Task-based robot interaction method, apparatus, device, and storage medium
By introducing exception handling procedures to handle user response anomalies in task-oriented robots, the design and maintenance of multi-turn dialogues are simplified, solving the problem of high complexity in existing technologies and achieving a clearer program architecture and easier-to-maintain interaction design.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHINA PING AN LIFE INSURANCE CO LTD
- Filing Date
- 2022-08-04
- Publication Date
- 2026-05-29
AI Technical Summary
Existing task-oriented robot interaction designs are complex, and multi-turn dialogues increase the difficulty of design and maintenance, especially when dealing with abnormal user responses, the program structure is cumbersome and inconvenient to maintain.
An exception procedure is introduced to handle abnormal situations in the robot's main program. When the user's answer does not match the dominant question and answer template, the system switches to the exception question and answer template to ask the abnormal question, resolves the abnormal answer, and returns to the corresponding stage in the main program. The exception procedure is designed and maintained independently to simplify the architecture.
It reduces the complexity of the internal architecture of task robots, simplifies the design and maintenance, and improves the transparency and maintainability of the robot's main program.
Smart Images
Figure CN115374260B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of artificial intelligence technology, and in particular to a task-oriented robot interaction method, apparatus, device, and storage medium. Background Technology
[0002] Task-oriented chatbots, also known as task-oriented conversational robots, engage in multi-turn dialogues to meet a user's specific needs. They determine the user's intent through understanding and clarification, and then complete the task by responding or calling APIs.
[0003] With the development of artificial intelligence, the application scope of task robots is becoming wider and wider, and they can be used in insurance purchase, outbound call robots, etc. Due to the complexity of user needs, it is usually necessary to set up multi-round interactive dialogues. Users continuously modify and improve their needs, and the robot helps users clarify their purpose by asking questions, clarifying and confirming. Multi-round interactive dialogues increase the complexity of task robots, and the design and maintenance of task robots are difficult. Summary of the Invention
[0004] The main objective of this disclosure is to provide a task-oriented robot interaction method, apparatus, device, and storage medium that can effectively reduce the complexity of the robot's internal architecture and make the design and maintenance of the task-oriented robot easier.
[0005] To achieve the above objectives, a first aspect of this disclosure provides a task-oriented robot interaction method, comprising:
[0006] The robot's main program uses a leading question-and-answer template to ask users leading questions.
[0007] Obtain the user's dominant answer to the dominant question;
[0008] When it is detected that the dominant answer does not match the preset answer in the dominant question and answer template, an abnormal procedure is entered and the main program of the robot is suspended.
[0009] An abnormal question is sent to the user through the abnormal question and answer template corresponding to the main answer in the abnormal procedure;
[0010] Obtain the user's abnormal response to the abnormal question, continue running the robot's main program and enter the stage in the robot's main program corresponding to the abnormal response, and then exit the abnormal program.
[0011] In some embodiments, prior to issuing a leading question to the user via a leading question-and-answer template in the robot's main program, the method further includes:
[0012] The robot's main program initiates an interaction request to the user.
[0013] When the user agrees to the interaction request, an interactive dialogue is established between the robot's main program and the user to send questions to the user or receive the user's answers.
[0014] In some embodiments, the step of entering an abnormal procedure and pausing the main robot program when it is detected that the dominant answer does not match the preset answer in the dominant question-and-answer template includes:
[0015] User keywords were obtained based on the analysis of the dominant responses.
[0016] When it is detected that the user keywords do not match the preset keywords in the dominant question and answer template, the abnormal program is entered and the main program of the robot is suspended. The preset keywords are extracted from the preset answers in the dominant question and answer template.
[0017] In some embodiments, issuing an abnormal question to the user through an abnormal question-and-answer template corresponding to the dominant answer in the abnormal procedure includes:
[0018] Select the abnormal question and answer template in the abnormal program that corresponds to the user keyword;
[0019] The error question template is used to send error questions to users.
[0020] In some embodiments, after obtaining the user's abnormal answer to the abnormal question, continuing to run the robot's main program and entering the stage in the robot's main program corresponding to the abnormal answer, and exiting the abnormal program, the method further includes:
[0021] The dominant question is asked again to the user using the dominant question-and-answer template, and the dominant answer from the user to the dominant question is obtained.
[0022] When it is detected that the dominant answer does not match the preset answer in the dominant question and answer template, after entering the abnormal procedure, an abnormal question is sent to the user through the abnormal question and answer template corresponding to the dominant answer in the abnormal procedure. Based on the user's abnormal answer to the abnormal question, the robot's main program continues to run and enters the stage in the robot's main program corresponding to the abnormal answer, and then exits the abnormal procedure.
[0023] In some embodiments, after obtaining the user's dominant answer to the dominant question, the method further includes:
[0024] When the dominant answer is identified as matching a preset answer in the dominant question-and-answer template, the process proceeds to the stage in the robot's main program corresponding to the dominant answer.
[0025] In some embodiments, the step of entering the stage corresponding to the dominant answer in the robot's main program when the dominant answer is identified as matching a preset answer in the dominant question-and-answer template includes:
[0026] When the dominant answer is identified as matching a preset answer in the dominant question-and-answer template, the user's business information is obtained;
[0027] Based on the business information and the dominant response, proceed to the corresponding stage in the robot's main program.
[0028] To achieve the above objectives, a second aspect of this disclosure provides a task-oriented robot interaction device, comprising:
[0029] The leading question-and-answer module is used to issue leading questions to the user through the leading question-and-answer template in the robot's main program;
[0030] The monitoring module is used to obtain the user's dominant answer to the dominant question;
[0031] The analysis module is used to enter an abnormal program and pause the robot's main program when it is detected that the dominant answer does not match the preset answer in the dominant question and answer template.
[0032] The abnormal question and answer module is used to send abnormal questions to the user through the abnormal question and answer template corresponding to the main answer in the abnormal program;
[0033] The jump module is used to obtain the user's abnormal answer to the abnormal question, continue running the robot's main program and enter the segment in the robot's main program corresponding to the abnormal answer, and exit the abnormal program.
[0034] To achieve the above objectives, a third aspect of the present disclosure provides an electronic device including at least one memory;
[0035] At least one processor;
[0036] At least one computer program;
[0037] The computer program is stored in the memory, and the processor executes the at least one computer program to achieve:
[0038] The task-oriented robot interaction method as described in any of the embodiments of the first aspect.
[0039] To achieve the above objectives, a fourth aspect of this disclosure also provides a computer-readable storage medium storing computer-executable instructions for causing a computer to perform:
[0040] The task-oriented robot interaction method as described in any of the embodiments of the first aspect.
[0041] The task-oriented robot interaction method, apparatus, device, and storage medium proposed in this disclosure embodiment utilize an exception program to handle abnormal situations in the robot's main program. During interactive dialogue with the user, after the robot's main program issues a leading question to the user using a leading question-and-answer template, if it detects that the user's answer to the leading question does not match the preset answer in the leading question-and-answer template (i.e., an answer is abnormal), the exception program is invoked. Based on the answer to the leading question, the corresponding abnormal question-and-answer template in the exception program is selected to issue an abnormal question to the user, thereby resolving the issue of the user's abnormal answer. Finally, based on the user's answer to the abnormal question, the system returns to the corresponding stage in the robot's main program to continue the interactive dialogue between the robot's main program and the user. The exception program and the robot's main program are separate programs that can be designed and maintained independently, thereby reducing the difficulty of design and maintenance. The robot's main program is easy to understand, and any stage in the robot's main program that detects an answer abnormality can be handled through the exception program, effectively reducing the complexity of the robot's internal architecture. Attached Figure Description
[0042] Figure 1 This is a flowchart of a task-oriented robot interaction method provided in an embodiment of this application;
[0043] Figure 2 yes Figure 1 A flowchart of the specific methods prior to step S100;
[0044] Figure 3 yes Figure 1 A flowchart illustrating the specific method of step S300;
[0045] Figure 4 yes Figure 1 A flowchart illustrating the specific method of step S400;
[0046] Figure 5 yes Figure 1 A flowchart of the specific methods following step S500;
[0047] Figure 6 yes Figure 1 A flowchart of the specific methods following step S200;
[0048] Figure 7 yes Figure 6 A flowchart illustrating the specific method of step S700 in the middle section;
[0049] Figure 8This is a schematic diagram of a specific application scenario of the task-oriented robot interaction method provided in the embodiments of this application;
[0050] Figure 9 This is a block diagram of the task-oriented robot interaction device provided in the embodiments of this application;
[0051] Figure 10 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation
[0052] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.
[0053] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.
[0054] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.
[0055] First, let's analyze some of the terms used in this application:
[0056] Artificial Intelligence (AI) is a new branch of computer science that studies, develops, and applies theories, methods, technologies, and systems to simulate, extend, and expand human intelligence. It aims to understand the essence of intelligence and produce intelligent machines that can react in a way similar to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing, and expert systems. AI can simulate the information processes of human consciousness and thought. Furthermore, AI utilizes digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceiving the environment, acquiring knowledge, and using that knowledge to achieve optimal results.
[0057] Natural Language Processing (NLP): NLP uses computers to process, understand, and utilize human language (such as Chinese and English). It is a branch of artificial intelligence and an interdisciplinary field of computer science and linguistics, often referred to as computational linguistics. NLP includes syntactic analysis, semantic analysis, and discourse understanding. It is commonly used in machine translation, handwritten and printed character recognition, speech recognition and text-to-speech conversion, information retrieval, information extraction and filtering, text classification and clustering, sentiment analysis, and opinion mining. It involves data mining, machine learning, knowledge acquisition, knowledge engineering, artificial intelligence research, and linguistic research related to language computation.
[0058] Task-oriented chatbots, also known as task-based conversational robots, are used to help users complete specific tasks, such as choosing insurance or using outbound calling robots. To meet a user's need, task-oriented chatbots engage in multi-turn dialogues. They understand, clarify, and confirm the user's intent, then complete the task by responding or calling APIs.
[0059] In practical applications, due to the complexity of user needs, multiple rounds of interactive dialogue are usually required to complete the corresponding tasks. Users continuously modify and improve their needs, and the robot helps users clarify their intentions by asking questions, clarifying, and confirming. For outbound call robots, the main process is generally clear, but exceptions may occur in each round of dialogue during the main process. For example, users may reply with "I can't hear you," "Listen again," "Wrong number," or "I don't want to listen."
[0060] In related technologies, by configuring a corresponding dialogue tree in each round of dialogue in the main process of a task-oriented robot, when an exception occurs in each round of dialogue, the exception can be directly handled through the dialogue tree set in that round of dialogue. However, the program structure of a task-oriented robot with this interactive design is complex and cannot see through the main process. Especially for the more rounds of dialogue, the more complex the program structure of the task-oriented robot with this interactive design becomes, the more tedious and complicated the interactive design of the dialogue is, and maintenance requires targeting the corresponding dialogue tree, which is inconvenient.
[0061] Based on this, the present disclosure provides a task-oriented robot interaction method, apparatus, device, and storage medium, which can effectively reduce the complexity of the robot's internal architecture and make the design and maintenance of the task-oriented robot easier.
[0062] This application provides a task-oriented robot interaction method, apparatus, device, and storage medium, which are specifically described through the following embodiments. First, the task-oriented robot interaction method in the embodiments of this disclosure is described.
[0063] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.
[0064] Foundational technologies for artificial intelligence generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.
[0065] The task-oriented robot interaction method provided in this application relates to the field of artificial intelligence technology, and particularly to the field of data mining technology. The task-oriented robot interaction method provided in this application can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, or smartwatch, etc.; the server can be an independent server or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms; the software can be an application implementing the task-oriented robot interaction method, but is not limited to the above forms.
[0066] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.
[0067] Firstly, please refer to Figure 1 , Figure 1 This is an optional flowchart of a task-oriented robot interaction method provided in some embodiments of this application. The task-oriented robot interaction method includes steps S100, S200, S300, S400, and S500. It should be understood that the task-oriented robot interaction method in the embodiments of this application includes, but is not limited to, steps S100 to S500. The following is in conjunction with... Figure 1 These six steps will be explained in detail.
[0068] Step S100: Send a leading question to the user through the leading question-and-answer template in the robot's main program.
[0069] Specifically, in step 100 of some embodiments, the robot main program has multiple stages, each with a corresponding leading question and answer template. Based on the stage of the robot main program, the robot sends a corresponding leading question to the user according to the leading question and answer template of that stage, enabling it to engage in a dialogue with the user.
[0070] When the task-oriented robot is an outbound call robot, after establishing an interactive dialogue with the user, it sends a leading question to the user through the leading question-and-answer template in the robot's main program. At the moment when the robot's main program has just established an interactive dialogue with the user, the leading question-and-answer template at this time is the opening dialogue template, and the leading question corresponding to the opening dialogue template is the opening dialogue voice, which is sent to the user.
[0071] It should be noted that after the main program of the robot establishes an interactive dialogue with the user, the main program of the robot can receive the user's voice and can also send voice messages to the user.
[0072] Step S200: Obtain the user's dominant answer to the dominant question.
[0073] Specifically, in step S200 of some embodiments, the user answers the main question via voice, the robot's main program acquires the user's voice, parses the user's voice, and obtains the user's main answer to the main question.
[0074] It should be noted that the dominant answer can represent the user's intent. The main program of the robot analyzes and identifies the dominant answer in order to determine the user's intent.
[0075] Step S300: When it is detected that the dominant answer does not match the preset answer in the dominant question and answer template, enter the exception procedure and suspend the operation of the robot's main program.
[0076] Specifically, in step S300 of some embodiments, after obtaining and parsing the dominant answer, in the corresponding stage of the robot's main program, the dominant question and answer template in this stage has a corresponding preset answer, which can also be regarded as a preset answer library. When it is identified that the dominant answer does not match the preset answer of the dominant question and answer template in this stage, that is, when it is identified that the dominant answer is not in the preset answer library of the dominant question and answer template in this stage, the abnormal program is entered and the robot's main program is suspended. At this time, the abnormal program establishes a dialogue interaction with the user, and the abnormal program can send voice to the user or receive the user's voice.
[0077] It should be noted that exception programs can be regarded as one part of the robot's main program, but exception programs can be stored, managed and modified independently, which can make the main process corresponding to the robot's main program clearer, and at the same time facilitate the design and maintenance of task-oriented robots.
[0078] Step S400: Send an exception question to the user through the exception question and answer template corresponding to the main answer in the exception procedure.
[0079] It should be noted that the exception program has multiple default exception question and answer templates. Different exception question and answer templates are used to deal with different exception situations to ensure that the task robot can perform the corresponding tasks smoothly. Exception situations are also called exceptional situations.
[0080] In step S400 of some embodiments, after entering the exception procedure, based on the user's dominant answer, an exception question-and-answer template corresponding to the dominant answer is selected to deal with the exception situation, and an exception question is issued to the user according to the corresponding exception question-and-answer template. When the user's dominant answer is "no time", the exception question corresponding to the exception question-and-answer template can be a reassuring voice.
[0081] Step S500: Obtain the abnormal response from the user to the abnormal question, continue running the robot's main program and enter the segment in the robot's main program corresponding to the abnormal response, and exit the abnormal program.
[0082] In step S500 of some embodiments, after sending an abnormal question to the user, the user answers the abnormal question via voice. The abnormal program can obtain the user's voice at this time, and by parsing the user's voice, it can obtain the user's abnormal answer to the abnormal question. Based on the parsed abnormal answer, the abnormal program selects the corresponding link in the robot's main program, continues to run the robot's main program, and enters the link in the robot's main program corresponding to the abnormal answer. At this time, the abnormal program's task is completed, and it exits. The robot's main program continues to run from the link corresponding to the abnormal answer to realize the corresponding main flow task. The link in the robot program corresponding to the abnormal program can be a link that has already been run. For example, if the user's dominant answer is "cannot hear clearly," after calming the user's emotions with a reassuring voice in the abnormal question, it can return to the previously run link where the user's dominant answer was "cannot hear clearly," and reissue the dominant question to the user according to the corresponding dominant question-and-answer template.
[0083] The task-oriented robot interaction method proposed in this application embodiment uses an exception procedure to handle abnormal situations in the robot's main program. During the interactive dialogue between the robot's main program and the user, after the robot's main program sends a leading question to the user through a leading question-and-answer template, if it detects that the user's answer to the leading question does not match the preset answer in the leading question-and-answer template (i.e., an answer is abnormal), the exception procedure is invoked. Based on the answer to the leading question, the corresponding abnormal question-and-answer template in the exception procedure is selected to send an abnormal question to the user, thereby resolving the problem of the user's abnormal answer. Finally, based on the user's answer to the abnormal question, the system returns to the corresponding stage in the robot's main program to continue the interactive dialogue between the robot's main program and the user. The exception procedure and the robot's main program are different programs and can be designed and maintained independently, thereby reducing the difficulty of design and maintenance. The robot's main program is easy to understand, and any stage in the robot's main program that detects an answer abnormality can be handled through the exception procedure, which can effectively reduce the complexity of the robot's internal architecture.
[0084] Please refer to Figure 2 , Figure 2 This is a flowchart illustrating the specific methods prior to step S100 in some embodiments of this application. In some embodiments of this application, steps S010 and S020 are included, but are not limited to, prior to step S100. The following describes these methods in conjunction with... Figure 2 These two steps will be explained in detail.
[0085] Step S010: Initiate an interaction request to the user through the robot's main program;
[0086] Step S020: When the user agrees to the interaction request, establish an interactive dialogue between the robot's main program and the user to send questions to the user or receive the user's answers.
[0087] Specifically, in this embodiment, the task robot can be an outbound call robot that initiates an interaction request to the user by dialing or other means. When the user answers the call, the user agrees to the interaction request, and an interactive dialogue is established between the robot's main program and the user. The task robot has natural language processing capabilities and can process speech. The robot's main program can send questions to the user or receive the user's answers.
[0088] It is understandable that the establishment of interactive dialogue between the robot's main program and the user is not limited to the robot's main program initiating an interaction request to the user. When the user actively initiates an interaction request to the robot's main program, the robot's main program can also establish an interactive dialogue with the user by accepting the user's interaction request, and subsequent steps can proceed normally.
[0089] Please refer to Figure 3 , Figure 3 This is a flowchart illustrating a specific method for step S300 provided in some embodiments of this application. In some embodiments of this application, step S300 also includes, but is not limited to, steps S310 and S320. The following is in conjunction with… Figure 3 These two steps will be explained in detail.
[0090] Step S310: Obtain user keywords based on the analysis of the dominant response;
[0091] Step S320: When it is detected that the user's keywords do not match the preset keywords in the main question and answer template, the abnormal program is entered and the robot's main program is paused. The preset keywords are extracted from the preset answers in the main question and answer template.
[0092] It is understood that the task-oriented robot in some embodiments of the task-oriented robot interaction method of this application is a task-oriented dialogue robot with natural language processing capabilities. Based on the user's voice answer to the leading question, it is converted into text through speech recognition to obtain the leading answer. The leading answer is analyzed according to a preset word library to obtain user keywords in the leading answer that match the preset word library. These user keywords can represent the user's intention. By recognizing and judging the user keywords, the next step to be entered can be obtained.
[0093] When a user's keywords are detected as not matching the preset keywords of the dominant question-and-answer template in the current stage, an anomaly is identified. The system then jumps from the main robot program to an exception handler to handle the anomaly, pausing the main robot program's operation; in other words, the main robot program is placed in standby mode. By using an exception handler independent of the main robot program, the difficulty of designing the robot's program architecture is reduced, and the maintenance of the abnormal question-and-answer template is performed within the exception handler.
[0094] Please refer to Figure 4, Figure 4 This is a detailed flowchart of step S400 provided in some embodiments of this application. In some embodiments, step S400 includes, but is not limited to, steps S410 and S420. The following is a detailed description of step S400 in conjunction with... Figure 4 These two steps will be described in detail.
[0095] Step S410: Select the exception question and answer template in the exception program that corresponds to the user's keywords;
[0096] Step S420: Send an exception question to the user using the exception question and answer template.
[0097] It should be noted that the exception program has multiple exception question and answer templates. Different exception question and answer templates are used to handle different exception situations. By managing the exception program, iterative updates and maintenance of exception situations can be achieved in a unified manner, which is convenient. In step S310, user keywords are obtained based on the analysis of the dominant answer; and in step S320, when it is found that the user keywords do not match the preset keywords in the dominant question and answer template, the exception program is entered and the robot's main program is paused; based on the obtained user keywords, the exception question and answer template corresponding to the user keywords in the exception program is selected, and an exception question is sent to the user through the exception question and answer template. The exception question is sent to the user in the form of voice.
[0098] Please refer to Figure 5 The flowchart below shows the specific method after step S500 in some embodiments of this application. In some embodiments, steps S610 and S620 are included, but are not limited to, after step S500. Figure 5 These two steps will be described in detail.
[0099] Step S610: Send the leading question to the user again through the leading question and answer template, and obtain the user's leading answer to the leading question;
[0100] Step S620: When it is detected that the dominant answer does not match the preset answer in the dominant question and answer template, after entering the abnormal procedure, an abnormal question is sent to the user through the abnormal question and answer template corresponding to the dominant answer in the abnormal procedure. Based on the user's abnormal answer to the abnormal question, the robot main program continues to run and enters the link corresponding to the abnormal answer in the robot main program, and then exits the abnormal procedure.
[0101] It is understood that, in the task-oriented robot interaction method provided in some embodiments of the application, after the first round of dialogue is completed through steps S100, S200, S300, S400 and S500, the exception procedure can also be used to handle other abnormal situations during the operation of the robot's main program.
[0102] Specifically, after obtaining an abnormal response from the user to an abnormal question, the robot's main program continues running and enters the stage corresponding to the abnormal response within the main program, then exits the abnormal program. The robot's main program continues running from the stage corresponding to the abnormal response, and based on the dominant question-and-answer template corresponding to the current stage, it sends a corresponding dominant question to the user. After the user answers the dominant question via voice, the dominant response is obtained by analyzing the user's voice. The dominant response is identified; if it does not match the preset response of the dominant question-and-answer template in the current stage, the abnormal program is entered and the robot's main program is paused. An abnormal question-and-answer template corresponding to the dominant response is selected in the abnormal program, and an abnormal question is sent to the user based on this template. After the user answers the abnormal question, the abnormal response is obtained by analyzing the user's voice. The stage corresponding to the abnormal response in the main program is selected, the robot's main program continues running and enters the stage corresponding to the abnormal response, and then exits the abnormal program.
[0103] In this way, exception handling procedures can be applied to all stages of the robot's main program. When an exception occurs in any stage of the robot's main program, it can be handled accordingly by entering the exception handling procedure. Task-oriented robots mainly consist of a main program and exception handling procedures. These procedures handle the main task flow and exceptions respectively. The overall internal architecture of the robot is simple, the design difficulty of task-oriented robots is low, and maintenance only requires maintaining the corresponding exception handling module within the exception handling procedure, making maintenance convenient.
[0104] Please refer to Figure 6 The flowchart below shows the specific method after step S200 in some embodiments of this application. In some embodiments, step S200 includes, but is not limited to, step S700. Figure 6 This step will be described in detail.
[0105] Step S700: When the dominant answer is identified as matching the preset answer in the dominant question and answer template, the robot enters the stage corresponding to the dominant answer in the main program.
[0106] Understandably, the different stages of the robot's main program are interconnected to implement the main task flow. The master question-and-answer template can have one or more preset answers, and different preset answers are used to associate with corresponding stages in the robot's main program to achieve the completion of different tasks. When a user answers the master question via voice, analyzing the user's voice response yields the master answer. When a master answer is identified as matching a preset answer in the master question-and-answer template for that stage, the robot enters the stage in the main program corresponding to that master answer, thereby advancing the task's progress.
[0107] Please refer to Figure 7 , Figure 7 This is a detailed flowchart of step S700 provided in some embodiments of this application. In some embodiments, step S700 includes, but is not limited to, steps S710 and S720. The following is a detailed description of step S700 in conjunction with... Figure 7 These two steps will be described in detail.
[0108] Step S710: When the dominant answer is identified as matching the preset answer in the dominant question and answer template, obtain the user's business information;
[0109] Step S720: Based on the business information and the main response, proceed to the corresponding stage in the robot's main program.
[0110] Specifically, through the natural language processing capabilities of the task-oriented robot, user keywords are obtained based on the analysis of the dominant answer. When the user keywords are identified as matching the preset keywords in the dominant question-and-answer template, the user's business information is obtained. This business information includes pre-existing order information, points information, etc. Based on the user's business information and the dominant answer, the robot enters the corresponding stage in the main program of the dominant answer and business information. In this stage, the robot sends the corresponding dominant question to the user or ends the dialogue through the dominant question-and-answer template.
[0111] Understandably, to improve the functionality of the robot's main program, when the dominant answer is identified as matching the preset answer in the dominant question-and-answer module of this stage, the user's business information is obtained, and the robot enters the corresponding stage in the main program based on the dominant answer and business information. This enables multi-condition association effects and allows for more complex task requirements to be met.
[0112] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards of the relevant countries and regions. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirects to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data for the proper functioning of the embodiments of this application obtained.
[0113] Please refer to Figure 8 , Figure 8 This is a schematic diagram illustrating a specific application scenario of the task-oriented robot interaction method provided in this application embodiment. In this application scenario, the task-oriented robot corresponds to an outbound call robot.
[0114] The following examples illustrate two different approaches to task-oriented robot interaction in this application scenario:
[0115] Scenario 1:
[0116] First, after the main robot program establishes an interactive dialogue with the user, the main robot program enters the opening remarks stage. It sends the opening remarks content to the user through the opening remarks Q&A module in the opening remarks stage. The opening remarks Q&A module is the main Q&A module in the opening remarks stage, and the opening remarks content is the main question asked by the opening remarks Q&A module.
[0117] Then, the user's dominant response to the opening remarks was "I don't know".
[0118] If an unclear answer is detected and does not match the preset answer in the opening Q&A template, the system will enter the abnormal operation procedure and suspend the main program of the robot.
[0119] Select the abnormal question and answer template corresponding to the unclear answer in the abnormal program as the relistening question and answer template. Send the relistening message to the user through the relistening question and answer template. The relistening message is the abnormal question in the relistening question and answer template.
[0120] If the user's abnormal response to the re-listening script is obtained as a re-listening response, the main robot program continues to run and enters the opening speech segment corresponding to the re-listening response in the main robot program, and then exits the abnormal program.
[0121] The robot's main program sends the opening remarks to the user again through the opening remarks Q&A module in the opening remarks section;
[0122] The user's primary response to the opening remarks is used to determine their identity.
[0123] The system identifies responses that confirm identity and determines if they match the preset responses in the opening Q&A template.
[0124] The system will determine your identity and answer accordingly, then proceed to the payment reminder section in the main robot program.
[0125] The payment content is sent to the user through the payment reminder section of the robot's main program. The payment reminder module is the main question and answer module in the payment reminder section, and the payment content is the main question asked by the payment reminder module.
[0126] The user's primary response regarding payment details is their "payment awareness" response.
[0127] The system identifies the answers to the payment notification questions and determines whether they match the preset answers in the payment Q&A template. It also retrieves the user's business information, which includes multiple insurance policies.
[0128] Based on the known answers and multiple policy information, proceed to the multiple policy stage in the robot's main program;
[0129] The robot's main program sends multi-policy status information to users through the multi-policy Q&A template in the multi-policy process. The multi-policy Q&A template is the dominant Q&A template in the multi-policy process, and the multi-policy status information is the dominant question in the multi-policy process.
[0130] The user's primary response to the content regarding the status of multiple insurance policies is the "multiple insurance policies awareness response";
[0131] The system identifies responses regarding awareness of multiple insurance policies and matches these responses with preset responses in the multiple insurance policy Q&A template.
[0132] Based on the answers provided by the multiple insurance policies, the system proceeds to the closing stage in the main robot program. After completing the closing stage, the interaction between the main robot program and the user ends.
[0133] Scenario 2:
[0134] In this application scenario, the task-oriented robot corresponds to the outbound call robot.
[0135] First, after the main robot program establishes an interactive dialogue with the user, the main robot program enters the opening remarks stage. It sends the opening remarks content to the user through the opening remarks Q&A module in the opening remarks stage. The opening remarks Q&A module is the main Q&A module in the opening remarks stage, and the opening remarks content is the main question asked by the opening remarks Q&A module.
[0136] Then, the dominant response from users to the opening remarks was a low-intention response;
[0137] The system identifies low-intention responses. If a low-intention response does not match the preset response to the opening remarks, it enters an abnormal operation procedure and suspends the main program of the robot.
[0138] Select the abnormal Q&A template corresponding to the low willingness response in the abnormal program as the low willingness Q&A template, and send the user a reassuring message through the low willingness Q&A template. The reassuring message is the abnormal question in the low willingness Q&A template.
[0139] If the user's abnormal response to the reassurance message is identified as an emotion-relieving response, the main robot program will continue running and proceed to the payment reminder step corresponding to the emotion-relieving response within the main robot program.
[0140] The payment content is sent to the user through the payment reminder section of the robot's main program. The payment reminder module is the main question and answer module in the payment reminder section, and the payment content is the main question asked by the payment reminder module.
[0141] The user's primary response to the payment details was "I don't know".
[0142] The system identifies unclear answers. If an unclear answer does not match the preset answer in the payment Q&A template, it enters the abnormal operation program and suspends the main program of the robot.
[0143] Select the abnormal question and answer template corresponding to the unclear answer in the abnormal program as the relistening question and answer template. Send the relistening message to the user through the relistening question and answer template. The relistening message is the abnormal question in the relistening question and answer template.
[0144] If the user's abnormal response to the repeated dialogue content is obtained as a repeated response, the main robot program will continue to run and enter the payment reminder step corresponding to the repeated response in the main robot program, and then exit the abnormal program.
[0145] The payment information will be sent to the user again through the payment reminder template in the main program of the robot.
[0146] The user's primary response regarding payment details is their "payment awareness" response.
[0147] The system identifies the answers to the payment notification questions and determines whether they match the preset answers in the payment Q&A template. It also retrieves the user's business information, which includes multiple insurance policies.
[0148] Based on the known answers and multiple policy information, proceed to the multiple policy stage in the robot's main program;
[0149] The robot's main program sends multi-policy status information to users through the multi-policy Q&A template in the multi-policy process. The multi-policy Q&A template is the dominant Q&A template in the multi-policy process, and the multi-policy status information is the dominant question in the multi-policy process.
[0150] The user's primary response to the content regarding the status of multiple insurance policies is the "multiple insurance policies awareness response";
[0151] The system identifies responses regarding awareness of multiple insurance policies and matches these responses with preset responses in the multiple insurance policy Q&A template.
[0152] Based on the answers provided by the multiple insurance policies, the system proceeds to the closing stage in the main robot program. After completing the closing stage, the interaction between the main robot program and the user ends.
[0153] Please refer to Figure 9 , Figure 9 This is a block diagram of a task-oriented robot interaction device 1000 provided in some embodiments of this application. In some embodiments, the task-oriented robot interaction device includes a leading question-and-answer module 1100, a monitoring module 1200, an analysis module 1300, an abnormal question-and-answer module 1400, and a jump module 1500.
[0154] The leading question-and-answer module 1100 is used to send leading questions to users through the leading question-and-answer template in the robot's main program;
[0155] The monitoring module 1200 is used to obtain the user's dominant answer to the dominant question;
[0156] Analysis module 1300 is used to enter an abnormal procedure and pause the robot's main program when it is detected that the dominant answer does not match the preset answer in the dominant question and answer template;
[0157] The exception question and answer module 1400 is used to send exception questions to users through the exception question and answer template corresponding to the main answer in the exception program;
[0158] Jump module 1500 is used to obtain the abnormal answer of the user to the abnormal question, continue to run the robot main program and enter the link in the robot main program corresponding to the abnormal answer, and exit the abnormal program.
[0159] The task-oriented robot interaction device of this application embodiment uses an exception program to handle exceptions in the robot's main program. During the interaction between the robot's main program and the user, after the robot's main program sends a leading question to the user through a leading question-and-answer template, if it detects that the user's answer to the leading question does not match the preset answer in the leading question-and-answer template (i.e., an answer is abnormal), the exception program is invoked. Based on the answer to the leading question, the corresponding abnormal question-and-answer template in the exception program is selected to send an abnormal question to the user, thereby resolving the problem of the user's abnormal answer. Finally, based on the user's answer to the abnormal question, the system returns to the corresponding stage in the robot's main program to continue the interaction between the robot's main program and the user. The exception program and the robot's main program are different programs and can be designed and maintained independently, thereby reducing the difficulty of design and maintenance. The robot's main program is easy to understand, and any stage in the robot's main program that detects an answer abnormality can be handled through the exception program, which can effectively reduce the complexity of the robot's internal architecture.
[0160] It should be noted that the task-oriented robot interaction device in this application corresponds to the aforementioned task-oriented robot interaction method. For specific interaction methods, please refer to the aforementioned task-oriented robot interaction method, which will not be elaborated here.
[0161] This application also provides an electronic device, including:
[0162] At least one memory;
[0163] At least one processor;
[0164] At least one program;
[0165] The program is stored in memory, and the processor executes at least one program to implement the task-oriented robot interaction method described above in this disclosure. The electronic device can be any smart terminal, including mobile phones, tablets, personal digital assistants (PDAs), in-vehicle computers, etc.
[0166] The electronic device in this application embodiment is used to execute the above-described task-oriented robot interaction method. An exception procedure is set up to handle exceptions in the robot's main program. During interactive dialogue with the user, after the robot's main program sends a leading question to the user using a leading question-and-answer template, if it detects that the user's answer to the leading question does not match the preset answer in the leading question-and-answer template (i.e., an answer is abnormal), the exception procedure is invoked. Based on the answer to the leading question, the corresponding abnormal question-and-answer template in the exception procedure is selected to send an abnormal question to the user, thereby resolving the issue of the user's abnormal answer. Finally, based on the user's answer to the abnormal question, the system returns to the corresponding stage in the robot's main program to continue the interactive dialogue between the robot's main program and the user. The exception procedure and the robot's main program are different programs and can be designed and maintained independently, thereby reducing the difficulty of design and maintenance. The robot's main program is easy to understand, and any stage in the robot's main program that detects an answer abnormality can be handled through the exception procedure, effectively reducing the complexity of the robot's internal architecture.
[0167] The following is combined with Figure 10 The electronic devices described in the embodiments of this application will be described in detail.
[0168] like Figure 10 , Figure 10 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes:
[0169] The processor 1600 can be implemented using a general-purpose central processing unit (CPU), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this disclosure.
[0170] The memory 1700 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 1700 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1700 and is called and executed by the processor 1600 to execute the task-oriented robot interaction method of the embodiments of this disclosure.
[0171] The input / output interface 1800 is used to implement information input and output.
[0172] The communication interface 1900 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).
[0173] Bus 2000 transmits information between various components of the device (e.g., processor 1600, memory 1700, input / output interface 1800, and communication interface 1900);
[0174] The processor 1600, memory 1700, input / output interface 1800 and communication interface 1900 are connected to each other within the device via bus 2000.
[0175] This disclosure also provides a storage medium, which is a computer-readable storage medium storing computer-executable instructions for causing a computer to perform the above-described task-oriented robot interaction method.
[0176] The storage medium of this application embodiment is used to execute the above-described task-oriented robot interaction method. An exception procedure is set up to handle exceptions in the robot's main program. During interactive dialogue with the user, after the robot's main program sends a leading question to the user using a leading question-and-answer template, if it detects that the user's answer to the leading question does not match the preset answer in the leading question-and-answer template (i.e., an answer is abnormal), the exception procedure is invoked. Based on the answer to the leading question, the corresponding abnormal question-and-answer template in the exception procedure is selected to send an abnormal question to the user, thereby resolving the issue of the user's abnormal answer. Finally, based on the user's answer to the abnormal question, the system returns to the corresponding stage in the robot's main program to continue the interactive dialogue between the robot's main program and the user. The exception procedure and the robot's main program are different programs and can be designed and maintained independently, thereby reducing the difficulty of design and maintenance. The robot's main program is easy to understand, and any stage in the robot's main program that detects an answer abnormality can be handled through the exception procedure, effectively reducing the complexity of the robot's internal architecture.
[0177] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.
[0178] The embodiments described in this disclosure are for the purpose of more clearly illustrating the technical solutions of this disclosure and do not constitute a limitation on the technical solutions provided by this disclosure. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by this disclosure are also applicable to similar technical problems.
[0179] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this disclosure, and may include more or fewer steps than shown, or combine certain steps, or different steps.
[0180] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.
[0181] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.
[0182] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.
[0183] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.
[0184] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.
[0185] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.
[0186] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.
[0187] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause an electronic device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0188] The preferred embodiments of the present disclosure have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present disclosure. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and spirit of the present disclosure shall be within the scope of the claims of the present disclosure.
Claims
1. A task-oriented robot interaction method, characterized in that, When the method is used in a multi-turn dialogue, the method includes: The robot sends a leading question to the user through a leading question-and-answer template corresponding to one of the multiple stages of the main program; Obtain the user's dominant answer to the dominant question; The dominant answer is analyzed based on a preset thesaurus to obtain user keywords in the dominant answer that match the preset thesaurus; When it is detected that the user keywords do not match the preset keywords in the main question and answer template, an abnormal program is entered and the main program of the robot is suspended. The preset keywords are extracted from the preset answers in the main question and answer template. An abnormal question is sent to the user through the abnormal question and answer template corresponding to the main answer in the abnormal procedure; Obtain the user's abnormal response to the abnormal question, continue running the robot's main program and enter the stage corresponding to the abnormal response in the multiple stages of the robot's main program, and exit the abnormal program. The stage corresponding to the abnormal response in the robot's main program includes stages that have already been run. The dominant question is asked again to the user using the dominant question-and-answer template, and the dominant answer from the user to the dominant question is obtained. When it is detected that the dominant answer does not match the preset answer in the dominant question and answer template, after entering the abnormal program, an abnormal question is sent to the user through the abnormal question and answer template corresponding to the dominant answer in the abnormal program. Based on the user's abnormal answer to the abnormal question, the robot main program continues to run and enters the multiple stages of the robot main program corresponding to the abnormal answer, and then exits the abnormal program. After exiting the abnormal program, the main robot program continues to run from the point in the main robot program that was entered that corresponds to the abnormal response.
2. The task-oriented robot interaction method according to claim 1, characterized in that, Before issuing a leading question to the user through the leading question-and-answer template in the robot's main program, the method further includes: The robot's main program initiates an interaction request to the user. When the user agrees to the interaction request, an interactive dialogue is established between the robot's main program and the user to send questions to the user or receive the user's answers.
3. The task-oriented robot interaction method according to claim 1, characterized in that, The step of issuing an abnormal question to the user through the abnormal question and answer template corresponding to the dominant answer in the abnormal procedure includes: Select the abnormal question and answer template in the abnormal program that corresponds to the user keyword; The error question template is used to send error questions to users.
4. The task-oriented robot interaction method according to claim 1, characterized in that, After obtaining the user's dominant answer to the dominant question, the method further includes: When the dominant answer is identified as matching a preset answer in the dominant question-and-answer template, the process proceeds to the stage in the robot's main program corresponding to the dominant answer.
5. The task-oriented robot interaction method according to claim 4, characterized in that, The step of entering the stage corresponding to the leading answer in the robot's main program when the leading answer is identified as matching a preset answer in the leading question-and-answer template includes: When the dominant answer is identified as matching a preset answer in the dominant question-and-answer template, the user's business information is obtained; Based on the business information and the dominant response, proceed to the corresponding stage in the robot's main program.
6. A task-oriented robot interaction device, characterized in that, When the device is used in a multi-turn dialogue, the device includes: The leading question-and-answer module is used to send leading questions to users through the leading question-and-answer template corresponding to one of the multiple stages of the robot's main program; The monitoring module is used to obtain the user's dominant answer to the dominant question; The analysis module is used to analyze the dominant answer according to a preset thesaurus to obtain user keywords in the dominant answer that match the preset thesaurus; when it is detected that the user keywords do not match the preset keywords in the dominant question and answer template, an abnormal program is entered and the main program of the robot is suspended, wherein the preset keywords are extracted from the preset answers in the dominant question and answer template; The abnormal question and answer module is used to send abnormal questions to the user through the abnormal question and answer template corresponding to the main answer in the abnormal program; The jump module is used to obtain the user's abnormal answer to the abnormal question, continue running the robot's main program and enter the stage corresponding to the abnormal answer in the multiple stages of the robot's main program, and exit the abnormal program; The device is also used for: The dominant question is asked again to the user using the dominant question-and-answer template, and the dominant answer from the user to the dominant question is obtained. When it is detected that the dominant answer does not match the preset answer in the dominant question and answer template, after entering the abnormal program, an abnormal question is sent to the user through the abnormal question and answer template corresponding to the dominant answer in the abnormal program. Based on the user's abnormal answer to the abnormal question, the robot main program continues to run and enters the multiple stages of the robot main program corresponding to the abnormal answer, and then exits the abnormal program. After exiting the abnormal program, the main robot program continues to run from the point in the main robot program that was entered that corresponds to the abnormal response.
7. An electronic device, characterized in that, include: At least one memory; At least one processor; At least one computer program; The computer program is stored in the memory, and the processor executes the at least one computer program to achieve: The task-oriented robot interaction method as described in any one of claims 1 to 5.
8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions for causing a computer to perform: The task-oriented robot interaction method as described in any one of claims 1 to 5.