CN112671984A - Service mode switching method and device, robot customer service and storage medium - Google Patents
Service mode switching method and device, robot customer service and storage medium Download PDFInfo
- Publication number
- CN112671984A CN112671984A CN202011388418.6A CN202011388418A CN112671984A CN 112671984 A CN112671984 A CN 112671984A CN 202011388418 A CN202011388418 A CN 202011388418A CN 112671984 A CN112671984 A CN 112671984A
- Authority
- CN
- China
- Prior art keywords
- customer service
- customer
- emotion
- robot
- service mode
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Manipulator (AREA)
Abstract
The embodiment of the application provides a service mode switching method and device, a robot customer service and a storage medium. The service mode switching method comprises the following steps: acquiring at least one characteristic data of a customer interacting with the robot customer service in a robot customer service mode; identifying an emotional state of the customer based on the at least one characteristic data; and if the emotional state is a preset emotional type, switching from the robot customer service mode to an artificial customer service mode. Therefore, the problem that in the prior art, service mode switching is complicated, and service quality is affected is solved.
Description
Technical Field
The embodiment of the application relates to the technical field of artificial intelligence, in particular to a service mode switching method and device, a robot customer service system and a storage medium.
Background
More and more Customer Service (Customer Service) systems support an artificial Customer Service mode and a robot Customer Service mode. In order to improve the customer service processing efficiency, the customer service system preferentially enters a robot customer service mode, the robot customer service is used for processing customer service work, and when the robot customer service cannot well meet the customer service requirement, a customer can request to switch to a manual customer service mode, and the manual customer service is used for processing the customer service work. The client requests to switch to the manual customer service mode, so that the operation burden of the user is increased, the client requirements cannot be solved in time, and the service quality is influenced.
Disclosure of Invention
The embodiment of the application provides a service mode switching method and device, a robot customer service and a storage medium, which are used for solving the technical problem that in the prior art, service mode switching is complicated, so that service quality is influenced.
The embodiment of the application provides a service mode switching method, which comprises the following steps:
acquiring at least one characteristic data of a customer interacting with the robot customer service in a robot customer service mode;
identifying an emotional state of the customer based on the at least one characteristic data;
and if the emotional state is a preset emotional type, switching from the robot customer service mode to an artificial customer service mode.
An embodiment of the present application further provides a service mode switching apparatus, including:
the system comprises an acquisition module, a service module and a service module, wherein the acquisition module is used for acquiring at least one characteristic data of a client interacting with the robot service in a robot service mode;
an identification module for identifying an emotional state of the customer based on the at least one characteristic data;
and the switching module is used for switching from the robot customer service mode to the manual customer service mode if the emotion state is a preset emotion type.
The embodiment of the application also provides a robot customer service, which comprises a storage component, a display component and a processing component; the storage component stores one or more computer program instructions; the one or more computer program instructions for invocation and execution by the processing component;
the processing component is to:
acquiring at least one characteristic data of a customer interacting with the robot customer service in a robot customer service mode;
identifying an emotional state of the customer based on the at least one characteristic data;
and if the emotional state is a preset emotional type, switching from the robot customer service mode to an artificial customer service mode.
Embodiments of the present application also provide a computer readable storage medium storing a computer program, which, when executed by a processor, causes the processor to implement the steps in the service mode switching method.
According to the service mode switching method and device, the robot customer service and the storage medium, the robot customer service can identify the emotion state of a customer interacting with the robot customer service, when the situation state of the customer is a preset emotion type, the robot customer service automatically requests to switch to the manual customer service mode, the customer service is timely served by the more humanized manual customer service, the customer complaint rate is reduced, and the customer service quality is guaranteed. Meanwhile, the client does not need to request to switch to the manual customer service mode, and the operation burden of the client is not increased.
These and other aspects of the present application will be more readily apparent from the following description of the embodiments.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a schematic method flow diagram illustrating an embodiment of a service mode switching method according to the present application;
FIG. 2 is a schematic diagram illustrating an embodiment of a customer service system provided herein;
fig. 3 is a schematic structural diagram illustrating an embodiment of a service mode switching apparatus provided in the present application;
FIG. 4 illustrates a schematic diagram of a robot customer service in accordance with one embodiment of the present application.
Detailed Description
In order to make the technical solutions better understood by those skilled in the art, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application.
In some of the flows described in the specification and claims of this application and in the above-described figures, a number of operations are included that occur in a particular order, but it should be clearly understood that these operations may be performed out of order or in parallel as they occur herein, the number of operations, e.g., 101, 102, etc., merely being used to distinguish between various operations, and the number itself does not represent any order of performance. Additionally, the flows may include more or fewer operations, and the operations may be performed sequentially or in parallel. It should be noted that, the descriptions of "first", "second", etc. in this document are used for distinguishing different messages, devices, modules, etc., and do not represent a sequential order, nor limit the types of "first" and "second" to be different.
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The technical solutions provided by the embodiments of the present application are described in detail below with reference to the accompanying drawings.
Fig. 1 is a schematic method flow diagram illustrating an embodiment of a service mode switching method provided in the present application. The service mode switching method provided in the embodiment of the present Application may be applied to APP (Application program), browser (browser), and applet, but is not limited thereto. The applet is a light-weight application program (light application for short) which can be used without downloading and installation, and the light application is usually characterized by easy implementation, simple and efficient functions, convenient use and the like. With the development of computer technology, a new application development mode is brought, for example, a provider of large platform application services provides an efficient development framework for developers besides providing own services for end users based on own platform applications, and a functional module of the platform itself or third-party service applications can be developed and deployed in a very efficient manner through the provided development framework and access to a platform main application with good compatibility and stability. The application developed in the development mode can realize the advantages, and can also realize the situation of mutual benefit among multiple parties by depending on the advantages of brands, users and the like of the application platform, for example, on one hand, the access of the application enriches the services of the platform, on the other hand, the popularization cost of the accessed application is saved, more importantly, the access time of the user is saved, and the reaching efficiency of the application is improved. The light application can run depending on the main application, that is, the main application is used as a running container to run, for example, the light application can be started from an interface provided in the main application, the main application distributes a running environment and the like, and meanwhile, certain independence is kept on functions, running characteristics and the like.
As shown in fig. 1, the service mode switching method may include the steps of:
The service mode switching method provided by the embodiment of the application can be applied to various customer service scenes, such as an e-commerce platform customer service scene, a takeaway platform customer service scene, a bank customer service scene and a telecommunication customer service scene, but is not limited to the above-mentioned exemplary scenes.
With the development of the artificial intelligence technology, the robot has more and more intelligent customer service, can help to process the customer service requirement of the customer, and improves the customer service processing efficiency. For example, when a customer has customer service requirements, the customer can interact with the robot customer service, for example, the customer consults a problem with the robot customer service, initiates a complaint, proposes a suggestion, and the like; and the robot serves the customer service requirement and returns a corresponding customer service processing result to the customer.
Due to individual differences of customers, answers of the robot customer service may not be satisfactory for some customers, and when the robot customer service cannot well meet customer service requirements, customer service experience of the customers is poor, and the customers have negative emotions such as anxiety, anger and the like. At this time, if the customer service mode is not switched to the manual customer service mode, the customer complaints are likely to occur.
In the embodiment of the application, when the robot customer service interacts with the customer, at least one characteristic data of the customer can be acquired to identify the emotional state of the customer. The at least one characteristic data of the client is, for example, but not limited to, voice data, text data, and historical behavior data.
The emotion of the client can be expressed by voice data in the process of interaction between the client and the robot. When the client is in different emotional states, speech characteristics such as speech speed, fundamental frequency characteristics and the like in the speech data are different. The pace of speech indicates the degree of moderation of the tone of speech, so that the pace of speech is increased when a person is angry and happy, the pace of speech is basically unchanged when the person is calm, and the pace of speech is decreased when the person is sad. The fundamental frequency features describe prosody change features of the voice, for example, the fundamental frequency change ranges of happy sentences and angry sentences are large, and the fundamental frequency change ranges of sad sentences and calm sentences are small. The average value of the fundamental frequency of the sad sentences is large, and the average value of the fundamental frequency of the happy sentences, the angry sentences and the calm sentences is small. Of course, the emotional state of the client can also be analyzed based on other voice characteristics, and more description about the emotional state of the client by voice characteristics is detailed in the related art.
The emotion of the client can be expressed by text data in the process of interaction between the client and the robot. When the client is in different emotional states, some words with emotional tendency may be included in the text data, for example, when the client is angry, the text data may include words such as "too bad", "no language", and the like, and when the client is happy, the text data may include words such as "too much", "good", and the like. Of course, semantic analysis can be performed on the text data to determine the emotional tendency corresponding to the text data.
Historical behavioral data of the customer may help characterize the customer's historical mood. The historical behavior data comprises, for example, the number of historical complaints and the number of historical artificial customer service requests, and the larger the number of historical complaints or the larger the number of historical artificial customer service requests is, the more negative the historical emotion of the customer is, the more artificial customer service is needed to serve the customer, so as to reduce the complaint rate of the customer.
And 102, identifying the emotional state of the client according to the at least one characteristic data.
And 103, if the emotion state is a preset emotion type, switching from the robot customer service mode to the manual customer service mode.
In the embodiment of the application, the emotion types can be divided into positive emotion, neutral emotion and negative emotion. Positive emotions are positive emotions, such as happiness. Negative emotions are one that is not positive, and include, for example, angry, sad, etc. Neutral mood is a mood that is intermediate between positive and negative mood, such as calmness. Of course, the mood class can be classified according to the actual situation.
The emotion types are divided into positive emotion, neutral emotion and negative emotion, and the preset emotion type is the negative emotion. In practical application, when the robot customer service identifies that a customer is in a negative emotion, the robot customer service indicates that the robot customer service may not be capable of well solving customer service requirements and the customer service experience of the customer is poor, and at the moment, the robot customer service automatically requests to switch to an artificial customer service mode, the customer is served by more humanized artificial customer service in time, the customer complaint rate is reduced, and the customer service quality is guaranteed.
According to the service mode switching method provided by the embodiment of the application, the robot customer service can identify the emotion state of the customer interacting with the robot customer service, when the situation state of the customer is a preset emotion type, the robot customer service automatically requests to switch to the manual customer service mode, the more humanized manual customer service timely serves the customer, the customer complaint rate is reduced, and the customer service quality is guaranteed. Meanwhile, the client does not need to request to switch to the manual customer service mode, and the operation burden of the client is not increased.
In some embodiments of the present application, one possible implementation manner of step 102 is: determining a current customer service scene; if the current customer service scene is a preset customer service scene, processing characteristic data of the client by adopting a first emotion recognition model so as to recognize the emotion state of the client; if the current customer service scene is not the preset customer service scene, processing characteristic data of the client by adopting a second emotion recognition model so as to recognize the emotion state of the client; wherein the recognition accuracy of the first emotion recognition model is higher than the recognition accuracy of the second emotion recognition model.
In practical application, different application scenes have different requirements on the satisfaction degree of a client, so that the judgment standards of negative emotions of the client in different application scenes can be different. For example, the requirement of the e-commerce platform customer service scene on the customer satisfaction is lower than that of the take-away platform customer service scene, the voice data of the same customer, namely My order fast-spot delivery, is judged to be neutral emotion in the e-commerce platform customer service scene, and is judged to be negative emotion in the take-away platform customer service scene.
Therefore, the robot customer service can combine at least one characteristic data of the customer and the current customer service scene to more accurately identify the emotional state of the customer, so that the robot customer service can more accurately request to switch to the manual customer service mode, and the customer service quality is ensured.
In the embodiment of the application, if the robot service is set with the service scene, the current service scene is the service scene set by the robot service. And if the customer service scene is not set by the robot customer service, performing semantic analysis on the text data of the interaction between the customer and the robot customer service to determine the current customer scene. For example, in an e-commerce platform customer service scenario, some textual data related to electronic shopping may appear from customer and robot customer service interactions; in a customer service scene of a take-away platform, some text data related to take-away delivery may appear in customer and robot service interaction; in a bank customer service scene, text data related to banking business may appear when a customer interacts with a robot customer service; in a teleservice scenario, some text data related to teleservice may appear when a customer interacts with a robot service.
When the recognition accuracy of the model is improved, the recognition speed of the model is reduced; when the recognition speed of the model is increased, the recognition accuracy of the model is reduced. Therefore, in the embodiment of the application, the first emotion recognition model is adopted to more accurately recognize the emotional state of the client aiming at the client scene with higher requirement on the satisfaction degree of the client. And aiming at a client scene with low requirement on the satisfaction degree of the client, the second emotion recognition model is adopted to recognize the emotion state of the client more quickly. Therefore, the second emotion recognition model with low recognition accuracy can meet the requirement of customer service processing efficiency of most customer scenes, the first emotion recognition model with high recognition accuracy can meet the customer service requirement of the customer service scene with high customer service satisfaction, the customer service system is controlled to be switched to the manual customer service mode more accurately, the customer service complaint rate is reduced, and the customer service quality is improved.
The preset customer service scene is set according to actual requirements, such as a takeout platform customer service scene, a bank customer service scene and a telecommunication customer service scene, and is a customer scene with high customer satisfaction requirement; the e-commerce platform customer service scene is a customer scene with low customer satisfaction requirements.
In the embodiment of the application, a first emotion recognition model is obtained by training a first training sample and a training label to which the first training sample belongs; and the second emotion recognition model is obtained by training a second training sample and the training label to which the second training sample belongs. Training labels are emotion categories, such as positive, neutral, negative emotion. More details of model training are described in the related art.
In practical application, the identification accuracy of the model can be improved by increasing the number of training samples. Therefore, as an alternative, the number of first training samples for training the first emotion recognition model is greater than the number of second training samples for training the second emotion recognition model, so that the recognition accuracy of the first emotion recognition model is higher than that of the second emotion recognition model.
As another alternative, the model structures used for the first emotion recognition model and the second emotion recognition model may be different, and the first emotion recognition model may use a model structure whose recognition accuracy is higher than that of the model structure of the second emotion recognition model. For example, the first emotion recognition model may employ a neural network model, and the second emotion recognition model may employ a linear model. The recognition accuracy of the neural network model is higher than that of the linear model. The linear model is, for example, a linear regression model.
Further, the robot customer service may also determine the current customer service scenario by using the attribute information of the customer, where the attribute information of the customer includes whether the customer is a vip (very important customer), gender, age, and the like. For example, for a VIP client, a female, or an elderly client with an age greater than a preset age, the current customer service scenario may be considered as a preset customer service scenario; and the non-VIP client, the male or the non-elderly client with the age not greater than the preset age may consider that the current customer service scenario is not the preset customer service scenario.
Further, the robot customer service can also determine whether to adopt the first emotion recognition model or the second emotion recognition model to recognize the emotional state of the customer by combining the current customer service scene and the attribute information of the customer. The attribute information of the client includes whether it is a vip (very important) client, gender, age, and the like. Whether the customer is easy to initiate complaints can be analyzed according to the attribute information of the customer, and a specific application can set which attribute is easy to initiate complaints according to the actual situation. For example, for a VIP client, a female client or an old client with an age greater than a preset age, whether the current customer service scene is a preset customer service scene or not, the first emotion recognition model may be used to recognize the emotional state of the client.
In some embodiments of the present application, yet another possible implementation manner of step 102 is: identifying an emotional state of the customer using a third emotion recognition model based on the at least one characteristic data; if the recognition result of the third emotion recognition model is the preset emotion type, recognizing the emotion state of the client by using a fourth emotion recognition model according to at least one characteristic datum; wherein the recognition accuracy of the third emotion recognition model is lower than the recognition accuracy of the second emotion recognition model; correspondingly, one implementation manner of step 103 is: and if the recognition result of the fourth emotion recognition model is the preset emotion type, switching from the robot customer service mode to the manual customer service mode.
When the recognition accuracy of the model is improved, the recognition speed of the model is reduced; when the recognition speed of the model is increased, the recognition accuracy of the model is reduced. Therefore, in order to give consideration to both the customer service processing efficiency and the service mode switching accuracy, the robot customer service firstly adopts the third emotion recognition model with low recognition accuracy to perform emotion recognition, and only when the customer is recognized to be in the preset emotion type, the fourth emotion recognition model is adopted to perform emotion recognition.
The third emotion recognition model is obtained by training a third training sample and a training label to which the third training sample belongs; and the fourth emotion recognition model is obtained by training a fourth training sample and the training label to which the fourth training sample belongs. Training labels are emotion categories, such as positive, neutral, negative emotion.
In practical application, the identification accuracy of the model can be improved by increasing the number of training samples. Therefore, as an alternative, the number of the fourth training samples for training the fourth emotion recognition model is greater than the number of the third training samples for training the third emotion recognition model, so that the recognition accuracy of the fourth emotion recognition model is higher than that of the third emotion recognition model.
As another alternative, the third emotion recognition model and the fourth emotion recognition model may have different model structures, and the fourth emotion recognition model may have a model structure whose recognition accuracy is higher than that of the model structure of the third emotion recognition model. For example, the fourth emotion recognition model may be a neural network model, and the third emotion recognition model may employ a linear model. The recognition accuracy of the neural network model is higher than that of the linear model. The linear model is, for example, a linear regression model.
In some embodiments of the present application, in order to determine the emotional state of the client more scientifically and objectively, one possible implementation of step 102 is: respectively processing the voice data, the text data and the historical behavior data to obtain a first emotion score, a second emotion score and a third emotion score of the client; weighting the first emotion score, the second emotion score and the third emotion score to obtain a total emotion score of the client; the emotional state of the customer is determined based on the customer's total emotional score.
Specifically, since the voice data includes a large number of emotional states, when the weight is assigned, the weight of the voice data is greater than that of the text data, the weight of the text data is greater than that of the historical behavior data, and the sum of the weights is 1. After obtaining the total emotional score of the customer, the total emotional score can be compared with the first threshold value and the second threshold value to judge the emotional state of the customer. Wherein the first threshold is greater than the second threshold. If the total emotion score is not less than the first threshold, the customer is in a positive emotion; if the total sentiment score is less than the first threshold but not less than the second threshold, the customer is in a neutral sentiment; if the total sentiment score is less than the second threshold, the customer is in a negative sentiment.
When the first emotion recognition model, the second emotion recognition model, the third emotion recognition model and the fourth emotion recognition model are used for processing voice data, text data and historical behavior data of a client, a first emotion score can be calculated based on the voice data, a second emotion score can be calculated based on the text data, a third emotion score can be calculated based on the historical behavior data, and the first emotion score, the second emotion score and the third emotion score are weighted to obtain a total emotion score of the client; the emotional state of the customer is determined based on the customer's total emotional score.
In some embodiments of the present application, after switching from the robot customer service mode to the manual customer service mode, service suggestion information corresponding to a preset emotion type may be further provided to the manual customer service for the manual customer service to view.
Specifically, the service recommendation information is, for example, some reference language information, which may help to placate the mood of the customer. For example, in the customer service scene of the take-away platform, the reference tactical information is "parent, we arrange to dispatch your order immediately, apology brings inconvenience to you", "apology brings inconvenience to you due to our work error", and the like. In addition, the service suggestion information can also comprise attribute information and historical behavior data of the client, so that the manual client can quickly master the relevant information of the client.
In some embodiments of the application, after the robot customer service mode is switched to the manual customer service mode, the emotion change trend of the customer after the switching to the manual customer service mode can be obtained; and evaluating the service quality of the manual customer service according to the emotion change trend of the customer. Therefore, the service quality of the artificial customer service is evaluated, the artificial customer service is helped to improve the customer service level of the customer service, and the emotion change trend of the customer can guide the artificial customer service to adjust the own skills.
