Conversation method and device, electronic equipment and storage medium
By switching to the dialogue interface in the AI dialogue scenario and using memory collection information to output the answer content, the problem of insufficient accuracy of the answer content is solved, and the accuracy and usability of question-and-answer of electronic devices are improved.
Patent Information
- Application Number
- CN202510453931.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-10
- Publication Date
- 2025-07-18
AI Technical Summary
In the AI-based dialogue scenarios in the prior art, the accuracy of the answer content is poor and it is impossible to effectively utilize the differences in user characteristics and demands.
By displaying the display interface of the target memory collection, switch to the dialogue interface after receiving the target operation, and output the answer content based on the memory information in the memory collection, and use a large language model to generate targeted answers.
It improves the accuracy of the answer content in the Q&A scenario, can better utilize stored memory information, and enhances the availability of electronic devices.
Smart Images
Figure CN120338112A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the technical field of electronic devices, and more particularly, to a dialogue method, apparatus, electronic device, and storage medium. Background Art
[0002] With the continuous development of Artificial Intelligence (AI) technology, there are more and more AI-based applications. Currently, in AI-based dialogue scenarios, it is often used to answer questions raised by users, but the accuracy of the answer content output in related technologies is not satisfactory. Summary of the Invention
[0003] This application provides a dialogue method, apparatus, electronic device, and storage medium, which can improve the accuracy of answer content in the question-and-answer scenario and can better utilize the stored memory information.
[0004] In a first aspect, an embodiment of this application provides a dialogue method, which includes: displaying a display interface of a target memory collection; switching the display interface to a dialogue interface when a target operation in the display interface is received; and when question information input in the dialogue interface is received, displaying answer content for the question information in the dialogue interface according to the question information and the memory information in the target memory collection.
[0005] In a second aspect, an embodiment of this application provides a dialogue apparatus, which includes: a first interface display module, a second interface display module, and an answer content display module. The first interface display module is configured to display a display interface of a target memory collection; the second interface display module is configured to switch the display interface to a dialogue interface when a target operation in the display interface is received; and the answer content display module is configured to display answer content for the question information in the dialogue interface according to the question information and the memory information in the target memory collection when question information input in the dialogue interface is received.
[0006] In a third aspect, an embodiment of this application provides an electronic device, including: one or more processors; a memory; one or more application programs, where the one or more application programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more application programs are configured to execute the dialogue method provided in the first aspect above.
[0007] In a fourth aspect, an embodiment of this application provides a computer-readable storage medium, in which program code is stored, and the program code can be called by a processor to execute the dialogue method provided in the first aspect above.
[0008] The solution provided by this application shows the display interface of the target memory collection. When a target operation in the display interface is received, the display interface is switched to a dialogue interface. When question information input in the dialogue interface is received, based on the question information and the memory information in the target memory collection, answer content for the question information is displayed in the dialogue interface. Thus, it can be realized that when the memory collection is displayed, it can trigger the entry into the dialogue interface, and based on the memory information recorded by the user in the memory collection, output answer content for the user's question, improving the accuracy of the answer content in the Q&A scenario, and can better utilize the stored memory information, enhancing the usability of the electronic device. BRIEF DESCRIPTION OF THE DRAWINGS
[0009] To more clearly illustrate the technical solutions in the embodiments of this application, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of this application. For those skilled in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.
[0010] Figure 1 FIG. shows a schematic flowchart of a dialogue method according to an embodiment of this application.
[0011] Figure 2 FIG. shows an interface schematic diagram provided by an embodiment of this application.
[0012] Figure 3 FIG. shows an interface schematic diagram provided by an embodiment of this application.
[0013] Figure 4 FIG. shows an interface schematic diagram provided by an embodiment of this application.
[0014] Figure 5 FIG. shows an interface schematic diagram provided by an embodiment of this application.
[0015] Figure 6 FIG. shows a schematic flowchart of a dialogue method according to another embodiment of this application.
[0016] Figure 7 FIG. shows a schematic flowchart of a dialogue method according to still another embodiment of this application.
[0017] Figure 8 FIG. shows a schematic flowchart of a dialogue method according to yet another embodiment of this application.
[0018] Figure 9 FIG. shows an interface schematic diagram provided by an embodiment of this application.
[0019] Figure 10Shows a schematic interface diagram provided by an embodiment of the present application.
[0020] Figure 11 Shows a schematic flowchart of a dialogue method according to another embodiment of the present application.
[0021] Figure 12 Shows a schematic interface diagram provided by an embodiment of the present application.
[0022] Figure 13 Shows a block diagram of a dialogue device according to an embodiment of the present application.
[0023] Figure 14 Is a block diagram of an electronic device for executing the dialogue method according to an embodiment of the present application.
[0024] Figure 15 Is a storage unit for storing or carrying program code for implementing the dialogue method according to an embodiment of the present application. Detailed implementation manners
[0025] In order to enable those skilled in the art to better understand the solutions of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application.
[0026] With the rapid development of AI technology, large language model (LLM) technology has gradually moved from theoretical research to large-scale applications, attracting extensive attention. In particular, language models based on deep learning have demonstrated excellent capabilities in tasks such as text generation, language understanding, and transformation. In the dialogue scenario, the application of large language models is particularly prominent. Through training with large-scale datasets, large language models can generate highly coherent and natural dialogue content, significantly improving the user experience.
[0027] In the dialogue scenario, large language models are often used to output answer content for user questions based on the found content sources. However, the answer content output in the related art is general. For different users, their user characteristics and needs are different. Therefore, the accuracy of the output answer content is insufficient.
[0028] To address the above problems, the inventors have proposed a dialogue method, device, electronic device, and storage medium provided by the embodiments of the present application, which can trigger entering the dialogue interface when displaying the memory collection, and output answer content for user questions based on the memory information in the memory collection, improving the accuracy of the answer content in the question-and-answer scenario, and can better utilize the stored memory information, improving the usability of the electronic device. Among them, the specific dialogue method will be described in detail in the subsequent embodiments.
[0029] Next, the dialogue method provided by the embodiments of the present application will be described in detail with reference to the accompanying drawings.
[0030] Please refer to Figure 1 , Figure 1 , which shows a schematic flowchart of the dialogue method provided by an embodiment of the present application. In a specific embodiment, the dialogue method is applied to a dialogue device 600 as shown in Figure 13 and an electronic device 100 configured with the dialogue device 600 ( Figure 14 ). Next, taking the electronic device as an example, the specific process of this embodiment will be described. Of course, it can be understood that the electronic device applied in this embodiment can be a smart phone, a tablet computer, a smart watch, an e-book, etc., which is not limited herein. Next, the process shown in Figure 1 will be elaborated in detail. The information storage method may specifically include the following steps:
[0031] Step S110: Display a display interface of a target memory collection.
[0032] Among them, the target memory collection can be any memory collection in the memory collections stored in the electronic device. The memory collection may refer to a set composed of memory information stored in the electronic device, and the memory collection may include multiple memory information. The memory information can be derived from various types of information, and the various types of information may include pictures, videos, audios, text information, etc. For example, the sources of the memory information may include: videos in the picture gallery, pictures in the picture gallery, web information (such as web addresses, etc.) collected into the system, text information recorded in applications (such as sticky notes, memos, etc.), recording information, itinerary information in the calendar, text information collected in third-party applications, pictures collected in third-party applications, videos collected in third-party applications, web information collected in third-party applications, etc. Of course, the sources of the memory information are not limited thereto. For example, it may also be derived from chat information in the chat interface, etc.
[0033] In the embodiments of the present application, the electronic device may display a display interface of the target memory collection, and the display interface is used to display the memory information in the target memory collection so that the user can view the memory information in the target memory collection.
[0034] In some embodiments, when the memory application is in the foreground running state, the electronic device may display a display interface of the target memory collection. The memory application may be an application for storing and managing memory information. Herein, the foreground running state refers to the state where the electronic device displays the application interface of the application on the screen. When the application is in the foreground running state, the user can see the application interface of the application on the screen; the background running state opposite to the foreground running state refers to the state where the electronic device does not display the application interface of the application on the screen and the user cannot interact with the application interface of the application, but the application still occupies the system resources of the electronic device when it is in the background running state.
[0035] In some embodiments, when constructing the above memory collection, the electronic device may display a creation interface of the memory collection in response to an operation detected for creating the memory collection; obtain the keyword of the memory collection input in the creation interface, and this keyword may be used as the name of the memory collection; through semantic understanding of the keyword, obtain a semantic understanding result, and may also deeply understand all the stored memory information according to the title and content of the memory information to obtain the memory understanding result of each memory information; then match the above semantic understanding result with the memory understanding result of the memory information to obtain a matching result; according to the matching result, the memory information belonging to the memory collection can be determined, wherein the memory understanding result of the memory information determined in the memory collection is matched with the above language understanding result.
[0036] In a possible embodiment, when a new memory information is added to the electronic device, the memory understanding result of the new memory information can be obtained by deeply understanding the new memory information according to the title and content of the memory information; then match the memory understanding result with the semantic understanding results corresponding to the keywords of each memory collection; if there is a semantic understanding result corresponding to the keyword of any memory collection that matches the memory understanding result of the memory information, then the memory information can be added to the memory collection.
[0037] In a possible embodiment, when a new memory information is added to the electronic device, the new memory information can also be matched with the memory information in each memory collection. If there is memory information in any memory collection that has a commonality with the new memory information, then the new memory information can be added to the memory collection.
[0038] Through the above embodiments, the memory collection can be continuously changed as the memory information is updated, ensuring the accuracy of the memory information in the memory collection.
[0039] In some embodiments, when constructing the above memory collections, the electronic device may also cluster the stored memory information to obtain a clustering result; according to each category in the clustering result, memory collections corresponding to each category may be created, and the memory collections include the memory information corresponding to the category.
[0040] In a possible embodiment, when new memory information is added to the electronic device, the memory information may be reclustered or the new memory information may be matched with the memory information in each memory collection, so that the new memory information can be added to the corresponding memory collection according to the clustering result or the matching result.
[0041] It should be noted that each memory information stored in the electronic device may exist in one memory collection or may exist in multiple memory collections at the same time. In the embodiments of the present application, the specific manner of constructing the memory collection may not be limited.
[0042] In some embodiments, the electronic device may display the application interface corresponding to the memory application, and the application interface is used to display the recorded memory information, and the application interface may include a control for triggering the display of the memory collection list; when a trigger operation on the control is detected, for example, a click operation or a press operation on the control is detected, the collection list may be displayed in the application interface, and the collection list may include the memory collections existing in the electronic device; when a selection operation on the target memory collection in the collection list is detected by the electronic device, the displayed interface may be switched to the display interface of the above target memory collection.
[0043] Exemplarily, please refer to Figure 2 , the electronic device may display the application interface A1 of the memory application. When displaying the application interface A1, it is default to display in the form of a list of the recorded memory information, that is, all the added memory information is displayed in the application interface; please refer to Figure 3 , after detecting an operation on the "Collection" option A2 in the application interface A1, options corresponding to each existing memory collection may be displayed in the application interface A1; please refer to Figure 4 , after detecting an operation on the option of the memory collection of "Travel Guide", the application interface A1 may be switched to the display interface A3 of the memory collection, and the memory information in the memory collection of "Travel Guide" may be displayed in the display interface A3.
[0044] Step S120: When a target operation in the display interface is received, switch the display interface to a dialogue interface.
[0045] In an embodiment of the present application, when the electronic device displays the display interface of the target memory collection, it can detect operations in the interface. Considering that when the user views the target memory collection, there may be needs for information query, content generation, content creation, information processing, etc. for the target memory collection. Therefore, when the electronic device detects the target operation in the above display interface, it can switch the above display interface to a dialogue interface so that the user can ask questions in the dialogue scenario, thereby meeting the user's needs.
[0046] In some embodiments, the above dialogue interface can be the dialogue interface corresponding to the assistant function called from the memory application. This assistant function can be understood as an intelligent assistant. An intelligent assistant is a software program that utilizes artificial intelligence technology and aims to help users complete various tasks, provide the required information and services. The user can ask questions to the intelligent assistant in the dialogue interface to complete the tasks required by the user.
[0047] Exemplarily, please refer to Figure 5 , after the electronic device receives the target operation in the above display interface, it can switch the displayed interface to the dialogue interface A4. In this way, the user can input questions in the dialogue interface A4 to complete the tasks required by the user.
[0048] In some embodiments, the above target operation can be an operation on the displayed interface element. For example, when the electronic device detects an operation on the target control in the display interface, it can switch the displayed interface to the dialogue interface. The above target operation can also be a touch operation that meets the preset touch conditions. For example, when the electronic device detects a sliding operation on the target area in the display interface that meets the target sliding conditions, it can switch the displayed interface to the dialogue interface. The target sliding conditions can include at least one of the sliding direction being the target sliding direction and the sliding distance being greater than the target sliding distance.
[0049] Of course, in the embodiment of the present application, the specific operation method for triggering the display of the above dialogue interface may not be limited.
[0050] Step S130: When receiving the question information input in the dialogue interface, display the answer content for the question information in the dialogue interface according to the question information and the memory information in the target memory collection.
[0051] In an embodiment of the present application, after the electronic device displays the above dialogue interface, it can detect operations in the dialogue interface. When it determines that the user has input question information based on the detected operations, it can display the answer content for the question information in the dialogue interface. Moreover, when displaying the answer content, it is based on the question information and the memory information in the above target memory collection to display the answer content, so that when answering the questions raised by the user, the data referred to by the electronic device is more targeted, avoiding the output answer content from being too general, and thus improving the accuracy of the displayed answer content.
[0052] In some embodiments, the electronic device can use a large language model and, based on the above question information and the memory information in the target memory collection, determine the answer content to be output. The large language model is an artificial intelligence technology based on deep learning. By training on a large amount of text data, it learns the statistical laws and semantic relationships of language, and thus can generate natural, fluent and human-language-habit-compliant content.
[0053] In some embodiments, the above question information can be determined based on the detected touch input operations in the interface, or can be determined based on the collected voice input by the user, so that the user can input question information either manually or by voice input.
[0054] The dialogue method provided by the embodiments of the present application can realize triggering to enter the dialogue interface when displaying the memory collection, and outputting the answer content for the user's question based on the memory information recorded by the user in the memory collection, so that when answering the questions raised by the user, the data referred to by the electronic device is more targeted, avoiding the output answer content from being too general, thereby improving the accuracy of the answer content in the question-and-answer scenario, and can better utilize the stored memory information, improving the usability of the electronic device; and there are usually multiple memory information in the memory collection. Using multiple memory information as a reference, information can be extracted across a single memory information and then output the answer content to the user.
[0055] Please refer to Figure 6 , Figure 6 which shows a schematic flowchart of a dialogue method provided by another embodiment of the present application. This dialogue method is applied to the above-mentioned electronic device. The following will elaborate in detail on the Figure 6 shown process. The dialogue method can specifically include the following steps:
[0056] Step S210: Display the display interface of the target memory collection.
[0057] Step S220: When receiving the target operation in the display interface, switch the display interface to a dialogue interface.
[0058] In the embodiment of the present application, steps S210 and S220 can refer to the content of other embodiments and will not be elaborated here.
[0059] Step S230: When the problem information input in the dialogue interface is received, match the memory information in the target memory set with the problem information to obtain a matching result.
[0060] In the embodiment of the present application, after the electronic device receives the problem information input in the above dialogue interface, when outputting the answer content for the problem information according to the received problem information and the memory information in the above target memory set, it can match each memory information in the target memory set with the problem information to obtain a matching result, so as to more accurately determine the content reference for determining the answer content according to the matching result in the subsequent process.
[0061] In some embodiments, when matching the memory information in the target memory set with the problem information, the keywords in the problem information can be determined. For example, if the problem information is "What is the scenery of Scenic Spot A?", then the keyword in the problem information can be "Scenic Spot A". Then, match the memory information in the target memory set with the keyword to obtain the matching results of each memory information in the target memory set.
[0062] In a possible embodiment, if the memory information is structured information, then the structured information can be directly matched with the above keywords to obtain the matching result of the memory information and the above keywords. Among them, structured information is a data organization form in which data is organized into a predefined format. Structured information can include clear fields or columns, as well as corresponding data types, etc. Such a data organization method can make the storage, retrieval, and analysis of data more efficient and accurate.
[0063] In the above embodiment, when structured information is used for memory information, the structured information can be in the form of triples or key-value pairs. Of course, the specific form of the structured information can be not limited. Exemplarily, when the structured information is in the form of triples, the structured information can include an entity, an entity attribute, and an entity attribute value. Among them, the entity can represent the subject, the entity attribute can represent the relationship between the entity and the entity attribute value, and the entity attribute value is the specific value of the entity attribute. For example, for the memory information of "The product was released on April 10th", the entity in the triple is "released product", the entity attribute is "time", and the entity attribute value is "April 10th".
[0064] In a possible implementation, if the memory information is unstructured information, then for the memory information, in-depth understanding can be carried out according to the title and content of the memory information to obtain a memory understanding result; then the memory understanding result is matched with the above keywords to obtain a matching result between the memory information and the above keywords. The memory information is stored in an unstructured form, which can provide necessary context information and knowledge background for the large language model, etc., and helps to optimize the response quality and decision-making logic of the large language model.
[0065] Step S240: Based on the matching result, display the answer content for the question information in the dialogue interface.
[0066] In the embodiments of the present application, after obtaining the matching result between the memory information in the target memory collection and the question information, based on the matching result, the answer content to be output can be determined, and the answer content is displayed in the dialogue interface.
[0067] In some embodiments, based on the matching result, the memory information in the target memory collection that matches the above question information can be determined, and based on the matching memory information, the answer content to be output to the user can be determined.
[0068] The dialogue method provided by the embodiments of the present application can realize triggering the entry into the dialogue interface when displaying the memory collection, and outputting the answer content for the user's question based on the memory information recorded by the user in the memory collection, so that when answering the question raised by the user, the data referred to by the electronic device is more targeted, avoiding the output answer content from being too general, thereby improving the accuracy of the answer content in the question-and-answer scenario, and can better utilize the stored memory information, improving the usability of the electronic device; and when outputting the answer content according to the memory information in the memory collection, by matching the memory information with the question information and outputting the answer content according to the matching result, the reference content for determining the answer content can be made more accurate, further improving the accuracy of the answer content.
[0069] Please refer to Figure 7 , Figure 7 which shows a schematic flowchart of a dialogue method provided by another embodiment of the present application. This dialogue method is applied to the above-mentioned electronic device, and the following will elaborate in detail on the Figure 7 shown process. The dialogue method may specifically include the following steps:
[0070] Step S310: Display the display interface of the target memory collection.
[0071] Step S320: When receiving the target operation in the display interface, switch the display interface to a dialogue interface.
[0072] Step S330: When the question information input in the conversation interface is received, match the memory information in the target memory set with the question information to obtain a matching result.
[0073] In the embodiments of the present application, for steps S310 to S330, reference may be made to the content of other embodiments, which will not be elaborated here.
[0074] Step S340: If the matching result indicates that there is target memory information in the target memory set that matches the question information, display the answer content for the question information in the conversation interface according to the target memory information.
[0075] In the embodiments of the present application, considering that although the user initiates a conversation process when viewing the target memory set, in some cases, there may be no memory information related to the question raised by the user in the target memory set. Therefore, it is impossible to match the corresponding memory information from the target memory set according to the question information input by the user. Therefore, after obtaining the above matching result, it can be determined whether there is memory information in the target memory set that matches the question information according to the matching result. If there is at least one memory information (as target memory information) in the target memory set that matches the above question information, then the answer content for the question information can be displayed in the conversation interface according to the target memory information.
[0076] In some embodiments, the above target memory information and question information can be input into a large language model so that the large language model outputs a result for answering the question information according to the target memory information. The result output by the large language model can be used as the answer content and displayed in the conversation interface.
[0077] In a possible embodiment, a prompt information can be generated according to the above target memory information and question information. The prompt information is used to prompt the large language model to generate the answer content for the question information according to the target memory information. Among them, the prompt information can be understood as a prompt (prompt), which is a piece of text or data specially designed to guide the large language model to generate a specific type or style of output and can be in the context provided to the large model (such as a deep learning model, a natural language processing model, etc.). The prompt information aims to activate the relevant knowledge and capabilities in the model so that it can respond or complete the task in the expected manner. The application of the prompt information in the large language model is extensive, especially in the field of natural language processing (Natural Language Processing, NLP).
[0078] Step S350: If the matching result indicates that there is no target memory information in the target memory collection that matches the question information, obtain content that matches the question information from other content stored in the electronic device other than the target memory collection.
[0079] In the embodiments of the present application, according to the above matching result, after determining whether there is memory information in the target memory collection that matches the question information, if there is no memory information in the target memory collection that matches the above question information, then in order to complete the task of answering the question information input by the user, the reference content for answering the question information can be determined from other content stored in the electronic device other than the target memory collection. Among them, other content may include: memory information other than the memory information in the target memory collection among all the memory information stored in the electronic device, non-memory information (such as photo albums, documents, etc.) stored in the electronic device, at least one of them.
[0080] In some embodiments, when matching the above other content with the question information, for the memory information in the other content, the foregoing method of matching the memory information with the question information can be referred to, which will not be elaborated here; for the information other than the memory information in the other content, in-depth understanding can be performed on this information to obtain an understanding result, and then the understanding result is matched with the question information to obtain the matching result of this information and the question information.
[0081] Step S360: If there is target storage content in the other content that matches the question information, display the answer content for the question information in the dialogue interface according to the target storage content.
[0082] In the embodiments of the present application, after the electronic device obtains content that matches the question information from the above other content stored, if there is target storage content in the other content that matches the question information, then the matched target storage content can be used as the reference content for answering the question information input by the user, so that the answer content for the question information can be displayed in the dialogue interface based on the target storage content.
[0083] In some embodiments, when outputting the answer content for the question information based on the target storage content, the above target storage content and the question information can also be input into a large language model, so that the large language model outputs a result for answering the question information according to the target storage content, and the result output by the large language model can be used as the answer content and displayed in the dialogue interface.
[0084] Step S370: If there is no target stored content in the other content that matches the problem information, then display, in the dialogue interface, the answer content for the problem information based on the online search content, where the online search content is obtained by performing an online search based on at least one of the collection information corresponding to the target memory collection and the problem information.
[0085] In an embodiment of the present application, after the electronic device obtains the content that matches the problem information from the above other stored content, if there is no stored content in the other content that matches the problem information, then the electronic device can also obtain, from the outside, reference content for answering the problem information input by the user. Among them, the electronic device can perform an online search based on at least one of the collection information corresponding to the above target memory collection and the problem information, and use the content found as the online search content, and then output, based on the online search content, the answer content for the above problem information.
[0086] It can be understood that when the user triggers a dialogue process to ask questions while viewing the target memory collection, the questions asked are usually related to the target memory collection. Then, performing an online search for content related to the target memory collection as the reference content for answering the problem information can ensure the accuracy of the reference content. In addition, directly performing an online search for content related to the problem information as the reference content for answering the problem information can also ensure the accuracy of the reference content. Therefore, an online search can be performed based on at least one of the collection information corresponding to the above target memory collection and the problem information, and the content found can be used as the reference content for answering the problem information.
[0087] Through the dialogue method provided by the embodiment of the present application, it is possible to output the answer content by referring to the memory information in the target memory collection, other content stored in the electronic device, and the online search content, and the memory information in the target memory collection, other content stored in the electronic device, and the online search content are used as reference content in decreasing order of priority, thereby improving the accuracy of the output answer content.
[0088] In some embodiments, the collection information of the above target memory collection can reflect the characteristics of the target memory collection. For example, the target memory collection may include at least one of the name of the target memory collection, tags, and summary and generalization information for the memory information in the target memory collection.
[0089] In some embodiments, when outputting the answer content for the problem information based on the online search content, the online search content and the problem information can also be input into a large language model, so that the large language model outputs, according to the online search content, the result for answering the problem information, and the result output by the large language model can be used as the answer content and displayed in the dialogue interface.
[0090] The dialogue method provided by the embodiment of the present application can trigger the entry into the dialogue interface when displaying the memory collection, and output the answer content to the user's question based on the memory information recorded by the user in the memory collection, so that when answering the question raised by the user, the memory information in the target memory collection is preferentially referred to, and other content stored in the electronic device and online search content are secondarily referred to. Thus, the reference content for determining the answer content can be made more accurate, avoiding the answer content being too general, further improving the accuracy of the answer content, and making better use of the stored memory information, thereby enhancing the usability of the electronic device.
[0091] Please refer to Figure 8 , Figure 8 which shows a schematic flowchart of the dialogue method provided by another embodiment of the present application. This dialogue method is applied to the above-mentioned electronic device. The following will elaborate in detail on the Figure 8 process shown. The dialogue method may specifically include the following steps:
[0092] Step S410: Display the display interface of the target memory collection.
[0093] Step S420: When receiving the target operation in the display interface, switch the display interface to the dialogue interface.
[0094] In the embodiment of the present application, for Step S410 and Step S420, reference may be made to the content of other embodiments, which will not be elaborated here.
[0095] Step S430: When receiving the question information input in the dialogue interface, based on the memory information in the target memory collection, other content stored in the electronic device except the target memory collection, and the online search content, display the answer content to the question information in the dialogue interface, where the online search content is obtained by online search based on at least one of the collection information corresponding to the target memory collection and the question information.
[0096] In the embodiment of the present application, after the electronic device receives the question information input in the above dialogue interface, when outputting the answer content to the question information, it may determine the answer content by referring to multiple data sources. Among them, the electronic device may, based on the memory information in the target memory collection, other content stored in the electronic device except the target memory collection, and the online search content, display the answer content to the question information in the dialogue interface, and the online search content is obtained by online search based on at least one of the collection information corresponding to the target memory collection and the question information.
[0097] In some embodiments, when outputting the answer content based on the memory information, other content, and online search content in the above-mentioned target memory collection, it can be based on the memory information in the above-mentioned target memory collection, the above-mentioned other content, the above-mentioned online search content, the reference weight corresponding to the above-mentioned target memory collection, the reference weight corresponding to the above-mentioned other content, and the reference weight corresponding to the above-mentioned online search content. The answer content for the question information is displayed in the dialogue interface. The reference weight corresponding to the target memory collection is greater than the reference weight corresponding to the other content, and the reference weight corresponding to the other content is greater than the reference weight corresponding to the online search content.
[0098] Among them, the above reference weight is used to represent the importance of the corresponding content when generating the answer content. The role of the reference weight can be understood as quantitatively evaluating the importance of the reference content of different data sources when generating the answer content, and dynamically adjusting its participation degree and influence in the answer generation process through this evaluation. The greater the reference weight, the greater the contribution of the reference content of this data source to the generation of the answer content. On the contrary, the smaller the reference weight, the smaller the contribution of the reference content of this data source to the generation of the answer content. Since the reference weight corresponding to the target memory collection, the reference weight corresponding to the other content, and the reference weight corresponding to the online search content decrease in sequence, the influence and contribution of the memory information, other content, and online search content in the target memory collection to the generated answer content also decrease in sequence. Thus, while being able to output the answer content by referring to multiple data sources, the memory information in the target memory collection can be preferentially referred to, and the other content and online search content stored in the electronic device can be referred to secondarily.
[0099] In some embodiments, when outputting the answer content based on the memory information, the above-mentioned other content, the above-mentioned online search content, the reference weight corresponding to the above-mentioned target memory collection, the reference weight corresponding to the above-mentioned other content, and the reference weight corresponding to the above-mentioned online search content, it can be to provide the above-mentioned question information, the memory information in the above-mentioned target memory collection, the above-mentioned other content, the above-mentioned online search content, the reference weight corresponding to the above-mentioned target memory collection, the reference weight corresponding to the above-mentioned other content, and the reference weight corresponding to the above-mentioned online search content to the large language model, and the definition of the reference weight can be provided to the large language model. Thus, the large language model can output the answer content for answering the above-mentioned question information according to the above reference weight and in combination with the memory information, the above-mentioned other content, and the above-mentioned online search content in the above-mentioned target memory collection.
[0100] Exemplarily, for the question information "What is the temperature in City A in the middle of this month" input by the user, the memory information in the target memory collection includes that the average temperature in City A in the current month is 20°C, the average temperature in City A in the current month recorded in the memo in the electronic device is 22°C, and the average temperature in City A in the current month found through online query is 24°C. And the reference weight corresponding to the target memory collection is 0.5, the reference weight corresponding to other content stored in the electronic device is 0.3, and the reference weight corresponding to the online search content is 0.2. Then, it can be determined that the average temperature in City A in the current month is: 20 * 0.5 + 22 * 0.3 + 24 * 0.2 = 21.4°C.
[0101] In a possible implementation manner, when the electronic device displays the answer content in the dialogue interface, it can also display the answer content gradually. Among them, the content output by the large language model can be gradually displayed, that is, the content is generated and displayed in the dialogue interface in the form of a continuous stream, thus achieving the effect of streaming on the screen and further improving the presentation effect of the result output by the large language model. Exemplarily, please refer to Figure 9 and Figure 10 , and can display Figure 9 and Figure 10 the interface A4 shown in sequence, so as to continuously display the content output by the large language model and achieve the effect of streaming on the screen.
[0102] In some implementation manners, after the electronic device outputs the answer content in the above dialogue interface each time, it can also store the answer content as memory information in the memory collection. That is to say, the answer content this time is related to the target memory collection. Therefore, the answer content this time can be used as new memory information, and the new memory information can be added to the target memory collection. In this way, when in the dialogue scenario corresponding to this memory collection subsequently, the answer content this time can be used as the reference content for the next question and answer. Moreover, if the answer content this time references the online search content, it can also realize supplementing the content related to the target memory collection from the Internet in the memory collection, thereby enriching the memory information in the memory collection. It should be noted that this implementation manner can also be applied to other embodiments. That is to say, in any embodiment of the present application, after the electronic device outputs the answer content for the question information, it can add the answer content to the memory collection, thereby enriching a certain type of memory content of the user (that is, the memory information in this memory collection).
[0103] The dialogue method provided by the embodiment of the present application can trigger entering the dialogue interface when displaying the memory collection, and output the answer content to the user's question based on the memory information recorded by the user in the memory collection, so that when answering the user's question, the memory information in the target memory collection, other content stored in the electronic device, and online search content are simultaneously referred to, so that the reference content for determining the answer content can be more comprehensive and accurate, further improving the accuracy of the answer content, and the stored memory information can be better utilized, improving the usability of the electronic device.
[0104] Please refer to Figure 11 , Figure 11 which shows a schematic flowchart of a dialogue method provided by another embodiment of the present application. This dialogue method is applied to the above-mentioned electronic device. The following will elaborate in detail on the Figure 11 process shown. The dialogue method can specifically include the following steps:
[0105] Step S510: Display the display interface of the target memory collection.
[0106] In the embodiment of the present application, step S510 can refer to the content of the foregoing embodiment and will not be elaborated here.
[0107] Step S520: When a trigger operation for the target interface element is received, switch the display interface to the dialogue interface.
[0108] Among them, the display interface of the target memory collection displayed by the electronic device may include a target interface element for triggering the display of the dialogue interface. The target interface element may be an interactive interface element. The interactive interface element may refer to an interface component that the user can operate and trigger a predefined action. These operations may include clicking, hovering, dragging, inputting text, selecting options, etc. The above target interface elements may include controls, service cards, hyperlinks, etc.
[0109] In the embodiment of the present application, after the electronic device displays the above display interface, it can detect the operations in the display interface. When a trigger operation for the above target interface element is detected, for example, a click operation or a press operation for the target interface element is detected, the display interface can be switched to the dialogue interface so that the user can ask questions in the dialogue scenario to meet the user's needs.
[0110] Exemplarily, please refer to Figure 12 , in the target interface element A5 displayed in the display interface A3, topic information about the memory information can be displayed, so as to guide the user to operate on the target interface element A5, thereby triggering a dialogue interaction with the user.
[0111] In some embodiments, when the electronic device displays the above target interface elements in the display interface, it may display the target interface elements in the display interface based on the memory information in the target memory collection. The target interface elements include topic prompt information generated based on the memory information, and the topic prompt information includes topic information for prompting the triggering of a conversation process. That is to say, topic information related to the memory information in the target memory collection can be presented in the target interface elements, thereby guiding the user to operate on the target interface elements, and then triggering the display of the conversation interface to achieve interaction with the user.
[0112] In a possible embodiment, since the memory information in the memory collection is constantly changing, the topic prompt information in the target interface elements displayed in the display interface can also change with the change of the memory information in the memory collection. For example, when new memory information is added to the memory collection, topic prompt information can be pushed in the target interface elements according to the newly added memory information.
[0113] In a possible embodiment, the above target interface elements may be in the form of service cards. After being triggered, the service cards can switch the displayed interface to the above conversation interface. And content can be displayed in the service cards, and the displayed content can include the above topic prompt information.
[0114] In a possible embodiment, the electronic device can extract the core elements of the memory information in the target memory collection. The core elements can include at least one of elements such as emotional value, scene characteristics, and user interests. The emotional value can include key emotions in the memory information (such as joy, touch, nostalgia, etc.), the scene characteristics can include key information such as time, place, people, and events, and the user interests can include the preference information of the user. According to the extracted core elements, topic information of types such as memory-triggering type, opinion-collision type, emotion-expression type, and call-to-action type can be generated.
[0115] Step S530: When the question information input in the conversation interface is received, according to the question information and the memory information in the target memory collection, display the answer content for the question information in the conversation interface.
[0116] In the embodiments of the present application, the content of step S530 can refer to the content of the foregoing embodiments and will not be elaborated here.
[0117] The dialogue method provided by the embodiments of the present application can trigger the entry into the dialogue interface when displaying the memory collection, and output the answer content to the user's question based on the memory information recorded by the user in the memory collection, so that when answering the user's question, the data referred to by the electronic device is more targeted, avoiding the output answer content from being too general, thereby improving the accuracy of the answer content in the question-and-answer scenario, and can better utilize the stored memory information, improving the usability of the electronic device; in addition, by displaying the topic information generated based on the memory information in the memory collection in the target interface element for triggering the dialogue when displaying the memory collection, it is possible to deeply explore the unmet needs of the user and actively explore the common demands of users with similar memory content, improving the user's interaction rate and also improving the usability of the electronic device.
[0118] Please refer to Figure 13 , which shows a structural block diagram of a dialogue device 600 provided by the embodiments of the present application. This dialogue device 600 is applied to the above-mentioned electronic device. The dialogue device 600 includes: a first interface display module 610, a second interface display module 620, and an answer content display module 630. Among them, the first interface display module 610 is used to display the display interface of the target memory collection; the second interface display module 620 is used to switch the display interface to the dialogue interface when receiving the target operation in the display interface; the answer content display module 630 is used to display the answer content to the question information in the dialogue interface according to the question information and the memory information in the target memory collection when receiving the question information input in the dialogue interface.
[0119] In some embodiments, the answer content display module 630 can be specifically used for: when receiving the question information, matching the memory information in the target memory collection with the question information to obtain a matching result; based on the matching result, displaying the answer content to the question information in the dialogue interface.
[0120] In a possible embodiment, the answer content display module 630 can be specifically used for: if the matching result indicates that there is target memory information in the target memory collection that matches the question information, then displaying the answer content to the question information in the dialogue interface according to the target memory information.
[0121] Optionally, the answer content display module 630 can also be used to: if the matching result indicates that there is no target memory information matching the question information in the target memory collection, obtain content matching the question information from other content stored in the electronic device other than the target memory collection; if there is target stored content matching the question information in the other content, display the answer content for the question information in the conversation interface according to the target stored content.
[0122] Optionally, the answer content display module 630 can also be used to: if there is no target stored content matching the question information in the other content, display the answer content for the question information in the conversation interface based on the online search content, where the online search content is obtained by online searching based on at least one of the collection information corresponding to the target memory collection and the question information.
[0123] In some embodiments, the answer content display module 630 can specifically be used to: when receiving the question information, display the answer content for the question information in the conversation interface based on the memory information in the target memory collection, other content stored in the electronic device other than the target memory collection, and the online search content, where the online search content is obtained by online searching based on at least one of the collection information corresponding to the target memory collection and the question information.
[0124] In a possible implementation, when receiving the question information, display the answer content for the question information in the conversation interface based on the memory information, the other content, the online search content, the reference weight corresponding to the target memory collection, the reference weight corresponding to the other content, and the reference weight corresponding to the online search content, where the reference weight is used to represent the importance of the corresponding content when generating the answer content, the reference weight corresponding to the target memory collection is greater than the reference weight corresponding to the other content, and the reference weight corresponding to the other content is greater than the reference weight corresponding to the online search content.
[0125] In some embodiments, the display interface includes a target interface element, and the second interface display module 620 can specifically be used to: when receiving a trigger operation for the target interface element, switch the display interface to a conversation interface.
[0126] In a possible implementation manner, the first interface display module 610 can also be configured to, before switching the display interface to a dialogue interface when receiving a trigger operation for the target interface element, display the target interface element in the display interface based on the memory information in the target memory collection. The target interface element includes topic prompt information generated based on the memory information, and the topic prompt information includes topic information for prompting the triggering of a dialogue process.
[0127] In a possible implementation manner, the dialogue device 600 may further include a memory storage module. The memory storage module is configured to, when receiving the question information input in the dialogue interface, after displaying the answer content for the question information in the dialogue interface according to the question information and the memory information in the target memory collection, store the answer content as memory information into the target memory collection.
[0128] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described devices and modules can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.
[0129] In several embodiments provided in the present application, the coupling between modules can be electrical, mechanical, or other forms of coupling.
[0130] In addition, in each embodiment of the present application, each functional module can be integrated in a processing module, or each module can exist physically alone, or two or more modules can be integrated in one module. The above-mentioned integrated modules can be implemented in the form of hardware or in the form of software functional modules.
[0131] In summary, the solution provided in the present application, by displaying the display interface of the target memory collection, switching the display interface to a dialogue interface when receiving a target operation in the display interface, and when receiving the question information input in the dialogue interface, displaying the answer content for the question information in the dialogue interface according to the question information and the memory information in the target memory collection. Thus, it can be realized that when displaying the memory collection, it can trigger the entry into the dialogue interface, and output the answer content for the user's question based on the memory information recorded by the user in the memory collection, improving the accuracy of the answer content in the question-and-answer scenario, and can better utilize the stored memory information, improving the usability of the electronic device.
[0132] Please refer to Figure 14, which shows a structural block diagram of an electronic device provided by an embodiment of the present application. The electronic device 100 may be an electronic device such as a smart phone, a tablet computer, an e-book, a notebook computer, a smart watch, etc. that can run application programs. The electronic device 100 in the present application may include one or more of the following components: a processor 110, a memory 120, a display screen 130, and one or more application programs, where one or more application programs may be stored in the memory 120 and configured to be executed by one or more processors 110, and one or more application programs are configured to execute the methods described in the foregoing method embodiments.
[0133] The processor 110 connects various parts within the entire electronic device 100 using various interfaces and lines, and by running or executing instructions, programs, code sets, or instruction sets stored in the memory 120, and by calling data stored in the memory 120, it executes various functions of the electronic device 100 and processes data.
[0134] The memory 120 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 120 may include a program storage area and a data storage area. Among them, the program storage area may store instructions for implementing an operating system, instructions for implementing at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the foregoing method embodiments, etc.
[0135] The display screen 130 is a display component for image display, and is usually arranged on the front panel of the electronic device 100.
[0136] Please refer to Figure 15 , which shows a structural block diagram of a computer-readable storage medium provided by an embodiment of the present application. Program code is stored in the computer-readable medium 800, and the program code can be called by a processor to execute the methods described in the foregoing method embodiments.
[0137] The computer-readable storage medium 800 may be an electronic memory such as a flash memory, an EEPROM (electrically erasable programmable read-only memory), an EPROM, a hard disk, or a ROM. Optionally, the computer-readable storage medium 800 includes a non-transitory computer-readable storage medium. The computer-readable storage medium 800 has a storage space for the program code 810 for executing any method step in the foregoing method. These program codes can be read out from or written into one or more computer program products. The program code 810 may be compressed in an appropriate form, for example.
[0138] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present application.
Claims
1. A dialogue method, characterized in that, The method includes: Displaying a display interface of a target memory collection; When receiving a target operation in the display interface, switching the display interface to a dialogue interface; When receiving question information input in the dialogue interface, displaying answer content for the question information in the dialogue interface according to the question information and the memory information in the target memory collection.
2. The method according to claim 1, characterized in that, The step of, when receiving question information input in the dialogue interface, displaying answer content for the question information in the dialogue interface according to the question information and the memory information in the target memory collection includes: When receiving the question information, matching the memory information in the target memory collection with the question information to obtain a matching result; Based on the matching result, displaying answer content for the question information in the dialogue interface.
3. The method according to claim 2, wherein The step of, based on the matching result, displaying answer content for the question information in the dialogue interface includes: If the matching result indicates that there is target memory information in the target memory collection that matches the question information, displaying answer content for the question information in the dialogue interface according to the target memory information.
4. The method according to claim 3, wherein The step of, based on the matching result, displaying answer content for the question information in the dialogue interface further includes: If the matching result indicates that there is no target memory information in the target memory collection that matches the question information, obtaining content that matches the question information from other content stored in the electronic device other than the target memory collection; If there is target stored content in the other content that matches the question information, displaying answer content for the question information in the dialogue interface according to the target stored content.
5. The method according to claim 4, wherein The step of, based on the matching result, displaying answer content for the question information in the dialogue interface further includes: If there is no target stored content in the other content that matches the question information, displaying answer content for the question information in the dialogue interface based on online search content, where the online search content is obtained by online search based on at least one of the collection information corresponding to the target memory collection and the question information.
6. The method according to claim 1, wherein The step of, when receiving question information input in the dialogue interface, displaying answer content for the question information in the dialogue interface according to the question information and the memory information in the target memory collection includes: When receiving the question information, displaying answer content for the question information in the dialogue interface based on the memory information in the target memory collection, other content stored in the electronic device other than the target memory collection, and online search content, where the online search content is obtained by online search based on at least one of the collection information corresponding to the target memory collection and the question information.
7. The method according to claim 6, characterized in that, In the case of receiving the problem information, based on the memory information in the target memory collection, other content stored in the electronic device other than the target memory collection, and online search content, display answer content for the problem information in the dialogue interface, including: In the case of receiving the problem information, based on the memory information, the other content, the online search content, the reference weight corresponding to the target memory collection, the reference weight corresponding to the other content, and the reference weight corresponding to the online search content, display answer content for the problem information in the dialogue interface, where the reference weight is used to represent the importance of the corresponding content when generating the answer content, the reference weight corresponding to the target memory collection is greater than the reference weight corresponding to the other content, and the reference weight corresponding to the other content is greater than the reference weight corresponding to the online search content.
8. The method according to any one of claims 1 to 7, characterized in that, The display interface includes a target interface element. In the case of receiving a target operation in the display interface, switching the display interface to a dialogue interface, including: In the case of receiving a trigger operation for the target interface element, switch the display interface to a dialogue interface.
9. The method according to claim 8, wherein Before switching the display interface to a dialogue interface in the case of receiving a trigger operation for the target interface element, the method further includes: Based on the memory information in the target memory collection, display the target interface element in the display interface, where the target interface element includes topic prompt information generated based on the memory information, and the topic prompt information includes topic information for prompting the triggering of the dialogue process.
10. The method according to any one of claims 1-7, characterized in that, After displaying answer content for the problem information in the dialogue interface according to the problem information and the memory information in the target memory collection in the case of receiving the problem information input in the dialogue interface, the method further includes: Store the answer content as memory information in the target memory collection.
11. A dialogue device, characterized in that, The device includes: a first interface display module, a second interface display module, and an answer content display module, where The first interface display module is used to display the display interface of the target memory collection; The second interface display module is used to switch the display interface to a dialogue interface in the case of receiving a target operation in the display interface; The answer content display module is used to display answer content for the problem information in the dialogue interface according to the problem information and the memory information in the target memory collection in the case of receiving the problem information input in the dialogue interface.
12. An electronic device, characterized in that, Including: One or more processors; A memory; One or more programs, where the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs are configured to execute the method according to any one of claims 1-10.
13. A computer-readable storage medium, characterized in that, Program code is stored in the computer-readable storage medium, and the program code can be called by the processor to execute the method according to any one of claims 1-10.