the question texts corresponding to the questions;
d3, intercepting the call-back voice information of each problem presented by the target AI answering port to the client from the extracted answering voice call record;
D4, extracting the return key words of the intercepted return voice information to obtain return key words corresponding to the problems extracted by the customer replied by the target AI answering port;
D5, matching the return key words corresponding to the problems presented by the target AI answering port replying clients with the problem key words corresponding to the problems, if the return key words corresponding to the problems presented by the target AI answering port replying clients are successfully matched with the problem key words corresponding to the problems, marking the return associated index corresponding to the problems as epsilon, otherwise marking the return associated index corresponding to the problems as epsilon';
and D6, counting the call relevance xi of the questions presented by the target AI answering port replying client according to the call relevance index corresponding to each question presented by the target AI answering port replying client, wherein the call relevance xi is expressed as the call relevance of the questions presented by the target AI answering port replying client, λk is expressed as the call relevance index corresponding to the kth question presented by the target AI answering port replying client, and the value of λk is epsilon or epsilon'.
8. The AI-based call center management platform of claim 1, wherein: the evaluation calculation formula of the target AI answering port to the comprehensive answering service quality coefficient of the current calling client is expressed as the comprehensive answering service quality coefficient of the target AI answering port to the current calling client, q is expressed as the answering service grading value of the client to the target AI answering port, q 0 is expressed as the highest grading value of answering service, and alpha 1, alpha 2, alpha 3 and alpha 4 are respectively expressed as the evaluation weight proportionality coefficient of answering call voice parameter matching degree, answering call timeliness, answering call relevance degree and answering service grading to the comprehensive answering service quality coefficient, and alpha 1+ alpha 2+ alpha 3+ alpha 4 = 1.
9. The AI-based call center management platform of claim 8, wherein: the corresponding magnitude relation of the alpha 1, the alpha 2, the alpha 3 and the alpha 4 is alpha 3> alpha 2> alpha 4> alpha 1.
10. The AI-based call center management platform of claim 1, wherein: the specific identification process for identifying the AI answering port with poor received service quality in the set time period according to the evaluation result comprises the following steps:
F1, comparing the comprehensive answering service quality coefficient corresponding to each time of answering the client call by each AI answering port in a set time period with the standard comprehensive answering service quality coefficient set in the management database, and counting the number of times of answering the client call by each AI answering port in the set time period, which is smaller than the standard comprehensive answering service quality coefficient;
F2, dividing the total number of client calls received by each AI receiving port in a set time period by the number of received client calls, in which the AI receiving port is smaller than the set standard comprehensive receiving service quality coefficient in the set time period, to obtain a corresponding poor receiving service duty ratio coefficient of each AI receiving port in the set time period;
and F3, comparing the corresponding answer service bad duty ratio coefficient of each AI answer port in the set time period with a set value, and if the corresponding answer service bad duty ratio coefficient of one AI answer port in the set time period is larger than the set value, marking the AI answer port as the answer service quality bad AI answer port.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111069737.5A CN113630505B (en) | 2021-09-13 | 2021-09-13 | AI-based call center management platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111069737.5A CN113630505B (en) | 2021-09-13 | 2021-09-13 | AI-based call center management platform |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113630505A CN113630505A (en) | 2021-11-09 |
| CN113630505B true CN113630505B (en) | 2024-06-04 |
Family
ID=78389829
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111069737.5A Active CN113630505B (en) | 2021-09-13 | 2021-09-13 | AI-based call center management platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113630505B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108965620A (en) * | 2018-08-24 | 2018-12-07 | 杭州数心网络科技有限公司 | A kind of artificial intelligence call center system |