Specifically, in the process of interaction between the artificial customer service and the customer, the voice data of the customer during the interaction with the artificial customer service can be acquired at intervals, the emotional state of the customer is identified according to the voice data of the customer at each interval, and the emotional change trend of the customer is determined according to the emotional state of the customer at each interval. If the emotion change trend of the client indicates that the switching time spent by the client for switching from negative emotion to neutral emotion or positive emotion is shorter than the preset switching time, the artificial customer service is considered to placate the emotion of the client in a shorter time, and the service quality of the artificial client is better. If the emotion change trend of the client indicates that the switching time spent by the client for switching from negative emotion to neutral emotion or positive emotion is not less than the preset switching time, the client emotion is pacified after the artificial customer service spends a long time, and the service quality of the artificial client is poor.
In practical application, the robot customer service can acquire the voice data of the customer when interacting with the artificial customer service, and the robot customer identifies the voice data of the customer to acquire the emotional state of the customer when interacting with the artificial customer service. And determining the emotion change trend according to the emotion states of the clients in the multiple time periods when the clients interact with the artificial customer service.
Fig. 2 is a schematic structural diagram illustrating an embodiment of a customer service system provided in the present application. The whole customer service system comprises a robot customer service 2 and a terminal device 3 of an artificial customer service. The client at the user terminal 1 side initiates a customer service demand to the robot customer service 2, the robot customer service 2 processes the customer service demand and generates a customer service processing result, and the customer service processing result is sent to the user terminal 1 so that the client can obtain the required customer service processing result. Meanwhile, the robot customer service 2 identifies the emotional state of the client, and when the emotional state of the client is identified to be in a negative emotion, sends notification information to the terminal device 3 of the artificial customer service to notify the artificial customer service to serve the client. In the manual customer service stage, the customer service appeal is sent to the terminal device 3 of the manual customer service through the robot customer service 2, and the terminal device 3 of the manual customer service sends a customer service processing result to the user terminal 1 side through the robot customer service 2.
Fig. 3 is a schematic structural diagram illustrating an embodiment of a service mode switching apparatus provided in the present application. The service mode switching device is an execution main body of the service mode switching method, can be realized by software and/or hardware, and can be configured in electronic equipment such as a mobile phone, a tablet computer, a wearable device and the like. As shown in fig. 3, the service mode switching means may include:
an obtaining module 10, configured to obtain, in a robot service mode, at least one feature data of a customer interacting with a robot service;
an identification module 20 for identifying an emotional state of the customer based on the at least one characteristic data;
and the switching module 30 is used for switching from the robot customer service mode to the manual customer service mode if the emotion state is a preset emotion type.
Further, the identification module 20 is specifically configured to:
determining a current customer service scene;
if the current customer service scene is a preset customer service scene, processing characteristic data of the client by adopting a first emotion recognition model so as to recognize the emotion state of the client;
if the current customer service scene is not the preset customer service scene, processing characteristic data of the client by adopting a second emotion recognition model so as to recognize the emotion state of the client;
wherein the recognition accuracy of the first emotion recognition model is higher than the recognition accuracy of the second emotion recognition model.
Further, the first emotion recognition model is obtained by training with the first training sample and the training label to which the first training sample belongs;
and the second emotion recognition model is obtained by training a second training sample and the training label to which the second training sample belongs.
Further, the identification module 20 is specifically configured to:
identifying an emotional state of the customer using a third emotion recognition model based on the at least one characteristic data;
if the recognition result of the third emotion recognition model is the preset emotion type, recognizing the emotion state of the client by using a fourth emotion recognition model according to at least one characteristic datum; wherein the recognition accuracy of the third emotion recognition model is lower than the recognition accuracy of the second emotion recognition model;
if the emotional state is a preset emotional type, switching from the robot customer service mode to the manual customer service mode comprises:
and if the recognition result of the fourth emotion recognition model is the preset emotion type, switching from the robot customer service mode to the manual customer service mode.
Further, the at least one characteristic data includes: voice data in the customer service interaction process of the customer and the robot, text data in the customer service interaction process of the customer and the robot and historical behavior data of the customer.
Further, the identification module 20 is specifically configured to:
respectively processing the voice data, the text data and the historical behavior data to obtain a first emotion score, a second emotion score and a third emotion score of the client;
weighting the first emotion score, the second emotion score and the third emotion score to obtain a total emotion score of the client;
the emotional state of the customer is determined based on the customer's total emotional score.
Further, the switching module 30, after switching from the robot customer service mode to the manual customer service mode, is further configured to:
and providing service suggestion information corresponding to the preset emotion type for the manual customer service to check.
Further, the obtaining module 10 is further configured to: after the robot customer service mode is switched to the manual customer service mode, acquiring the emotion change trend of the client after the switching to the manual customer service mode is performed; and evaluating the service quality of the manual customer service according to the emotion change trend of the customer.
