Interaction method and device, wearable equipment, server and storage medium
By interacting between wearable devices and servers, using AI models to process user commands, and combining historical interaction information and object descriptions, the problem of insufficient intelligence in virtual characters is solved, thus improving the user experience.
Patent Information
- Application Number
- CN202411022480.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-29
- Publication Date
- 2024-11-12
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The lack of intelligence in virtual characters in existing wearable devices leads to a poor user experience.
By interacting between wearable devices and servers, AI models are used to process user commands and combine historical interaction information between users and smart objects with object descriptions to generate response information that better meets user needs.
It improves the intelligence level of virtual characters, enhances the user's interactive experience with virtual characters, and provides a more vivid and interesting interaction method.
Smart Images

Figure CN118939120A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure belongs to the technical field of electronic devices, and more specifically, relates to an interaction method and device, a wearable device, a server, and a storage medium. Background Art
[0002] Virtual characters or images are set in wearable devices (such as smart watches, smart glasses, etc.). These virtual characters (or smart objects) interact with users and can provide users with a variety of services such as information display, entertainment, and health management. For example, virtual characters can serve as users' virtual assistants to display basic information such as time, weather, and schedules; virtual characters can interact with users through voice and action, and provide entertainment content such as games, music, and videos to increase user fun and interactivity.
[0003] As user demands continue to increase, it is necessary to further improve the intelligence of virtual characters to enhance the user experience of wearable devices. Summary of the invention
[0004] The purpose of the present disclosure is to provide an interaction method and apparatus, a wearable device, a server, and a storage medium, which can improve the user experience of the wearable device.
[0005] According to a first aspect of an embodiment of the present disclosure, there is provided an interaction method, which is applied to a first device and includes: In response to a first user instruction sent by a first user and directed to a first smart object, sending the first user instruction to a second device; The first user instruction is used to instruct the second device to request reply information from the AI model based on the first user instruction and first historical information corresponding to the first smart object; The first historical information corresponding to the first smart object belongs to historical interaction information generated by the interaction between the first user and the first smart object; In response to the first response information returned by the second device, the first response information is returned to the first user; the first response information is information determined by the second device based on the reply information.
[0006] A second aspect of an embodiment of the present disclosure provides an interaction method, which is applied to a second device and includes: In response to a first user instruction sent by the first device, requesting reply information from the AI model based on the first user instruction and first historical information corresponding to the first smart object; the first user instruction is a first user instruction sent by the first user and directed to the first smart object; The first historical information corresponding to the first smart object belongs to historical interaction information generated by the interaction between the first user and the first smart object; First response information is determined based on the reply information, and the first response information is sent to the first device to instruct the first device to return the first response information to the first user.
[0007] According to a third aspect of an embodiment of the present disclosure, an interaction apparatus is provided, which is applied to a first device and includes: A first sending module, configured to respond to a first user instruction sent by a first user and directed to a first smart object, and send the first user instruction to a second device; The first user instruction is used to instruct the second device to request reply information from the AI model based on the first user instruction and first historical information corresponding to the first smart object; The first historical information corresponding to the first smart object belongs to historical interaction information generated by the interaction between the first user and the first smart object; The first returning module is used to return the first response information to the first user in response to the first response information returned by the second device; the first response information is information determined by the second device based on the reply information.
[0008] According to a fourth aspect of the embodiments of the present disclosure, an interaction device is provided, which is applied to a second device and includes: A first response module, configured to respond to a first user instruction sent by the first device and request reply information from the AI model based on the first user instruction and first historical information corresponding to the first smart object; the first user instruction is a first user instruction sent by the first user and directed to the first smart object; The first historical information corresponding to the first smart object belongs to historical interaction information generated by the interaction between the first user and the first smart object; The first processing module is configured to determine first response information based on the reply information, and send the first response information to the first device to instruct the first device to return the first response information to the first user.
[0009] According to a fifth aspect of an embodiment of the present disclosure, a wearable device is provided, comprising a first memory, a first processor, and a computer program stored in the first memory and running on the first processor, wherein the first processor implements the steps of the interaction method of the first aspect when executing the computer program.
[0010] In a sixth aspect of an embodiment of the present disclosure, a server is provided, comprising a second memory, a second processor, and a computer program stored in the second memory and running on the second processor, wherein the second processor implements the steps of the interaction method of the second aspect when executing the computer program.
[0011] According to a seventh aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the interaction method of the first aspect are implemented.
[0012] According to an eighth aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the interaction method of the second aspect are implemented.
[0013] The interactive method and apparatus, wearable device, server, and storage medium provided by the embodiments of the present disclosure have the following beneficial effects: In the embodiment of the present disclosure, the first device is a wearable device, the second device may be a server, and a plurality of smart objects are arranged in the first device. When the first user issues a first user instruction to the first smart object, the first device may send the first user instruction to the second device, the second device may call the AI model to respond to the first user instruction, and determine the first response information based on the reply information of the AI model, and send the first response information to the first device. The first device outputs the first response information to the first user to complete the response to the first user instruction.
[0014] Among them, when the second device requests reply information from the AI model, the request information includes historical interaction information generated by the interaction between the first user and the first smart object. The AI model can generate reply information based on the historical interaction information to make the reply information more in line with the needs of the first user.
[0015] Through the interaction between the first device, the second device and the AI model, the disclosed embodiment can fully utilize the semantic understanding ability of the AI model, improve the intelligence level of the smart object, and thus improve the user experience of the first device. BRIEF DESCRIPTION OF THE DRAWINGS In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0016] Figure 1 A schematic diagram of an application scenario of an interaction method provided by an embodiment of the present disclosure; Figure 2 A flowchart of an interactive method provided by an embodiment of the present disclosure; Figure 3 A flowchart of another interactive method provided by an embodiment of the present disclosure; Figure 4 A structural block diagram of an interactive device provided by an embodiment of the present disclosure; Figure 5 A structural block diagram of another interactive device provided by an embodiment of the present disclosure; Figure 6 A schematic block diagram of a wearable device provided in one embodiment of the present disclosure; Figure 7 A schematic block diagram of a server provided in accordance with an embodiment of the present disclosure. DETAILED DESCRIPTION
[0017] In the following description, specific details such as specific system structures and technologies are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present disclosure. However, it should be clear to those skilled in the art that the present disclosure may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obstructing the description of the present disclosure with unnecessary details.
[0018] In order to make the purpose, technical solutions and advantages of the present disclosure more clear, specific embodiments will be described below in conjunction with the accompanying drawings.
[0019] Please refer to Figure 1 , Figure 1 This is a schematic diagram of an application scenario of an interactive method provided in an embodiment of the present disclosure. The method of this embodiment can be implemented by an application in a wearable device 120. The wearable device 120 can be smart glasses or a smart bracelet, etc. The wearable device 120 receives a user instruction issued by a user 110. At the same time, the wearable device 120 can communicate with the server 130 through an app in a smart terminal. The smart terminal can be a mobile phone or a tablet computer, etc. After receiving the user instruction issued by the user 110, the wearable device 120 can forward the user instruction to the server 130, and the server 130 calls the AI model 140 to assist in responding to the user instruction.
[0020] in, Figure 1 The wearable device 120 and the smart terminal are only some of the devices cited as examples. In the present disclosure, the terminal is not limited to the wearable device 120 and the smart terminal. Figure 1The server 140 mentioned above can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, vehicle-road collaboration, content delivery networks (CDNs), and big data and artificial intelligence platforms.
[0021] Please refer to Figure 2 , Figure 2 A flowchart of an interaction method provided in an embodiment of the present disclosure is provided. The method is applied to a first device and includes: S201: In response to a first user instruction sent by a first user and directed to a first smart object, the first user instruction is sent to a second device.
[0022] The first user instruction is used to instruct the second device to request reply information from the AI model based on the first user instruction and the first historical information corresponding to the first smart object.
[0023] The first historical information corresponding to the first smart object belongs to historical interaction information generated by the first user interacting with the first smart object.
[0024] In this embodiment, the first device is a wearable device, the second device may be a server, and multiple smart objects are arranged in the first device. When the first user issues a first user instruction to a smart object (for example, the first smart object), the first device may send the first user instruction to the second device.
[0025] Among them, the communication between the first device and the second device can be achieved by setting a wifi / 4G communication module in the first device, or by installing an app on a smart terminal such as a mobile phone or a tablet computer, and realizing the communication between the first device and the smart terminal through the app. At the same time, the smart terminal can communicate with the second device, thereby realizing the communication between the first device and the second device.
[0026] After the first device sends the first user instruction to the second device, the second device can call the AI model to respond to the first user instruction. The AI model can process and understand natural language text to accurately understand the first user's instruction. At the same time, the AI model can mine valuable information and knowledge from massive data to generate more accurate reply information.
[0027] Preferably, when the second device requests reply information from the AI model, the request information includes historical interaction information generated by the interaction between the first user and the first intelligent object. The AI model can generate reply information based on the historical interaction information to make the reply information more in line with the needs of the first user.
[0028] S202: In response to the first response information returned by the second device, return the first response information to the first user; the first response information is information determined by the second device based on the reply information.
[0029] In this embodiment, the second device can determine the first response information based on the reply information of the AI model, send the first response information to the first device, and the first device outputs the first response information to the first user to complete the response to the first user's instruction.
[0030] From the above, it can be concluded that the embodiments of the present disclosure can fully utilize the semantic understanding ability of the AI model through the interaction between the first device, the second device and the AI model, improve the intelligence level of the smart object, and thus improve the user experience of the first device.
[0031] In one embodiment of the present disclosure, the first user instruction is specifically used to instruct: The second device generates a first prompt word corresponding to the first user based on the first historical information; and requests reply information from the AI model based on the first prompt word and the first user instruction.
[0032] In this embodiment, after the second device receives the first user instruction, it can generate a first prompt word corresponding to the first user based on the first historical information, input the first prompt word and the first user instruction into the AI model, and the AI model comprehensively considers the first prompt word and the first user instruction to generate reply information.
[0033] The first historical information is usually historical conversation information. The second device can first use natural language processing technologies such as term frequency-inverse document frequency (TF-IDF) and TextRank to extract key words or phrases from the first historical information as keywords; based on the above keywords, use topic models such as latent Dirichlet allocation (LDA) to discover potential topics in historical conversations, and generate relevant third prompt words based on topic information. Then, considering the context of the conversation, extract words closely related to the first user's instructions as fourth prompt words; judge the emotional tendency in the conversation, such as positive, negative or neutral, and generate corresponding fifth prompt words according to emotional characteristics; identify patterns that often appear in historical conversations, such as specific question types, answer structures, etc., and generate sixth prompt words based on these patterns. Finally, the third prompt word, fourth prompt word, fifth prompt word and sixth prompt word are integrated to obtain the first prompt word.
[0034] Similarly, for the first user instruction, the second device can also use the same method to generate a prompt word corresponding to the first user instruction, and request reply information from the AI model based on the first prompt word corresponding to the first historical information and the prompt word corresponding to the first user instruction.
[0035] In this embodiment, a corresponding first prompt word is generated based on the first historical information, and reply information is requested from the AI model based on the first prompt word. This allows the AI model to better understand user needs and generate reply information that is more in line with the first user's expectations.
[0036] In one embodiment of the present disclosure, the first user instruction is specifically used to instruct: The second device generates a first prompt word corresponding to the first user based on the first historical information, generates a second prompt word corresponding to the first smart object based on the object description corresponding to the first smart object, and requests reply information from the AI model based on the first prompt word, the second prompt word and the first user instruction.
[0037] The first response information is a reply information.
[0038] In this embodiment, there may be multiple smart objects set in the first device, and the object description of the smart object may include character background information, personality, work and life situation, values, knowledge field, etc. Multiple smart objects have different object descriptions (or role settings) and can be personalized according to user preferences and needs. By creating multiple smart objects, a richer and more diverse user experience can be provided. When a user issues a first user instruction to a first smart object, in addition to generating a corresponding first prompt word based on the first historical information, a second prompt word can also be generated based on the object description of the first smart object, wherein the first prompt word can represent the habits, interests, preferences and other characteristics of the first user, and the second prompt word can represent the character background information, personality, work and life situation, values and knowledge areas of the smart object. Therefore, requesting reply information from the AI model based on the first prompt word, the second prompt word and the first user instruction can make the reply information more in line with the role setting of the first smart object.
[0039] On this basis, after receiving the reply information, the second device directly sends the reply information as the first response information to the first device, and the first device returns the first response information to the first user. The above interaction process enables the user to experience a more vivid and interesting experience when interacting with the first device.
[0040] In one embodiment of the present disclosure, the first response information is information obtained by the second device converting the reply information based on the object description corresponding to the first smart object.
[0041] In this embodiment, after the first device sends the first user instruction to the second device, the second device may also request reply information from the AI model based only on the first prompt word and the first user instruction. The AI model does not need to consider the difference between smart objects, and only generates reply information based on the characteristics of the first user and the first user instruction. After the reply information is sent to the second device, the second device may convert the reply information based on the object description (such as timbre, expression habits, etc.) corresponding to the first smart object to obtain the first response information. For example, the object description corresponding to the first smart object is a hero. When the second device receives the reply information of the AI model, it can convert the reply information based on the timbre and expression habits of the hero to obtain the first response information. After the first response information is sent to the first device, the first device can broadcast the first response information to the first user in the timbre and expression habits of the hero.
[0042] In this embodiment, the generation and conversion of the reply information are performed separately, the generation process is performed by the AI model, and the conversion process is performed by the server. The AI model can only focus on the first user and the first user's instruction itself, which is conducive to reducing the calculation amount of the AI model. The conversion method for each smart object can be pre-stored in the second device, which is conducive to improving the conversion efficiency of the reply information, thereby increasing the generation speed of the first response information and realizing a rapid response to the first user's instruction.
[0043] In one embodiment of the present disclosure, the historical interaction information generated by the interaction between the first user and the first smart object further includes: second historical information.
[0044] The first historical information is the historical interaction information within a first time period before the current moment; the second historical information is the historical interaction information within a second time period before the current moment; and the second time period is greater than the first time period.
[0045] The second history information is used to instruct the second device to update the object description and / or the first prompt word based on the second history information.
[0046] In this embodiment, the first historical information is equivalent to short-term conversation history information, and the second historical information is equivalent to medium-term conversation history information. By combining the first historical information and the second historical information, more information about the first user's habits, interests, preferences, etc. can be learned. Updating the first prompt word based on the second historical information is conducive to obtaining a more accurate first prompt word.
[0047] In addition, the second historical information can also be used to learn more about the first smart object's background information, personality, work and life situation, values, and knowledge fields. As the first user's interaction needs continue to change, the second historical information will change accordingly, and the object description and first prompt word corresponding to the second historical information will also change. The object description of the first smart object is updated based on the second historical information, so that the first smart object is continuously updated as the first user's interaction needs change, thereby improving the user's interaction experience.
[0048] For example, if in the past period of time (one month), the first user often asks the first smart object for knowledge about health management, and the second historical record between the first user and the first smart object contains a lot of knowledge about health management, based on the second historical record, a description of health management can be added to the first prompt word corresponding to the first user, and a description of health management can also be added to the object description of the first smart object. For another example, in the recent period of time (one month), the first user often asks the first smart object for knowledge about financial management, and the second historical record between the first user and the first smart object contains a lot of knowledge about financial management, based on the second historical record, a description of financial management can be added to the first prompt word corresponding to the first user, and a description of financial management can also be added to the object description of the first smart object.
[0049] The character background information, personality, work and life situation, and values of the first intelligent object are relatively fixed, and can serve as a long-term unchanging object description of the first intelligent object. The long-term unchanging object description forms a multi-level memory mechanism with the first historical information and the second historical information, which can ensure the continuity of the dialogue and the stability of the intelligent object, and provide a more natural and smooth interactive experience.
[0050] In addition, in order to reduce the storage burden, the second historical information can be summarized and organized, and the summarized second historical information can be stored. The process of summarization and organization can be described in detail as follows: using natural language processing technology to perform lexical analysis, syntactic analysis and semantic analysis on the text in the second historical information to understand the meaning of each sentence; extracting the main topics in the second historical information through algorithms such as Dirichlet allocation LDA or deep learning-based methods; identifying and extracting important information in the second historical information, such as time, place, people, events, etc.; classifying the second historical information into different categories according to its theme and content, such as work, life, entertainment, etc.; integrating the extracted key information and classification results to form a preliminary inductive framework; based on the integrated information, using a generative model to generate a concise and clear summary to summarize the main content of the second historical information.
[0051] In one embodiment of the present disclosure, before sending the first user instruction to the second device, the interaction method further includes: The intent information of the first user instruction is identified, the format of the identified intent information is converted, and the first user instruction is updated to the intent information after the format conversion.
[0052] In this embodiment, the intent information of the first user instruction can be identified by extracting keywords. On the basis of obtaining the intent information, the format of the intent information can be converted through a pre-set template. The pre-set template can be in the form of a table. The fields in the table can include knowledge fields, specific processing matters, explicitly excluded options, etc. The corresponding intent information is filled into the pre-set template, and then the template is sent to the second device as the first user instruction. The second device can supplement the template based on the first historical information and the second historical information. The supplementation process can be to refine the content corresponding to each field, or to add new fields, and finally obtain a complete template. Sending the complete template as a prompt word to the AI model can provide a clear direction and focus for the AI model, which is conducive to the AI model outputting reply information that meets user expectations.
[0053] For example, the first user instruction is: Please recommend some healthy recipes to me. Based on the first user instruction, the first device extracts the keywords "health" and "recipes", fills in the pre-set template according to the keywords, and obtains that the knowledge field corresponding to the template is the health field, and the specific processing item is to query recipes. The first device sends the template to the second device, and the second device obtains that the first user does not like to eat celery based on the first historical information and the second historical information. The second device fills celery into the explicitly excluded options of the template accordingly, and the second device sends the template to the AI model, and the AI model can recommend healthy recipes that do not contain celery.
[0054] In one embodiment of the present disclosure, the interaction method further includes: In response to a second user instruction, a second smart object is generated; the second user instruction includes object description information of the second smart object.
[0055] In this embodiment, the user can create a new smart object through the app of the smart terminal and set the object description according to his or her needs.
[0056] Corresponding to the interaction method of the above embodiment, Figure 3 A flowchart of another interaction method provided in an embodiment of the present disclosure, the method is applied to a second device, including: S301: In response to a first user instruction sent by a first device, request reply information from the AI model based on the first user instruction and first historical information corresponding to the first smart object; the first user instruction is a first user instruction sent by a first user for the first smart object.
[0057] The first historical information corresponding to the first smart object belongs to historical interaction information generated by the first user interacting with the first smart object.
[0058] In this embodiment, when the second device receives the first user instruction, the AI model can be called to respond to the first user instruction. The AI model can process and understand natural language text to achieve accurate understanding of the first user instruction. At the same time, the AI model can mine valuable information and knowledge from massive data to generate more accurate reply information.
[0059] Preferably, when the second device requests reply information from the AI model, the request information includes historical interaction information generated by the interaction between the first user and the first intelligent object. The AI model can generate reply information based on the historical interaction information to make the reply information more in line with the needs of the first user.
[0060] S302: Determine first response information based on the reply information, and send the first response information to the first device to instruct the first device to return the first response information to the first user.
[0061] In this embodiment, the second device can determine the first response information based on the reply information of the AI model, send the first response information to the first device, and the first device outputs the first response information to the first user to complete the response to the first user's instruction.
[0062] From the above, it can be concluded that the embodiments of the present disclosure can fully utilize the semantic understanding ability of the AI model through the interaction between the first device, the second device and the AI model, improve the intelligence level of the smart object, and thus improve the user experience of the first device.
[0063] In one embodiment of the present disclosure, requesting reply information from the AI model based on the first user instruction and the first historical information corresponding to the first smart object includes: The second device generates a first prompt word corresponding to the first user based on the first historical information; and requests reply information from the AI model based on the first prompt word and the first user instruction.
[0064] In this embodiment, after the second device receives the first user's instruction, it can generate a first prompt word corresponding to the first user based on the first historical information, input the first prompt word and the first user's instruction into the AI model, and the AI model comprehensively considers the first prompt word and the first user's instruction to generate reply information, so that the AI model can better understand user needs and the generated reply information is more in line with the first user's expectations.
[0065] In one embodiment of the present disclosure, requesting reply information from the AI model based on the first user instruction and the first historical information corresponding to the first smart object includes: The second device generates a first prompt word corresponding to the first user based on the first historical information, generates a second prompt word corresponding to the first smart object based on the object description corresponding to the first smart object, and requests reply information from the AI model based on the first prompt word, the second prompt word and the first user instruction.
[0066] Determining first response information based on the reply information includes: The reply information is determined as the first response information.
[0067] In this embodiment, there may be multiple smart objects set in the first device, and the object description of the smart object may include character background information, personality, work and life situation, values, knowledge field, etc. Multiple smart objects have different object descriptions (or role settings) and can be personalized according to user preferences and needs. By creating multiple smart objects, a richer and more diverse user experience can be provided.
[0068] When a user issues a first user instruction to a first smart object, in addition to generating a corresponding first prompt word based on the first historical information, a second prompt word can also be generated based on the object description of the first smart object, wherein the first prompt word can represent the habits, interests, preferences and other characteristics of the first user, and the second prompt word can represent the character background information, personality, work and life situation, values and knowledge areas of the smart object. Therefore, requesting reply information from the AI model based on the first prompt word, the second prompt word and the first user instruction can make the reply information more in line with the role setting of the first smart object.
[0069] On this basis, after receiving the reply information, the second device directly sends the reply information as the first response information to the first device, and the first device returns the first response information to the first user. The above interaction process enables the user to experience a more vivid and interesting experience when interacting with the first device.
[0070] In one embodiment of the present disclosure, determining first response information based on reply information includes: The reply information is converted based on the object description corresponding to the first smart object to obtain the first response information.
[0071] In this embodiment, when receiving the first user instruction sent by the first device, the second device can also request reply information from the AI model based only on the first prompt word and the first user instruction. The AI model does not need to consider the difference of smart objects, and can generate reply information only according to the characteristics of the first user and the first user instruction. After the reply information is sent to the second device, the second device can convert the reply information based on the object description corresponding to the first smart object (such as timbre, expression habits, etc.) to obtain the first response information.
[0072] In this embodiment, the generation and conversion of the reply information are performed separately, the generation process is performed by the AI model, and the conversion process is performed by the server. The AI model can only focus on the first user and the first user's instruction itself, which is conducive to reducing the calculation amount of the AI model. The conversion method for each smart object can be pre-stored in the second device, which is conducive to improving the conversion efficiency of the reply information, thereby increasing the generation speed of the first response information and realizing a rapid response to the first user's instruction.
[0073] In one embodiment of the present disclosure, the historical interaction information generated by the interaction between the first user and the first smart object further includes: second historical information; The first historical information is the historical interaction information within a first time period before the current moment; the second historical information is the historical interaction information within a second time period before the current moment; and the second time period is greater than the first time period.
[0074] Interaction methods also include: The object description and / or the first prompt word are updated based on the second historical information.
[0075] In this embodiment, the first historical information is equivalent to short-term conversation history information, and the second historical information is equivalent to medium-term conversation history information. By combining the first historical information and the second historical information, more information about the first user's habits, interests, preferences, etc. can be learned. Updating the first prompt word based on the second historical information is conducive to obtaining a more accurate first prompt word.
[0076] In addition, the second historical information can also be used to learn more about the first smart object's background information, personality, work and life situation, values, and knowledge fields. As the first user's interaction needs continue to change, the second historical information will change accordingly, and the object description and first prompt word corresponding to the second historical information will also change. The object description of the first smart object is updated based on the second historical information, so that the first smart object is continuously updated as the first user's interaction needs change, thereby improving the user's interaction experience.
[0077] The character background information, personality, work and life situation, and values of the first intelligent object are relatively fixed, and can serve as a long-term unchanging object description of the first intelligent object. The long-term unchanging object description forms a multi-level memory mechanism with the first historical information and the second historical information, which can ensure the continuity of the dialogue and the stability of the intelligent object, and provide a more natural and smooth interactive experience.
[0078] In one embodiment of the present disclosure, the first user instruction is an updated first user instruction, and the updating process is: The first device identifies the intent information of the first user instruction, converts the format of the identified intent information, and updates the first user instruction to the intent information after the format conversion.
[0079] In this embodiment, the intent information of the first user instruction can be identified by extracting keywords. On the basis of obtaining the intent information, the format of the intent information can be converted through a pre-set template. The pre-set template can be in the form of a table. The fields in the table can include knowledge fields, specific processing matters, explicitly excluded options, etc. The corresponding intent information is filled into the pre-set template, and then the template is sent to the second device as the first user instruction. The second device can supplement the template based on the first historical information and the second historical information. The supplementation process can be to refine the content corresponding to each field, or to add new fields, and finally obtain a complete template. Sending the complete template as a prompt word to the AI model can provide a clear direction and focus for the AI model, which is conducive to the AI model outputting reply information that meets user expectations.
[0080] Corresponding to the interaction method of the above embodiment, Figure 4 This is a structural block diagram of an interactive device provided by an embodiment of the present disclosure. For ease of explanation, only the parts related to the embodiment of the present disclosure are shown. Figure 4 The interaction device 40 is applied to the first device, and includes: a first sending module 41 and a first returning module 42. The first sending module 41 is used to send the first user instruction to the second device in response to the first user instruction sent by the first user and directed to the first smart object.
[0081] The first user instruction is used to instruct the second device to request reply information from the AI model based on the first user instruction and the first historical information corresponding to the first smart object.
[0082] The first historical information corresponding to the first smart object belongs to historical interaction information generated by the first user interacting with the first smart object.
[0083] The first returning module 42 is used to return the first response information to the first user in response to the first response information returned by the second device; the first response information is information determined by the second device based on the reply information. In one embodiment of the present disclosure, the first user instruction is specifically used to instruct: The second device generates a first prompt word corresponding to the first user based on the first historical information; and requests reply information from the AI model based on the first prompt word and the first user instruction.
[0084] In one embodiment of the present disclosure, the first user instruction is specifically used to instruct: The second device generates a first prompt word corresponding to the first user based on the first historical information, generates a second prompt word corresponding to the first smart object based on the object description corresponding to the first smart object, and requests reply information from the AI model based on the first prompt word, the second prompt word and the first user instruction.
[0085] The first response information is a reply information.
[0086] In one embodiment of the present disclosure, the first response information is information obtained by the second device converting the reply information based on the object description corresponding to the first smart object.
[0087] In one embodiment of the present disclosure, the historical interaction information generated by the interaction between the first user and the first smart object further includes: second historical information.
[0088] The first historical information is the historical interaction information within a first time period before the current moment; the second historical information is the historical interaction information within a second time period before the current moment; and the second time period is greater than the first time period.
[0089] The second history information is used to instruct the second device to update the object description and / or the first prompt word based on the second history information.
[0090] In one embodiment of the present disclosure, the first sending module 41 is further configured to: The intent information of the first user instruction is identified, the format of the identified intent information is converted, and the first user instruction is updated to the intent information after the format conversion.
[0091] Corresponding to the interaction method of the above embodiment, Figure 5 This is a structural block diagram of an interactive device provided by an embodiment of the present disclosure. For ease of explanation, only the parts related to the embodiment of the present disclosure are shown. Figure 5 The interaction device 50 is applied to the second device, and includes: a first response module 51 and a first processing module 52.
[0092] Among them, the first response module 51 is used to respond to the first user instruction sent by the first device, and request reply information from the AI model based on the first user instruction and the first historical information corresponding to the first smart object; the first user instruction is the first user instruction sent by the first user for the first smart object.
[0093] The first historical information corresponding to the first smart object belongs to historical interaction information generated by the first user interacting with the first smart object.
[0094] The first processing module 52 is configured to determine first response information based on the reply information, and send the first response information to the first device to instruct the first device to return the first response information to the first user.
[0095] In one embodiment of the present disclosure, the first response module 51 is specifically used to: A first prompt word corresponding to the first user is generated based on the first historical information; and reply information is requested from the AI model based on the first prompt word and the first user instruction.
[0096] In one embodiment of the present disclosure, the first response module 51 is specifically used to: A first prompt word corresponding to the first user is generated based on the first historical information, a second prompt word corresponding to the first smart object is generated based on the object description corresponding to the first smart object, and reply information is requested from the AI model based on the first prompt word, the second prompt word and the first user instruction.
[0097] Determining first response information based on the reply information includes: The reply information is determined as the first response information.
[0098] In one embodiment of the present disclosure, the first response module 51 is specifically used to: The reply information is converted based on the object description corresponding to the first smart object to obtain the first response information.
[0099] In one embodiment of the present disclosure, the historical interaction information generated by the interaction between the first user and the first smart object further includes: second historical information; The first historical information is the historical interaction information within a first time period before the current moment; the second historical information is the historical interaction information within a second time period before the current moment; and the second time period is greater than the first time period.
[0100] The first response module 51 is specifically used for: The object description and / or the first prompt word are updated based on the second historical information.
[0101] In one embodiment of the present disclosure, the first user instruction is an updated first user instruction, and the updating process is: The first device identifies the intent information of the first user instruction, converts the format of the identified intent information, and updates the first user instruction to the intent information after the format conversion.
[0102] See also Figure 6 , Figure 6 A schematic block diagram of a wearable device provided by an embodiment of the present disclosure. Figure 6 The wearable device 600 in the embodiment shown may include: one or more first processors 601, one or more input devices 602, one or more output devices 603 and one or more memories 604. The first processor 601, the first input device 602, the output device 603 and the memory 604 communicate with each other through a first communication bus 605. The memory 604 is used to store computer programs, and the computer programs include program instructions. The first processor 601 is used to execute the program instructions stored in the memory 604. Among them, the first processor 601 is configured to call the program instructions to execute the functions of each module / unit in the above-mentioned device embodiments, such as Figure 4 The functions of the modules 41 to 42 are shown.
[0103] It should be understood that in the embodiment of the present disclosure, the so-called first processor 601 may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.
[0104] The first input device 602 may include a touch panel, a fingerprint collection sensor (for collecting fingerprint information of the user and fingerprint direction information), a microphone, etc. The first output device 603 may include a display (LCD, etc.), a speaker, etc.
[0105] The memory 604 may include a read-only memory and a random access memory, and provide instructions and data to the first processor 601. A portion of the first memory 604 may also include a non-volatile random access memory. For example, the first memory 604 may also store information about the device type.
[0106] In a specific implementation, the first processor 601, the first input device 602, and the first output device 603 described in the embodiments of the present disclosure can execute the implementation methods described in the first and second embodiments of the interaction method provided in the embodiments of the present disclosure, and can also execute the implementation methods of the wearable device described in the embodiments of the present disclosure, which will not be repeated here.
[0107] See also Figure 7 , Figure 7 A schematic block diagram of a server provided in an embodiment of the present disclosure. Figure 7 The server 700 in the embodiment shown may include: one or more second processors 701, one or more second input devices 702, one or more second output devices 703 and one or more second memories 704. The second processor 701, the second input device 702, the second output device 703 and the second memory 704 communicate with each other via a second communication bus 705. The second memory 704 is used to store computer programs, and the computer programs include program instructions. The second processor 701 is used to execute the program instructions stored in the second memory 704. The second processor 701 is configured to call the program instructions to execute the functions of each module / unit in the above-mentioned device embodiments, such as Figure 5 The functions of modules 51 to 52 are shown.
[0108] It should be understood that in the embodiment of the present disclosure, the so-called second processor 701 may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.
[0109] The second input device 702 may include a touch panel, a fingerprint collection sensor (for collecting fingerprint information of the user and fingerprint direction information), a microphone, etc. The second output device 703 may include a display (LCD, etc.), a speaker, etc.
[0110] The second memory 704 may include a read-only memory and a random access memory, and provide instructions and data to the second processor 701. A portion of the second memory 704 may also include a non-volatile random access memory. For example, the second memory 704 may also store information about the device type.
[0111] In a specific implementation, the second processor 701, the second input device 702, and the second output device 703 described in the embodiments of the present disclosure can execute the implementation methods described in the first and second embodiments of the interaction method provided in the embodiments of the present disclosure, and can also execute the implementation methods of the wearable device described in the embodiments of the present disclosure, which will not be repeated here.
[0112] In another embodiment of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by the processor, all or part of the processes in the above-mentioned embodiment method are implemented, and the computer program can also be completed by instructing the relevant hardware through the computer program. The computer program can be stored in a computer-readable storage medium, and when the computer program is executed by the processor, the steps of each of the above-mentioned method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc.
[0113] The computer-readable storage medium may be an internal storage unit of the wearable device of any of the foregoing embodiments, such as a hard disk or memory of the wearable device. The computer-readable storage medium may also be an external storage device of the wearable device, such as a plug-in hard disk equipped on the wearable device, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. Furthermore, the computer-readable storage medium may also include both an internal storage unit of the wearable device and an external storage device. The computer-readable storage medium is used to store computer programs and other programs and data required by the wearable device. The computer-readable storage medium may also be used to temporarily store data that has been output or is to be output.
[0114] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this disclosure.
[0115] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the wearable device and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.
[0116] In the several embodiments provided in the present application, it should be understood that the disclosed wearable devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces or units, or it can be an electrical, mechanical or other form of connection.
[0117] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present disclosure.
[0118] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0119] The above are only specific embodiments of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present disclosure, and these modifications or replacements should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.
Claims
1. An interaction method, applied to a first device, characterized in that: include: In response to a first user instruction sent by a first user and directed to a first smart object, sending the first user instruction to a second device; The first user instruction is used to instruct the second device to request reply information from the AI model based on the first user instruction and first historical information corresponding to the first smart object; The first historical information corresponding to the first smart object belongs to historical interaction information generated by the interaction between the first user and the first smart object; In response to the first response information returned by the second device, returning the first response information to the first user; The first response information is information determined by the second device based on the reply information.
2. The interactive method according to claim 1, characterized in that: The first user instruction is specifically used to instruct: The second device generates a first prompt word corresponding to the first user based on the first historical information; and requests reply information from the AI model based on the first prompt word and the first user instruction.
3. The interactive method according to claim 1, characterized in that: The first user instruction is specifically used to instruct: The second device generates a first prompt word corresponding to the first user based on the first historical information, generates a second prompt word corresponding to the first smart object based on the object description corresponding to the first smart object, and requests reply information from the AI model based on the first prompt word, the second prompt word and the first user instruction; The first response information is the reply information.
4. The interactive method according to claim 2, characterized in that: The first response information is information obtained by the second device converting the reply information based on the object description corresponding to the first smart object.
5. The interactive method according to claim 3 or 4, characterized in that: The historical interaction information generated by the interaction between the first user and the first smart object further includes: second historical information; The first historical information is the historical interaction information within a first time period before the current moment; the second historical information is the historical interaction information within a second time period before the current moment; the second time period is greater than the first time period; The second historical information is used to instruct the second device to update the object description and / or the first prompt word based on the second historical information.
6. The interactive method according to claim 1, characterized in that: Before sending the first user instruction to the second device, the method further includes: Identify the intent information of the first user instruction, convert the format of the identified intent information, and update the first user instruction to the intent information after the format conversion.
7. An interaction method, applied to a second device, characterized in that: include: In response to a first user instruction sent by the first device, requesting reply information from the AI model based on the first user instruction and first historical information corresponding to the first smart object; the first user instruction is a first user instruction sent by the first user and directed to the first smart object; The first historical information corresponding to the first smart object belongs to historical interaction information generated by the interaction between the first user and the first smart object; First response information is determined based on the reply information, and the first response information is sent to the first device to instruct the first device to return the first response information to the first user.
8. The interactive method according to claim 7, characterized in that: The requesting reply information from the AI model based on the first user instruction and the first historical information corresponding to the first smart object includes: A first prompt word corresponding to the first user is generated based on the first historical information; and reply information is requested from the AI model based on the first prompt word and the first user instruction.
9. The interactive method according to claim 7, characterized in that: The requesting reply information from the AI model based on the first user instruction and the first historical information corresponding to the first smart object includes: Generate a first prompt word corresponding to the first user based on the first historical information, generate a second prompt word corresponding to the first smart object based on the object description corresponding to the first smart object, and request reply information from the AI model based on the first prompt word, the second prompt word and the first user instruction; The determining the first response information based on the reply information includes: The reply information is determined as the first response information.
10. The interactive method according to claim 8, characterized in that: The determining the first response information based on the reply information includes: The first response information is obtained by converting the reply information based on the object description corresponding to the first smart object.
11. The interactive method according to claim 9 or 10, characterized in that: The historical interaction information generated by the interaction between the first user and the first smart object further includes: second historical information; The first historical information is the historical interaction information within a first time period before the current moment; the second historical information is the historical interaction information within a second time period before the current moment; the second time period is greater than the first time period; The interaction method further comprises: The object description and / or the first prompt word are updated based on the second historical information.
12. The interactive method according to claim 7, characterized in that: The first user instruction is an updated first user instruction, and the updating process is: The first device identifies the intent information of the first user instruction, converts the format of the identified intent information, and updates the first user instruction to the intent information after the format conversion.
13. An interactive device, applied to a first device, characterized in that: include: A first sending module, configured to respond to a first user instruction sent by a first user and directed to a first smart object, and send the first user instruction to a second device; The first user instruction is used to instruct the second device to request reply information from the AI model based on the first user instruction and first historical information corresponding to the first smart object; The first historical information corresponding to the first smart object belongs to historical interaction information generated by the interaction between the first user and the first smart object; A first returning module, configured to return the first response information to the first user in response to the first response information returned by the second device; The first response information is information determined by the second device based on the reply information.
14. An interactive device, applied to a second device, characterized in that: include: A first response module, configured to respond to a first user instruction sent by the first device and request reply information from the AI model based on the first user instruction and first historical information corresponding to the first smart object; the first user instruction is a first user instruction sent by the first user and directed to the first smart object; The first historical information corresponding to the first smart object belongs to historical interaction information generated by the interaction between the first user and the first smart object; The first processing module is configured to determine first response information based on the reply information, and send the first response information to the first device to instruct the first device to return the first response information to the first user.
15. A wearable device, comprising a first memory, a first processor, and a computer program stored in the first memory and running on the first processor, characterized in that: When the first processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.
16. A server comprising a second memory, a second processor, and a computer program stored in the second memory and running on the second processor, characterized in that: When the second processor executes the computer program, the steps of the method according to any one of claims 7 to 12 are implemented.
17. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the processor executes the steps of the method according to any one of claims 1 to 6, or executes the steps of the method according to any one of claims 7 to 12.
Citation Information
Patent Citations
Man-machine interaction method and device based on artificial intelligence
CN106503156A
Assistance in service provision
CN111800451A
Interaction method and system for intelligent man-machine conversation and electronic equipment
CN112948534A
NPC interaction method and device and storage medium
CN117122927A
Question and answer interaction method and device, electronic equipment and storage medium
CN117332067A
Cited By
Task execution program, task execution method, and information processing device
WO2026181311A1
