WO2017088593A1

WO2017088593A1 - Service seat selection method, apparatus and system

Info

Publication number: WO2017088593A1
Application number: PCT/CN2016/101577
Authority: WO
Inventors: 唐朝辉; 韦思宁; 庄乃峰
Original assignee: 华为技术有限公司
Priority date: 2015-11-24
Filing date: 2016-10-09
Publication date: 2017-06-01
Also published as: CN106791235B; CN106791235A

Abstract

Provided are a service seat selection method, apparatus and system, which relate to the technical field of communications, can allocate a most suitable service seat to a user, and improve a satisfaction degree of the user for the service of a call centre. The method comprises: an intelligent decision device receiving a request message for selecting a service seat for an accessed user currently accessing a call centre; the intelligent decision device acquiring an identifier of the service seat decided according to a matching degree of the accessed user and the service seat according to the request message, the matching degree being obtained through calculation based on a voice feature of the accessed user; the intelligent decision device sending the acquired identifier of the service seat to a route controller, wherein the identifier of the service seat sent to the route controller is used as a reference for the accessed user selecting the service seat.

Description

Method, device and system for selecting service seat

This application claims priority to Chinese Patent Application No. 201510821125.5, entitled "A Method, Apparatus and System for Selecting Service Agents", which is filed on November 24, 2015, the entire contents of which are incorporated by reference. In this application.

Technical field

The present invention relates to the field of communications technologies, and in particular, to a method, device, and system for selecting a service agent.

Background technique

In the era of mobile Internet, the demand for superior user service experience is becoming more and more urgent. As an entry point for operators and enterprise Internet services, the call center provides a service experience that is directly related to the user's satisfaction with operators and enterprise Internet services.

Among them, whether the user is satisfied with the service of the call center and which service agent is currently serving the service has a great relationship. For example, the service agent A provides services for the user B having specific characteristics (such as habit, voice, emotion, etc.). The service habits, voices, and ways of handling the crisis of the service agent A will affect the satisfaction of the user B to the service experience. Therefore, how to select a suitable service agent for an access user to provide services is directly related to the user's final satisfaction.

At present, the call center system generally allocates a service agent to the user by using a random selection method, or selects a service agent that provides the service for the user last time as the service agent that provides the service for the user. The above method is used to select a service agent for the user, and the user cannot be assigned a suitable service agent for its service, resulting in lower satisfaction of the user to the service of the call center.

Summary of the invention

The embodiments of the present invention provide a method, an apparatus, and a system for selecting a service agent, which solve the problem that the user cannot be allocated a suitable service agent for the service, and the user's satisfaction with the service of the call center is low.

In order to achieve the above object, embodiments of the present invention adopt the following technical solutions:

An embodiment of the present invention provides a method for selecting a service agent. After receiving the request message for selecting an service agent for an access user currently accessing the call center, the smart decision device according to the language of the access user The tone feature obtains the matching degree between the access user and the service agent, and determines the identifier of the service agent according to the matching degree between the access user and the service agent. Then, the intelligent decision device sends the acquired service agent identifier to the route. The controller is configured to select a service agent for the access user according to the identifier of the service agent sent by the intelligent decision device.

In the embodiment of the present invention, the identifier of the service agent determined by the intelligent decision unit is obtained based on the voice feature of the access user and the matching degree between the access user and the service agent. Therefore, the method is used to select the service agent selected by the access user. It is more suitable to meet the needs of users, and can also improve the satisfaction of access users to the services of the call center.

Before the smart decision device receives the request message for selecting a service agent for the access user, the method further includes:

The intelligent decision device analyzes the historical call record in advance, and obtains a calculation model for calculating the matching degree between the voice feature and the service agent.

Further, the intelligent decision device analyzes the historical call record in advance, and obtains a calculation model for calculating the matching degree between the voice feature and the service agent, and specifically includes:

The smart decision device selects a part of the call record from the historical call record as sampling data, and each call record includes a recording of the call, an identifier of the service agent of the call, and an identifier of the user of the call, and then The intelligent decision making device performs voice analysis on the recording in each call record in the sampled data to obtain voice feature data corresponding to each call record, and the voice feature data and the identifier of the service agent in the corresponding call record are formed. Data pairing Finally, the intelligent decision device uses the preset algorithm to train the obtained data pair to obtain a calculation model for calculating the matching degree between the voice feature and the service agent.

The intelligent decision device itself maintains a calculation model for calculating the matching degree between the voice feature and the service agent, so that the intelligent decision device can use the calculation model to calculate the user and the smart decision device when receiving the request for selecting a service agent for a certain user. The matching degree of the service agent is determined according to the calculated matching degree as the identifier of the service agent whose matching degree is within the preset range.

The calculation model for calculating the matching degree between the voice feature and the service agent maintained by the intelligent decision device itself is obtained by the intelligent decision device based on the selected part of the call record selected in the historical call record, and the historical call record can clearly know the user to the service agent. Satisfaction, therefore, the calculation model obtained by analyzing historical call records can more accurately represent the matching relationship between users and service agents. In one step, it is also more accurate to use the calculation model to obtain the matching degree between the current call user and the service agent.

Optionally, the request message received by the intelligent decision device carries the voice feature data of the access user, so that the smart decision device obtains the service agent determined according to the matching degree between the access user and the service agent. Identification, including:

The intelligent decision device inputs the voice feature data of the access user into the calculation model, and the calculation model calculates the matching degree between the voice feature of the access user and the service agent, and obtains the connection with the service agent. The identifier of the service agent whose matching degree of the user's voice feature is within the preset range.

Further, after the smart decision device obtains the calculation model, the method further includes:

First, the intelligent decision device analyzes the voice feature data of each user in the historical call record; the smart decision device inputs the analyzed voice feature data of the user into the calculation model for calculation, and obtains the The identifier of the service agent of the analyzed user's voice feature is within a preset range; finally, the smart decision device saves the identity of the analyzed user and the matching degree with the analyzed user's voice feature The correspondence between the identifiers of the service agents within the preset range.

In addition to the calculation model for calculating the matching degree between the voice feature and the service agent, the intelligent decision device in the embodiment of the present invention may also store the correspondence between the identifier of the analyzed user and the identifier of the service agent, and the corresponding relationship is adopted. Obtained using a computational model. In this way, the intelligent decision device can obtain the identifier of the service agent whose matching degree with the access user is within the preset range, regardless of whether the voice feature of the access user of the current call can be obtained.

Optionally, the request message received by the intelligent decision device carries the identifier of the access user, so that the smart decision device obtains the identifier of the service agent determined according to the matching degree between the access user and the service agent. Specifically, including:

The smart decision device searches for the identifier of the service agent corresponding to the identifier of the access user in the corresponding relationship according to the identifier of the access user.

Specifically, the intelligent decision device in the embodiment of the present invention selects a part of the call record from the historical call record as sampling data, and the smart decision device selects a quality call from the historical call record as a satisfactory call record. As sampled data.

In this case, the identifier of the service agent whose matching degree is within the preset range in the embodiment of the present invention is the identifier of the service agent with the highest matching degree.

The calculation model and the corresponding relationship in the embodiment of the present invention are generated according to part of the historical call record. If the intelligent decision device selects the quality evaluation as a satisfactory call record and analyzes the call record with satisfactory quality evaluation, a calculation model is generated and In the corresponding relationship, the intelligent decision device can obtain the identifier of the service agent with the highest matching degree with the access user according to the calculation model or the corresponding relationship.

Specifically, the intelligent decision device in the embodiment of the present invention selects a part of the call record from the historical call record as sampling data, and the smart decision device selects a call whose quality evaluation is not satisfactory from the historical call record. Record as sampled data.

In this scenario, the identifier of the service agent whose matching degree is within the preset range in the embodiment of the present invention is the identifier of the service agent whose matching degree is greater than or equal to the preset threshold.

Similar to the sixth possible implementation manner of the first aspect, the smart decision device in the embodiment of the present invention may also select a call record whose quality evaluation is not satisfactory, and analyze the call record whose quality evaluation is not satisfactory, and generate a call record. The calculation of the model or the corresponding relationship may be performed by the intelligent decision device according to the calculation model or the corresponding relationship, and the identifier of the service agent whose matching degree with the access user is greater than or equal to the preset threshold is obtained.

Further, in the method for selecting a service agent provided by the embodiment of the present invention, after the call of the access user ends, the method further includes:

The intelligent decision device collects the call record of the current call, and updates the calculation model by using the collected call record.

The intelligent decision-making device realizes the further update of the calculation model by analyzing the new call record, and improves the accuracy of the calculation model.

Another embodiment of the present invention also provides a method for selecting a service agent. First, the routing controller sends a request message to the intelligent decision device, requesting the smart decision device to select a service for the access user currently accessing the call center. An agent, and then the routing controller receives an identifier of the service agent that is determined by the intelligent decision device based on a matching degree between the access user and the service agent; and finally, the routing controller refers to the identifier of the received service agent. And selecting an identifier of a service agent serving the access user.

In the embodiment of the present invention, when the route controller allocates a route to the access user, the route controller selects the identifier of the service agent serving the access user according to the identifier of the service agent sent by the smart decision device, and the smart decision device sends the identifier. The identity of the service agent to the routing controller is based on the intelligent decision device. The matching degree between the access user and the service agent is determined. Therefore, when the routing controller allocates a route to the access user, the routing controller indirectly considers the matching degree between the access user and the service agent, and selects the matching degree according to the matching degree. The service agent is more accurate, can meet the needs of the access users, and improve the satisfaction of the access users to the service agents.

Optionally, in the embodiment of the present invention, the routing controller receives the identifier of the service agent that has the highest matching degree with the voice feature of the access user, and the routing controller refers to the identifier of the received service agent, and selects The process of identifying the identity of the service agent serving the user is that the routing controller determines the identity of the received service agent as the identity of the service agent serving the access user.

Optionally, in the embodiment of the present invention, the routing controller receives the identifier of the service agent whose matching degree with the voice feature of the access user is greater than or equal to a preset threshold, and the routing controller refers to the received service. The process of selecting the identity of the agent and selecting the identity of the service agent that accesses the user service is as follows:

The routing controller selects, from the identifier of the existing service agent, an identifier of any service agent other than the identifier of the received service agent as the identifier of the service agent serving the access user.

Corresponding to the previous embodiment, in the embodiment of the present invention, the identifier of the service agent received by the routing controller may be the identifier of the service agent with the highest matching degree with the voice feature of the access user, or may be the voice of the access user. The identifier of the service agent whose matching degree is greater than or equal to the preset threshold. In different application scenarios, the routing controller adopts different ways for the access user to select a service agent.

An embodiment of the present invention provides an intelligent decision device, including: a receiving unit, an obtaining unit, and a sending unit.

Specifically, the functions implemented by the respective unit modules provided by the embodiments of the present invention are specifically as follows:

a receiving unit, configured to receive a request message for selecting a service agent for an access user currently accessing the call center;

An obtaining unit, configured to calculate, according to the request message received by the receiving unit, a matching degree based on a voice feature of the access user, and determine an identifier of the service agent;

And a sending unit, configured to send the identifier of the service agent acquired by the acquiring unit to the routing controller, so that the routing controller refers to when the access user selects a service agent.

For the technical effects of the intelligent decision device provided by the embodiment of the present invention, refer to the technical effects of the smart decision device described in the method for selecting a service agent performed by the intelligent decision device in the foregoing embodiment, where No longer.

Further, the intelligent decision device further includes a generating unit,

The generating unit is configured to analyze the historical call record in advance before the receiving unit receives the request message for selecting the service agent for the access user, to obtain a calculation model for calculating the matching degree between the voice feature and the service agent.

Further, the intelligent decision device further includes a selection unit,

The selecting unit is configured to select a part of the call record from the historical call record as sampling data, where each call record includes a recording of the call, an identifier of the service agent of the call, and an identifier of the user of the call.

In this application scenario, the generating unit is specifically configured to perform voice analysis on the recording in each of the call records selected by the selecting unit to obtain voice feature data corresponding to each call record. The voice feature data forms a data pair with the identifier of the service agent in the corresponding call record, and is used to train the data pair by using a preset algorithm to obtain a calculation model for calculating the matching degree between the voice feature and the service agent. .

Optionally, the request message received by the receiving unit in the embodiment of the present invention carries the voice feature data of the access user.

In this application scenario, the acquiring unit is specifically configured to input voice feature data of the access user into the computing model obtained by the generating unit, and calculate, by the computing model, the access user. Matching the voice feature with the service agent, and obtaining the identifier of the service agent whose matching degree with the voice feature of the access user is within a preset range.

Further, after the generating unit obtains the computing model, the generating unit is further configured to:

Analyzing the voice feature data of each user in the historical call record;

Inputting the voice feature data of the analyzed user into the calculation model, respectively, to obtain an identifier of the service agent whose matching degree with the analyzed user's voice feature is within a preset range;

Corresponding relationship between the identifier of the analyzed user and the identifier of the service agent whose matching degree with the analyzed user's voice feature is within a preset range is saved.

In this application scenario, the acquiring unit is specifically configured to: search, according to the identifier of the access user, a service agent corresponding to the identifier of the access user in the corresponding relationship generated by the generating unit. Logo.

Optionally, the selecting unit in the embodiment of the present invention is specifically configured to select, from the historical call record, a call record whose quality is evaluated as satisfactory as sampling data.

In this way, in the embodiment of the present invention, the identifier of the service agent whose matching degree is within the preset range is the identifier of the service agent with the highest matching degree.

Optionally, the selecting unit in the embodiment of the present invention is specifically configured to select, from the historical call record, a call record whose quality is evaluated as unsatisfactory as sampling data.

In this way, in the embodiment of the present invention, the identifier of the service agent whose matching degree is within the preset range is the identifier of the service agent whose matching degree is greater than or equal to the preset threshold.

In the same manner as the method for selecting a service agent performed by the intelligent decision device in the above embodiment, the identifier of the service agent selected by the selected unit in the embodiment of the present invention may be the identifier of the service agent with the highest matching degree, or may be the matching degree greater than or equal to the preset. The identifier of the service agent for the threshold.

Further, the intelligent decision device further includes an acquisition unit,

The collecting unit is configured to collect a call record of the current call.

Further, the intelligent decision device further includes a processing unit,

The processing unit is configured to update the calculation model by using a call record collected by the collection unit.

For technical effects, reference may be made to the technical effects of the intelligent decision device described in the method for selecting a service agent performed by the intelligent decision device in the foregoing embodiment, and details are not described herein again.

Another embodiment of the present invention provides a routing controller, including a receiving unit, a sending unit, and a processing unit.

a sending unit, configured to send a request message to the smart decision device, requesting the smart decision device to select a service agent for an access user currently accessing the call center;

The receiving unit is further configured to receive the smart decision device according to the access user and the service agent The identifier of the service agent determined by the matching degree, where the matching degree is calculated based on the voice feature of the access user;

And a processing unit, configured to refer to an identifier of the service agent received by the receiving unit, and select an identifier of a service agent serving the access user.

Optionally, in the embodiment of the present invention, the receiving unit receives the identifier of the service agent that has the highest matching degree with the voice feature of the access user.

In this way, the processing unit in the embodiment of the present invention is specifically configured to determine an identifier of a service agent received by the receiving unit as an identifier of a service agent serving the access user.

For the technical effects of the routing controller provided by the embodiment of the present invention, refer to the technical effects of the routing controller described in the method for selecting a service agent performed by the routing controller in the foregoing embodiment, and details are not described herein again.

Optionally, the receiving unit in the embodiment of the present invention receives the identifier of the service agent whose matching degree with the voice feature of the access user is greater than or equal to a preset threshold.

In this way, the processing unit is specifically configured to select, from the identifier of the existing service agent, an identifier of any service agent other than the identifier of the received service agent as the identifier of the service agent serving the access user.

For technical effects, refer to the technical effects of the routing controller described in the method for selecting a service agent performed by the routing controller in the foregoing embodiment, and details are not described herein again.

A further embodiment of the present invention provides a system for selecting a service agent, comprising the intelligent decision device according to any one of the preceding embodiments, and the routing controller according to any one of the above embodiments, the routing control The device is connected to the intelligent decision device.

Further, the system provided by the embodiment of the present invention further includes a computer telephony integrated CTI, and the CTI is connected to the routing controller.

The CTI is configured to receive a call request initiated by an access user, and send, according to the call request, a route assignment message that allocates a route to the access user to the routing controller, and is configured to receive the route controller sent by the routing controller. An identifier and a target queue of the service agent serving the access user, and connecting the access user to the service agent serving the access user.

Further, the system provided by the embodiment of the present invention further includes an interactive voice response IVR, the IVR Connected to the CTI;

The IVR is configured to acquire voice feature data of the access user, and send the voice feature data of the access user to the CTI.

For the technical effects of the system for selecting a service agent provided by the embodiment of the present invention, refer to the technical effects of the intelligent decision device described in the method for selecting a service agent performed by the intelligent decision device in the foregoing embodiment, and the selection performed by the routing controller in the foregoing embodiment. The technical effect of the routing controller described in the method of serving the agent is not described here.

DRAWINGS

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the description of the prior art will be briefly described below. Obviously, the drawings in the following description are only It is a certain embodiment of the present invention, and other drawings can be obtained from those skilled in the art without any creative work.

1 is a schematic structural diagram of a conventional call center system;

2 is a schematic flow chart of an existing method for selecting a service agent;

3 is a schematic structural diagram of a call center system according to an embodiment of the present invention;

4 is a schematic flowchart of a method for generating a calculation model according to an embodiment of the present invention;

FIG. 5 is a schematic flowchart of a method for generating a correspondence relationship according to an embodiment of the present invention;

6 is a schematic flowchart 1 of a method for selecting a service agent according to an embodiment of the present invention;

FIG. 7 is a second schematic flowchart of a method for selecting a service agent according to an embodiment of the present invention; FIG.

8 is a schematic flowchart 3 of a method for selecting a service agent according to an embodiment of the present invention;

9 is a schematic structural diagram 1 of an intelligent decision device according to an embodiment of the present invention;

FIG. 10 is a schematic structural diagram 2 of an intelligent decision device according to an embodiment of the present invention;

11 is a schematic structural diagram 1 of a routing controller according to an embodiment of the present invention;

12 is a schematic structural diagram 3 of an intelligent decision device according to an embodiment of the present invention;

FIG. 13 is a schematic structural diagram 2 of a routing controller according to an embodiment of the present invention;

FIG. 14 is a schematic structural diagram 1 of a system for selecting a service agent according to an embodiment of the present invention; FIG.

FIG. 15 is a schematic structural diagram 2 of a system for selecting a service agent according to an embodiment of the present invention.

detailed description

The technical solutions in the embodiments of the present invention are clearly and completely described in the following with reference to the accompanying drawings in the embodiments of the present invention. It is obvious that the described embodiments are only a part of the embodiments of the present invention, but not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

The terms "first", "second", "third" and "fourth" and the like in the specification and claims of the present invention and the above drawings are used to distinguish different objects, and are not intended to describe a specific order. Furthermore, the terms "comprises" and "comprising" and "comprising" are intended to cover a non-exclusive inclusion. For example, a process, method, system, product, or device that comprises a series of steps or units is not limited to the listed steps or units, but optionally also includes steps or units not listed, or alternatively Other steps or units inherent to these processes, methods, products or equipment.

In the following description, for purposes of illustration and description However, it will be apparent to those skilled in the art that the present invention may be practiced in other embodiments without these specific details. In other instances, detailed descriptions of well-known devices, circuits, and methods are omitted so as not to obscure the description of the invention.

In addition, the term "and/or" herein is merely an association relationship describing an associated object, indicating that there may be three relationships, for example, A and/or B, which may indicate that A exists separately, and A and B exist at the same time. There are three cases of B alone. In addition, the character "/" in this article generally indicates that the contextual object is an "or" relationship.

Automatic Call Distribution (ACD) is a telephone device that arranges and processes incoming calls based on the number being called and the associated command processing database.

Computer Telephony Integration (CTI), also known as computer telephony, is used to describe computers to manage telephone calls. It refers to a computerized call center, for example, to connect a user's call directly to the right place.

1 is a schematic structural diagram of a conventional call center system, in which the call center system includes an ACD, an interactive voice response (IVR) connected to the ACD, a CTI connected to the IVR, and a CTI respectively. Service agent and routing controller, where CTI root Different skill queues are set according to different service types and user levels, and multiple service agents are checked in each skill queue. The user terminal interacts with the call center system through a transport network.

The user terminal may be a wireless terminal or a wired terminal, and the wireless terminal may be a device that provides voice and/or data connectivity to the user, a handheld device with wireless connectivity, or other processing device connected to the wireless modem. The wireless terminal can communicate with one or more core networks via a radio access network (eg, RAN, Radio Access Network), which can be a mobile terminal, such as a mobile phone (or "cellular" phone) and with a mobile terminal. The computer, for example, can be a portable, pocket, handheld, computer built-in or in-vehicle mobile device that exchanges language and/or data with the wireless access network. A wireless terminal may also be called a system, a Subscriber Unit, a Subscriber station, a Mobile Station, a Mobile, a Remote Station, an Access Point, Remote Terminal, Access Terminal, User Terminal, User Agent, or User Equipment.

With the traditional call center system, a user call is queued to enter the IVR self-service process after accessing the call center. If the user requires a human service, the CTI then assigns a service agent to the user via the routing controller, wherein the routing controller determines the rules for assigning the service agent to the user.

Specifically, the manner in which the CTI selects a service agent via the routing controller is:

Mode 1: After the call arrives, the call is queued in a certain skill queue. After the queue is finished, an idle service agent is selected from the service agents that are checked in to the queue to serve the access user.

Manner 2: After the call arrives, the CTI obtains the service agent that served the user last time, and directly puts the call into the skill queue checked in by the service agent. After the queue is finished, the call enters the private queue of the service agent to be queued. Finally, the service agent provides services to the user.

As shown in FIG. 2, if the manner in which the CTI selects the service agent via the routing controller is the second mode, the specific method for selecting the service agent for the user is:

1. The user terminal sends a call request to the CTI.

2. The CTI sends a route request message to the routing controller.

The routing request message includes at least a user identity number (ID, Identity), and the user ID is generally a user's phone number.

3. The routing controller obtains the historical service data, and queries the historical service data from the historical service data to query the service agent A1 that last served the user.

4. The routing controller selects a target skill queue Q1 that matches the service attribute level of the user from the skill queue checked in by the service agent A1.

5. The routing controller returns the acquired target skill queue Q1 and the ID of the service agent A1 to the CTI.

6. The CTI queues the user's call into the skill queue Q1.

7. When the call is dequeued from the skill queue Q1, the CTI then queues the user's call into the private queue of the service agent A1.

8. When the call is dequeued from the private queue of the service agent A1, the CTI connects the service agent A1 with the call session of the user.

It can be understood that if the way in which the CTI selects the service agent via the routing controller is mode one, then an idle service agent is selected from the skill queue to provide services for the user, but the selection process is random, and the service agent is not considered. It is suitable for providing services to current users, such as: the habits, sounds, intonations and other factors of the service agent are in line with the user's preference.

If the way in which the CTI selects the service agent via the routing controller is mode 2, the service agent that served the user last time is obtained according to the historical call information, and the service agent that last served the user is specified to provide the service for the user, although The method of selecting a service agent is to specify the service agent according to the rules of the last service. However, the service agent that provides the service to the access user last time may not be suitable for providing services to the current user, and cannot achieve intelligent matching between the service agent and the user. Accurate matching between the user and the service agent results in lower user satisfaction with the call center's services.

In view of the above problems, the present invention provides a method, device and system for selecting a service agent, which are realized by adding an intelligent decision device storing a calculation model for calculating the matching degree of a voice feature and a service agent in a conventional call center system. The indirect access of the routing controller determines the identifier of the service agent based on the matching degree between the access user and the service agent that initiates the call, and selects a service agent for the access user according to the determined service agent identifier, so that the selected service agent is selected. It is more suitable to meet the needs of users, and can also improve the satisfaction of access users to the services of the call center.

FIG. 3 is a schematic structural diagram of a call center system according to an embodiment of the present invention. Referring to FIG. 3, the call center system adds an intelligent decision device to a traditional call center system, wherein the smart decision device and the route controller are connected through a network.

Optionally, the routing controller in the embodiment of the present invention may be a device that is separately set up, or may be a processor that is separately set up in the CTI, or may be integrated and implemented on a processor in the CTI. The processor described herein may be a Central Processing Unit (CPU), or an Application Specific Integrated Circuit (ASIC), or one or more integrated systems configured to implement embodiments of the present invention. Circuit.

It should be noted that, if the routing controller in the embodiment of the present invention is configured in the CTI, the steps performed by the routing controller in the embodiment of the present invention are all performed by CTI, and the interaction steps between the CTI and the routing controller are both CTIs. The corresponding module inside is executed.

Embodiment 1

In the method for selecting a service agent provided by the embodiment of the present invention, the intelligent decision device analyzes the historical call record in advance, and obtains a calculation model for calculating the matching degree between the voice feature and the service agent.

In the process of selecting a service agent for the user, the service agent serving the user is selected according to the matching degree between the voice feature representing the user's emotion and the service agent, so that the service agent selected for the user is more suitable for the user, and the user is also improved. Satisfaction with call center services.

The intelligent decision device may analyze the historical call record in advance, or the intelligent decision device may analyze the call record that is satisfactory in quality evaluation selected from the historical call record, or may be the quality selected by the intelligent decision device in advance from the historical call record. The evaluation was analyzed for unsatisfactory call records.

It can be easily understood that the call record with satisfactory quality evaluation in the embodiment of the present invention can also be understood as a successful call record in the application scenario of the call center active outbound marketing. Correspondingly, the call record with unsatisfactory quality evaluation can be understood as the unsuccessful call record in the application scenario of the call center active outbound marketing.

Specifically, as shown in FIG. 4, the method for generating the calculation model in the embodiment of the present invention is:

S101. The intelligent decision device selects part of the call record from the historical call record as sample data, and each call record includes a record of the call, an identifier of the service agent of the call, and an identifier of the user of the call.

Specifically, the intelligent decision device acquires a historical call record from the CTI, and selects a part of the call record from the historical call record as sampling data.

Optionally, the intelligent decision device in the embodiment of the present invention may select quality from the historical call record. The call record that is evaluated as satisfactory is used as sampling data, and the call record whose quality is evaluated as unsatisfactory may also be selected from the historical call record as sampling data.

It can be understood that after the user talks with the service agent of the call center, the service agent who provides the service for the user in the call is generally evaluated for quality, and the user may choose to be satisfied or unsatisfied. Therefore, the historical call record directly carries the information of the quality evaluation, and the intelligent decision-making device can easily extract the call record whose quality evaluation is satisfactory or the quality evaluation is not satisfactory according to the information.

Optionally, the identifier of the user in the embodiment of the present invention may be a user ID, or may be an International Mobile Subscriber Identification Number (IMSI) of the user terminal used by the user.

S102: The intelligent decision device performs voice analysis on the call recording in each call record in the sampled data, and obtains voice feature data corresponding to each call record, wherein the voice feature data and the identifier of the service agent in the corresponding call record Form a data pair.

Specifically, the intelligent decision device can perform voice analysis on the call recording in each call record in the sampled data by using a voice recognition technology, obtain voice feature data corresponding to each call record, and obtain voice feature data and a corresponding call. The data pair formed by the identification of the service agent in the record.

The voice feature data includes at least loudness, call duration, speech rate, and emotional characteristics.

It should be noted that the smart decision device may use any existing voice recognition technology to obtain the voice feature data in each call record in the sample data, which is not specifically limited in this embodiment of the present invention.

Exemplarily, the intelligent decision device processes the call recording of each call record in the sampled data, and uses voice recognition technology to obtain the following voice feature data from the call recording: frequency (a1), loudness (a2), speech rate (a3) ), emotional characteristics (a4), call duration (a5), and silence duration (a6).

Further, the intelligent decision device may discretize the voice feature data of each call record according to a preset discretization rule, and obtain the discretized processed voice feature data and the identifier of the service agent in the corresponding call record. The resulting data pair.

Illustratively, the intelligent decision device discretizes the frequency, loudness, speech rate, call duration, and mute duration in each call record in the sampled data using the following discretization rules:

Frequency: 1: [20-30) 2: [30, 50) 3: [50, 90), and so on;

Loudness: 1:[0-10)2:[10,20)3:[20,30), and so on;

Call duration (seconds): 1: [0-10) 2: [10, 20) 3: [20, 30), and so on;

Silence duration (seconds): 1: [0-10) 2: [10, 20) 3: [20, 30), and so on;

Emotional characteristics: 1: irritability, 2: impatience, 3: sedation;

Speech rate (words/second): 1:[0-0.5)2:[0.5-1)3:[1-1.5), and so on.

If the mute duration is 11 seconds, the mute duration in the call record is indicated by 2.

The intelligent decision device discriminates the voice feature data of each call record according to the discretization rule, and acquires a data pair formed by the discretized processed voice feature data and the identifier of the service agent in the corresponding call record.

Exemplarily, the intelligent decision device discriminates the voice feature data of the call record according to the discretization rule, and obtains a data pair formed by the discretized processed voice feature data and the identifier of the service agent in the corresponding call record. :{[frequency: 15]^5, [loudness: 3]^5, [call duration: 6]^1, [mute duration: 2]^2, [emotional feature: 2]^10, [speech rate: 3 ]^5, [Service Agent ID: 120]}, where the number after "^" represents the weight of the field.

S103. The intelligent decision device uses a preset algorithm to train the data pair to obtain a calculation model for calculating the matching degree between the voice feature and the service agent.

The preset algorithm in the embodiment of the present invention may be a naive Bayes algorithm, or may be a K-Nearest Neighbor Algorithm (KNN) algorithm, or may be another association rule algorithm, which is not specifically described in this embodiment of the present invention. limited.

The following is an example of intelligent data processing equipment using the Naive Bayes algorithm to train data pairs.

If the voice feature data of the call record can be expressed as: x=(a1, a2, a3, a4, a5, a6), in this feature vector, the weight of each data item is different, in units of 1, and the weights are: 1, 1, 1, 2, 1, 2. The intelligent decision device uses the weight to extend the feature vector to form an 8-dimensional feature vector: x'=(b1, b2, b3, b4, b5, b6, b7, b8), where b1=a1, b2=a2 , b3 = a3, b4 = a3, b5 = a4, b6 = a5, b7 = a6, b8 = a6. Each b is a feature attribute of x'.

(1) Let x' = (b1, b2, ..., bm), each b being a characteristic attribute of x'.

(2) All the agents y existing in the call center system form a classification set C, C= (y1, y2, y3, y4, ..., yn).

(3) The intelligent decision device calculates P(y1) according to x'=(b1, b2, b3, b4, b5, b6, b7, b8) and C=(y1, y2, y3, y4, ..., yn) respectively. |x'), P(y2|x'),...,P(yn|x').

(4) If P(yk|x') = max{P(y1|x'), P(y2|x'), ..., P(yn|x')}, then x' belongs to the class yk.

Among them, the calculation method of step (3) is as follows:

a. Find a set of items to be classified that are known to be classified. This set is called a training sample set.

b. Statistics obtain the conditional probability estimates of each feature attribute under each category. which is:

P(b1|y1), P(b2|y1),...,P(b1|y2), P(b2|y2),...,P(b1|yn),P(b2|yn),...,P( Bm|yn).

c. If each feature attribute is conditionally independent, then the Bayesian theorem is derived as follows:

In the above formula, the denominator is constant for all categories, and the result of (4) can be obtained by maximizing the molecule.

In the embodiment of the present invention, the feature data x' is used as the classification item in the step (1), and the agent set C is used as the category set in the step (2), thereby calculating the classification (agent) corresponding to each feature data x'.

After the calculation model is obtained, the user's voice feature can be directly input into the calculation model for calculation, so as to obtain the identifier of the service agent whose matching degree with the user is within the preset range. Then, according to the obtained identification of the service agent, the service agent that provides the service to the user is selected.

It should be noted that, in some scenarios, for example, for a non-ASR call request, the request message received by the intelligent decision device may only include the identifier of the user, and the voice feature data of the user may not be carried.

Accordingly, another embodiment of the present invention provides a way to obtain a user's voice characteristics using a user's historical call history. When the method is used, the intelligent decision device can calculate the voice features of each user of the historical call record after obtaining the calculation model, and obtain the identifier of the service agent whose matching degree of the voice feature of the user is within a preset range, and then, The correspondence between the identifiers of the service agents in which the matching degree of each user and the voice features of each user is within a preset range is saved. In subsequent access to these users When calling, you can directly obtain the identifier of the corresponding service agent by querying the corresponding relationship.

Specifically, a method for generating a correspondence relationship in the embodiment of the present invention is shown in FIG. 5.

S201-S203 in FIG. 5 is the same as S101-S103 in FIG. 4, and details are not described herein again. After executing S203, the intelligent decision device performs S204-S206.

S204. The intelligent decision device analyzes voice feature data of each user in the historical call record.

For each call record in the historical call record, the intelligent decision device processes according to the method described in S102, and obtains the voice feature data corresponding to the user in each call record.

S205. The intelligent decision device inputs the analyzed voice feature data of the user into the calculation model, and obtains an identifier of the service agent whose matching degree with the analyzed user's voice feature is within a preset range.

If the calculation model is generated according to the quality record that is satisfactory for the call record, the identifier of the service agent whose matching degree with the analyzed user's voice feature is within the preset range is the highest matching degree with the analyzed user's voice feature. The identity of the service agent.

If the calculation model is generated according to the call record whose quality evaluation is unsatisfactory, the identifier of the service agent whose matching degree with the analyzed user's voice feature is within the preset range is the matching degree with the analyzed user's voice feature. The ID of the service agent that is greater than or equal to the preset threshold.

S206. The intelligent decision device saves a correspondence between the identifier of the analyzed user and the identifier of the service agent whose matching degree with the analyzed user's voice feature is within a preset range.

The intelligent decision device records the identifier of the analyzed user and the service agent identifier that matches the voice feature of the analyzed user within a preset range, generates an identifier including at least one analyzed user, and generates at least one The corresponding relationship of the identifiers of the at least one service agent corresponding to the identifier of each of the analyzed users in the identified user's identity.

Optionally, the correspondence may further include a matching degree between each analyzed user identifier and an identifier of each service agent corresponding thereto.

Optionally, the correspondence is represented by a mapping relationship table. As shown in Table 1, the identifier includes at least one analyzed user identifier and an identifier of each analyzed user in the identifier of the at least one analyzed user. Corresponding relationship of the identifiers of at least one service agent corresponding to each other.

Table 1

It should be noted that Table 1 only shows, in the form of a table, an identifier of at least one analyzed user and an identifier of at least one service agent corresponding to the identifier of each analyzed user in the identifier of at least one analyzed user. The correspondence relationship is not limited to the embodiment of the corresponding relationship, and the present invention does not limit this.

In the embodiment of the present invention, the intelligent decision device obtains a calculation model for calculating the matching degree between the voice feature and the service agent by analyzing the historical call record, and further obtains the correspondence between the analyzed user identifier and the matching degree of the service agent. Therefore, the intelligent decision device can use the computing model or the corresponding relationship to obtain a service agent that can be referred to when the service user selects the service agent, and then the routing controller obtains the service agent identity obtained by the intelligent decision device as the access user. Select the identity of the service agent. By using this method, the service agent selected by the access user can more closely meet the needs of the access user, and improve the satisfaction of the access user to the service of the call center.

Embodiment 2

The embodiment of the present invention provides a method for selecting a service agent. In the embodiment of the present invention, the following two different application scenarios are used to select a service agent for the access user:

I. Application scenario in which the access user actively accesses the call center.

In this application scenario, the following two different situations exist in the embodiment of the present invention to select a service agent for an access user:

1. As shown in FIG. 6, the access user initiates a normal call request (not an ASR call request), and the When the call request is changed to a manual service, the service agent is selected for the access user.

2. As shown in Figure 7, when the access user initiates an ASR call request and the call request is converted to a manual service, the service agent is selected for the access user.

Second, the call center active outbound marketing application scenario.

As shown in FIG. 8, the CTI in the call center system actively initiates an outbound call request (not an ASR call request), and the outbound call requesting user is an access user, and when the outbound call request is connected, the service agent is selected for the access user.

When the current call request is an ASR call request, the method for selecting a service agent provided by the embodiment of the present invention is performed based on the smart decision device obtaining the calculation model in advance. When the current call is a non-ASR call request, the method for selecting a service agent provided by the embodiment of the present invention is performed based on the smart decision device pre-obtaining the corresponding relationship.

Specifically, for the first scenario and the second application scenario of the foregoing first application scenario, the method for selecting a service agent provided by the embodiment of the present invention is based on that the smart decision device has obtained the identifier of the collected user and the identifier of the service agent in advance. The corresponding relationship is performed. Therefore, in the first scenario and the second application scenario of the first application scenario, the method for selecting a service agent provided by the embodiment of the present invention is performed after step S206 in the first embodiment. .

For the second aspect of the foregoing first application scenario, the method for selecting a service agent provided by the embodiment of the present invention is performed based on an intelligent decision device that has been previously obtained for use in a calculation model, and therefore, in the first application scenario described above. In the second case, the method for selecting a service agent provided by the embodiment of the present invention is performed after step S103 in the first embodiment.

The process of pre-generating a computing model or a corresponding relationship by an intelligent decision device is not described in detail here.

Specifically, when the access user initiates the non-ASR call request, as shown in FIG. 6, the method for selecting a service agent according to the embodiment of the present invention includes:

S301. The access user initiates a non-ASR call request to the CTI by using the user terminal.

Specifically, the user terminal sends a non-ASR call request to the CTI in response to the call operation of the access user. The non-ASR call request includes at least an identifier of the access user.

Optionally, the identifier of the access user in the embodiment of the present invention may be an access user ID, or may be an IMSI of the user terminal used by the access user.

It can be understood that the daily user initiated call request means that the user initiates a call request by using the user terminal, so that the user terminal can make a call request to the CTI.

S302. The CTI sends a route assignment message that carries the route to the access user that carries the identifier of the access user to the routing controller.

When the CTI allocates a service agent to an access user, it needs to be allocated according to the rules of the service agent determined by the routing controller. Therefore, after receiving the call request message sent by the user equipment or the CTI receives the response message sent by the user equipment, the CTI sends a route assignment message including the identifier of the access user to the route controller.

S303. The routing controller sends a request message for selecting a service agent to the access user to the intelligent decision device according to the route allocation message.

The route assignment message includes an identifier of the access user.

Optionally, the intelligent decision device and the routing controller in the embodiment of the present invention may communicate through the Http protocol (Hypertext Transfer Protocol), or communicate through the Socket protocol, and may also adopt other related protocols. The communication is not limited in this embodiment of the present invention.

S304: The intelligent decision device searches for the identifier of the service agent corresponding to the identifier of the access user in the correspondence according to the identifier of the access user.

Specifically, for the method for generating the corresponding relationship in the embodiment of the present invention, refer to the description in the first embodiment.

Optionally, if the calculation model in the embodiment of the present invention is generated by analyzing the call record in which the quality evaluation in the history call record is satisfactory, and the correspondence relationship is generated according to the calculation model, the correspondence in the embodiment of the present invention is The relationship may record the identity of each analyzed user and the identity of the service agent that has the highest degree of matching with the voice characteristics of each analyzed user.

Specifically, after receiving the request message for selecting the service agent for the access user, the intelligent decision device searches for the identifier of the service agent corresponding to the identifier of the access user in the corresponding relationship according to the identifier of the access user. The identifier of the service agent is the identifier of the service agent with the highest matching degree with the voice feature of the access user.

Optionally, if the calculation model in the embodiment of the present invention is generated by analyzing the call record in which the quality evaluation in the historical call record is not satisfactory, and the corresponding relationship is generated according to the calculation model, therefore, in the embodiment of the present invention, Correspondence can record the identity of each analyzed user and each analyzed The matching degree of the user's voice feature is greater than or equal to the identifier of the service agent of the preset threshold.

Specifically, after receiving the request message for selecting the service agent for the access user, the intelligent decision device searches for the identifier of the service agent corresponding to the identifier of the access user in the corresponding relationship according to the identifier of the access user. The identifier of the service agent is an identifier of the service agent whose matching degree with the voice feature of the access user is greater than or equal to a preset threshold.

S305. The intelligent decision device sends the obtained identifier of the service agent to the routing controller.

After obtaining the identifier of the service agent corresponding to the identifier of the access user, the intelligent decision device sends the identifier to the routing controller, so that the routing controller routes the call request according to the identifier of the service agent.

S306. The routing controller selects an identifier of the service agent serving the access user according to the preset selection rule and the identifier of the received service agent.

In the embodiment of the present invention, the identifier of the service agent obtained by the intelligent decision device according to the correspondence relationship and the identifier of the access user is the identifier of the service agent with the highest matching degree with the voice feature of the access user, that is, the information obtained by the intelligent decision device. The identifier of the service agent is the identifier of the service agent that is most satisfactory to the access user, and the routing controller determines the service agent identifier that is received by the routing controller as the identifier of the service agent serving the access user.

It should be noted that, in this case, if the service agent identifier received by the routing controller is empty, the routing controller selects the identifier of the service agent for the access user by using the existing method of selecting the service agent, that is, if the route is If the first service agent identifier received by the controller is empty, the routing controller obtains the identifier of the service agent that last served the access user according to the user identifier of the access user, and determines the identifier of the service agent as the current time. The identifier of the service agent that provides the service to the access user, or randomly identifies the identity of a service agent as the identifier of the service agent that provides the service to the access user.

Exemplarily, if the intelligent decision device generates a correspondence for the satisfactory call record by analyzing the quality evaluation, the user A calls to access the call center, and the routing controller receives the identifier of the service agent sent by the intelligent decision device is empty, then the route control The device obtains the service agent B that last served the user A, and determines the service agent B as the service agent that serves the user A this time.

In the embodiment of the present invention, the identifier of the service agent obtained by the intelligent decision device according to the corresponding relationship and the identifier of the access user is the identifier of the service agent whose matching degree with the voice feature of the access user is greater than or equal to the preset threshold, that is, the smart The identifier of the service agent obtained by the decision device is the most dissatisfied to the access user. The identifier of the plurality of service agents, the routing controller selects, from the identifiers of the existing service agents, the identifier of any service agent other than the identifiers of the plurality of service agents received as the service agent serving the access user. Logo.

Exemplarily, if the intelligent decision device generates a correspondence by analyzing the quality evaluation to the unsatisfactory call record, the user A calls to access the call center, and the routing controller receives the identifier of the service agent sent by the intelligent decision device as the service agent B and In the service agent C, the routing controller selects the service agent A as the service agent that provides the service for the user A from the identifiers of the existing service agents (the service agent A, the service agent B, the service agent C, and the service agent D).

S307. The routing controller determines the identifier of the target queue according to the identifier of the service agent serving the access user and the subscription attribute of the access user.

The subscription attribute of the access user in the embodiment of the present invention refers to the service type and service level information of the access user. For example, the subscription attribute of user A is a global service, a VIP user.

S308. The routing controller sends the identifier of the service agent serving the access user and the identifier of the target queue to the CTI.

After the routing controller determines the identifier of the service agent serving the access user, the routing controller determines the identifier of the target queue according to the identifier of the service agent serving the access user and the subscription attribute of the access user by using an existing method. The identity of the service agent serving the access user and the identity of the target queue are sent to the CTI.

S309. The CTI places the call request of the access user into the target queue for queuing.

The CTI allocates a route to the access user according to the identifier of the service agent that is sent by the routing controller for the access user and the identifier of the target queue.

S310. When the call request of the access user is dequeued from the target queue, the CTI places the call request of the access user into a private queue of the service agent serving the access user.

S311: When the call request of the access user is dequeued from the private queue of the service agent serving the access user, the CTI connects to the call session between the service agent serving the access user and the access user.

When the call request of the access user is dequeued from the private queue of the service agent serving the access user, the CTI connects the call session between the service agent serving the access user and the user terminal used by the access user.

S312. The call ends, and the call data of the call is generated.

When the call ends, the access user hangs up and the user terminal disconnects from the call center. At the same time, the CTI generates and records the call data of the call.

The S308-S312 in the embodiment of the present invention is the same as the method in the prior art, and details are not described herein again.

S313. The intelligent decision device collects the call record of the current call.

Specifically, the intelligent decision device collects the newly generated call record in the preset time period from the CTI.

The preset time period in the embodiment of the present invention may be a daily or a weekly, which is not specifically limited in the embodiment of the present invention. For example, when the smart decision device updates the correspondence, the call record recorded every day may be collected. .

S314. The intelligent decision device updates the correspondence by using the collected call record.

The intelligent decision device analyzes the collected call records by the above method, and updates the corresponding relationship by using the collected call records.