The specific manner in which each module and unit of the service mode switching device in the above embodiments perform operations has been described in detail in the embodiments related to the method, and will not be described in detail here.
FIG. 4 illustrates a schematic diagram of a robot customer service in accordance with one embodiment of the present application. As shown in fig. 4, the robot service includes: a storage component 11, a display component 13 and a processing component 12; the storage component 11 stores one or more computer program instructions; one or more computer program instructions for invocation and execution by the processing component 12;
the processing assembly 12 is configured to:
acquiring at least one characteristic data of a customer interacting with the robot in a robot customer service mode;
identifying an emotional state of the customer based on the at least one characteristic data;
and if the emotional state is a preset emotional type, switching from the robot customer service mode to the manual customer service mode.
Further, the processing component 12 identifies, based on the at least one characteristic data, an emotional state of the customer, in particular:
determining a current customer service scene;
if the current customer service scene is a preset customer service scene, processing characteristic data of the client by adopting a first emotion recognition model so as to recognize the emotion state of the client;
if the current customer service scene is not the preset customer service scene, processing characteristic data of the client by adopting a second emotion recognition model so as to recognize the emotion state of the client;
wherein the recognition accuracy of the first emotion recognition model is higher than the recognition accuracy of the second emotion recognition model.
Further, the first emotion recognition model is obtained by training with the first training sample and the training label to which the first training sample belongs;
and the second emotion recognition model is obtained by training a second training sample and the training label to which the second training sample belongs.
Further, the processing component 12 identifies, based on the at least one characteristic data, an emotional state of the customer, in particular:
identifying an emotional state of the customer using a third emotion recognition model based on the at least one characteristic data;
if the recognition result of the third emotion recognition model is the preset emotion type, recognizing the emotion state of the client by using a fourth emotion recognition model according to at least one characteristic datum; wherein the recognition accuracy of the third emotion recognition model is lower than the recognition accuracy of the second emotion recognition model;
if the emotional state is a preset emotional type, switching from the robot customer service mode to the manual customer service mode comprises:
and if the recognition result of the fourth emotion recognition model is the preset emotion type, switching from the robot customer service mode to the manual customer service mode.
Further, the at least one characteristic data includes: voice data in the customer service interaction process of the customer and the robot, text data in the customer service interaction process of the customer and the robot and historical behavior data of the customer.
Further, the processing component 12 identifies, based on the at least one characteristic data, an emotional state of the customer, in particular:
respectively processing the voice data, the text data and the historical behavior data to obtain a first emotion score, a second emotion score and a third emotion score of the client;
weighting the first emotion score, the second emotion score and the third emotion score to obtain a total emotion score of the client;
the emotional state of the customer is determined based on the customer's total emotional score.
Further, the processing component 12, after switching from the robot customer service mode to the manual customer service mode, is further configured to:
and providing service suggestion information corresponding to the preset emotion type for the manual customer service to check.
Further, the processing component 12, after switching from the robot customer service mode to the manual customer service mode, is further configured to: acquiring the emotion change trend of the client after switching to the manual customer service mode; and evaluating the service quality of the manual customer service according to the emotion change trend of the customer.
Wherein the processing component 12 may include one or more processors to execute computer instructions to perform all or some of the steps of the methods described above. Of course, the processing component 12 may also be implemented as one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, micro-controllers, microprocessors or other electronic components configured to perform the above-described methods.
The storage component 11 is configured to store various types of data to support operations at the terminal. The storage component 11 may be implemented by any type or combination of volatile or non-volatile memory devices, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.
The display element 13 may be an Electroluminescent (EL) element, a liquid crystal display or a micro display having a similar structure, or a laser scanning type display in which the retina can directly display or the like.
Of course, a computing device may also necessarily include other components, such as input/output interfaces, communication components, and so forth.
The input/output interface provides an interface between the processing component 12 and peripheral interface modules, which may be output devices, input devices, etc.
The communication component is configured to facilitate wired or wireless communication between the computing device and other devices, and the like.
An embodiment of the present application further provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a computer, the service mode switching method according to the embodiment shown in fig. 1 may be implemented.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods of the various embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions in the embodiments of the present application.
Claims (11)
1. A method for service mode switching, comprising:
acquiring at least one characteristic data of a customer interacting with the robot customer service in a robot customer service mode;
identifying an emotional state of the customer based on the at least one characteristic data;
and if the emotional state is a preset emotional type, switching from the robot customer service mode to an artificial customer service mode.
2. The method of claim 1, wherein identifying an emotional state of the customer based on the at least one characteristic data comprises:
determining a current customer service scene;
if the current customer service scene is a preset customer service scene, processing the characteristic data of the customer by adopting a first emotion recognition model so as to recognize the emotion state of the customer;
if the current customer service scene is not the preset customer service scene, processing the characteristic data of the customer by adopting a second emotion recognition model so as to recognize the emotion state of the customer;
wherein the recognition accuracy of the first emotion recognition model is higher than the recognition accuracy of the second emotion recognition model.
