CN111611357A

CN111611357A - Configuration method of man-machine conversation system, multi-turn conversation configuration platform and electronic equipment

Info

Publication number: CN111611357A
Application number: CN201910141210.5A
Authority: CN
Inventors: 庞胜
Original assignee: Beijing Didi Infinity Technology and Development Co Ltd
Current assignee: Beijing Didi Infinity Technology and Development Co Ltd
Priority date: 2019-02-25
Filing date: 2019-02-25
Publication date: 2020-09-01
Anticipated expiration: 2039-02-25
Also published as: CN111611357B

Abstract

The invention belongs to the technical field of human-computer interaction, and particularly relates to a configuration method of a human-computer conversation system, a multi-turn conversation configuration platform and electronic equipment. The configuration method comprises the following steps: receiving a configuration request input by a user; responding to the configuration request to display a configuration interface; receiving a first configuration operation input by the user on the configuration interface and aiming at a target task, wherein the first configuration operation is used for configuring the functional attributes of the state points of the finite-state machine and the functional attributes of directed state edges between the state points; and responding to the first configuration operation to complete the finite state machine-based dialog management logic configuration of the man-machine dialog system. The configuration method enables a user to establish the state points in the finite state machine and the directed state edges among the state points in an interactive mode to complete the dialogue management logic configuration of the man-machine dialogue system, so that the dialogue scene development is more convenient and visual, and the flexibility is good.

Description

Configuration method of man-machine conversation system, multi-turn conversation configuration platform and electronic equipment

Technical Field

The invention belongs to the technical field of human-computer interaction, and particularly relates to a configuration method of a human-computer conversation system, a multi-turn conversation configuration platform and electronic equipment.

Background

With the development of mobile intelligent terminals and information network technologies, people use human-computer interactive applications in more and more scenes. The customer service is extremely important for the service industry, the demand for the customer service can be increased along with the increase of users of a product, the efficiency of solving the problem by the customer service directly influences the user experience, the problem of customer service consultation by the user often has a large number of repeated problems, for some complex scenes, an intelligent conversation robot takes place, the conversation answering robot faces to tasks (problems), the problem of customer service consultation by the user is changed along with the product dynamic state, inconvenience is caused to the development of the conversation robot, and the rapid development of the conversation scene has important significance for the operation of the user and the improvement of the satisfaction degree of the user.

Disclosure of Invention

In view of this, embodiments of the present invention provide a configuration method for a human-computer dialog system, a multi-round dialog configuration platform, and an electronic device, which configure a service logic by creating a finite state machine in an interactive manner, so that development of a dialog scene is more convenient and intuitive.

The embodiment of the invention is realized by the following steps:

in a first aspect, an embodiment of the present invention provides a method for configuring a human-computer dialog system, including: receiving a configuration request input by a user; responding to the configuration request to display a configuration interface; receiving a first configuration operation, which is input by the user on the configuration interface and aims at a target task, wherein the first configuration operation is used for configuring functional attributes of state points of a finite state machine and functional attributes of directed state edges between the state points, the state points represent dialog state points of a man-machine dialog system for completing the target task, the directed state edges represent dialog states of the man-machine dialog system and can be transferred from one end dialog state point of the directed state edges to the other pointed end dialog state point, and the functional attributes of the directed state edges are used for limiting conditions meeting state transfer; and responding to the first configuration operation to complete the finite state machine-based dialog management logic configuration of the man-machine dialog system. In the embodiment of the application, the state points in the finite state machine and the directed state edges between the state points are established in an interactive mode to complete the dialogue management logic configuration of the man-machine dialogue system, so that the development of a dialogue scene is more convenient and visual, and the flexibility is good.

With reference to a possible implementation manner of the embodiment of the first aspect, the receiving a first configuration operation for a target task, which is input by the user on the configuration interface, includes: receiving a dialog state point creating operation of the target task input on the configuration interface by the user; responding to the creating operation of the conversation state point, generating the conversation state point and displaying the conversation state point; receiving function attribute configuration operation of the user for each dialog state point; receiving directed state edge creation operation input by the user; generating and displaying directed state edges between state points corresponding to the directed state edge creation operation based on the directed state edge creation operation; and receiving configuration operation of the directed state edges input by the user, wherein the configuration operation of the directed state edges is used for configuring transition conditions of the directed state edges. In the embodiment of the application, when the service logic of the man-machine conversation system is configured in an interactive mode, the service logic configuration of the man-machine conversation system for a new scene can be quickly completed through the configuration modes of creating the conversation state point, defining the function attribute of the conversation state point, creating the directed state edge of the conversation state point, defining the configuration of the directed state edge, such as the transition condition to the state edge, and calling a function or API after the transition condition is met, and in addition, the method can complete the complicated service logic and conversation logic configuration through the function or API.

With reference to a possible implementation manner of the embodiment of the first aspect, the receiving a first configuration operation for a target task, which is input by the user on the configuration interface, further includes: and receiving the configuration operation of the function call or the API call which is input by the user and meets the transfer condition for each directed state edge. In the embodiment of the application, the user can also configure the function call or the API call after each directed state edge meets the transition condition, so as to improve the flexibility of developing the dialog scene.

With reference to a possible implementation manner of the embodiment of the first aspect, when the configuration operation on the directed state edge is used to configure a third-party API call after a branch condition is satisfied, the method further includes: displaying an API configuration interface, and receiving the name, address, request mode and return parameter settings of the third-party API input by the user on the API configuration interface; and configuring the third-party API based on the settings of the name, the address, the request mode and the return parameters. In the embodiment of the application, a user can configure the third-party API according to actual needs, so that the method is convenient for scene developers to develop new scenes and has the flexibility based on a code configuration platform.

With reference to a possible implementation manner of the embodiment of the first aspect, when the configuration operation on the directed state edge is used to configure a function call after a branch condition is satisfied, the method further includes: displaying a function configuration interface, and receiving the settings of the name, the function content and the return parameters of the function, which are input by the user on the function configuration interface; and configuring the function based on the name of the function, the content of the function and the setting of the return parameter. In the embodiment of the application, a user can configure the required function according to actual needs, so that the method is convenient for scene developers to develop new scenes and has the flexibility based on a code configuration platform.