The method for selecting a service agent provided by the embodiment of the present invention is further applicable to: when the call request initiated by the access user is an ASR call request, and the call request is converted into a manual service, the service agent is selected for the access user.

Specifically, as shown in FIG. 7, the method for selecting a service agent provided by the embodiment of the present invention includes:

S401c: The user terminal initiates an ASR call request to the IVR.

Similarly, the user terminal here initiates an ASR call request to the IVR to send the ASR call request information to the CTI in response to the ASR call operation of the access user.

S401d and IVR obtain voice feature data of the access user.

Specifically, the IVR uses voice recognition technology to acquire voice feature data of the access user.

The voice recognition technology is an existing mature technology. The IVR in the embodiment of the present invention can use any existing voice recognition technology to obtain the voice feature data of the access user, which is not specifically limited in this embodiment of the present invention.

Further, the IVR may further discretize the voice feature data of the access user according to a preset discretization rule, and obtain the voice feature data of the access user after the discretization process.

Exemplarily, IVR uses voice recognition technology to obtain the following voice feature data of the access user: frequency: 25, loudness: 15, speech rate: 1.2, and emotional characteristics: irritable, IVR uses the following discretization rules for accessing users. Discretization of speech feature data:

Frequency: 1: [20-30) 2: [30, 50) 3: [50, 90), and so on;

Loudness: 1:[0-10)2:[10,20)3:[20,30), and so on;

Call duration (seconds): 1: [0-10) 2: [10, 20) 3: [20, 30), and so on;

Silence duration (seconds): 1: [0-10) 2: [10, 20) 3: [20, 30), and so on;

Emotional characteristics: 1: irritability, 2: impatience, 3: sedation;

Speech rate (words/second): 1:[0-0.5)2:[0.5-1)3:[1-1.5), and so on.

The IVR uses the above discretization rule to discretize the voice feature data of the access user, and then obtains the discretized processed voice feature data of the access user: {[frequency: 15], [loudness: 3], [call duration) :6], [mute duration: 2], [emotional feature: 2]}.

S401e: When the call request is transferred to the manual service, the IVR sends the voice feature data of the access user to the CTI.

When the call request is transferred to the manual service, the IVR carries the voice feature data of the access user in the associated data to the CTI.

If the IVR obtains the discretized processed speech feature data of the access user, the IVR sends the discretized processed speech feature data of the access user to the CTI when the call request is transferred to the manual service.

S402. The CTI sends a route assignment message that carries the route to the access user that carries the voice feature data of the access user to the route controller.

S403. The routing controller sends a request message for selecting a service agent to the access user to the intelligent decision device according to the route allocation message.

The route assignment message includes voice feature data of the access user.

S404: The intelligent decision device inputs the voice feature data of the access user to the calculation model, and obtains the identifier of the service agent whose matching degree with the voice feature of the access user is within a preset range.

After receiving the voice feature data of the access user, the intelligent decision device inputs the voice feature data into the calculation model, and the calculation model calculates the matching degree between the voice feature of the access user and the service agent, and Obtaining an identifier of the service agent whose matching degree with the voice feature of the access user is within a preset range.

Specifically, if the calculation model in the embodiment of the present invention is generated by analyzing the quality record in the historical call record as satisfactory, the intelligent decision device inputs the voice feature data of the access user into the calculation model, and the calculation model The matching degree between the voice feature of the access user and the service agent is calculated, and the identifier of the service agent with the highest matching degree with the voice feature of the access user is obtained.

If the calculation model in the embodiment of the present invention is generated by analyzing the call record in which the quality evaluation in the historical call record is not satisfactory, the intelligent decision device inputs the voice feature data of the access user into the calculation model, and the calculation model calculates The matching degree between the voice feature of the user and the service agent, and the identifier of the service agent whose matching degree with the voice feature of the access user is greater than or equal to a preset threshold.

The method provided by the embodiment of the present invention, after executing S404, executes S405-S413. The S405-S413 in the embodiment of the present invention is the same as the above-mentioned S305-S313, and details are not described herein again.

After the step S413 is performed, the method provided by the embodiment of the present invention further performs S414.

S414. The intelligent decision device updates the calculation model by using the collected call record.

Specifically, as shown in FIG. 8 , when the CTI in the call center system actively initiates an outbound call, the method for selecting a service agent provided by the embodiment of the present invention includes:

S501a. The CTI initiates a call request to the user equipment.

S501b. The user equipment sends a response message to the CTI when the user connects the call request.

After the execution of S501b, S502-S514 is performed, wherein S502-S514 in the embodiment of the present invention is the same as S302-S314 described above, and details are not described herein again.

As shown in the above description, in the method for selecting a service agent, the smart decision device obtains a match according to the request message according to the request message after receiving the request message for selecting the service agent for the access user. Determine the identity of the service agent and send the obtained service agent ID to the routing controller. The routing controller selects a service agent for the access user according to the identifier of the received service agent. Because the routing controller selects the service agent for the access user, it is indirectly selected according to the matching degree between the access user and the service agent. Therefore, the service agent selected by this method is more suitable for the user's needs. It can improve the satisfaction of the access user to the service of the call center.

Embodiment 3

The embodiment of the invention provides an intelligent decision device, which is used to execute the steps performed by the intelligent decision device in the above method. The intelligent decision device may include a module corresponding to the corresponding step. As shown in FIG. 9, the intelligent decision device includes:

The receiving unit 10 is configured to receive a request message for selecting a service agent for the access user, where the access user is a user currently accessing the call center.

The obtaining unit 11 is configured to obtain, according to the request message received by the receiving unit 10, an identifier of a service agent determined according to a matching degree between the access user and a service agent, where the matching degree is based on the The voice characteristics of the incoming user are calculated.

The sending unit 12 is configured to send the identifier of the service agent acquired by the acquiring unit 11 to the routing controller, where the identifier of the service agent sent to the routing controller is used as the access user selection service. The reference of the agent.

Further, as shown in FIG. 10, the intelligent decision device further includes a generating unit 13,

The generating unit 13 is configured to analyze the historical call record in advance before the receiving unit 10 receives the request message for selecting the service agent for the access user, and obtain a calculation model for calculating the matching degree between the voice feature and the service agent. .

Further, as shown in FIG. 10, the intelligent decision device further includes a selecting unit 14,

The selecting unit 14 is configured to select a partial call record as sample data from the historical call record, where each call record includes a recording of the call, an identifier of the service agent of the call, and an identifier of the user of the call.

Further, the generating unit 13 is specifically configured to perform voice analysis on the recordings in each of the call records selected by the selecting unit 14 to obtain voice feature data corresponding to each call record. The voice feature data forms a data pair with the identifier of the service agent in the corresponding call record, and is used to train the data pair by using a preset algorithm to obtain a calculation model for calculating the matching degree between the voice feature and the service agent.

Further, the request message carries the voice feature data of the access user.

Further, the acquiring unit 11 is specifically configured to input the voice feature data of the access user into the computing model to the generating unit 13, and calculate the access user by using the computing model. The matching of the voice feature to the service agent, and obtaining the identifier of the service agent whose matching degree with the voice feature of the access user is within a preset range.

Further, after the obtaining the calculation model, the generating unit 13 is further configured to:

Analyzing the voice feature data of each user in the historical call record;

Further, the request message carries an identifier of the access user.

Further, the obtaining unit 11 is configured to: search, according to the identifier of the access user, the identifier of the service agent corresponding to the identifier of the access user in the corresponding relationship generated by the generating unit 13 .

Further, the selecting unit 14 is specifically configured to select, from the historical call record, a call record whose quality is evaluated as satisfactory as sampling data;

The identifier of the service agent whose matching degree is within the preset range is the identifier of the service agent with the highest matching degree.

Further, the selecting unit 14 is specifically configured to select, from the historical call record, a call record whose quality is evaluated as unsatisfactory as sampling data;

The identifier of the service agent whose matching degree is within the preset range is the identifier of the service agent whose matching degree is greater than or equal to the preset threshold.

Further, as shown in FIG. 10, the intelligent decision device further includes an acquisition unit 15,

The collecting unit 15 is configured to collect a call record of the current call.

Further, as shown in FIG. 10, the intelligent decision device further includes a processing unit 16,

The processing unit 16 is configured to update the calculation model by using a call record collected by the collection unit.

It can be understood that the intelligent decision device of this embodiment may correspond to the smart decision device in the method for selecting a service agent of the embodiment described in any one of the above-mentioned FIG. 4, and the smart of the embodiment. The division and/or function of each module in the decision device is to implement the method flow shown in any one of FIG. 4 to FIG. 8. For brevity, no further details are provided herein.

The embodiment of the invention provides an intelligent decision device, which mainly comprises a receiving unit, an obtaining unit and a sending unit. When the access user initiates a call request to the computer telephony integrated CTI, the routing controller sends a request message for selecting the service agent to the access user to the intelligent decision device according to the route assignment message sent by the CTI; the intelligent decision device obtains the basis according to the request message. The identifier of the service agent determined by the matching degree between the access user and the service agent, wherein the matching degree is calculated based on the voice feature of the access user; the intelligent decision device sends the acquired identifier of the service agent to the routing controller, so that The routing controller selects a service agent for the access user according to the identifier of the service agent sent by the intelligent decision device.

Through the foregoing solution, the identifier of the service agent determined by the intelligent decision unit is obtained based on the voice feature of the access user and the matching degree between the access user and the service agent. Therefore, the service agent selected by the method for the access user is more suitable. Access to the user's needs can also increase the satisfaction of the access user's service to the call center.

Embodiment 4

An embodiment of the present invention provides a routing controller, where the routing controller is configured to perform the steps performed by the routing controller in the above method. The routing controller may include a module corresponding to the corresponding step. As shown in FIG. 11, the routing controller includes:

The sending unit 20 is configured to send a request message to the smart decision device, requesting the smart decision device to select a service agent for an access user currently accessing the call center.

The receiving unit 21 is further configured to receive, by the smart decision device, an identifier of the service agent that is determined according to the matching degree between the access user and the service agent, where the matching degree is calculated based on the voice feature of the access user.

The processing unit 22 is configured to select an identifier of the service agent served by the access user by referring to the identifier of the service agent received by the receiving unit 21.

Further, the receiving unit 21 receives the identifier of the service agent with the highest matching degree with the voice feature of the access user.

Correspondingly, the processing unit 22 is specifically configured to determine an identifier of a service agent that has the highest degree of matching with a voice feature of the access user as an identifier of a service agent serving the access user.

Further, the receiving unit 21 receives the identifier of the service agent whose matching degree with the voice feature of the access user is greater than or equal to a preset threshold.

Correspondingly, the processing unit 22 is specifically configured to: select, from the identifier of the existing service agent, an identifier of any service agent other than the identifier of the received service agent as a service agent serving the access user. Logo.

It can be understood that the routing controller in this embodiment may correspond to the routing controller in the method for selecting a service agent in the foregoing embodiment of any one of the embodiments of FIG. 4, and in the routing controller of this embodiment. The division and/or function of each module are to implement the method flow shown in any one of FIG. 4 to FIG. 8. For brevity, no further details are provided herein.

The embodiment of the invention provides a routing controller, which mainly comprises a receiving unit, a sending unit and a processing unit. When the computer telephony integrated CTI receives the call request initiated by the access user, the CTI sends a route assignment message for assigning the route to the access user to the routing controller; the routing controller sends the selective service agent for the access user according to the route assignment message. The request message is sent to the intelligent decision device; after receiving the identifier of the service agent determined by the smart decision device according to the matching degree between the access user and the service agent, the routing controller selects the service as the access user according to the identifier of the received service agent. The identifier of the service agent, and determining the target queue according to the identifier of the service agent serving the access user and the subscription attribute of the access user; the routing controller notifying the CTI according to the identifier of the service agent serving the access user and the target queue Access users and service agents that serve access users.

Through the foregoing solution, when the routing controller allocates a route to the access user, the routing controller selects the identifier of the service agent serving the access user according to the identifier of the service agent sent by the smart decision device, and the intelligent decision device sends the route to the route. The identifier of the service agent of the controller is determined by the smart decision device according to the matching degree between the access user and the service agent. Therefore, when the routing controller allocates a route for the access user, the access controller considers the access user indirectly. The matching degree with the service agent is more accurate according to the matching degree, which can meet the needs of the access user and improve the satisfaction of the access user to the service agent.

Embodiment 5

An embodiment of the present invention provides an intelligent decision device. As shown in FIG. 12, the smart decision device includes an interface circuit 30, a processor 31, a memory 32, and a system bus 33.

The interface circuit 30, the processor 31, and the memory 32 pass through the system. The bus 33 is connected and communicates with each other.

When the intelligent decision device is in operation, the intelligent decision device performs the method of selecting a service agent of the embodiment as described in any one of FIGS. 4-8. For a specific method for selecting a service agent, refer to the related description in the foregoing embodiment shown in any one of FIG. 4 to FIG. 8 , and details are not described herein again.

Specifically, the processor 31 may be a CPU, and may also be other general-purpose processors, other programmable logic devices or transistor logic devices, discrete hardware components, and the like. The general purpose processor may be a microprocessor or the processor or any conventional processor or the like.

Specifically, the memory 32 may include a volatile memory such as a RAM; the memory 32 may also include a non-volatile memory such as a ROM, a flash memory, an HDD or an SSD; the memory 32 may also include the above categories a combination of memory.

The system bus 33 can include a data bus, a power bus, a control bus, and a signal status bus. For the sake of clarity in the present embodiment, various buses are illustrated as the system bus 33 in FIG.

The identity of the service agent determined by the intelligent decision-making unit is obtained based on the voice characteristics of the access user and the matching degree between the access user and the service agent. Therefore, the service agent selected by the method for the access user is more suitable for accessing the user. Demand can also increase the satisfaction of access users to the services of the call center.

Embodiment 6

The embodiment of the present invention provides a routing controller. As shown in FIG. 13, the routing controller includes an interface circuit 40, a processor 41, a memory 42, and a system bus 43.

The interface circuit 40, the processor 41 and the memory 42 are connected by the system bus 43 and complete communication with each other.

When the routing controller is running, the routing controller performs the method of selecting a service agent of the embodiment of any of FIGS. 4-8. For a specific method for selecting a service agent, refer to the related description in the foregoing embodiment shown in any one of FIG. 4 to FIG. 8 , and details are not described herein again.

Specifically, the processor 41 may be a CPU, and may be other general-purpose processors, other programmable logic devices or transistor logic devices, discrete hardware components, and the like. The general purpose processor may be a microprocessor or the processor or any conventional processor or the like.

Specifically, the memory 42 may include a volatile memory such as a RAM; the memory 42 may also include a non-volatile memory such as a ROM, a flash memory, an HDD or an SSD; 42 may also include a combination of the above types of memory.

The system bus 43 can include a data bus, a power bus, a control bus, and a signal status bus. For the sake of clarity in the present embodiment, various buses are illustrated as the system bus 33 in FIG.

When the routing controller allocates a route to the access user, the routing controller selects the identifier of the service agent serving the access user according to the identifier of the service agent sent by the smart decision device, and the smart decision device sends the service to the routing controller. The identifier of the agent is determined by the smart decision device according to the matching degree between the access user and the service agent. Therefore, in the embodiment of the present invention, when the routing controller allocates a route to the access user, the access user and the service agent are indirectly considered. Matching degree, the service agent selected according to the matching degree is more accurate, can meet the needs of the access user, and improve the satisfaction of the access user to the service agent.

Example 7

An embodiment of the present invention provides a system for selecting a service agent. As shown in FIG. 14, the system includes an intelligent decision device 1 and a routing controller 2. The routing controller 2 and the intelligent decision device 1 are connected through a network.

In the system for selecting a service agent provided by the embodiment of the present invention, the intelligent decision device 1 and the routing controller 2 respectively perform the method flow for selecting a service agent according to the embodiment described in any one of FIG. 4-8. The corresponding steps complete the method for selecting a service agent in the embodiment of the present invention.

Further, as shown in FIG. 15, the system for selecting a service agent provided by the embodiment of the present invention further includes a computer telephony integration CTI3, and the CTI3 is connected to the routing controller 2 through a network.

Further, as shown in FIG. 15, the system for selecting a service agent provided by the embodiment of the present invention further includes an IVR4, and the IVR4 and the CTI3 are connected through a network.

The IVR4 is configured to acquire voice feature data of the access user by using a voice recognition technology, and perform discretization processing on the voice feature data of the access user according to a preset discretization rule, and The voice feature data discretized by the access user is sent to the CTI3.

It should be noted that the CTI3 and the routing controller 2 in the system for selecting a service agent provided by the embodiment of the present invention may be the same entity, and may be different entities. If the routing controller in the embodiment of the present invention is configured in the CTI, the steps performed by the routing controller in the embodiment of the present invention are all performed by the CTI, and the interaction steps between the CTI and the routing controller are performed by corresponding modules in the CTI. .

It will be apparent to those skilled in the art that for the convenience and brevity of the description, only the above The division of each functional module is illustrated. In practical applications, the above function assignment can be completed by different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. For the specific working process of the system, the device and the unit described above, reference may be made to the corresponding process in the foregoing method embodiments, and details are not described herein again.

In the several embodiments provided by the present application, it should be understood that the disclosed system, apparatus, and method may be implemented in other manners. For example, the device embodiments described above are merely illustrative. For example, the division of modules or units is only a logical function division. In actual implementation, there may be another division manner, for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in an electrical, mechanical or other form.

The units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

A person skilled in the art can understand that all or part of the steps of implementing the above method embodiments may be completed by using hardware related to the program instructions. The foregoing program may be stored in a computer readable storage medium, and the program is executed when executed. The foregoing steps include the steps of the foregoing method embodiments; and the foregoing storage medium includes: a medium that can store program codes, such as a ROM, a RAM, a magnetic disk, or an optical disk.

The above is only the specific embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily think of changes or substitutions within the technical scope of the present invention. It is within the scope of the invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

A method for selecting a service agent, characterized by comprising:

The intelligent decision device receives a request message for selecting a service agent for the access user, where the access user is a user currently accessing the call center;

And the smart decision device obtains, according to the request message, an identifier of the service agent that is determined according to the matching degree between the access user and the service agent, where the matching degree is calculated based on the voice feature of the access user;

The smart decision device sends the obtained identifier of the service agent to the routing controller; wherein the identifier of the service agent sent to the routing controller is used as a reference for the access user to select a service agent.
The method according to claim 1, wherein before the smart decision device receives the request message for selecting a service agent for the access user, the method further includes:

The intelligent decision device analyzes the historical call record in advance, and obtains a calculation model for calculating the matching degree between the voice feature and the service agent.
The method according to claim 2, wherein the intelligent decision device analyzes the historical call record in advance, and obtains a calculation model for calculating the matching degree between the voice feature and the service agent, and specifically includes:

The smart decision device selects a part of the call record from the historical call record as sampling data, and each call record includes a recording of the call, an identifier of the service agent of the call, and an identifier of the user of the call;

The intelligent decision device performs voice analysis on the recording in each of the call records to obtain voice feature data corresponding to each call record, and the voice feature data and the service agent in the corresponding call record The identification forms a data pair;

The intelligent decision device uses the preset algorithm to train the data pair to obtain a calculation model for calculating the matching degree between the voice feature and the service agent.
The method according to claim 3, wherein the request message carries voice feature data of the access user;

The smart decision device obtains the identifier of the service agent that is determined according to the matching degree between the access user and the service agent, and specifically includes:

The intelligent decision device inputs voice feature data of the access user to the calculation model The calculation model calculates the matching degree between the voice feature of the access user and the service agent, and obtains the identifier of the service agent whose matching degree with the voice feature of the access user is within a preset range.
The method according to claim 3, wherein after the smart decision device obtains the calculation model, the method further comprises:

The intelligent decision device analyzes voice feature data of each user in the historical call record;

The smart decision device inputs the voice feature data of the analyzed user into the calculation model for calculation, and obtains an identifier of the service agent whose matching degree with the analyzed user's voice feature is within a preset range;

The smart decision device saves a correspondence between the identifier of the analyzed user and the identifier of the service agent whose matching degree with the analyzed user's voice feature is within a preset range.
The method according to claim 5, wherein the request message carries an identifier of the access user;

The smart decision device obtains, according to the request message, the identifier of the service agent that is determined according to the matching degree between the access user and the service agent, and specifically includes:

The smart decision device searches for the identifier of the service agent corresponding to the identifier of the access user in the corresponding relationship according to the identifier of the access user.
The method according to any one of claims 3-6, wherein the intelligent decision device selects a partial call record from the historical call record as sampling data, including: the intelligent decision device from the history Select the call record with satisfactory quality evaluation as the sampling data in the call record;

The identifier of the service agent whose matching degree is within the preset range is the identifier of the service agent with the highest matching degree.
The method according to any one of claims 3-6, wherein the intelligent decision device selects a partial call record from the historical call record as sampling data, including: the intelligent decision device from the history In the call record, a call record whose quality evaluation is unsatisfactory is selected as sampling data;

The identifier of the service agent whose matching degree is within the preset range is the identifier of the service agent whose matching degree is greater than or equal to the preset threshold.
The method according to any one of claims 1-8, wherein after the call of the access user ends, the method further comprises:

The intelligent decision device collects a call record of the current call;

The intelligent decision device updates the computing model using the collected call records.
A method for selecting a service agent, characterized by comprising:

The routing controller sends a request message to the intelligent decision device, requesting the smart decision device to select a service agent for the access user currently accessing the call center;

The routing controller receives an identifier of the service agent that is determined by the smart decision device according to the matching degree between the access user and the service agent, and the matching degree is calculated based on the voice feature of the access user;

The routing controller selects an identifier of the service agent serving the access user by referring to the identifier of the received service agent.
The method according to claim 10, wherein the routing controller selects the identifier of the service agent that serves the access user service by referring to the identifier of the received service agent, and specifically includes:

The routing controller determines the identifier of the received service agent as the identifier of the service agent serving the access user.
The method according to claim 10, wherein the routing controller selects the identifier of the service agent that serves the access user service by referring to the identifier of the received service agent, and specifically includes:

The routing controller selects, from the identifier of the existing service agent, an identifier of any service agent other than the identifier of the received service agent as the identifier of the service agent serving the access user.
An intelligent decision device, comprising:

a receiving unit, configured to receive a request message for selecting a service agent for the access user, where the access user is a user currently accessing the call center;

An obtaining unit, configured to obtain, according to the request message received by the receiving unit, an identifier of a service agent that is determined according to a matching degree between the access user and a service agent, where the matching degree is based on the access user The speech characteristics are calculated;

a sending unit, configured to send, to the routing controller, an identifier of the service agent acquired by the acquiring unit, where the identifier of the service agent sent to the routing controller is used as the access user to select a service agent reference.
The intelligent decision device according to claim 13, wherein the intelligent decision device further comprises a generating unit,

The generating unit is configured to analyze the historical call record in advance before the receiving unit receives the request message for selecting the service agent for the access user, to obtain a calculation model for calculating the matching degree between the voice feature and the service agent.
The intelligent decision device according to claim 14, wherein the intelligent decision device further comprises a selection unit,

The selecting unit is configured to select a part of the call record from the historical call record as sampling data, where each call record includes a recording of the call, an identifier of the service agent of the call, and an identifier of the user of the call;

The generating unit is configured to perform voice analysis on the recording in each of the call records selected by the selecting unit, to obtain voice feature data corresponding to each call record, and the voice feature data and The identifier of the service agent in the corresponding call record forms a data pair, and is used to train the data pair by using a preset algorithm to obtain a calculation model for calculating the matching degree between the voice feature and the service agent.
The intelligent decision device according to claim 15, wherein the request message carries voice feature data of the access user;

The acquiring unit is specifically configured to input voice feature data of the access user into the computing model obtained by the generating unit, and calculate, by the computing model, a voice feature of the access user and a service agent. The degree of matching is obtained, and the identifier of the service agent whose matching degree with the voice feature of the access user is within a preset range is obtained.
The intelligent decision-making device according to claim 15, wherein the generating unit, after obtaining the calculation model, is further configured to:

Analyzing the voice feature data of each user in the historical call record;

Inputting the voice feature data of the analyzed user into the calculation model, respectively, to obtain an identifier of the service agent whose matching degree with the analyzed user's voice feature is within a preset range;

Corresponding relationship between the identifier of the analyzed user and the identifier of the service agent whose matching degree with the analyzed user's voice feature is within a preset range is saved.
The intelligent decision device according to claim 16, wherein the request message carries an identifier of the access user;

The obtaining unit is specifically configured to search for, according to the identifier of the access user, the identifier of the service agent corresponding to the identifier of the access user in the corresponding relationship generated by the generating unit.
The intelligent decision device according to any one of claims 15-18, characterized in that

The selecting unit is specifically configured to select, from the historical call record, a call record whose quality is evaluated as satisfactory as sampling data;

The identifier of the service agent whose matching degree is within the preset range is the identifier of the service agent with the highest matching degree.
The intelligent decision device according to any one of claims 15-18, characterized in that

The selecting unit is specifically configured to select, from the historical call record, a call record whose quality is evaluated as unsatisfactory as sampling data;

The identifier of the service agent whose matching degree is within the preset range is the identifier of the service agent whose matching degree is greater than or equal to the preset threshold.
The intelligent decision device according to any one of claims 13 to 20, characterized in that

The intelligent decision device further includes an acquisition unit,

The collecting unit is configured to collect a call record of the current call;

The intelligent decision device further includes a processing unit,

The processing unit is configured to update the calculation model by using a call record collected by the collection unit.
A routing controller, comprising:

a sending unit, configured to send a request message to the smart decision device, requesting the smart decision device to select a service agent for an access user currently accessing the call center;

The receiving unit is further configured to receive, by the smart decision device, an identifier of the service agent that is determined according to the matching degree between the access user and the service agent, where the matching degree is calculated based on the voice feature of the access user;

And a processing unit, configured to refer to an identifier of the service agent received by the receiving unit, and select an identifier of a service agent serving the access user.
The routing controller of claim 22 wherein:

The processing unit is specifically configured to determine an identifier of the service agent received by the receiving unit as an identifier of a service agent serving the access user.
A routing controller according to claim 23, wherein

The processing unit is specifically configured to: select, from the identifier of the existing service agent, an identifier of any service agent other than the identifier of the received service agent as a service seat serving the access user. The logo of the seat.
A system for selecting a service agent, comprising: the intelligent decision device according to any one of claims 13 to 21, and the routing controller according to any one of claims 22-24, wherein The routing controller is connected to the intelligent decision device.
The system of claim 25, wherein said system further comprises a computer telephony integrated CTI, said CTI being coupled to said routing controller;

The CTI is configured to receive a call request initiated by an access user, and send, according to the call request, a route assignment message that allocates a route to the access user to the routing controller, and is configured to receive the route controller sent by the routing controller. An identifier and a target queue of the service agent serving the access user, and connecting the access user to the service agent serving the access user.
The system of claim 26, wherein the system further comprises an interactive voice response IVR, the IVR being coupled to the CTI;

The IVR is configured to acquire voice feature data of the access user, and send the voice feature data of the access user to the CTI.