| CN111294468A (en) * | 2020-02-07 | 2020-06-16 | 普强时代(珠海横琴)信息技术有限公司 | Tone quality detection and analysis system for customer service center calling |
| CN112217934A (en) * | 2019-07-09 | 2021-01-12 | 杭州巧卡科技有限公司 | Voice recognition processing method and system in real-time communication system |
| CN112885376A (en) * | 2021-01-23 | 2021-06-01 | 深圳通联金融网络科技服务有限公司 | Method and device for improving voice call quality inspection effect |
| US11057519B1 (en) * | 2020-02-07 | 2021-07-06 | Open Text Holdings, Inc. | Artificial intelligence based refinement of automatic control setting in an operator interface using localized transcripts |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6847711B2 (en) * | 2003-02-13 | 2005-01-25 | Sbc Properties, L.P. | Method for evaluating customer call center system designs |
-
2021
- 2021-09-13 CN CN202111069737.5A patent/CN113630505B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108965620A (en) * | 2018-08-24 | 2018-12-07 | 杭州数心网络科技有限公司 | A kind of artificial intelligence call center system |
| CN112217934A (en) * | 2019-07-09 | 2021-01-12 | 杭州巧卡科技有限公司 | Voice recognition processing method and system in real-time communication system |
| CN111294468A (en) * | 2020-02-07 | 2020-06-16 | 普强时代(珠海横琴)信息技术有限公司 | Tone quality detection and analysis system for customer service center calling |
| US11057519B1 (en) * | 2020-02-07 | 2021-07-06 | Open Text Holdings, Inc. | Artificial intelligence based refinement of automatic control setting in an operator interface using localized transcripts |
| CN112885376A (en) * | 2021-01-23 | 2021-06-01 | 深圳通联金融网络科技服务有限公司 | Method and device for improving voice call quality inspection effect |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113630505A (en) | 2021-11-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10083691B2 (en) | Computer-implemented system and method for transcription error reduction | |
| US8798255B2 (en) | Methods and apparatus for deep interaction analysis | |
| US8774372B2 (en) | Telephone call inbox | |
| CN109767791A (en) | A kind of voice mood identification and application system conversed for call center | |
| CN110619506A (en) | Post portrait generation method, post portrait generation device and electronic equipment | |
| CN110135669A (en) | A kind of service quality inspection management method and system | |
| CN110633912A (en) | Method and system for monitoring service quality of service personnel | |
| CN109726275B (en) | Intelligent legal knowledge query method applied to legal consultation service | |
| CN108924371B (en) | Method for identifying user number through incoming call number in electric power customer service process | |
| CN108257594A (en) | A kind of conference system and its information processing method | |
| CN113297365B (en) | User intention judging method, device, equipment and storage medium | |
| CN102402717A (en) | Data analysis facility and method | |
| CN113435627A (en) | Work order track information-based electric power customer complaint prediction method and device | |
| CN110705309A (en) | Service quality evaluation method and system | |
| CN111062564A (en) | Method for calculating power customer appeal sensitive value | |
| CN113630505B (en) | AI-based call center management platform | |
| WO2022041403A1 (en) | Sales order prediction method | |
| CN112052686A (en) | Voice learning resource pushing method for user interactive education | |
| CN114490992A (en) | Intelligent answer system for legal affair consultation | |
| CN110874533B (en) | Intention identification method based on keyword matching and response path after voice identification | |
| Coletto et al. | The accuracy of public polls in provincial elections | |
| CN110853623A (en) | Automatic financial collection urging method and device based on voice question answering | |
| CN113345419B (en) | Dialect accent based speech translation method, system and readable storage medium | |
| CN116628206B (en) | Enterprise credit analysis management system based on data analysis | |
| CN115499545A (en) | Telephone customer service performance scoring method, device, equipment and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20240506 Address after: Room 16, 6th floor, Room 101, 1st to 8th floors, Building 2, No. 186 South Fourth Ring West Road, Fengtai District, Beijing, 100000 Applicant after: Beijing Green News Technology Co.,Ltd. Country or region after: China Address before: 518109 room 801, 8 / F, block C, Gongcun new town commercial center (Huihai Plaza), No. 19, Sanlian Chuangye Road, Sanlian community, Longhua street, Longhua District, Shenzhen, Guangdong Applicant before: Shenzhen Zhongtou Internet Information Technology Co.,Ltd. Country or region before: China |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |