Dialogue method, electronic device, storage medium, and product

Abstract

The present disclosure relates to the technical field of computers, and relates to a dialogue method, an electronic device, a storage medium, and a product. The dialogue method comprises: displaying a dialogue between a user and a first agent; on the basis of the dialogue, generating setting information of a second agent to be created; on the basis of the setting information, creating the second agent, the second agent being configured to participate in the dialogue between the user and the first agent on the basis of the setting information of the second agent; and displaying a dialogue among the user, the first agent, and the second agent.

Description

Dialogue method, electronic device, storage medium and product TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of computer, in particular, to a dialogue method, an electronic device, a storage medium and a product. BACKGROUND

[0002] With the development of artificial intelligence technology, a user can have a dialogue with an agent through a user device. An agent is a concept in computer science, which refers to an entity capable of autonomously performing tasks in a specific environment. In response to obtaining a message sent by a user, the agent can generate a corresponding message and send it to the user. The agent can interact with the user as an intelligent customer service, an intelligent assistant, etc.

[0003] SUMMARY

[0004] This summary is provided to introduce a selection of concepts, which are further described below in the detailed description. This summary is not intended to identify key or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.

[0005] According to some embodiments of the present disclosure, a dialogue method is provided, comprising: displaying a dialogue between a user and a first agent; generating setting information of a second agent to be created according to the dialogue; creating the second agent according to the setting information, the second agent being configured to participate in the dialogue between the user and the first agent based on the setting information of the second agent; and displaying the dialogue between the user, the first agent and the second agent.

[0006] According to some embodiments of the present disclosure, an electronic device is provided, comprising: a memory; and a processor coupled to the memory, the processor being configured to execute a dialogue method of any of the embodiments described in the present disclosure based on instructions stored in the memory.

[0007] According to some embodiments of the present disclosure, a computer-readable storage medium is provided, having stored thereon a computer program, which, when executed by a processor, performs a dialogue method of any of the embodiments described in the present disclosure.

[0008] According to some embodiments of the present disclosure, a computer program is provided, comprising: instructions which, when executed by a processor, cause the processor to perform a dialogue method of any of the embodiments described in the present disclosure.

[0009] Other features, aspects, and advantages of the present disclosure will become apparent from the following detailed description of the exemplary embodiments with reference to the following accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0010] Preferred embodiments of the present disclosure are described below with reference to the accompanying drawings. The drawings described herein are for illustration purposes only and thus are not intended to limit the present disclosure in any regard. Descriptions and details of the drawings have been simplified to illustrate elements that are relevant for a clear understanding of the present disclosure, while eliminating, for purposes of clarity, other elements that are of little or no consequence to the understanding of the present disclosure. It should be understood that the drawings are not necessarily to scale and that, unless otherwise indicated, one of ordinary skill in the art would understand that the drawings are not intended to portray the relative αxial dimensions of the inventive concepts. It should be further understood that the drawings are not necessarily mutually consistent and that, unless otherwise indicated, the various features of the drawings are not necessarily to scale.

[0011] FIG. 1 shows a flowchart of a dialogue method according to some embodiments of the present disclosure.

[0012] FIG. 2 shows a flowchart of a method of generating setting information according to some embodiments of the present disclosure.

[0013] FIG. 3 shows a flowchart of a method of generating setting information according to some other embodiments of the present disclosure.

[0014] FIG. 4 shows a flowchart of a method of creating a third intelligent agent according to some embodiments of the present disclosure.

[0015] FIGS. 5A and 5B show schematic diagrams of a dialogue interface according to some embodiments of the present disclosure.

[0016] FIG. 6 shows a schematic diagram of a creation interface of an intelligent agent according to some embodiments of the present disclosure.

[0017] FIG. 7 shows a schematic diagram of a structure of a dialogue device according to some embodiments of the present disclosure.

[0018] FIG. 8 shows a schematic diagram of a structure of an electronic device according to some embodiments of the present disclosure.

[0019] FIG. 9 shows a schematic diagram of a structure of a computer system according to some embodiments of the present disclosure.

[0020] It should be understood that the dimensions of the various portions shown in the drawings are not necessarily to scale. Identical or similar reference numerals are used consistently throughout the several drawings to designate the same or similar components. Thus, where a component is described as being "shown" in a drawing, it will be understood that such component is not necessarily shown in every drawing. DETAILED DESCRIPTION

[0021] The technical solutions in the embodiments of the present disclosure will be described clearly and completely below with reference to the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only some of the embodiments of the present disclosure, but not all the embodiments. The description of the embodiments below is actually only illustrative, and should not be construed as any limitation on the present disclosure and its application or use. It should be understood that the present disclosure can be implemented in various forms, and should not be construed as being limited to the embodiments described herein.

[0022] It should be understood that various steps in the method implementations of the present disclosure can be performed in different order and / or in parallel. Additionally, the method implementations can include additional steps and / or omit performing the steps shown. The scope of the present disclosure is not limited in this regard. The relative arrangement of components and steps, numerical expressions, and numerical values set forth in these examples are to be interpreted as examples only and not limiting of the scope of the present disclosure unless otherwise specifically stated.

[0023] The term "include," and derivations thereof, means the term "comprise" or "contain" and variations as an open-ended term such that when the phrase "includes (comprises, contains)" is used, it means at least the stated elements, but not excluding others. In addition, the term "comprise" and variations thereof as used in the present disclosure means the term "comprise" or "contain" and variations as an open-ended term such that when the phrase "comprises (contains)" is used, it means at least the stated elements, but not excluding others. Thus, include and comprise are synonymous. The term "based on" means "based, at least in part, on."

[0024] Reference throughout this specification to "an embodiment", "some embodiments" or "embodiments" means that a particular feature, structure or characteristic described in connection with the embodiment is included in at least one embodiment of the application. The appearances of the phrases "in one embodiment", "in some embodiments" or "in embodiments" in various places in the specification are not necessarily all referring to the same embodiment, although they can. Furthermore, the terms "a" or "an", as used herein, mean "one or more" when used in the context of expressing an quantity of objects unless otherwise indicated.

[0025] It should be noted that the terms "first", "second", and so on used in the present disclosure are merely used to distinguish different devices, modules or units, and do not imply the order or sequence of the functions of the devices, modules or units. Unless otherwise specified, the terms "first", "second", and so on are not intended to imply a given order or sequence of the objects described.

[0026] It should be noted that the terms "one", "multiple", and "plurality" used in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise explicitly specified in the context, they should be understood as "one or more".

[0027] The names of the messages or information exchanged between the plurality of devices in the embodiments of the present disclosure are merely used for illustrative purposes, and are not intended to limit the scope of the messages or information.

[0028] Embodiments of the present disclosure will be described in detail below with reference to the drawings, but the present disclosure is not limited to these specific embodiments. The following specific embodiments can be combined with each other, and for the same or similar concepts or processes, some embodiments can not be described again. In addition, in one or more embodiments, specific features, structures or characteristics can be combined by any suitable means from the present disclosure that will be clear to those of ordinary skill in the art.

[0029] First, some concepts related to the present disclosure are explained.

[0030] Agent technology: Agent is a concept in computer science, referring to an entity that can autonomously perform tasks in a specific environment. In a multi-agent system, multiple agents collaborate to solve complex problems. Agents can generate content corresponding to other subjects in a dialogue scene based on the dialogue sent by the subjects, such as users or agents participating in the dialogue. It can be implemented in software, hardware, or a combination of software and hardware. Agents can also be referred to as digital humans, robots, virtual agents of machine learning models. Agents can rely on machine learning models, such as large language models (LLM) or foundation models. Machine learning models can be generative models.

[0031] Chatbot technology: Chatbot is a software that communicates with humans through natural language processing (NLP). They are often used in customer service, information query and entertainment fields. Chatbots can be considered a special kind of agent.

[0032] Multi-agent system (MAS): In a MAS, each agent has its own goals and behaviors, and they work together to achieve common goals through communication and collaboration.

[0033] Context awareness: Context awareness refers to the ability of agents to understand the environment and context information in which the dialogue takes place, in order to provide more relevant and personalized responses.

[0034] Generative model: The generative model is used to output target content based on input information. The information input by the generative model includes the basis for processing in the generation process of the generative model, such as messages sent by other subjects in the dialogue, requirements for the output content, and the like. The generative model includes, for example, a model for generating based on text or a model for generating based on images, and the output of the generative model can include text, images, or a combination of both. Of course, the input or output of the generative model can also be data of other modalities, such as audio, video, or a combination of multiple types of data. The generative model can be a single-modal model, such as a model for generating text based on text (referred to as a "text-to-text model"), a model for generating images based on images (referred to as an "image-to-image model"), or a cross-modal model, i.e., a model in which the input and output belong to different modalities, such as a model for generating images based on text (referred to as a "text-to-image model"). Alternatively, the input of the generative model can include multiple modalities, and the output can also include multiple modalities.

[0035] In the related art, a user can have a one-on-one dialogue with an agent, i.e., in one dialogue interface, the user can have a dialogue with one agent, such as chatting with the agent, asking the agent for help, sending a command to the agent, and the like. Some agents are configured with setting information, a knowledge base, and the like, so for specific dialogue content, the agent can give accurate feedback. Accordingly, when the content of the dialogue relates to a field with low relevance to the agent, the agent can not be able to efficiently and accurately send a response to the user. At this time, the user can search for other agents, or the user can give up this approach of having a dialogue with the agent and instead use a search engine, ask others, or other approaches to obtain information. Therefore, in the related art, when the agent having a dialogue with the user cannot accurately respond to the user, the efficiency of the user's information acquisition is reduced.

[0036] To at least partially solve the above technical problems, embodiments of the present disclosure provide a dialogue method, an electronic device, a storage medium, and a product. In embodiments of the present disclosure, during a dialogue between a user and a first agent, a second agent is created and added to the dialogue, forming a multi-subject dialogue, or a "group chat", of the user, the first agent, and the second agent, so that the second agent can make up for the missing information of the first agent and improve the efficiency of the user's information acquisition. Embodiments of the dialogue method of the present disclosure are described below with reference to FIG. 1.

[0037] FIG. 1 shows a flowchart of a dialogue method according to some embodiments of the present disclosure. As shown in FIG. 1, the dialogue method of this embodiment includes steps S102 to S108.

[0038] In step S102, a dialogue between a user and a first agent is displayed.

[0039] The conversation between the user and the first agent includes one or more messages sent by at least one of the user and the first agent, and each message can include text, voice, image, video, or a link, etc. The first agent can be created by the application platform and provided for use by the users, or can be created by the user in step S102, or can be created by other users and authorized for use by other users.

[0040] The participants of the conversation between the user and the first agent can include only the user and the first agent, or can further include other users, or can further include other agents. The disclosure does not limit the number of subjects participating in the conversation in step S102.

[0041] The user can input the message sent to the first agent through the user device, for example, through a keyboard, a microphone, a touch screen, etc. The message sent by the first agent to the user can be displayed in a visual manner on the screen of the user device, and in the case of need, the sound corresponding to the message can also be played through a speaker or the like.

[0042] In step S104, setting information of the second agent to be created is generated according to the conversation.

[0043] Step S104 can be triggered by the user or automatically triggered according to the content of the conversation. In some embodiments, in the case of automatic triggering, a prompt can be sent by the application system or a prompt can be sent by the first agent through the conversation to confirm with the user whether to confirm the creation of the agent. Before creating a new agent each time, the user can be asked and the creation process can be performed after the confirmation of the user. Alternatively, in the case where the user has authorized the automatic creation of the agent in the application, the user can not be confirmed one by one for each creation process.

[0044] The setting information of the second agent is obtained by processing the conversation. In the case where the conversation includes text or voice (which can be converted into text), a generative model capable of receiving text as input can be used to obtain the setting information; in the case where the conversation text includes sound, a generative model capable of receiving sound as input can be used; in the case where the conversation text includes image or video, a generative model capable of receiving image as input can be used. Of course, a generative model with multi-modal input can also be used.

[0045] In some embodiments, a classification model or a topic analysis model can also be used to determine the category or topic of the conversation, and according to the correspondence between the preset category or topic and the setting information, the information of the second agent corresponding to the conversation can be obtained.

[0046] In step S106, a second agent is created according to the setting information, the second agent being configured to participate in the conversation between the user and the first agent based on the setting information of the second agent.

[0047] The creation of the second agent can be an interface for generating the second agent. Thus, in response to the second agent being triggered, for example, in response to the second agent being in the conversation, the interface can be invoked to obtain the conversation output by the second agent.

[0048] In the case where the second agent relies on a model of machine learning, the second agent can be regarded as a client or agent of the model, but the client or agent has a specific setting, the content of which includes the generated setting information. The setting is, for example, a system prompt. Thus, the machine learning model can output the conversation generated by the second agent based on the conversation generated by other objects in the conversation scenario in which the second agent is located, in combination with the setting information.

[0049] Alternatively, the setting information can be stored in data corresponding to the second agent. In response to the running of the control logic of the second agent, the setting information is read, and the running result is derived according to the setting information.

[0050] Thus, the information output by the created second agent in the conversation process can be associated with the content of the conversation between the user and the first agent, improving the relevance of the second agent to the current conversation scenario.

[0051] In some embodiments, the second agent is created according to the setting information and the history of the conversation. Thus, the message sent by the created second agent can match the history of the conversation, and the user or the first agent does not need to repeatedly input the conversation content that has been sent before the second agent is created in the process of the conversation. Therefore, the second agent can efficiently provide information consistent with the scenario of the conversation. Of course, the second agent can also be created without using the history of the conversation according to needs, and those skilled in the art can select as needed. By maintaining and updating the chat context of the user and the agent, the newly added agent can understand the background of the current conversation and generate coherent and relevant replies according to the context.

[0052] In step S108, the conversation between the user, the first agent and the second agent is displayed.

[0053] After the second agent is created, the second agent can be pulled into the conversation between the user and the first agent to form a group chat among the user, the first agent, and the second agent. After the user or the first agent sends a message in the conversation, the second agent can determine whether the second agent needs to speak (i.e., send a message) according to the obtained message. If so, the second agent further processes the obtained message according to the set information to output the conversation content of the second agent, i.e., send a message from the second agent.

[0054] The above embodiment can automatically create a second agent during the conversation between the user and the first agent according to the conversation content, and form a group chat among the user, the first agent, and the second agent. Thus, the second agent can automatically participate in the conversation between the user and the first agent, and provide more information for the user, improving the efficiency of information acquisition of the user.

[0055] Moreover, in the above process, the user can still remain in the original conversation interface, and only an agent is newly added to the conversation. In this way, the user does not need to perform complex operations to obtain more information through the newly added agent, improving the efficiency of information acquisition.

[0056] In the above embodiment, steps S102 and S108 can be performed in the user device, e.g., in an application of the user device, which can be an application with the function of conversing with an agent. Part or all of steps S104 and S106 can be performed in the user device (e.g., in the application of the user device), or part or all of steps S104 and S106 can be performed in another device other than the user device, e.g., in a server. Those skilled in the art can determine the execution subject of the step according to the performance or computing requirement of the user device.

[0057] The set information of the agent can be determined according to the content of the conversation, e.g., the topic of the conversation. Since the conversation between the user and the agent can involve one topic or multiple topics, the set information of the agent can be determined according to a target topic among one or more topics. The following describes an embodiment of generating set information based on a target topic with reference to FIG. 2.

[0058] FIG. 2 shows a flowchart of a method of generating set information according to some embodiments of the present disclosure. As shown in FIG. 2, the method of generating set information according to the embodiment includes steps S202 to S204.

[0059] In step S202, a target topic is extracted from the conversation.

[0060] The target topic can be a topic recently discussed or a topic of interest to the user, or a topic of a pre-specified type. For example, the dialog can be processed using a topic analysis model to obtain one or more topics, and the target topic can be extracted therefrom. For another example, the dialog and the requirement for the target topic can be input into a generative model to obtain the target topic output by the generative model.

[0061] The target topic can be determined based on at least one of an order of generation of the one or more topics extracted from the dialog, and an information amount. For example, a specified number of topics generated most recently can be taken as the target topic according to the order of generation of the one or more topics, so that the target topic is a topic involved in the most recent dialog of the user with the first agent. For another example, a specified number of topics with the largest information amount can be taken as the target topic, so that the target topic is a topic of more interest to the user in the dialog, the information amount can be represented by the number of words, the frequency of occurrence, or the like, or can be obtained by processing the dialog in each topic using a semantic analysis model. For yet another example, the order of generation of the topics and the information amount can be combined, for example, a specified number of topics generated most recently and with an information amount greater than a threshold can be taken as the target topic. Thus, the target topic can be matched to a topic of interest to the user.

[0062] The target topic can also be a topic predicted based on the dialog. That is, the target topic can be a topic not involved in the current dialog, but likely to be generated subsequently. For example, the next topic involved in the dialog can be predicted based on one or more topics in historical dialogs. A sequence-based machine learning model such as a recurrent neural network (RNN), a long short-term memory (LSTM), or the like can be used in the prediction, or the historical dialogs can be directly input into a generative model, and the generative model can be instructed to generate the content of the dialog to be generated. Thus, a topic of interest to the user can be predicted in advance, and effective information can be sent to the user by the second agent in a timely manner.

[0063] In addition, in a case where the dialog of the user with the first agent is developed based on a specified story background, the target topic to be generated subsequently can be determined based on the story background and the content of the current dialog. For example, the user and the first agent can perform role-playing and develop a dialog based on a specific story background (e.g., a script, a plot), and the target topic can be the next plot in the story background.

[0064] In step S204, setting information of the second agent to be created is generated according to the target topic.

[0065] The setting information of the second agent is associated with the target topic, which can be derivative information, related information, or the like of the target topic. Several methods of generating setting information according to the target topic are described below. In some embodiments, the target topic can be input into a generative model to obtain generated setting information matching the target topic; or the setting information corresponding to the target topic can be determined from a pre-set correspondence between topics and setting information.

[0066] In some embodiments, generating the setting information of the second agent includes generating a summary of the dialogue; and generating the setting information of the second agent to be created according to the summary of the dialogue and the difference information of the target topic and the setting information of the first agent.

[0067] The summary of the dialogue is used to summarize the main content involved in the dialogue. The summary of the dialogue can be more comprehensive than the target topic. Thus, when generating the setting information according to the target topic, the key information in the dialogue context will not be lost. Thus, the generated second agent is more matched with the current dialogue scenario.

[0068] In some embodiments, the dialogue can be input into a generative model, and a processing instruction of generating a summary can be input into the generative model, so as to generate the summary of the dialogue. The input dialogue can be all the dialogue between the user and the first agent, or part of the dialogue. For example, the input dialogue can be the dialogue within a specified time length recently, or the dialogue with a specified amount of information (such as the number of words, etc.) recently.

[0069] The difference between the target topic and the setting information of the first agent reflects the information that the first agent cannot cover in the current dialogue. For example, the field not involved by the first agent. Therefore, the created second agent can make up for the missing information of the first agent to provide more comprehensive information for the user.

[0070] For example, the first agent is an "English translation assistant" to answer the user's questions about translation. When the user talks more about traveling to the United Kingdom, a "UK travel assistant" can be created as a second agent and joined in the group chat to respond to the user in a timely manner in the current dialogue interface, improving the user's efficiency of obtaining information.

[0071] In some embodiments, the setting information includes at least one of an attribute of the second agent and a relationship between the second agent and the first agent. Generating the setting information of the second agent to be created includes: generating a summary of the dialogue; and generating at least one of the attribute of the second agent and the relationship between the second agent and the first agent according to the summary of the dialogue and the target topic.

[0072] The attribute of the second agent describes basic information of the second agent, such as gender, occupation, hobby, birthday, or specialty, so that the generated second agent has more specific characteristics. The relationship between the second agent and the first agent reflects the association between the two, such as relatives, friends, teachers, and the like, so that the process of the second agent joining the dialogue is relatively smooth and natural.

[0073] Both the attribute and the relationship can be generated with reference to the summary of the dialogue and the target theme. Thus, the generated second agent can adapt to the current context and can provide more information matching the target theme.

[0074] In some embodiments, the setting information includes at least one of the attribute of the second agent, the relationship between the second agent and the first agent, and background information of the second agent. Generating the setting information of the second agent to be created includes: determining at least one of the attribute of the second agent and the relationship between the second agent and the first agent; and generating the background information of the second agent according to the at least one of the attribute of the second agent and the relationship between the second agent and the first agent and the summary of the dialogue.

[0075] In this embodiment, the attribute and the relationship of the second agent can be automatically generated, for example, generated in the manner of the foregoing embodiments. Alternatively, the attribute and the relationship of the second agent can be input by the user or edited by the user on the basis of the automatically generated result. After the attribute or the relationship of the second agent is determined, the background information can be further generated on the basis of the summary of the dialogue, so as to provide the second agent with more abundant setting information. The background information can be displayed to the user or can not be displayed to the user but used as a basis for subsequent generation of the dialogue. The background information can describe what has happened to the virtual second agent in the past (for example, whether the second agent has traveled around the world or has been working and living in a city). Thus, the dialogue generated by the second agent can be more matched with the background information of the second agent.

[0076] In some embodiments, the setting information includes a knowledge base of the second agent. Generating the setting information of the second agent to be created includes: generating the knowledge base of the second agent according to the knowledge base of the first agent and the knowledge base involved in the target theme.

[0077] The knowledge base is used to provide basis for the second agent to generate the dialogue. For example, when generating the dialogue, the second agent can search for data related to or corresponding to the input in the knowledge base, and process the data to generate the content of the dialogue of the second agent. The knowledge base can be a local database, an online database, a search engine, etc. Configuring as many knowledge bases as possible for the agent can improve the information coverage of the agent. However, correspondingly, the computational cost and computational time of the agent in processing will also be increased, which may affect the performance of the agent. By configuring appropriate knowledge bases for the agent, the processing efficiency of the agent can be improved while providing effective and accurate information for the user.

[0078] The knowledge base of the second agent needs to include a knowledge base related to the target topic, so as to be able to provide effective information for the user. In addition, the knowledge base of the second agent can also be determined according to the knowledge base of the first agent, for example, it is required to cover the knowledge base of the first agent, so that the second agent has more abundant information sources than the first agent; of course, according to the knowledge base of the first agent, the knowledge base of the second agent can also be configured not to include the knowledge base of the first agent, so that the first agent and the second agent provide knowledge in different fields respectively, so as to improve the processing efficiency.

[0079] In the process of determining the role of the second agent, it can be generated from the roles associated with the first agent. In some embodiments, one or more candidate associated roles are determined for the first agent based on the setting information of the first agent; and the second agent is any one of the agents to be created, which is determined from the candidate associated roles according to the relevance of each candidate associated role to the dialogue. Thus, the created second agent can have relevance with the first agent, so that the process of the second agent joining the dialogue is more smooth.

[0080] In the process of the user's dialogue with the first agent, whether to generate an agent can be automatically decided according to the progress of the dialogue. FIG. 3 shows a flowchart of a method for generating setting information according to some other embodiments of the present disclosure. As shown in FIG. 3, the generating method of this embodiment includes steps S302 to S304.

[0081] In step S302, the matching degree of the first agent to the dialogue is determined according to the dialogue and the setting information of the first agent.

[0082] The matching degree is used to represent whether the dialogue between the user and the first agent deviates from the information that the first agent can provide. For example, according to the setting information of the first agent, the specialty of the first agent is English. If the dialogue between the user and the first agent involves mathematical problems, the matching degree of the first agent to the current dialogue is low. The matching degree can be represented by a specific numerical value, a plurality of levels, or whether it is matched.

[0083] In some embodiments, a similarity between the dialogue and the setting information can be calculated, and the similarity can be taken as the matching degree. When calculating the matching degree, the matching degree can be determined based on feature vectors of the dialogue and the setting information. The features of the dialogue and the setting information can be obtained by a feature extraction model.

[0084] In step S304, in response to the matching degree being lower than the threshold, the setting information of the second agent to be created is generated.

[0085] In the case where the matching degree is lower than the threshold, it indicates that the first agent is difficult to continue to provide effective information for the user. At this time, the generation process of the setting information of the second agent can be triggered, and the creation process of the second agent is further triggered. That is, in response to the matching degree being lower than the threshold, the creation of the second agent is triggered, and the second agent is controlled to participate in the dialogue between the user and the first agent.

[0086] The above embodiments can automatically create and introduce the second agent to join the dialogue according to the support capability of the first agent for the dialogue content in the process of the dialogue between the user and the first agent. Therefore, when the first agent cannot continue to provide effective information for the user, the information can be provided in time through the second agent, and the efficiency of the user to obtain information is improved.

[0087] In the case where the creation of the second agent is triggered automatically, the process of determining whether to create the second agent (or determining whether to generate the setting information of the second agent) can be triggered at a specified period, or the process of determining whether to create the second agent can be triggered after each new message is generated in the dialogue. For example, whether to create a new agent can be determined according to the dialogue, in response to determining to create a new agent, the second agent is created to display the dialogue among the user, the first agent, and the second agent.

[0088] After the second agent joins the dialogue, whether to continue to create a new agent can also be determined according to the above logic. Embodiments of a method of determining whether to create a new agent are described below with reference to FIG. 4.

[0089] FIG. 4 shows a flowchart of a method of creating a third agent according to some embodiments of the present disclosure. As shown in FIG. 4, the creation method of this embodiment includes steps S402 to S406.

[0090] In step S402, whether to create a new agent is determined according to the dialogue. For example, whether to create a new agent can be determined based on the matching degree according to the foregoing embodiments. In addition, the creation of a new agent can also be triggered in other ways. For example, in the case where the message sent by the user indicates that the user hopes to invite a new agent to join the dialogue, it is determined to create a new agent. The specific meaning of the message sent by the user can be determined by a semantic analysis model.

[0091] In step S404, in response to determining to create a new agent, a third agent is created, the third agent being configured to participate in the conversation between the user, the first agent, and the second agent based on the setting information of the third agent. The creation method of the third agent can refer to the second agent, which will not be described here.

[0092] In step S406, the conversation between the user, the first agent, the second agent, and the third agent is displayed.

[0093] In some embodiments, in response to determining not to create a new agent, the next agent that sends the conversation content is determined according to the conversation. For example, should the first agent, the second agent, or other agents (if any) send the message. For another example, if it is judged that the user should continue to speak at this time, no agent can be instructed to send the conversation content.

[0094] In determining whether to speak by an agent or to create a new agent, the prediction of the next token (Next Token Prediction) capability of the machine learning model can be used, i.e., the model predicts whether the next action is performed by an agent that has participated in the conversation or a new agent, thereby making a decision.

[0095] The interaction of multiple agents is described below.

[0096] The multiple agents are all in the same virtual environment, share the virtual environment, and the environment contains global state information, and the agents can interact and update information between them. Each agent assumes a corresponding role positioning and task according to the setting information. In order to control the interaction between different agents, a controller can be used to make decisions. The controller can be a model (such as an LLM model) or a pre-defined rule, responsible for switching between different agents and task stages, managing the action sequence of the agents, such as controlling which agent sends the message currently. In addition, the agent also needs to be "memory maintained", in the multi-agent framework, the memory not only includes the interaction history of the user and the agent, but also includes the internal state of each agent and the interaction history between the agents, to ensure the coherence of the conversation. The agent calls the model to perform a specific action according to the instruction of the controller, such as generating conversation content, calling tool plugins, etc. After the agent performs the action, the output information is sent to the conversation interface or updated to the public environment shared by the agents for use by other agents.

[0097] Through such an interactive process, the multi-agent system can complete efficient and complex tasks while keeping the interaction between the roles natural and logical. Users can enter a multi-role interactive mode by selecting different roles, implement group chat functions, and achieve a rich dialogue experience created by multiple agents.

[0098] Through the above embodiments, the third agent can also be added to the dialogue as the dialogue progresses. That is, embodiments of the present disclosure support the creation of multiple agents multiple times during the dialogue and support the dialogue of the created multiple agents with the user.

[0099] The dialogue method of the embodiments of the present disclosure will be described below in conjunction with some dialogue interfaces and related interface diagrams.

[0100] FIGS. 5A and 5B show schematic diagrams of dialogue interfaces according to some embodiments of the present disclosure. As shown in FIG. 5A, in the dialogue interface 5, the user is in dialogue with the agent “Wise Old Man”. The messages sent by the agent are displayed left-aligned, and the messages sent by the user are displayed right-aligned.

[0101] In the dialogue, the agent “Wise Old Man” sends the message 51 “Young man, I heard that you have started using some auxiliary tools to improve your English learning”. After receiving the message 51, the user sends the message 52 “Yes, Mr. Old Man, I have started some basic learning”, indicating that the user already has some basis for English learning. After judging, the content in the message 52 sent by the user exceeds the information that the agent “Wise Old Man” can cover. Therefore, based on the current topic of discussion, the agent “English Teacher” is automatically created and added to the dialogue between the user and “Wise Old Man”. As shown in FIG. 5B, the “English Teacher” sends the message 53 “I am glad to hear that you have taken action. I have prepared some special exercises for you in some online resources” based on the last message 52 sent by the user. Thereafter, the user, the “English Teacher”, and the “Wise Old Man” continue to interact through messages such as 54 to 56. Other agents can also be invited to join the group chat as needed.

[0102] In some embodiments, before creating the second agent, the setting information of the second agent can also be displayed through the creation interface for the user to modify or confirm. FIG. 6 shows a schematic diagram of an agent creation interface according to some embodiments of the present disclosure. As shown in FIG. 6, in the creation interface 6, the setting information of the generated second agent “English Teacher” is displayed, such as including the avatar 61, the name 62, the setting description 63, the public permission 64, and the like. The user can modify one or more of them as needed. After the user confirms the setting information displayed in the interface 6, the creation process of the second agent can be completed by triggering the creation control 65.

[0103] The creation interface 6 is created to input the box bearing part of the setting information, and the initial content of the setting information can be automatically generated. According to needs, the user can also select from some alternative information. In some embodiments, the setting information of the second agent to be created is displayed, and the setting information includes one or more alternative information; a selection operation of the user on the alternative information is received; and the second agent is created based on the information selected by the user. The one or more alternative information can be generated according to the method in the foregoing various embodiments, and in the case of generating multiple alternative information, it can be indicated that the model generating the setting information outputs multiple results or uses multiple different models to output results respectively. By displaying one or more alternative information for the user to select, the efficiency of the user operation can be improved while allowing the user to customize the agent. In some embodiments, the user can further edit the selected information to meet the user's individual requirements after selecting the alternative information.

[0104] The creation interface 6 can be displayed in response to determining that the second agent needs to be created, for example, after the dialogue interface 5 shown in FIG. 5A. In response to the user triggering the creation control 65, the dialogue interface 5 can be returned, as shown in FIG. 5B. The creation interface 6 can be displayed full screen on the user device, or can be displayed in the upper layer of the dialogue interface 5 in the form of a floating layer or the like.

[0105] Embodiments of the present disclosure can be interacted by the front end, the back end, the control logic (based on algorithms) and the model. The front end is responsible for providing an interactive interface to the user for the user to interact with the agent, as well as creating and setting the role. The back end is used to process the input of the user, store and integrate the role information into the dialogue system. The back end is also responsible for calling the control logic and the model to generate natural and fluent dialogue content, and to make the interaction between the agent and the user consistent with the preset logic and background story. The algorithm plays a coordinating role in the multi-agent system, for example, according to the input of the user and the progress of the dialogue, dynamically adjusting and optimizing the behavior of the agent. The algorithm is responsible for allocating roles and tasks between different models, so that each agent can participate in the dialogue according to its setting information.

[0106] The interaction process between the user interaction interface, the front end, and the back end is described below. The user inputs the setting information of the agent created by the user through the interaction interface, and the front end receives the setting information and formats and stores it. The back end receives the formatted setting information of the agent and stores it. Then, the back end integrates the stored setting information of the agent into the dialogue environment, and then displays the setting result of the agent through the front end, so as to display the information of the agent created by the user. Then, the user can select the created agent for dialogue. Of course, the agent can be created by the user, by other users, or provided by the application. After the front end receives the selection result of the user, the agent is integrated into the dialogue environment, the front end calls the formatted setting information of the agent involved in the dialogue environment, and sends it to the back end. The back end calls the algorithm to assign tasks to each agent. For example, if the agent needs to speak, the model generates a message, and the model generates the dialogue content of the agent according to the setting information of the agent. The generated dialogue content is output to the chat environment, that is, sent to the dialogue interface. For another example, if a new agent needs to be created, a new agent is generated and added to the dialogue environment. The algorithm coordinates the interaction between multiple agents. Thus, whether it is the interaction between the original agents or the interaction between the original agents and the newly created agents, the original dialogue content can be continued.

[0107] The above describes various method embodiments of the present disclosure. The following describes an apparatus of the present disclosure for performing the above-described various embodiments.

[0108] FIG. 7 shows a structural schematic diagram of a dialogue apparatus according to some embodiments of the present disclosure. As shown in FIG. 7, the dialogue apparatus 7 of this embodiment includes a first display module 701 configured to display the dialogue between the user and the first agent; a generation module 702 configured to generate the setting information of a second agent to be created according to the dialogue; a creation module 703 configured to create the second agent according to the setting information, the second agent being configured to participate in the dialogue between the user and the first agent based on the setting information of the second agent; and a second display module 704 configured to display the dialogue between the user, the first agent, and the second agent.

[0109] In some embodiments, the generation module 702 is further configured to extract a target topic from the dialogue; and generate the setting information of the second agent to be created according to the target topic.

[0110] In some embodiments, the generation module 702 is further configured to generate a summary of the dialogue; and generate the setting information of the second agent to be created according to the summary of the dialogue, and the difference information of the target topic and the setting information of the first agent.

[0111] In some embodiments, the setting information includes at least one of the attribute of the second agent, the relationship between the second agent and the first agent, the generation module 702 is further configured to: generate a summary of the dialogue; and generate at least one of the attribute of the second agent, the relationship between the second agent and the first agent according to the summary of the dialogue and the target topic.

[0112] In some embodiments, the setting information includes at least one of the attribute of the second agent, the relationship between the second agent and the first agent, and the second generation module 702 is further configured to: determine at least one of the attribute of the second agent, the relationship between the second agent and the first agent; and generate the background information of the second agent according to at least one of the attribute of the second agent, the relationship between the second agent and the first agent and the summary of the dialogue.

[0113] In some embodiments, the setting information includes the knowledge base of the second agent, and the generation module 702 is further configured to: generate the knowledge base of the second agent according to the knowledge base of the first agent and the knowledge base involved by the target topic.

[0114] In some embodiments, the target topic is determined based on at least one of the generation order, the amount of information of one or more topics extracted from the dialogue; or the target topic is a topic predicted based on the dialogue.

[0115] In some embodiments, the generation module 702 is further configured to: determine the matching degree of the first agent and the dialogue according to the dialogue and the setting information of the first agent; and generate the setting information of the second agent to be created in response to the matching degree being lower than a threshold.

[0116] In some embodiments, the creation module 703 is further configured to: display the setting information of the second agent to be created, the setting information including one or more alternative information; receive a selection operation of the user on the alternative information; and create the second agent based on the information selected by the user.

[0117] In some embodiments, the dialogue device 705 further includes: a first determination module 705 configured to determine one or more alternative associated roles for the first agent based on the setting information of the first agent; and determine the second agent to be created from the alternative associated roles according to the association degree of each alternative associated role and the dialogue, wherein the second agent is any one of the agents to be created.

[0118] In some embodiments, the creating module 703 is further configured to: determine, according to the dialogue, whether to create a new agent; and in response to determining to create the new agent, create a third agent, the third agent being configured to participate in the dialogue between the user, the first agent, the second agent based on setting information of the third agent; and the second displaying module 704 is further configured to display the dialogue between the user, the first agent, the second agent and the third agent.

[0119] In some embodiments, the dialogue device 705 further includes a second determining module 706 configured to: in response to determining not to create the new agent, determine, according to the dialogue, a next agent to send the dialogue content.

[0120] In some embodiments, the creating module 703 is further configured to: create the second agent according to the setting information and a history of the dialogue.

[0121] It should be noted that the above-mentioned various units are only logical modules according to the specific functions they implement, and are not intended to limit the specific implementation manner, for example, they can be implemented in software, hardware or a combination of software and hardware. In actual implementation, the above-mentioned various units can be implemented as independent physical entities, or can also be implemented by a single entity (for example, a processor (CPU or DSP, etc.), an integrated circuit, etc.). In addition, the above-mentioned various units are indicated by dashed lines in the drawings, indicating that these units can not actually exist, and the operations / functions they implement can be implemented by the processing circuit itself.

[0122] In addition, although not shown, the device can also include a memory, which can store various information generated by the device, the units included in the device in operation, programs and data for operation, data to be sent by the communication unit, etc. The memory can be a volatile memory and / or a non-volatile memory. For example, the memory can include but is not limited to random access memory (RAM), dynamic random access memory (DRAM), static random access memory (SRAM), read-only memory (ROM), flash memory. Of course, the memory can also be located outside the device. Alternatively, although not shown, the device can also include a communication unit, which can be used for communication with other devices. In one example, the communication unit can be implemented in a suitable manner known in the art, for example, including communication components such as antenna array and / or radio frequency link, various types of interfaces, communication units, etc. Here will not be described in detail. In addition, the device can also include other components not shown, such as radio frequency link, baseband processing unit, network interface, processor, controller, etc. Here will not be described in detail.

[0123] Some embodiments of the present disclosure also provide an electronic device. FIG. 8 shows a structural schematic diagram of an electronic device according to some embodiments of the present disclosure. For example, in some embodiments, the electronic device 8 can be various types of devices, for example, can include but is not limited to mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablets), PMPs (portable multimedia players), vehicle terminals (for example, car navigation terminals), and the like, and fixed terminals such as digital TVs, desktop computers, and the like. For example, the electronic device 8 can include a display panel for displaying data and / or execution results utilized in the schemes according to the present disclosure. For example, the display panel can be various shapes, for example, a rectangular panel, an oval panel, or a polygonal panel, and the like. In addition, the display panel can not only be a flat panel, but also a curved panel, or even a spherical panel.

[0124] As shown in FIG. 8, the electronic device 8 of this embodiment includes a memory 81 and a processor 82 coupled to the memory 81. It should be noted that the components of the electronic device 8 shown in FIG. 8 are only exemplary and are not limiting, and the electronic device 8 can also have other components according to actual application needs. The processor 82 can control other components in the electronic device 8 to perform desired functions.

[0125] In some embodiments, the memory 81 is configured to store one or more computer readable instructions. When the processor 82 executes the computer readable instructions, the computer readable instructions are executed by the processor 82 to implement the method according to any of the above embodiments. For specific implementation of each step of the method and related explanations, please refer to the above embodiments, and repeated parts will not be described here.

[0126] For example, the processor 82 and the memory 81 can directly or indirectly communicate with each other. For example, the processor 82 and the memory 81 can communicate through a network. The network can include a wireless network, a wired network, and / or any combination of a wireless network and a wired network. The processor 82 and the memory 81 can also communicate with each other through a system bus, and the present disclosure does not limit this.

[0127] For example, the processor 82 can be embodied as various appropriate processors, processing devices, and the like, such as a central processing unit (CPU), a graphics processing unit (GPU), a network processing unit (NP), and the like; also can be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component. The central processing unit (CPU) can be an X86 or ARM architecture, and the like. For example, the memory 81 can include any combination of various forms of computer-readable storage media, such as a volatile memory and / or a non-volatile memory. The memory 81 may, for example, include a system memory, which stores, for example, an operating system, application programs, a boot loader, a database, and other programs, and the like. Various application programs and various data, and the like, can also be stored in the storage medium.

[0128] In addition, according to some embodiments of the present disclosure, various operations / processes according to the present disclosure, in the case of being implemented by software and / or firmware, programs constituting the software can be installed from a storage medium or a network to a computer system having a dedicated hardware structure, such as the computer system 90 shown in FIG. 9, which, when various programs are installed, is capable of performing various functions, including functions such as those described above, and the like. FIG. 9 shows a structural schematic diagram of a computer system according to some embodiments of the present disclosure.

[0129] In FIG. 9, the central processing unit (CPU) 901 performs various processes according to programs stored in the read-only memory (ROM) 902 or programs loaded from the storage portion 908 to the random access memory (RAM) 903. In the RAM 903, data required when the CPU 901 performs various processes, and the like, is also stored as needed. The central processing unit is merely exemplary, and can also be other types of processors, such as various processors described above. The ROM 902, the RAM 903, and the storage portion 908 can be various forms of computer-readable storage media, as described below. It should be noted that, although the ROM 902, the RAM 903, and the storage device 908 are shown separately in FIG. 9, one or more of them can be combined or located in the same or different memory or storage module.

[0130] The CPU 901, the ROM 902, and the RAM 903 are connected to each other via a bus 904. An input / output interface 905 is also connected to the bus 904.

[0131] The following components are connected to the input / output interface 905: an input portion 906, such as a touch panel, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, and the like; an output portion 907, including a display, such as a cathode ray tube (CRT), a liquid crystal display (LCD), a speaker, a vibrator, and the like; a storage portion 908, including a hard disk, a magnetic tape, and the like; and a communication portion 909, including a network interface card, such as a LAN card, a modem, and the like. The communication portion 909 allows communication processing to be performed via a network, such as the Internet. It is easily understood that, although the respective devices or modules in the computer system 90 are shown as communicating through the bus 904 in FIG. 9, they can also communicate through a network or other means, where the network can include a wireless network, a wired network, and / or any combination of a wireless network and a wired network.

[0132] The drive 910 is also connected to the input / output interface 905 as necessary. A removable medium 911, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, and the like, is attached to the drive 910 as necessary, so that a computer program read therefrom is installed into the storage portion 908 as necessary.

[0133] In the case where the above series of processes are implemented by software, the program constituting the software can be installed from a network such as the Internet or a storage medium such as the removable medium 911.

[0134] According to an embodiment of the present disclosure, the processes described above with reference to the flowcharts can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product including a computer program carried on a computer-readable medium, the computer program containing program codes for executing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from the network by the communication device 909, or installed from the storage device 908, or installed from the ROM 902. When the computer program is executed by the CPU 901, the above-described functions defined in the methods of the embodiments of the present disclosure are executed.

[0135] Note that in the context of the present disclosure, a computer-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium or any combination thereof. The computer-readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the computer-readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present disclosure, a computer-readable storage medium can be any tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. In the present disclosure, a computer-readable signal medium can include a computer-readable program code transmitted in baseband or as part of a carrier wave over a transmission medium, in which the computer-readable program code can be embodied. Such a transmitted computer-readable signal medium can take a variety of forms, including but not limited to electro-magnetic, optical, or any suitable combination thereof. The computer-readable signal medium can also be any computer-readable medium that can be used to carry or transport a computer-readable program code for use by or in connection with an instruction execution system, apparatus, or device. The program code contained in the computer-readable medium can be transmitted using any suitable medium, including but not limited to wire, cable, RF (radio frequency), etc., or any suitable combination of the foregoing.

[0136] The above computer-readable medium can be included in the above electronic device; or can exist separately from the electronic device.

[0137] In some embodiments, a computer program is also provided, comprising instructions which, when executed by a processor, cause the processor to perform the method of any one of the above embodiments. For example, the instructions can be embodied in computer program code.

[0138] Computer program code for carrying out operations of the present disclosure can be written in any one or more of a variety of programming languages or combinations of languages including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0139] The flow diagrams and the block diagrams in the drawings are illustrations of architectures, functionalities, and operations of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flow diagrams or block diagrams can represent a module, a procedure, or a part of code, which comprises one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or in the reverse order, depending on the functionality involved. It is also noted that each block in the block diagrams and / or flow diagrams and combinations of blocks in the block diagrams and / or flow diagrams can be implemented by special-purpose hardware-based systems that perform the specified functions or operations, or combinations of special-purpose hardware and computer instructions.

[0140] The modules, components or units described in the embodiments of the present disclosure can be implemented by software or by hardware. In some cases, the name of the module, component or unit does not constitute a limitation on the module, component or unit itself.

[0141] The functions described above in the detailed description of embodiments of the present disclosure can be implemented in one or more hardware logic components or by any combination of hardware logic components and computer instructions. For example, non-limiting examples of hardware logic components include field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SOCs), complex programmable logic devices (CPLDs), etc.

[0142] The above description merely provides an overview of some embodiments of the present disclosure and the inventive concept thereof. The disclosure should not be limited by the specific illustrated embodiments, which can vary in many ways. For example, the above described embodiments and terminology are used in a non-limiting sense to provide a general description of the principles of the present disclosure. The scope of the disclosure is not limited to the specific embodiments described herein, but only to what the claims shall ultimately claim. It should be appreciated that those skilled in the art can devise other embodiments and modifications that fall within the scope and spirit of the disclosure. For example, features shown in the above described embodiments can be used in combination with each other or in other embodiments, or replaced by other features having the same or similar function. Any such modifications and embodiments are intended to fall within the scope of the present disclosure.

[0143] In the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the application can be practiced without these specific details. In other instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.

[0144] In addition, while operations are depicted in a particular, chronological sequence in this specification, this should not be understood as requiring such order unless specifically specified. To the contrary, it is understood that additional tasks can be performed and / or the described tasks could be performed in a different order. Moreover, as is customary in user of "for instance," "e.g.," "such as," and the like, the terms are meant to convey example, not limitation. Many of the materials and methods used are set forth in detail in the description of the application. It is understood that where expense is involved, commercial materials and methods can be used. In addition, to the extent used, terms such as "adapted," "modified," and the like can refer to use of a commercially available device or material without modification or can refer to a device or material that has been modified but is functionally equivalent to the unmodified device or material.

[0145] While certain embodiments of the disclosure have been described herein, other embodiments will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure. Therefore, the disclosure is not limited to these embodiments, but instead has a scope as defined by the appended claims. In addition, various modifications can be made in detail to the embodiments of the present disclosure without changing the overall scope and spirit of the present disclosure. The scope of the present disclosure should be defined by the claims rather than the detailed description, and all differences within the scope equivalent to or within the scope of the claims will be construed as being included in the present disclosure.

Claims

1. A dialogue method, comprising: Displays the dialogue between the user and the first intelligent agent; Based on the dialogue, generate the configuration information for the second intelligent agent to be created; Based on the setting information, a second intelligent agent is created, and the second intelligent agent is configured to participate in the dialogue between the user and the first intelligent agent based on the setting information of the second intelligent agent. Displays the dialogue between the user, the first agent, and the second agent.

2. The dialogue method according to claim 1, wherein, The setting information for generating the second intelligent agent to be created based on the dialogue includes: Extract the target topic from the dialogue; Based on the target topic, generate the configuration information for the second intelligent agent to be created.

3. The dialogue method according to claim 2, wherein, The setting information for generating the second intelligent agent to be created based on the target topic includes: Generate a summary of the dialogue; Based on the summary of the dialogue and the differences between the target topic and the setting information of the first agent, the setting information of the second agent to be created is generated.

4. The dialogue method according to claim 2, wherein, The configuration information includes at least one of the attributes of the second agent and the relationship between the second agent and the first agent. The step of generating the configuration information for the second agent to be created based on the target topic includes: Generate a summary of the dialogue; Based on the summary of the dialogue and the target topic, at least one of the following is generated: the attributes of the second agent and the relationship between the second agent and the first agent.

5. The dialogue method according to claim 2, wherein, The configuration information includes at least one of the attributes of the second agent, the relationship between the second agent and the first agent, and the background information of the second agent. The step of generating the configuration information for the second agent to be created based on the target topic includes: Determine at least one of the following: the attributes of the second agent, and the relationship between the second agent and the first agent; Background information of the second agent is generated based on at least one of the attributes of the second agent, the relationship between the second agent and the first agent, and a summary of the dialogue.

6. The dialogue method according to claim 2, wherein, The configuration information includes the knowledge base of the second intelligent agent, and the configuration information for generating the second intelligent agent to be created based on the target topic includes: The knowledge base of the second intelligent agent is generated based on the knowledge base of the first intelligent agent and the knowledge base related to the target topic.

7. The dialogue method according to any one of claims 2 to 6, wherein: The target topic is determined based on at least one of the following: the generation order of one or more topics extracted from the dialogue, or the amount of information; or The target topic is a topic predicted based on the dialogue.

8. The dialogue method according to claim 1, wherein, The setting information for generating the second intelligent agent to be created based on the dialogue includes: Based on the dialogue and the settings of the first agent, determine the matching degree between the first agent and the dialogue; In response to the matching degree being lower than a threshold, setting information for a second agent to be created is generated.

9. The dialogue method according to claim 1, wherein, Creating the second intelligent agent based on the set information includes: Display the setting information of the second intelligent agent to be created, the setting information including one or more alternative information; Receive the user's selection operation for the alternative information; Based on the information selected by the user, a second intelligent agent is created.

10. The dialogue method according to claim 1, further comprising: Based on the configuration information of the first intelligent agent, one or more alternative associated roles are determined for the first intelligent agent; Based on the relevance of each candidate associated role to the dialogue, an agent to be created is determined from the candidate associated roles, wherein the second agent is any agent to be created.

11. The dialogue method according to claim 1, further comprising: Based on the dialogue, determine whether to create a new intelligent agent; In response to determining to create a new intelligent agent, a third intelligent agent is created, the third intelligent agent being configured to participate in the dialogue between the user, the first intelligent agent, and the second intelligent agent based on the setting information of the third intelligent agent; Displays the dialogue between the user, the first agent, the second agent, and the third agent.

12. The dialogue method according to claim 11, further comprising: In response to the determination not to create a new agent, the next agent to send dialogue content is determined based on the dialogue.

13. The dialogue method according to claim 1, wherein, Creating the second intelligent agent based on the aforementioned settings includes: Based on the settings and the history of the dialogue, a second intelligent agent is created.

14. An electronic device, comprising: Memory; as well as A processor coupled to the memory, the processor being configured to execute the dialogue method as described in any one of claims 1 to 13 based on instructions stored in the memory.

15. A computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the dialogue method of any one of claims 1 to 13.

16. A computer program product, when run on a computer, causes the computer to implement the dialogue method of any one of claims 1 to 13.

17. A computer program comprising: Instructions, which, when executed by a processor, cause the processor to perform the dialogue method according to any one of claims 1 to 13.

Images