3. The method according to claim 2, wherein the first emotion recognition model is obtained by training with a first training sample and a training label to which the first training sample belongs;
and the second emotion recognition model is obtained by training a second training sample and a training label to which the second training sample belongs.
4. The method of claim 1, wherein identifying an emotional state of the customer based on the at least one characteristic data comprises:
identifying an emotional state of the customer using a third emotion recognition model based on the at least one characteristic data;
if the recognition result of the third emotion recognition model is a preset emotion type, recognizing the emotion state of the client by using a fourth emotion recognition model according to the at least one characteristic data; wherein the recognition accuracy of the third emotion recognition model is lower than the recognition accuracy of the second emotion recognition model;
if the emotional state is a preset emotional type, switching from the robot customer service mode to the manual customer service mode comprises:
and if the recognition result of the fourth emotion recognition model is a preset emotion type, switching from the robot customer service mode to an artificial customer service mode.
5. The method according to any one of claims 1 to 4, wherein the at least one characteristic data comprises: voice data in the customer service interaction process with the robot, text data in the customer service interaction process with the robot, and historical behavior data of the customer.
6. The method of claim 5, wherein identifying an emotional state of the customer based on the at least one characteristic data comprises:
processing the voice data, the text data and the historical behavior data respectively to obtain a first emotion score, a second emotion score and a third emotion score of the customer;
weighting the first emotion score, the second emotion score and the third emotion score to obtain a total emotion score of the client;
determining an emotional state of the customer based on the customer's total emotional score.
7. The method of any of claims 1 to 4, further comprising, after switching from the robotic customer service mode to a manual customer service mode:
and providing the service suggestion information corresponding to the preset emotion type for the manual customer service to check.
8. The method of any of claims 1 to 4, further comprising, after switching from the robotic customer service mode to a manual customer service mode:
acquiring the emotion change trend of the client after switching to the manual customer service mode;
and evaluating the service quality of the artificial customer service according to the emotion change trend of the customer.
9. A service mode switching apparatus, comprising:
the system comprises an acquisition module, a service module and a service module, wherein the acquisition module is used for acquiring at least one characteristic data of a client interacting with the robot service in a robot service mode;
an identification module for identifying an emotional state of the customer based on the at least one characteristic data;
and the switching module is used for switching from the robot customer service mode to the manual customer service mode if the emotion state is a preset emotion type.
10. The robot customer service system is characterized by comprising a storage component, a display component and a processing component; the storage component stores one or more computer program instructions; the one or more computer program instructions for invocation and execution by the processing component;
the processing component is to:
acquiring at least one characteristic data of a customer interacting with the robot customer service in a robot customer service mode;
identifying an emotional state of the customer based on the at least one characteristic data;
and if the emotional state is a preset emotional type, switching from the robot customer service mode to an artificial customer service mode.
11. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, causes the processor to carry out the steps of the method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011388418.6A CN112671984B (en) | 2020-12-01 | 2020-12-01 | Service mode switching method and device, robot customer service and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011388418.6A CN112671984B (en) | 2020-12-01 | 2020-12-01 | Service mode switching method and device, robot customer service and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112671984A true CN112671984A (en) | 2021-04-16 |
| CN112671984B CN112671984B (en) | 2022-09-23 |
Family
ID=75400817
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011388418.6A Active CN112671984B (en) | 2020-12-01 | 2020-12-01 | Service mode switching method and device, robot customer service and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112671984B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114971652A (en) * | 2022-05-16 | 2022-08-30 | 山东浪潮爱购云链信息科技有限公司 | Online customer service method, equipment and medium for enterprise purchasing platform |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102270450A (en) * | 2010-06-07 | 2011-12-07 | 株式会社曙飞电子 | System and method of multi model adaptation and voice recognition |
| CN107818798A (en) * | 2017-10-20 | 2018-03-20 | 百度在线网络技术（北京）有限公司 | Customer service quality evaluating method, device, equipment and storage medium |
| CN108122552A (en) * | 2017-12-15 | 2018-06-05 | 上海智臻智能网络科技股份有限公司 | Voice mood recognition methods and device |