With reference to one possible implementation manner of the embodiment of the first aspect, before receiving a first configuration operation for a target task input by the user on the configuration interface, the method further includes: receiving a target task creation request input on the configuration interface by the user; and completing the creation of the target task according to the conversation field, the conversation intention and the conversation slot position carried in the target task creation request.

With reference to one possible implementation manner of the embodiment of the first aspect, after the displaying a configuration interface in response to the configuration request, the method further includes: receiving a second configuration operation input by the user on the configuration interface, wherein the second configuration operation is used for configuring NLU parameters of a natural language understanding module of the man-machine conversation system, and the NLU parameters comprise a conversation field, a conversation intention and a conversation slot position; and responding to the second configuration operation to complete the NLU parameter configuration of the natural language understanding module of the man-machine conversation system. In the embodiment of the application, a user can add an intention (intent) and a slot (slot) in a created field to complete the configuration of the natural language understanding module of the man-machine conversation system, and the method does not need to consider the problem of switching of multiple fields due to the fact that the fields are also considered, so that the configuration program is simplified.

With reference to one possible implementation manner of the embodiment of the first aspect, after completing the NLU parameter configuration of the natural language understanding module of the human-computer dialog system in response to the second configuration operation, the method further includes: responding to a test instruction which is input on the configuration interface by the user and aims at the natural language understanding module, and generating a test address; and issuing the test address.

With reference to one possible implementation manner of the embodiment of the first aspect, after the finite state machine-based dialog management logic configuration of the human-computer dialog system is completed in response to the first configuration operation, the method further includes: receiving a third configuration operation input by the user on the configuration interface, wherein the third configuration operation is used for configuring natural language generation parameters of the man-machine conversation system; and responding to the third configuration operation to complete the natural language generation parameter configuration of the man-machine conversation system.

With reference to one possible implementation manner of the embodiment of the first aspect, a component selection list is displayed on the configuration interface, where the component selection list includes a plurality of selection components with different purposes; receiving a third configuration operation input by the user on the configuration interface, wherein the third configuration operation comprises: receiving a component determination instruction input by the user in the component selection list, and displaying a corresponding component configuration interface; and receiving the component configuration operation input on the configuration interface by the user.

With reference to a possible implementation manner of the embodiment of the first aspect, after configuring a dialog management module, a natural language understanding module, and a natural language generation module in a human-computer dialog system corresponding to at least two of the target tasks, respectively, the method further includes: and issuing a uniform conversation interface through which a user can carry out conversation with any one of the man-machine conversation systems respectively corresponding to the at least two target tasks. In the embodiment of the application, the man-machine conversation system applied to the non-use scenes can be issued through the same conversation interface, namely, developers only need to configure one conversation interface, all application scenes configured by the developers can be triggered as long as the developers access the conversation interface, and then the corresponding scenes can be corresponded through field recognition, so that the developers do not need to consider the problem of switching a plurality of scenes when configuring, and the automatic switching of a plurality of scenes is realized.

In a second aspect, an embodiment of the present invention further provides a multi-round dialog configuration platform, including: the configuration module is used for receiving a configuration request input by a user; the system is also used for responding to the configuration request and displaying a configuration interface; the system is also used for receiving a first configuration operation aiming at a target task, which is input on a configuration interface by a user, wherein the first configuration operation is used for configuring a function attribute of a state point aiming at a finite state machine and a function attribute of a directed state edge between the state points, the state point represents a conversation state point of a man-machine conversation system for completing the target task, the directed state edge represents that a conversation state of the man-machine conversation system can be transferred from the conversation state point at one end of the directed state edge to the conversation state point at the other end, and the function attribute of the directed state edge is used for limiting a condition meeting state transfer; and is also used for responding to the first configuration operation to complete the finite state machine-based dialog management logic configuration of the man-machine dialog system.

With reference to a possible implementation manner of the embodiment of the second aspect, the configuration module is further configured to: receiving a dialog state point creating operation of the target task input on the configuration interface by the user; responding to the creating operation of the conversation state point, generating the conversation state point and displaying the conversation state point; receiving function attribute configuration operation of the user for each dialog state point; receiving directed state edge creation operation input by the user; generating and displaying directed state edges between state points corresponding to the directed state edge creation operation based on the directed state edge creation operation; and receiving configuration operation of the directed state edges input by the user, wherein the configuration operation of the directed state edges is used for configuring transition conditions of the directed state edges.

With reference to a possible implementation manner of the embodiment of the second aspect, the configuration module is further configured to: and receiving the configuration operation of the function call or the API call which is input by the user and meets the transfer condition for each directed state edge.

With reference to a possible implementation manner of the embodiment of the second aspect, when the configuration operation on the directed state edge is used to configure a third-party API call after the branch condition is satisfied, the configuration module is further configured to: receiving the name, address, request mode and return parameter setting of the third party API input by the user on an API configuration interface; and configuring the third-party API based on the settings of the name, the address, the request mode and the return parameters.

With reference to a possible implementation manner of the embodiment of the second aspect, when the configuration operation on the directed state edge is used to configure a function call after the branch condition is satisfied, the configuration module is further configured to: receiving the name, the function content and the setting of the return parameters of the function input by the user on a function configuration interface; and configuring the function based on the name of the function, the content of the function and the setting of the return parameter.

With reference to a possible implementation manner of the embodiment of the second aspect, the configuration module is further configured to: receiving a target task creation request input on the configuration interface by the user; and completing the creation of the target task according to the conversation field, the conversation intention and the conversation slot position carried in the target task creation request.

With reference to a possible implementation manner of the embodiment of the second aspect, the configuration module is further configured to: receiving a second configuration operation input by the user on the configuration interface, wherein the second configuration operation is used for configuring NLU parameters of a natural language understanding module of the man-machine conversation system, and the NLU parameters comprise a conversation field, a conversation intention and a conversation slot position; and responding to the second configuration operation to complete the NLU parameter configuration of the natural language understanding module of the man-machine conversation system.

With reference to a possible implementation manner of the embodiment of the second aspect, the configuration module is further configured to: responding to a test instruction which is input on the configuration interface by the user and aims at the natural language understanding module, and generating a test address; and issuing the test address.

With reference to a possible implementation manner of the embodiment of the second aspect, the configuration module is further configured to: receiving a third configuration operation input by the user on the configuration interface, wherein the third configuration operation is used for configuring natural language generation parameters of the man-machine conversation system; and responding to the third configuration operation to complete the natural language generation parameter configuration of the man-machine conversation system.

With reference to a possible implementation manner of the embodiment of the second aspect, a component selection list is displayed on the configuration interface, where the component selection list includes a plurality of selection components with different purposes, and the configuration module is further configured to: receiving a component determination instruction input by the user in the component selection list, and displaying a corresponding component configuration interface; and receiving the component configuration operation input on the configuration interface by the user.

With reference to a possible implementation manner of the embodiment of the second aspect, after configuring the human-computer interaction systems corresponding to at least two of the target tasks, the configuration module is further configured to: and issuing a uniform conversation interface through which a user can carry out conversation with any one of the man-machine conversation systems respectively corresponding to the at least two target tasks.

In a third aspect, an embodiment of the present invention further provides an electronic device, including: a processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the electronic device runs, the processor and the storage medium communicate with each other through the bus, and the processor executes the machine-readable instructions to perform the steps of the method as described in the embodiment of the first aspect and/or any one of the possible implementation manners of the embodiment of the first aspect.

Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by the practice of the embodiments of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and drawings.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without creative efforts. The above and other objects, features and advantages of the present invention will become more apparent from the accompanying drawings. Like reference numerals refer to like parts throughout the drawings. The drawings are not necessarily to scale, emphasis instead being placed upon illustrating the principles of the invention.

Fig. 1 is a flowchart illustrating a configuration method of a human-computer dialog system according to an embodiment of the present invention.

Fig. 2 shows a configuration diagram of a finite state machine according to an embodiment of the present invention.

FIG. 3 is a diagram illustrating an interface for task creation provided by an embodiment of the present invention.

FIG. 4 is a schematic diagram of an interface for configuring third party APIs provided by an embodiment of the invention.

Fig. 5 is a schematic interface diagram illustrating a configuration function according to an embodiment of the present invention.

FIG. 6 is a schematic diagram of an interface for configuring a natural language generation module according to an embodiment of the present invention.

Fig. 7 is a schematic interface diagram of a natural language generation module according to an embodiment of the present invention.

Fig. 8 is a functional structure diagram of a multi-turn dialog configuration platform provided in an embodiment of the present invention.

Fig. 9 shows a schematic structural diagram of an electronic device provided in an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be described below with reference to the drawings in the embodiments of the present invention. It is to be understood that the drawings herein are for purposes of illustration and description only and are not intended as a definition of the limits of the application. It should be understood that the schematic drawings are not necessarily to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of the present application. It should be understood that the operations of the flow diagrams may be performed out of order, and steps without logical context may be performed in reverse order or simultaneously. One skilled in the art, under the guidance of this application, may add one or more other operations to, or remove one or more operations from, the flowchart.

It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.

In the description of the present invention, it should be noted that the terms "first", "second", "third", and the like are used only for distinguishing the description, and are not intended to indicate or imply relative importance. Further, the term "and/or" in the present application is only one kind of association relationship describing the associated object, and means that three kinds of relationships may exist, for example, a and/or B may mean: a exists alone, A and B exist simultaneously, and B exists alone.

For the development of the multi-turn multi-phone robot, two development modes are available at present, one is based on code configuration and hard coding, and the other is based on platform dialogue configuration. The development mode based on code configuration is flexible, and developers can complete complex business logic and conversation logic in programs. However, such a development mode is not favorable for optimization and fast iteration of the dialog scene, the scene developer is also limited to the program developer, and as the business requirement increases, it is difficult to develop a new scene completely based on the development mode to meet the requirements of fast iteration and expansion of the dialog scene. The platform-based configuration can facilitate scene developers to develop new scenes, is low in development cost, and generally completes the callback by sequentially inquiring and clarifying the conversation slots. Although platform-type development can be convenient for product managers or operators to develop and reduce development cost, flexibility is sacrificed, for some complex scenes, only fixed mode inquiry slot positions or clarification slot positions are used for fixing conversation templates, and for scenes with requirements for calling third-party interfaces, the existing platform configuration mode is difficult to support.

It should be noted that the defects existing in the above solutions are the results obtained after the inventors have practiced and studied carefully, and therefore, the discovery process of the above problems and the solutions proposed by the following embodiments of the present invention to the above problems should be the contribution of the inventors to the present invention in the course of the present invention.

In view of this, the embodiment of the present invention provides a new multi-turn dialog configuration platform, which integrates the advantages of code configuration (or development) and platform configuration, that is, retains the flexibility of a code development manner, and makes multi-turn new scene development more convenient and intuitive. The multi-round conversation configuration platform provides a third party API calling interface, a code compiling interface, multi-field switching, an online interface and a testing interface, and a user can complete multi-round conversation new scene development and online in a one-stop mode.

The multi-round conversation configuration platform provided by the embodiment of the application is divided into three modules from the framework: a front-end, a back-end and a dialog engine. The front end is used for directly interacting with a user, and data operated by the user can be synchronized to the rear end; the back-end can control the dialogue engine to load the dialogue logic and other information configured by the user. For example, the data flow of a user operating a certain dialog scenario is: the user operates at the front-end- > back-end- > dialog engine. The architecture may be a CS (Client/Server) architecture or a BS (Browser/Server) architecture, which is not further limited herein.

The dialog engine is divided into three main modules: an NLU (Natural Language Understanding) module, a DM (Dialog Management) module, and an NLG (Natural Language generation) module. The NLU module is mainly responsible for scene (field) identification, intention identification and slot extraction; the DM module is responsible for conversation management and controls the process of man-machine conversation, the DM determines the reaction to the user at the moment according to the history information of the conversation, the most common application is a task-driven multi-turn conversation, the user has definite purposes such as meal ordering, ticket ordering and the like, the user needs to be complicated usually, and has a plurality of limiting conditions, and the user possibly needs to state in multiple turns. The DM module takes the output of the NLU module as input, outputs action and updates context, and the NLG module takes the action and generates natural language according to the action to be displayed to a user.

The user can complete the development and online of the multi-turn dialogue new scene man-machine dialogue system based on the multi-turn dialogue configuration platform according to the self requirement. The configured conversation process can be analyzed by a conversation engine, and a conversation robot capable of interacting according to the conversation process is generated. Among them, the conversation robot is technically called a conversation agent, which is defined as a software program that interprets and responds to statements made by a user in a common natural language. The following will describe the procedure of the configuration method of the man-machine dialog system in conjunction with the flowchart shown in fig. 1.

Step S101: a configuration request input by a user is received.

The configuration request is used for enabling the back end of the multi-turn conversation configuration platform to return to a configuration interface, and the configuration interface is used for being configured by a user according to needs so as to meet the needs of the user.

The configuration request includes account information and/or IP (Internet Protocol) address information of the user. So as to return the corresponding configuration interface through the account information and/or the IP address information.

Step S102: and responding to the configuration request to display a configuration interface.

The user operates on the front end line, data can be synchronized to the back end, and the back end responds to the configuration request and displays the configuration interface after receiving the configuration request input by the user.

Optionally, the configuration interface is an interface in which various functional components are preset, and for example, the configuration interface displays components such as a task list, an interface list, a function list, an action configuration, and an NLU configuration. The user can display the corresponding interface by selecting different functional components, if a task list is selected, each task created by the user is displayed in the task list, and if the task list is not created, the task list is empty. And selecting one task, namely checking the basic information of the task, and entering a logic diagram editing interface if clicking the logic diagram, wherein a user can complete the creation of a finite state machine and complete the dialogue management logic configuration aiming at the task on the interface. For another example, if the interface list is selected, a corresponding API configuration interface is displayed, and the user inputs a required parameter in an input box, that is, the interface is configured, where the interface is used to configure a third party API request, and the input parameter may be session information (intention, slot position, etc.) in a state point in a finite state machine, intermediate information (output generated by other API calls or output of function calls), or some static information. The API request supports common request modes (such as POST, GET, PUT and the like). For another example, selecting a function list may display processing functions built in the list platform (the platform may have some commonly used information processing functions built therein, but may not cover all scenes), and of course, a user may customize a desired function, for example, a python function may be written to define the desired function. For another example, if an action configuration is selected, a corresponding action configuration interface is displayed, and a selection list of each component is displayed in the interface, so that the user can perform function configuration on the interface, and can also customize a required function component. For another example, selecting the NLU configuration may enter an NLU configuration editing interface where the user may implement the configuration of the conversation domain, conversation intent, and conversation slot.

Step S103: receiving a first configuration operation input on the configuration interface by the user and aiming at a target task.

The first configuration operation is an operation triggered by a user according to option information displayed on the configuration interface.

The first configuration operation is to configure functional attributes of state points of a finite state machine, functional attributes of directed state edges between the state points. The finite state machine is a mathematical model representing finite states and behaviors such as transition and action among the states, and the finite state machine mainly plays a role in describing a state sequence which an object experiences in the life cycle of the object and how to respond to various events from the outside. The state points represent the conversation state points of a man-machine conversation system used for completing the target task, namely the conversation state points of the conversation robot. The directed state edge represents that the dialog state of the man-machine dialog system can be transferred from the dialog state point at one end of the directed state edge to the dialog state point at the other end, namely the directed state edge is used for representing the jump relation between the state points, and if the dialog state is transferred to a secondary state according to the direction of the directed state edge from the current state, the dialog state can still keep the original state, wherein the current state refers to the current state, the secondary state refers to a new state to be transferred after the condition is met, and the secondary state is relative to the current state, and is converted into the new 'current state' once being activated. The functional attribute of the directed state edge is used for limiting conditions for satisfying state transition, and when a condition is satisfied, an action is triggered or state transition is executed once. That is, the dialogue robot uses the logic to perform the conversation, that is, according to the statement input by the user and the current state, it is judged that the condition of state jump is not satisfied, and if the condition is satisfied, the dialogue robot jumps to the next dialogue state point. Optionally, receiving a first configuration operation for a target task input by the user on the configuration interface includes: receiving a logic diagram editing instruction input on the configuration interface by the user, and displaying a logic diagram editing interface; receiving a dialog state point creating operation of the target task input by the user on the logic diagram editing interface; responding to the creating operation of the dialog state point, generating the dialog state point and displaying the dialog state point on the logic diagram editing interface; receiving function attribute configuration operation of the user for each dialog state point; receiving directed state edge creation operation input by the user on the logic diagram editing interface; generating and displaying directed state edges between state points corresponding to the directed state edge creation operation based on the directed state edge creation operation; and receiving configuration operation of the directed state edges input by the user, wherein the configuration operation of the directed state edges is used for configuring transition conditions of the directed state edges. That is, the user may complete the creation of the state points in the finite state machine through the visual interface, then configure the created state points, then create the directed state edges, and then configure the created directed state edges, for example, configure the transition conditions characterizing the jump relationship between the state points.

Optionally, receiving a first configuration operation for a target task input by the user on the configuration interface, further includes: and receiving the configuration operation of the function call or the API call which is input by the user and meets the transfer condition for each directed state edge. That is, the user may also configure function calls and API calls that satisfy the transition condition to configure the output action of the directed state edge.

Step S104: and responding to the first configuration operation to complete the finite state machine-based dialog management logic configuration of the man-machine dialog system.

Receiving a first configuration operation aiming at a target task and input by a user on the configuration interface, and responding to the first configuration operation to complete the finite-state-machine-based dialog management logic configuration of the man-machine dialog system, namely to complete the management logic configuration of a dialog management module of the dialog robot, namely to judge that the condition of state jump is not met according to a statement input by the user and the current state, and to jump to the next dialog state point if the condition is met.

As an optional implementation, the completing the management logic configuration of the dialogue management module of the dialogue robot in response to the first configuration operation includes: responding to a logic diagram editing instruction input on the configuration interface by the user, and displaying a logic diagram editing interface; responding to the creation operation of the dialog state points input by the user on the logic diagram editing interface, generating the dialog state points and displaying the dialog state points on the logic diagram editing interface; responding to the function attribute configuration operation of the user aiming at each conversation state point, and finishing the corresponding function attribute configuration of the conversation robot; responding to directed state edge creation operation input on the logic diagram editing interface by the user, and generating and displaying directed state edges between state points corresponding to the directed state edge creation operation; and responding to the configuration operation of the directed state edges input by the user, and completing the corresponding function attribute configuration of the dialogue robot, wherein the directed state edge configuration instruction is used for configuring the transfer condition of each directed state edge, and the function call or the API call after the transfer condition is met.

Optionally, when configuring the function attribute of the directed state edge of each dialog state point, the logic diagram editing interface displays a callback list, and a selection menu of the callback list includes a plurality of built-in APIs and functions. At this time, responding to the configuration operation of the directed state edge input by the user comprises: responding to the transfer condition configuration operation of the directed state edge input by the user, and completing the transfer condition configuration of the corresponding directed state edge; and responding to the call instruction which is input in the callback list by the user and aims at each directed state edge after meeting the transfer condition, and completing the function call or API call of the corresponding directed state edge after meeting the transfer condition. That is, when configuring the transfer condition of each directed state edge, and the function call or API call that satisfies the transfer condition, the user may configure the transfer condition in the logic diagram editing interface of the corresponding directed state edge, and then select the function call or API call that satisfies the transfer condition from the callback list.

For ease of understanding, the configuration diagram shown in fig. 2 may be used to describe the above. When configuring the branch condition of the state edge, the "condition" button in fig. 2 may be clicked, that is, the configuration of the branch condition may be realized, wherein fig. 2 illustrates an exemplary diagram of the unconfigured branch condition. When configuring the function call or the API call which meets the transfer condition, the existing API call or the custom function call in the callback list is selected.

The process of step S104, also referred to as a configuration process of the dialogue management module, when the user wants to configure the management logic configuration of the dialogue management module of the dialogue robot for the target task, on the multi-turn dialog configuration platform, a configurator can select a target task that has already been created, since the domain of the target task is updated, therefore, a plurality of versions can be provided, after the target task is created, a domain version needs to be created, for each domain version, a dialog state graph (composed of state points and directed state edges) corresponds to the domain version, a configurator can connect the required dialog state points by creating anchor points between the dialog state points and dragging the dialog state points, a directed edge (also called a directed state edge) is arranged after the connected dialog state points, the conditions for branching, function calls or API calls after the branching conditions are satisfied may be configured on the edge.

The rectangular block in fig. 2 is a dialog state point, a small point in the dialog state point is an anchor point, and the connecting line attached with an arrow in fig. 2 is a directed state edge.

After the configuration of the dialogue management module is completed, the user can modify, delete and issue the dialogue management module. For example, if the user selects a test instruction for the dialog management module on the configuration interface, the back end may generate a test address in response to the test instruction for the dialog management module input by the user on the configuration interface, so as to issue the test address to the test environment for testing. The release may be to a test environment or to an online environment. Wherein, it must be released to the testing environment before being released to the online environment.

Optionally, before receiving a first configuration operation for a target task input by the user on the configuration interface, the method further includes: receiving a target task creation request input by the user on the configuration interface, and completing creation of the target task according to the conversation field, the conversation intention and the conversation slot position carried in the target task creation request. That is, a user may enter the task list interface after clicking the task list, and at this time, the user may create a task in the task interface, for example, after clicking a button similar to "create task" or "+", the user may enter the task creation interface, and the user may input parameters in the interface, for example, the user may create a field in the interface, and when creating the field, all intents (entries) and slots (slots) that will be involved in the field need to be filled in, and after completing the filling, finally, after clicking a button similar to "determine" or save, the creation of the target task may be completed. The process of task creation can be seen in fig. 3.

And when the configuration operation on the state edge is used for configuring the third-party API call which meets the transfer condition, the API call selected from the callback list is set in advance. Optionally, the third party API may be configured in the following manner: receiving the name, address, request mode and return parameter setting of the third party API input by the user on the API configuration interface; and configuring the third-party API based on the settings of the name, the address, the request mode and the return parameters. And then, inputting parameters such as the name, the address, the request mode, the return parameters and the like of the third party API on the API configuration interface so as to complete the configuration of the third party API based on the parameters. The specific configuration process may refer to the configuration interface shown in fig. 4, where "query weather" in the figure is the name of the API, the request mode is POST, the request parameter is "cityid", and the address is "http: // 100.90.144.143: 8004/kbqa ".

When the configuration operation on the state edge is used to configure a function call that satisfies a transition condition, the function call selected from the callback list is also set in advance, and optionally, the function may be configured in the following manner: receiving the name, the function content and the setting of a return parameter of the function, which are input by a user on a function configuration interface; and configuring the function based on the name of the function, the content of the function and the setting of the return parameter. The user clicks the function list on the configuration interface to enter the function configuration interface, the name of the set function, the function content, the return parameter and other contents are input on the interface, and the back end completes the configuration of the function based on the parameters. The specific configuration process can refer to the configuration interface shown in fig. 5, and the "generated number" in the figure is the name of the function.

Optionally, after the displaying a configuration interface in response to the configuration request, the method further includes: receiving a second configuration operation input by the user on the configuration interface, wherein the second configuration operation is used for configuring NLU parameters of a natural language understanding module of the conversation robot, and the NLU parameters comprise a conversation field, a conversation intention and a conversation slot position; and completing NLU parameter configuration of a natural language understanding module of the dialogue robot in response to the second configuration operation. That is, the user can configure the natural language understanding module, that is, the NLU module, of the dialogue robot. The user can create fields under the natural language understanding module, and for each created field, an intention (intent) and a slot (slot) can be added in the field. When the intention is created, the user is required to add the basic corpus of the intention, and the more corpuses are added, the more accurate the intention identification is. Some slot positions built in the platform, such as time slots, place slots and the like, can be selected when the slot positions are created, and such common slot positions are added by selecting a built-in slot position option. The natural language understanding module also supports self-defined slot positions which have two modes: the slot positions matched through the regular expressions and the slot positions matched through the complete expressions.

Optionally, after completing the NLU parameter configuration of the natural language understanding module of the conversation robot in response to the second configuration operation, the method further comprises: responding to a test instruction which is input on the configuration interface by the user and aims at the natural language understanding module, and generating a test address; and issuing the test address so as to issue the natural language understanding module to a test environment for testing. When the user has added all intentions and slots in a domain, the system can directly train the intent classification of the domain. Because the NLU configuration module is an independent module, after training is completed, the NLU configuration module can be directly issued to a test environment or an online environment through a platform. Wherein, the user must first issue the test environment test and then issue the test environment test to the online environment.

Optionally, after completing the finite state machine based dialog management logic configuration of the human-machine dialog system in response to the first configuration operation, the method further comprises: receiving a third configuration operation input by the user on the configuration interface, wherein the third configuration operation is used for configuring natural language generation parameters of the man-machine conversation system; and responding to the third configuration operation to complete the natural language generation parameter configuration of the conversation robot. That is, after the user completes the management logic configuration of the dialogue management module of the dialogue robot, the natural language generation module of the dialogue robot is configured, and as the multi-turn dialogue robot interacts with the user mainly through the form of the dialogue + card information, some protocols related to front-end display exist in the NLG module, and the protocols can be abstracted into some specific components, such as a list selection component, an order selection component, a progress viewing component and the like. That is, a user clicks an action configuration and action configuration interface, where a component selection list is displayed on the interface, where the component selection list includes a plurality of selection components with different purposes, and the user may configure functions of the selection components, and at this time, a third configuration operation input by the user on the configuration interface is received, where the third configuration operation includes: receiving a component determination instruction input by the user in the component selection list, and displaying a corresponding component configuration interface; and receiving the component configuration operation input on the configuration interface by the user. The specific configuration process may be described with reference to the schematic diagrams shown in fig. 6 and fig. 7, after the task created by the user is a weather query task, after the management logic configuration of the dialog management module of the dialog robot is completed by the method shown in fig. 1, when the user clicks on the action configuration, the action configuration interface displays the component list shown in fig. 6, and when the user selects "request _ date" in fig. 6, the interface shown in fig. 7 is displayed, and at this time, the user may define the function of the component on the interface.

After the configuration of each module is completed based on the target task, the platform can synchronize the configuration information to the test environment and the online environment. For example, zookeeper (which is a distributed application coordination service) is used to maintain IP addresses of both the test machine and the online machine, both of which have a command server (HTTP service) that registers the IP address of the running machine with zookeeper at startup. The command server is used for receiving an instruction sent by the rear end of the multi-turn conversation platform, when a user operates on line at the front end of the platform, the rear end receives an on-line request, then an IP address list registered on a target environment is read in the zookeeper, the instruction is sent to the corresponding command server, and finally the machine where the command server is located completes state synchronization.

Optionally, after configuring the human-computer dialog systems corresponding to the at least two target tasks respectively, that is, after configuring the dialog management module, the natural language understanding module, and the natural language generating module in the dialog robot corresponding to the at least two target tasks respectively, the method further includes: and issuing a uniform conversation interface through which a user can carry out conversation with any one of the man-machine conversation systems respectively corresponding to the at least two target tasks. That is, after the multi-round dialog configuration platform provided by the embodiment of the present application is configured to be applied to the dialog robots in a plurality of different scenes, the dialog robots applied to different scenes can be released through the same dialog interface, that is, a developer only needs to configure one dialog interface, and can start all application scenes configured by the developer as long as accessing the dialog interface, and then can correspond to corresponding scenes through field recognition, so that the developer does not need to consider the problem of switching a plurality of scenes when configuring, and automatic switching of multiple scenes is achieved.

An embodiment of the present invention further provides a multi-round dialog configuration platform, as shown in fig. 8, the multi-round dialog configuration platform includes: a conversation robot and a configuration module, wherein the conversation robot comprises: the device comprises a dialogue management module, a natural language understanding module and a natural language generating module. The configuration module is used for configuring the conversation management module, the natural language understanding module and the natural language generating module so as to realize development and online of the conversation robot.

The configuration module is used for receiving a configuration request input by a user when the conversation management module is configured; the system is also used for responding to the configuration request and displaying a configuration interface; the system is also used for receiving a first configuration operation aiming at a target task, which is input on a configuration interface by a user, wherein the first configuration operation is used for configuring the functional attributes of state points of a finite state machine and the functional attributes of directed state edges between the state points, the state points represent the dialog state points of a man-machine dialog system for completing the target task, the directed state edges represent that the dialog state of the man-machine dialog system can be transferred from the dialog state point at one end of the directed state edge to the dialog state point at the other end, and the functional attributes of the directed state edges are used for limiting the condition of meeting the state transfer; and the management logic configuration of the dialogue management module of the dialogue robot is completed in response to the first configuration operation.

Optionally, the configuration module is further configured to: receiving a logic diagram editing instruction input on the configuration interface by the user, and displaying a logic diagram editing interface; receiving a dialog state point creating operation of the target task input by the user on the logic diagram editing interface; responding to the creating operation of the dialog state point, generating the dialog state point and displaying the dialog state point on the logic diagram editing interface; receiving function attribute configuration operation of the user for each dialog state point; receiving directed state edge creation operation input by the user on the logic diagram editing interface; generating and displaying directed state edges between state points corresponding to the directed state edge creation operation based on the directed state edge creation operation; and receiving configuration operation of the directed state edges input by the user, wherein the configuration operation of the directed state edges is used for configuring transition conditions of the directed state edges.

The configuration module is further configured to: and receiving the configuration operation of the function call or the API call which is input by the user and meets the transfer condition for each directed state edge.

Optionally, when the configuration operation on the directed state edge is used to configure a third-party API call after the transfer condition is satisfied, the configuration module is further configured to: receiving the name, address, request mode and return parameter setting of the third party API input by the user on an API configuration interface; and configuring the third-party API based on the settings of the name, the address, the request mode and the return parameters.

Optionally, when the configuration operation on the directed state edge is used to configure a function call after the branch condition is satisfied, the configuration module is further configured to: receiving the name, the function content and the setting of the return parameters of the function input by the user on a function configuration interface; and configuring the function based on the name of the function, the content of the function and the setting of the return parameter.

Optionally, the configuration module is further configured to: receiving a target task creation request input on the configuration interface by the user; and completing the creation of the target task according to the conversation field, the conversation intention and the conversation slot position carried in the target task creation request.

Wherein, when the natural language understanding module is configured, the configuration module is further configured to: receiving a second configuration operation input by the user on the configuration interface, wherein the second configuration operation is used for configuring NLU parameters of a natural language understanding module of the conversation robot, and the NLU parameters comprise a conversation field, a conversation intention and a conversation slot position; and completing NLU parameter configuration of a natural language understanding module of the dialogue robot in response to the second configuration operation.

Optionally, the configuration module is further configured to: responding to a test instruction which is input on the configuration interface by the user and aims at the natural language understanding module, and generating a test address; and issuing the test address.

Wherein, when the natural language generation module is configured, the configuration module is further configured to: receiving a third configuration operation input by the user on the configuration interface, wherein the third configuration operation is used for configuring NLG parameters of a natural language generation module of the conversation robot; and responding to the third configuration operation to complete the NLG parameter configuration of the natural language generation module of the dialogue robot.

Optionally, a component selection list is displayed on the configuration interface, the component selection list includes a plurality of selection components with different purposes, and the configuration module is further configured to: receiving a component determination instruction input by the user in the component selection list, and displaying a corresponding component configuration interface; and receiving the component configuration operation input on the configuration interface by the user.

After configuring the dialog management module, the natural language understanding module, and the natural language generating module in the dialog robot corresponding to at least two of the target tasks, respectively, and when issuing the dialog robot, the configuration module is further configured to: and issuing a uniform conversation interface through which a user can have a conversation with any one of conversation robots respectively corresponding to at least two of the target tasks.

The multi-turn dialog configuration platform provided by the embodiment of the invention has the same implementation principle and technical effect as the method embodiment, and for brief description, reference may be made to the corresponding contents in the method embodiment for non-mention.

Fig. 9 is a schematic diagram of exemplary hardware and software components of an electronic device provided by an embodiment of the present invention, as shown in fig. 9.

The electronic device 100 may be a general-purpose computer or a special-purpose computer, both of which may be used to implement the method of configuring the human-machine dialog system of the present application. Although only a single computer is shown, for convenience, the functions described herein may be implemented in a distributed fashion across multiple similar platforms to balance processing loads.

For example, the electronic device 100 may include a network port 110 connected to a network, one or more processors 120 for executing program instructions, a communication bus 130, and a storage medium 140 of different form, such as a disk, ROM, or RAM, or any combination thereof. Illustratively, the computer platform may also include program instructions stored in ROM, RAM, or other types of non-transitory storage media, or any combination thereof. The method of the present application may be implemented in accordance with these program instructions. The electronic device 200 also includes an Input/Output (I/O) interface 150 between the computer and other Input/Output devices (e.g., keyboard, display screen).

For ease of illustration, only one processor is depicted in electronic device 100. However, it should be noted that the electronic device 100 in the present application may also comprise a plurality of processors, and thus the steps performed by one processor described in the present application may also be performed by a plurality of processors in combination or individually. For example, if the processor of the electronic device 100 executes steps a and B, it should be understood that steps a and B may also be executed by two different processors together or separately in one processor. For example, a first processor performs step a and a second processor performs step B, or the first processor and the second processor perform steps a and B together.

It should be noted that, in the present specification, the embodiments are all described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments may be referred to each other.

The embodiment of the present application further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the computer program performs the steps of the configuration method of the human-computer dialog system according to the above-mentioned method embodiment. For specific implementation, reference may be made to the method embodiment, which is not described herein again.

In particular, the storage medium can be a general-purpose storage medium, such as a removable disk, a hard disk, or the like, and the program code on the storage medium can be executed when being executed to perform the configuration method of the human-computer interaction system shown in the above embodiments.

In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented in other ways. The apparatus embodiments described above are merely illustrative, and for example, the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

In addition, the functional modules in the embodiments of the present invention may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a notebook computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes. It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for configuring a human-computer dialog system, comprising:

receiving a configuration request input by a user;

responding to the configuration request to display a configuration interface;

receiving a first configuration operation, which is input by the user on the configuration interface and aims at a target task, wherein the first configuration operation is used for configuring functional attributes of state points of a finite state machine and functional attributes of directed state edges between the state points, the state points represent dialog state points of a man-machine dialog system for completing the target task, the directed state edges represent dialog states of the man-machine dialog system and can be transferred from one end dialog state point of the directed state edges to the other pointed end dialog state point, and the functional attributes of the directed state edges are used for limiting conditions meeting state transfer;

and responding to the first configuration operation to complete the finite state machine-based dialog management logic configuration of the man-machine dialog system.

2. The method of claim 1, wherein receiving a first configuration operation for a target task input by the user on the configuration interface comprises:

receiving a dialog state point creating operation of the target task input on the configuration interface by the user;

responding to the creating operation of the conversation state point, generating the conversation state point and displaying the conversation state point;

receiving function attribute configuration operation of the user for each dialog state point;

receiving directed state edge creation operation input by the user;

generating and displaying directed state edges between state points corresponding to the directed state edge creation operation based on the directed state edge creation operation;

and receiving configuration operation of the directed state edges input by the user, wherein the configuration operation of the directed state edges is used for configuring transition conditions of the directed state edges.

3. The method of claim 1, wherein receiving a first configuration operation for a target task input by the user on the configuration interface further comprises:

and receiving the configuration operation of the function call or the API call which is input by the user and meets the transfer condition for each directed state edge.

4. The method of claim 3, wherein when the configuration operation on the directed state edge is used to configure a third party API call after the branch condition is satisfied, the method further comprises:

an API configuration interface is displayed on the display,

receiving the name, address, request mode and return parameter setting of the third party API input by the user on the API configuration interface;

and configuring the third-party API based on the settings of the name, the address, the request mode and the return parameters.

5. The method of claim 3, wherein when the configuration operation on the directed state edge is used to configure a function call after a branch condition is satisfied, the method further comprises:

a function configuration interface is displayed and, when the function configuration interface is displayed,

receiving the name, function content and return parameter settings of the function input by the user on the function configuration interface;

and configuring the function based on the name of the function, the content of the function and the setting of the return parameter.

6. The method of claim 1, wherein prior to receiving a first configuration operation for a target task entered by the user on the configuration interface, the method further comprises:

receiving a target task creation request input on the configuration interface by the user;

and completing the creation of the target task according to the conversation field, the conversation intention and the conversation slot position carried in the target task creation request.

7. The method of claim 1, wherein after said displaying a configuration interface in response to said configuration request, said method further comprises:

receiving a second configuration operation input by the user on the configuration interface, wherein the second configuration operation is used for configuring NLU parameters of a natural language understanding module of the man-machine conversation system, and the NLU parameters comprise a conversation field, a conversation intention and a conversation slot position;

and responding to the second configuration operation to complete the NLU parameter configuration of the natural language understanding module of the man-machine conversation system.

8. The method of claim 7, wherein after completing the NLU parameter configuration of the natural language understanding module of the human-computer dialog system in response to the second configuration operation, the method further comprises:

responding to a test instruction which is input on the configuration interface by the user and aims at the natural language understanding module, and generating a test address;

and issuing the test address.

9. The method of claim 1, wherein after completing a finite state machine based dialog management logic configuration of a human machine dialog system in response to the first configuration operation, the method further comprises:

receiving a third configuration operation input by the user on the configuration interface, wherein the third configuration operation is used for configuring natural language generation parameters of the man-machine conversation system;

and responding to the third configuration operation to complete the natural language generation parameter configuration of the man-machine conversation system.

10. The method of claim 9, wherein a component selection list is displayed on the configuration interface, the component selection list comprising a plurality of different purpose selection components; receiving a third configuration operation input by the user on the configuration interface, wherein the third configuration operation comprises:

receiving a component determination instruction input by the user in the component selection list, and displaying a corresponding component configuration interface;

and receiving the component configuration operation input on the configuration interface by the user.

11. The method of claim 1, wherein after configuring the human-machine dialog system corresponding to each of at least two of the target tasks, the method further comprises:

and issuing a uniform conversation interface through which a user can carry out conversation with any one of the man-machine conversation systems respectively corresponding to the at least two target tasks.

12. A multi-turn dialog configuration platform, comprising: a configuration module for configuring the operation of the mobile terminal,

the configuration module is used for receiving a configuration request input by a user;

the system is also used for responding to the configuration request and displaying a configuration interface;

the system is also used for receiving a first configuration operation aiming at a target task, which is input on a configuration interface by a user, wherein the first configuration operation is used for configuring the functional attributes of state points of a finite state machine and the functional attributes of directed state edges between the state points, the state points represent the dialog state points of a man-machine dialog system for completing the target task, the directed state edges represent that the dialog state of the man-machine dialog system can be transferred from the dialog state point at one end of the directed state edge to the dialog state point at the other end, and the functional attributes of the directed state edges are used for limiting the condition of meeting the state transfer;

and is also used for responding to the first configuration operation to complete the finite state machine-based dialog management logic configuration of the man-machine dialog system.

13. The platform of claim 12, wherein the configuration module is further to:

receiving directed state edge creation operation input by the user;

14. The platform of claim 12, wherein the configuration module is further to: and receiving the configuration operation of the function call or the API call which is input by the user and meets the transfer condition for each directed state edge.

15. The platform of claim 14, wherein, when the configuration operation on the directed state edge is to configure a third-party API call that satisfies a branching condition, the configuration module is further to:

receiving the name, address, request mode and return parameter setting of the third party API input by the user on an API configuration interface;

16. The platform of claim 14, wherein when the configuration operation on the directed state edge is to configure a function call that satisfies a branch condition, the configuration module is further to:

receiving the name, the function content and the setting of the return parameters of the function input by the user on a function configuration interface;

17. The platform of claim 12, wherein the configuration module is further to:

18. The platform of claim 12, wherein the configuration module is further to:

19. The platform of claim 18, wherein the configuration module is further configured to:

and issuing the test address.

20. The platform of claim 18, wherein the configuration module is further configured to:

21. The platform of claim 20, wherein the configuration interface has a component selection list displayed thereon, the component selection list including a plurality of different purpose selection components, and wherein the configuration module is further configured to:

22. The platform of claim 12, wherein after configuring the human-machine dialog systems corresponding to the at least two target tasks, the configuration module is further configured to:

23. An electronic device, comprising: a processor, a storage medium and a bus, the storage medium storing machine-readable instructions executable by the processor, the processor and the storage medium communicating via the bus when the electronic device is operating, the processor executing the machine-readable instructions to perform the steps of the method for configuring a human-machine dialog system according to any of claims 1 to 11.

24. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a computer program which, when being executed by a processor, carries out the steps of the method for configuration of a human-computer dialog system according to any of claims 1 to 11.