| CN108922564A (en) * | 2018-06-29 | 2018-11-30 | 北京百度网讯科技有限公司 | Emotion identification method, apparatus, computer equipment and storage medium |
| CN109033257A (en) * | 2018-07-06 | 2018-12-18 | 中国平安人寿保险股份有限公司 | Talk about art recommended method, device, computer equipment and storage medium |
| CN109784414A (en) * | 2019-01-24 | 2019-05-21 | 出门问问信息科技有限公司 | Customer anger detection method, device and electronic equipment in a kind of phone customer service |
| CN109887525A (en) * | 2019-01-04 | 2019-06-14 | 平安科技（深圳）有限公司 | Intelligent customer service method, apparatus and computer readable storage medium |
| WO2019144926A1 (en) * | 2018-01-26 | 2019-08-01 | 上海智臻智能网络科技股份有限公司 | Intelligent interaction method and apparatus, computer device and computer-readable storage medium |
| CN110223714A (en) * | 2019-06-03 | 2019-09-10 | 杭州哲信信息技术有限公司 | A kind of voice-based Emotion identification method |
| CN110472023A (en) * | 2019-07-10 | 2019-11-19 | 深圳追一科技有限公司 | Customer service switching method, device, computer equipment and storage medium |
| CN111767385A (en) * | 2020-07-09 | 2020-10-13 | 腾讯科技（深圳）有限公司 | Intelligent question and answer method and device |
-
2020
- 2020-12-01 CN CN202011388418.6A patent/CN112671984B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102270450A (en) * | 2010-06-07 | 2011-12-07 | 株式会社曙飞电子 | System and method of multi model adaptation and voice recognition |
| CN107818798A (en) * | 2017-10-20 | 2018-03-20 | 百度在线网络技术（北京）有限公司 | Customer service quality evaluating method, device, equipment and storage medium |
| CN108122552A (en) * | 2017-12-15 | 2018-06-05 | 上海智臻智能网络科技股份有限公司 | Voice mood recognition methods and device |
| WO2019144926A1 (en) * | 2018-01-26 | 2019-08-01 | 上海智臻智能网络科技股份有限公司 | Intelligent interaction method and apparatus, computer device and computer-readable storage medium |
| CN108922564A (en) * | 2018-06-29 | 2018-11-30 | 北京百度网讯科技有限公司 | Emotion identification method, apparatus, computer equipment and storage medium |
| CN109033257A (en) * | 2018-07-06 | 2018-12-18 | 中国平安人寿保险股份有限公司 | Talk about art recommended method, device, computer equipment and storage medium |
| CN109887525A (en) * | 2019-01-04 | 2019-06-14 | 平安科技（深圳）有限公司 | Intelligent customer service method, apparatus and computer readable storage medium |
| CN109784414A (en) * | 2019-01-24 | 2019-05-21 | 出门问问信息科技有限公司 | Customer anger detection method, device and electronic equipment in a kind of phone customer service |
| CN110223714A (en) * | 2019-06-03 | 2019-09-10 | 杭州哲信信息技术有限公司 | A kind of voice-based Emotion identification method |
| CN110472023A (en) * | 2019-07-10 | 2019-11-19 | 深圳追一科技有限公司 | Customer service switching method, device, computer equipment and storage medium |
| CN111767385A (en) * | 2020-07-09 | 2020-10-13 | 腾讯科技（深圳）有限公司 | Intelligent question and answer method and device |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114971652A (en) * | 2022-05-16 | 2022-08-30 | 山东浪潮爱购云链信息科技有限公司 | Online customer service method, equipment and medium for enterprise purchasing platform |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112671984B (en) | 2022-09-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11868732B2 (en) | System for minimizing repetition in intelligent virtual assistant conversations | |
| EP3525438B1 (en) | Artificial intelligence based service implementation | |
| US10171659B2 (en) | Customer portal of an intelligent automated agent for a contact center | |
| AU2014233357B2 (en) | Intelligent automated agent for a contact center | |
| CN108476230B (en) | Optimal routing of machine learning based interactions to contact center agents | |
| US10135989B1 (en) | Personalized support routing based on paralinguistic information | |
| KR20140042994A (en) | Machine learning based of artificial intelligence conversation system using personal profiling information to be extracted automatically from the conversation contents with the virtual agent | |
| CN108682420A (en) | A kind of voice and video telephone accent recognition method and terminal device | |
| US12003668B1 (en) | Virtual assistant architecture for natural language understanding in a customer service system | |
| US11743378B1 (en) | Intelligent agent assistant for natural language understanding in a customer service system | |
| WO2022095377A1 (en) | Online self-service dialogue method and apparatus, computer device, and computer readable medium | |
| CN113360622A (en) | User dialogue information processing method and device and computer equipment | |
| KR20190117840A (en) | Method and computer readable recording medium for, during a customer consulting by a conversation understanding ai system, passing responsibility of proceeding with subsequent customer consulting to a human consultant | |
| CN112671984B (en) | Service mode switching method and device, robot customer service and storage medium | |
| CN115017289A (en) | Method and device for serving customers based on knowledge base | |
| EP1761015A1 (en) | Self-adaptive user interface for dialogue systems | |
| WO2023027832A1 (en) | Determination and display of estimated hold durations for calls | |
| US11196864B1 (en) | Analyzing voice response to telephone call to assign appropriate agent | |
| US20230025655A1 (en) | Intelligent Voice Interface as a Personal or Social Network Entity | |
| CN117786079A (en) | Dialogue method and system based on self-adaptive learning and context awareness AI proxy | |
| CN115455164A (en) | Method and device for identifying will level, computer equipment and storage medium | |
| EP4393143A1 (en) | Determination and display of estimated hold durations for calls | |
| WO2023017532A1 (en) | A system and method for user ranking and adaptations thereof for voice based interactions |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |