Method and device for constructing family metadata, method and device for generating control page

Abstract

The application discloses a family metadata construction method and device and a control page generation method and device. The method is applied to a terminal device and includes the following steps: obtaining a voice instruction sent by a target object and converting the voice instruction into a first text by using a voice recognition technology or obtaining a second text sent by the target object; analyzing the first text or the second text by using a large language model to obtain a plurality of first keywords in the first text or the second text, wherein the plurality of first keywords correspond to a plurality of spaces in a target area where the target object is located or a plurality of Internet of Things devices in the plurality of spaces in a one-to-one manner; and constructing family metadata according to the plurality of first keywords in the case that the plurality of first keywords match target keywords in a database of a cloud server successfully, wherein the family metadata is used to represent at least a positional relationship between the plurality of Internet of Things devices and the plurality of spaces.

Description

Technical Field

[0001] This application relates to the field of communications, and more specifically, to a method and apparatus for constructing home metadata, and a method and apparatus for generating control pages. Background Technology

[0002] In the development of smart home systems and related IoT technologies, the management and maintenance of home metadata has always been a key aspect of improving user experience and technical efficiency. Home metadata includes detailed information about physical spaces in the home environment (such as rooms), IoT devices (such as smart light bulbs, air conditioners, and security cameras), and the relationships between devices and spaces.

[0003] However, existing smart home systems typically rely on users manually inputting metadata such as home layout, room names, and device information. Users need to enter the name of each room and detailed information about all devices within it, including device ID, type, and brand. This is not only time-consuming but can also degrade the user experience due to the tedious input process. During manual input, users may enter incorrect data due to spelling errors or inaccurate memory of device information, affecting device control and system operation. Furthermore, different users may use different names for the same device, such as "table lamp," "bedside lamp," or "study lamp," and this inconsistency makes it difficult for the system to uniformly manage and identify devices.

[0004] Furthermore, in related technologies, a single speech recognition technology may fail to accurately understand the user's intent, resulting in a significant reduction in the accuracy of speech recognition. Even if speech recognition technology can convert speech into text, existing technologies may still be unable to understand the user's vague or complex control intentions when parsing text commands due to a lack of advanced natural language processing capabilities.

[0005] In the relevant technologies, manually entering family metadata is not only time-consuming and labor-intensive, but also prone to errors, and no effective solution has yet been proposed. Summary of the Invention

[0006] This application provides a method and apparatus for constructing family metadata, and a method and apparatus for generating control pages, to at least solve the problem in related technologies that manually inputting family metadata is not only time-consuming and laborious, but also prone to errors.

[0007] According to one embodiment of this application, a method for constructing family metadata is provided, applied to a terminal device, comprising: acquiring a voice command issued by a target object, and converting the voice command into first text using speech recognition technology, or acquiring second text sent by the target object; parsing the first text or the second text using a large language model to acquire multiple first keywords in the first text or the second text, wherein the multiple first keywords correspond one-to-one with multiple spaces in a target area where the target object is located or multiple IoT devices in the multiple spaces; and constructing family metadata based on the multiple first keywords when the multiple first keywords successfully match target keywords in a database of a cloud server, wherein the family metadata is at least used to characterize the location relationship between the multiple IoT devices and the multiple spaces.

[0008] In an exemplary embodiment, after constructing family metadata based on the plurality of first keywords, the method further includes: determining a plurality of second keywords corresponding to the plurality of spaces and a plurality of third keywords corresponding to a plurality of IoT devices in the plurality of spaces; generating a plurality of first controls corresponding to the plurality of second keywords and a plurality of second controls corresponding to the plurality of third keywords through front-end page generation rules; determining a second positional relationship between the plurality of first controls and the plurality of second controls based on a first positional relationship between the plurality of spaces and the plurality of IoT devices; and generating a first control page for the terminal device based on the plurality of first controls, the plurality of second controls, and the second positional relationship.

[0009] In an exemplary embodiment, after generating the first control page of the terminal device based on the plurality of first controls, the plurality of second controls, and the second positional relationship, the method further includes: receiving an adjustment instruction sent by the target object, and obtaining a fourth keyword corresponding to the adjustment instruction through an IoT interface, wherein the adjustment instruction is used to add or delete the fourth keyword in the family metadata; and adjusting the control corresponding to the fourth keyword on the first control page through the control interface of the front-end page according to the adjustment instruction, wherein the front-end page is used to display the first control page.

[0010] In an exemplary embodiment, before parsing the first text or the second text using a large language model to obtain multiple first keywords from the first text or the second text, the method further includes: changing the format of the third text to obtain a standard format third text, wherein the third text is the first text or the second text; performing noise filtering on the standard format third text to obtain a cleared third text; and extracting multiple fifth keywords from the cleared third text, wherein the multiple fifth keywords include at least one of the following: the multiple first keywords and a sixth keyword, wherein the sixth keyword is used to indicate the time or action corresponding to the third text.

[0011] In an exemplary embodiment, after constructing family metadata based on the plurality of first keywords, the method further includes: obtaining a scene control instruction issued by the target object from the first text or the second text, wherein the scene control instruction is used to control the target IoT device to perform a target action; determining a target space from the plurality of spaces based on the scene control instruction, and determining a second control page corresponding to the target space; and displaying the control instruction on the second control page.

[0012] In an exemplary embodiment, after constructing the family metadata based on the plurality of first keywords, the method further includes: obtaining the status of a plurality of IoT devices in the plurality of spaces through an IoT interface; if the status of a first IoT device among the plurality of IoT devices is that it is not in use, deleting the third keyword corresponding to the first IoT device from the family metadata; if a second IoT device is added to the plurality of spaces, updating the family metadata based on the seventh keyword corresponding to the second IoT device.

[0013] In an exemplary embodiment, after parsing the first text or the second text using a large language model to obtain multiple first keywords in the first text or the second text, the method further includes: if an eighth keyword among the multiple first keywords fails to match multiple keywords in the database of the cloud server, receiving an error message sent by the cloud server, wherein the error message includes the reason why the eighth keyword fails to match the multiple keywords.

[0014] According to another embodiment of the present application, a device for constructing family metadata is also provided, comprising: an acquisition module, configured to acquire a voice command issued by a target object and convert the voice command into first text using speech recognition technology, or acquire a second text sent by the target object; a parsing module, configured to parse the first text or the second text using a large language model to acquire a plurality of first keywords in the first text or the second text, wherein the plurality of first keywords correspond one-to-one with a plurality of spaces in a target area where the target object is located or a plurality of IoT devices in the plurality of spaces; and a construction module, configured to construct family metadata based on the plurality of first keywords when the plurality of first keywords successfully match target keywords in a database of a cloud server, wherein the family metadata is at least used to characterize the location relationship between the plurality of IoT devices and the plurality of spaces.

[0015] According to one embodiment of this application, a method for generating a control page is provided, applied to a terminal device, comprising: acquiring a voice command issued by a target object, and converting the voice command into first text using speech recognition technology, or acquiring second text sent by the target object; parsing the first text or the second text using a large language model to acquire multiple first keywords in the first text or the second text, wherein the multiple first keywords correspond one-to-one with multiple spaces in a target area where the target object is located or multiple IoT devices in the multiple spaces; if the multiple first keywords successfully match target keywords in a database of a cloud server, constructing family metadata based on the multiple first keywords, wherein the family metadata is at least used to characterize the positional relationship between the multiple IoT devices and the multiple spaces; generating a target control corresponding to the family metadata using front-end page generation rules, and generating a first control page based on the target control.

[0016] According to another embodiment of this application, a control page generation device is also provided, comprising: an acquisition module, configured to acquire a voice command issued by a target object and convert the voice command into first text using speech recognition technology, or acquire second text sent by the target object; a parsing module, configured to parse the first text or the second text using a large language model to acquire multiple first keywords in the first text or the second text, wherein the multiple first keywords correspond one-to-one with multiple spaces in a target area where the target object is located or multiple IoT devices in the multiple spaces; a construction module, configured to construct family metadata based on the multiple first keywords when the multiple first keywords successfully match target keywords in a database of a cloud server, wherein the family metadata is at least used to characterize the positional relationship between the multiple IoT devices and the multiple spaces; and a generation module, configured to generate a target control corresponding to the family metadata according to front-end page generation rules, and generate a first control page based on the target control.

[0017] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, wherein a computer program is stored in the computer program, and the computer program is configured to execute the above-described method at runtime.

[0018] According to another aspect of the embodiments of this application, an electronic device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the above-described method through the computer program.

[0019] In this embodiment, a voice command issued by the target object is obtained, and the voice command is converted into first text using speech recognition technology, or a second text sent by the target object is obtained. The first or second text is parsed using a large language model to obtain multiple first keywords, wherein each of the multiple first keywords corresponds one-to-one with multiple spaces or multiple IoT devices in the target area where the target object is located. If the multiple first keywords successfully match target keywords in a cloud server database, family metadata is constructed based on the multiple first keywords, wherein the family metadata at least represents the positional relationship between multiple IoT devices and multiple spaces. By adopting the above scheme, multiple first keywords in the voice command issued by the target object are accurately matched with target keywords in the database using a large language model, and family metadata is generated based on the multiple first keywords. This solves the problem in related technologies where manually inputting family metadata is time-consuming, labor-intensive, and prone to errors, effectively improving the efficiency of family metadata generation and achieving accurate recognition of the target object's ambiguous intent. Attached Figure Description

[0020] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0021] Figure 1 This is a hardware structure block diagram of a terminal device for a method of constructing family metadata according to an embodiment of this application;

[0022] Figure 2 This is a flowchart of a method for constructing family metadata according to an embodiment of this application;

[0023] Figure 3 This is a structural block diagram of a control page generation system according to an embodiment of this application;

[0024] Figure 4 This is a first schematic diagram of a method for constructing family metadata according to an embodiment of this application;

[0025] Figure 5 This is a second schematic diagram of a method for constructing family metadata according to an embodiment of this application;

[0026] Figure 6 This is a structural block diagram of a family metadata construction apparatus according to an embodiment of this application;

[0027] Figure 7 This is a flowchart of a method for generating a control page according to an embodiment of this application;

[0028] Figure 8 This is a structural block diagram of a control page generation device according to an embodiment of this application. Detailed Implementation

[0029] To enable those skilled in the art to better understand 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. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0030] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0031] The methods and embodiments provided in this application can be executed in a terminal device or a similar computing device. Taking running on a terminal device as an example, Figure 1 This is a hardware structure block diagram of a terminal device for a method of constructing family metadata according to an embodiment of this application. For example... Figure 1 As shown, the terminal device may include one or more ( Figure 1 Only one is shown in the diagram. A processor 102 (which may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.) and a memory 104 for storing data are also shown. In one exemplary embodiment, the terminal device may further include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the terminal device described above. For example, the terminal device may also include components that are more... Figure 1 The more or fewer components shown, or having the same Figure 1 Equivalent functions or ratios shown Figure 1 The functions shown have more different configurations.

[0032] The memory 104 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the method in this embodiment. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, thus implementing the above-described method. The memory 104 may include high-speed random access memory and non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to terminal devices via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0033] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the terminal device. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a Radio Frequency (RF) module used for wireless communication with the Internet.

[0034] This embodiment provides a method for constructing family metadata, applied to a terminal device. Figure 2 This is a flowchart of a method for constructing family metadata according to an embodiment of this application, the process including the following steps:

[0035] Step S202: Obtain the voice command issued by the target object, and convert the voice command into first text through voice recognition technology, or obtain the second text sent by the target object;

[0036] Step S204: Parse the first text or the second text using a large language model to obtain multiple first keywords in the first text or the second text, wherein the multiple first keywords correspond one-to-one with multiple spaces in the target area where the target object is located or multiple IoT devices in the multiple spaces;

[0037] Step S206: If the plurality of first keywords successfully match the target keywords in the database of the cloud server, construct family metadata based on the plurality of first keywords, wherein the family metadata is used at least to characterize the location relationship between the plurality of IoT devices and the plurality of spaces.

[0038] It's important to clarify that the aforementioned cloud server stores all user input information uploaded by terminal devices. This information is processed and transformed into structured data, including but not limited to user preferences, historical control commands, home layout information, device lists, and their status. This centralized data management helps to uniformly maintain and update home metadata, ensuring the consistency and accuracy of information across all devices and spaces.

[0039] When a user sends control commands via voice or text, the terminal device first performs preliminary speech recognition or text analysis, then uploads the extracted key information (i.e., multiple primary keywords) to the cloud server. The large language model on the cloud server further analyzes these keywords to understand the user's true intent and request details. For example, if a user says, "There is a ceiling light, two grid lights, and two recessed spotlights in the living room," the large language model will parse out a room named "living room" and information about five specific lighting products within that room. The large language model then matches the parsed results with target keywords in the database. If a match is successful, the aforementioned primary metadata is stored in the cloud server's database, and home metadata is constructed based on these primary keywords.

[0040] Through the above steps, the voice commands issued by the target object are obtained, and the voice commands are converted into first text using speech recognition technology, or second text sent by the target object is obtained. The first or second text is parsed using a large language model to obtain multiple first keywords, where each of the multiple first keywords corresponds one-to-one with multiple spaces or multiple IoT devices within the target area where the target object is located. If the multiple first keywords successfully match target keywords in the cloud server's database, family metadata is constructed based on the multiple first keywords. This family metadata at least represents the locational relationships between multiple IoT devices and multiple spaces. This method solves the problem in related technologies where manually inputting family metadata is not only time-consuming and labor-intensive but also prone to errors.

[0041] Optionally, suppose a user says to the terminal device: "My house has two bedrooms, one living room, one master bedroom, one secondary bedroom, one living room, and one bathroom. The living room has one ceiling light, two grille lights, and two recessed spotlights. The secondary bedroom has four downlights. The master bedroom has one ceiling light and four spotlights. The bathroom has one ceiling light." After receiving the user's voice command, the terminal device converts it into first text and parses it using a large language model to obtain multiple primary keywords, such as "master bedroom," "secondary bedroom," "living room," "bathroom," "ceiling light," "grout light," "spotlight," and "downlight." During parsing, the large language model not only identifies these primary keywords but also understands the relationships between them (e.g., which devices belong to which rooms, and how many devices are in each room), and extracts this information using a dedicated prompt or instruction set.

[0042] The aforementioned primary keywords are matched against target keywords in the database. A successful match indicates that the primary keywords exist in the database, where the target keywords are a pre-defined list of device and space names. Then, family metadata is constructed based on these primary keywords. For example: Living room: Ceiling light {quantity: 1}, Grille light {quantity: 2}, Spotlight {quantity: 2}; Second bedroom: Downlight {quantity: 4}; Master bedroom: Ceiling light {quantity: 1}, Spotlight {quantity: 4}; Bathroom: Ceiling light {quantity: 1}.

[0043] In an exemplary embodiment, after constructing family metadata based on the plurality of first keywords, the method further includes: determining a plurality of second keywords corresponding to the plurality of spaces and a plurality of third keywords corresponding to a plurality of IoT devices in the plurality of spaces; generating a plurality of first controls corresponding to the plurality of second keywords and a plurality of second controls corresponding to the plurality of third keywords through front-end page generation rules; determining a second positional relationship between the plurality of first controls and the plurality of second controls based on a first positional relationship between the plurality of spaces and the plurality of IoT devices; and generating a first control page for the terminal device based on the plurality of first controls, the plurality of second controls, and the second positional relationship.

[0044] Optionally, assuming multiple primary keywords are master bedroom, secondary bedroom, living room, bathroom, ceiling light, grille light, spotlight, and downlight, multiple secondary keywords are determined from these, and multiple third keywords are ceiling light, grille light, spotlight, and downlight. Multiple primary and secondary controls corresponding to these secondary and third keywords are rendered using front-end page generation rules. The positional relationships of these primary and secondary controls are determined based on the actual spatial layout. For example, if the living room has one ceiling light, two grille lights, and two recessed spotlights, then a first control page is generated based on the primary control corresponding to the living room and the secondary controls corresponding to the ceiling light, grille light, and spotlight.

[0045] In an exemplary embodiment, after generating the first control page of the terminal device based on the plurality of first controls, the plurality of second controls, and the second positional relationship, the method further includes: receiving an adjustment instruction sent by the target object, and obtaining a fourth keyword corresponding to the adjustment instruction through an IoT interface, wherein the adjustment instruction is used to add or delete the fourth keyword in the family metadata; and adjusting the control corresponding to the fourth keyword on the first control page through the control interface of the front-end page according to the adjustment instruction, wherein the front-end page is used to display the first control page.

[0046] Optionally, assuming a user tells the terminal device: "Use a crystal chandelier instead of a ceiling light in the living room," the large language model parses the adjustment instruction, determines the fourth keywords as "crystal chandelier" and "ceiling light," and identifies the living room as the space requiring adjustment. Assuming the unadjusted family metadata is: Living room: Ceiling light {quantity: 1}, Grille light {quantity: 2}, Spotlight {quantity: 2}, and the keyword "chandelier" exists in the database, the adjusted family metadata would be: Living room: Chandelier {quantity: 1}, Grille light {quantity: 2}, Spotlight {quantity: 2}. The adjustment process generates corresponding operation records. Based on these records, the control corresponding to the ceiling light is deleted, and the control corresponding to the chandelier is added, thus updating the first control page for the living room.

[0047] In an exemplary embodiment, before parsing the first text or the second text using a large language model to obtain multiple first keywords from the first text or the second text, the method further includes: changing the format of the third text to obtain a standard format third text, wherein the third text is the first text or the second text; performing noise filtering on the standard format third text to obtain a cleared third text; and extracting multiple fifth keywords from the cleared third text, wherein the multiple fifth keywords include at least one of the following: the multiple first keywords and a sixth keyword, wherein the sixth keyword is used to indicate the time or action corresponding to the third text.

[0048] Suppose a user enters the following text into a terminal device: "I need to turn off the living room lights at 9 PM." The terminal device first receives the raw text entered by the user, i.e., the third-party text. However, the actual user input may contain non-standard formats, such as multiple ways of writing date and time ("9 PM", "9pm", "21:00", etc.), or use non-technical terms (such as "I need"). Therefore, the terminal device needs to modify the format of the third-party text, converting all non-standard expressions into a standard format. For example, "9 PM" is uniformly converted to "21:00," which facilitates subsequent parsing and the execution of time-triggered operations.

[0049] Meanwhile, the text entered by the user may contain noise information unrelated to the control operation, such as the user's emotional expressions, interjections, and redundant descriptions. Noise filtering can remove or ignore these words that are not directly related to the smart home control task, thus making the text clearer and reducing interference during the parsing process. For example, filtering out expressions like "I need" and retaining only the description directly related to control, such as "turn off the living room lights."

[0050] After formatting and noise filtering the text, the terminal device extracts several fifth keywords from the cleared third text. The fifth keywords include the first keywords (i.e., information related to a specific device or space, such as "living room" or "light") and the sixth keywords (words indicating time or action, such as "9 p.m." or "off").

[0051] In an exemplary embodiment, after constructing family metadata based on the plurality of first keywords, the method further includes: obtaining a scene control instruction issued by the target object from the first text or the second text, wherein the scene control instruction is used to control the target IoT device to perform a target action; determining a target space from the plurality of spaces based on the scene control instruction, and determining a second control page corresponding to the target space; and displaying the control instruction on the second control page.

[0052] Optionally, in one embodiment, suppose a user issues a scene control command to the terminal device: turn on the ceiling light in the living room every morning at 9:00 AM. The terminal device determines the target space as the living room from the scene control command and identifies the corresponding second control page for the living room, displaying the user's control command on the second control page. The user can modify the control command on the second control page.

[0053] In another embodiment, suppose a user says to the terminal device: "Set Goodnight Mode, turn off the bedroom lights." The large language model parses this control command, determining the scene mode (Goodnight Mode), the target space (bedroom), and the target IoT devices within the target space (all lights). After parsing, the terminal device determines the second control page corresponding to the bedroom and displays the scene mode on the second control page. In subsequent use, the user only needs to issue the voice command corresponding to the scene mode to the terminal device to implement the control command corresponding to that scene mode. For example, when the Goodnight Mode is available on the second control page, the user only needs to say the keyword containing "Goodnight Mode" to turn off the bedroom lights.

[0054] In an exemplary embodiment, after constructing the family metadata based on the plurality of first keywords, the method further includes: obtaining the status of a plurality of IoT devices in the plurality of spaces through an IoT interface; if the status of a first IoT device among the plurality of IoT devices is that it is not in use, deleting the third keyword corresponding to the first IoT device from the family metadata; if a second IoT device is added to the plurality of spaces, updating the family metadata based on the seventh keyword corresponding to the second IoT device.

[0055] Terminal devices continuously acquire the operational status of IoT devices (such as smart light bulbs, air conditioners, security cameras, etc.) deployed in various spaces of the home and their corresponding IoT interfaces. Operational status may include whether the device is online, whether it is in working condition, its battery level, and its fault status. When the IoT interface detects a device in a specific space, such as a smart air conditioner in the living room (i.e., the first IoT device), and its status changes to "not in use" (possibly due to physical removal, prolonged inactivity, or manual setting by the user), the terminal device automatically identifies the corresponding device in the living room from the home metadata and removes the third keyword related to the air conditioner. When the user adds a new device to the home, such as a smart security camera in the master bedroom (the second IoT device), the terminal device matches the camera with multiple keywords in the database. If a match is successful, the terminal device updates the home metadata based on the seventh keyword of the second IoT device.

[0056] In an exemplary embodiment, after parsing the first text or the second text using a large language model to obtain multiple first keywords in the first text or the second text, the method further includes: if an eighth keyword among the multiple first keywords fails to match multiple keywords in the database of the cloud server, receiving an error message sent by the cloud server, wherein the error message includes the reason why the eighth keyword fails to match the multiple keywords.

[0057] Suppose a user says to their terminal device, "Please turn on the smart fan on the balcony." Here, "balcony" and "smart fan" are the first keywords extracted from the user's command. The large language model compares these keywords with the cloud server's database to verify whether they correspond to known spaces or devices. The cloud server's database contains information on all registered devices and space names to ensure that the user's command can be executed correctly. If no keyword matching "balcony" or "smart fan" is found in the database (i.e., the eighth keyword), it means that the device or space information has not yet been registered. In this case, the terminal device cannot execute the control operation in the user's command. To help the user understand the problem, the cloud server generates and sends an error message to the user's terminal device. The error message includes the specific reason for the matching failure, such as: No space or device information matching "balcony" or "smart fan" found. Please check if the device is registered and correctly named, or confirm that the space name is accurate. After receiving the error message, the user can check if the balcony has been defined in the terminal device, if the smart fan has been added to the device list in the home metadata, and ensure that the device name is consistent with the naming rules in the database.

[0058] To better understand the process of constructing the above-mentioned family metadata, the following description, in conjunction with optional embodiments, further illustrates the method of constructing the above-mentioned family metadata, but is not intended to limit the technical solutions of the embodiments of this application.

[0059] Figure 3 This is a structural block diagram of a control page generation system according to an embodiment of this application, such as... Figure 3 As shown, the specific steps include:

[0060] Step S302: Obtain the voice command of the target object or the text directly input by the target object (i.e., the second text) through the user voice / text input module.

[0061] Step S304: Using the speech recognition / text preprocessing module, the speech command obtained in step S302 is converted into first text, and text preprocessing is performed on the first text or the second text. Text preprocessing includes, but is not limited to, formatting changes and noise filtering. Finally, keywords are extracted from the preprocessed first text or the second text.

[0062] Step S306: The keywords extracted in step S304 are parsed by the large language model parsing module to determine multiple first keywords that correspond one-to-one with multiple spaces or multiple IoT devices in multiple spaces in the target area where the target object is located.

[0063] Step S308: The decision-making and matching module matches the multiple primary keywords from step S306 with multiple keywords in the cloud server's database. If the target keyword for matching the primary keyword exists in the database, proceed to step 310; if some of the primary keywords fail to match with multiple keywords in the database, it means that these keywords are not registered in the database. In this case, the cloud server will generate and send an error message to the user's terminal device. The error message will include the specific reason for the matching failure, such as: unrecognized control name or product matching error.

[0064] Step S310: Construct family metadata based on multiple primary keywords using the metadata generation module. Optionally, suppose the user says to the terminal device: "My home has two bedrooms and one living room, one master bedroom, one secondary bedroom, one living room, and one bathroom. The living room has one ceiling light, two grille lights, and two recessed spotlights. The secondary bedroom has four downlights. The master bedroom has one ceiling light and four spotlights. The bathroom has one ceiling light." The large language model parsing module in step S306 can determine multiple primary keywords including master bedroom, secondary bedroom, living room, bathroom, ceiling light, grille light, spotlight, and downlight. Simultaneously, the large language model will also understand the relationships between these primary keywords during parsing, such as which devices belong to which room and how many devices are in each room, and extract this information using a dedicated Prompt or instruction set. The final family metadata is as follows: Living room: ceiling light {quantity: 1}, grille light {quantity: 2}, spotlight {quantity: 2}; Second bedroom: downlight {quantity: 4}; Master bedroom: ceiling light {quantity: 1}, spotlight {quantity: 4}; Bathroom: ceiling light {quantity: 1}.

[0065] Meanwhile, the terminal device continuously acquires the operational status of IoT devices deployed in various spaces of the home and their corresponding IoT interfaces. When the IoT interface detects that a device in a certain space, such as a smart air conditioner in the living room (i.e., the first IoT device), has changed its status to "not in use," the terminal device automatically identifies the corresponding device in the living room from the home metadata and deletes keywords related to the air conditioner. When the user adds a new device to the home, such as a smart security camera in the master bedroom (the second IoT device), the terminal device matches the camera with multiple keywords in the database. If a match is successful, the terminal device updates the home metadata based on the keywords of the second IoT device.

[0066] Step S312: The control page generation module determines multiple second keywords corresponding to multiple spaces contained in multiple first keywords, and multiple third keywords corresponding to multiple IoT devices in multiple spaces. Multiple first controls corresponding to the multiple second keywords and multiple second controls corresponding to the multiple third keywords are generated according to the front-end page generation rules. The second positional relationship between the multiple first controls and the multiple second controls is determined based on the first positional relationship between the multiple spaces and the multiple IoT devices. The control page for the terminal device is generated based on the multiple first controls, the multiple second controls, and the second positional relationship.

[0067] When a user needs to adjust the control page, they can issue an adjustment command to the terminal device. For example, they can change the ceiling light in the living room to a chandelier. The large language model will retrieve the operation record of the family metadata, which records the changes to the devices in the living room. For example, the unadjusted family metadata is: Living Room: Ceiling Light {Quantity: 1}, Grille Light {Quantity: 2}, Spotlight {Quantity: 2}. If the keyword "chandelier" exists in the database, the adjusted family metadata will be: Living Room: Chandelier {Quantity: 1}, Grille Light {Quantity: 2}, Spotlight {Quantity: 2}. Based on the operation record, the terminal device will delete the control corresponding to the ceiling light and add the control corresponding to the chandelier, thereby updating the control page for the living room.

[0068] When a user issues a scene control command, the system determines the target space corresponding to the command and the corresponding control page. The scene control command is then displayed on the control page corresponding to the target space. Optionally, in one embodiment, assuming a user issues a scene control command to the terminal device: "Turn on the ceiling light in the living room at 9:00 AM every morning," the terminal device determines the target space as the living room from the scene control command, determines the corresponding control page, and displays the user's command on the living room control page. In another embodiment, assuming a user says to the terminal device: "Set goodnight mode, turn off the bedroom lights," the large language model parses the control command, determining the scene mode (goodnight mode), the target space (bedroom), and the target IoT devices (all lights) within the target space. After parsing, the terminal device determines the control page corresponding to the bedroom and displays the scene mode on the bedroom control page. In subsequent use, the user only needs to issue the voice command corresponding to the scene mode to the terminal device to implement the control command corresponding to that scene mode. For example, when the bedroom control page has a goodnight mode, the user only needs to say the keyword containing "goodnight mode" to turn off the bedroom lights.

[0069] Figure 4 This is a first schematic diagram of a method for constructing family metadata according to an embodiment of this application, as shown below. Figure 4 As shown, it specifically includes the following:

[0070] Optionally, suppose a user issues a voice command to the terminal device: "My house has two bedrooms, one living room, one master bedroom, one secondary bedroom, one living room, and one bathroom. The living room has one ceiling light, two grid lights, and two recessed spotlights. The secondary bedroom has four downlights. The master bedroom has one ceiling light and four spotlights. The bathroom has one ceiling light." The user's voice command is converted into first text using speech recognition technology, or the user can directly input second text into the input box. After the first or second text is entered, the user clicks "Generate Design Scheme," and the large language model begins parsing the first or second text to identify multiple primary keywords within the text.

[0071] Figure 5 This is a second schematic diagram of a method for constructing family metadata according to an embodiment of this application, as shown below. Figure 5 As shown, it specifically includes the following:

[0072] Optionally, assuming multiple primary keywords are master bedroom, secondary bedroom, living room, bathroom, ceiling light, grille light, spotlight, and downlight, multiple secondary keywords are determined from these, and multiple third keywords are ceiling light, grille light, spotlight, and downlight. Multiple primary and secondary controls corresponding to the aforementioned secondary and third keywords are rendered using front-end page generation rules. The positional relationship of the multiple primary and secondary controls is determined based on the actual spatial layout. For example, assuming the master bedroom has four downlights, the primary controls corresponding to the secondary keywords (master bedroom, secondary bedroom, living room, bathroom, etc.) are displayed on the master bedroom control page. Based on the correspondence between the master bedroom and the downlights, the secondary controls for the downlights (i.e., downlight 1, downlight 2, downlight 3, and downlight 4 in the diagram) are retrieved from the cloud server's database and displayed on the master bedroom control page.

[0073] Furthermore, the control page can also display user-defined scenarios and automation commands. In one optional embodiment, suppose a user issues a voice command to the terminal device: "Set home mode, turn on the living room lights"; or "Set away mode, turn off all lights in all rooms." The large language model parses the control command to determine the scenario mode, the target space, and the target IoT device within the target space. After parsing, the terminal device determines the control page corresponding to the target space and displays the scenario mode on the control page. In subsequent use, the user only needs to issue the voice command corresponding to the scenario mode to the terminal device to implement the control command corresponding to that scenario mode. For example, when the control page has a home mode, the user only needs to say the keyword containing "home mode" to turn on the living room lights. In another optional embodiment, suppose a user issues a scene control command to the terminal device: "Turn on the master bedroom light at 9:00 AM every morning." The terminal device determines the target space as the master bedroom from the scene control command and determines the control page corresponding to the master bedroom, displaying the user's control command on the control page.

[0074] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of this application.

[0075] Figure 6 This is a structural block diagram of a family metadata construction apparatus according to an embodiment of this application; as shown below. Figure 6 As shown, it includes:

[0076] The acquisition module 62 is used to acquire voice commands issued by the target object and convert the voice commands into first text through voice recognition technology, or acquire second text sent by the target object;

[0077] The parsing module 64 is used to parse the first text or the second text through a large language model to obtain multiple first keywords in the first text or the second text, wherein the multiple first keywords correspond one-to-one with multiple spaces in the target area where the target object is located or multiple IoT devices in the multiple spaces;

[0078] The construction module 66 is used to construct family metadata based on the multiple first keywords when the multiple first keywords are successfully matched with target keywords in the database of the cloud server, wherein the family metadata is used to at least characterize the location relationship between the multiple IoT devices and the multiple spaces.

[0079] The aforementioned device acquires voice commands issued by the target object and converts these commands into first text using speech recognition technology, or acquires second text sent by the target object. A large language model is used to parse the first or second text to obtain multiple first keywords. These first keywords correspond one-to-one with multiple spaces or multiple IoT devices within the target area where the target object is located. If the multiple first keywords successfully match target keywords in the cloud server's database, family metadata is constructed based on these first keywords. This family metadata at least characterizes the locational relationships between multiple IoT devices and multiple spaces. This method solves the problem in related technologies where manually inputting family metadata is time-consuming, labor-intensive, and prone to errors.

[0080] In an exemplary embodiment, the construction module 66 is further configured to: determine a plurality of second keywords corresponding to the plurality of spaces included in the plurality of first keywords, and a plurality of third keywords corresponding to the plurality of IoT devices in the plurality of spaces; generate a plurality of first controls corresponding to the plurality of second keywords and a plurality of second controls corresponding to the plurality of third keywords through front-end page generation rules; determine a second positional relationship between the plurality of first controls and the plurality of second controls based on a first positional relationship between the plurality of spaces and the plurality of IoT devices; and generate a first control page of the terminal device based on the plurality of first controls, the plurality of second controls, and the second positional relationship.

[0081] In an exemplary embodiment, the construction module 66 is further configured to receive an adjustment instruction sent by the target object, and obtain a fourth keyword corresponding to the adjustment instruction through an IoT interface, wherein the adjustment instruction is used to add or delete the fourth keyword in the family metadata; according to the adjustment instruction, the control corresponding to the fourth keyword is adjusted on the first control page through the control interface of the front-end page, wherein the front-end page is used to display the first control page.

[0082] In an exemplary embodiment, the parsing module 64 is further configured to: change the format of the third text to obtain a standard format third text, wherein the third text is either the first text or the second text; perform noise filtering on the standard format third text to obtain a cleared third text; and extract a plurality of fifth keywords from the cleared third text, wherein the plurality of fifth keywords includes at least one of the following: the plurality of first keywords and a sixth keyword, wherein the sixth keyword is used to indicate the time or action corresponding to the third text.

[0083] In an exemplary embodiment, the construction module 66 is further configured to obtain a scene control instruction issued by the target object from the first text or the second text, wherein the scene control instruction is used to control the target IoT device to perform a target action; determine a target space from the plurality of spaces according to the scene control instruction, and determine a second control page corresponding to the target space; and display the control instruction on the second control page.

[0084] In one exemplary embodiment, the construction module 66 is further configured to obtain the status of multiple IoT devices in the multiple spaces through an IoT interface; if the status of a first IoT device among the multiple IoT devices is that it is not in use, delete the third keyword corresponding to the first IoT device from the home metadata; if a second IoT device is added to the multiple spaces, update the home metadata according to the seventh keyword corresponding to the second IoT device.

[0085] In an exemplary embodiment, the parsing module 64 is further configured to receive error information sent by the cloud server when an eighth keyword among the plurality of first keywords fails to match with multiple keywords in the database of the cloud server, wherein the error information includes the reason why the eighth keyword fails to match with the plurality of keywords.

[0086] This embodiment also provides a method for generating a control page, applied to a terminal device. Figure 7 This is a flowchart of a method for generating a control page according to an embodiment of this application. The process includes the following steps:

[0087] Step S702: Obtain the voice command issued by the target object, and convert the voice command into first text through voice recognition technology, or obtain the second text sent by the target object;

[0088] Step S704: Parse the first text or the second text using a large language model to obtain multiple first keywords in the first text or the second text, wherein the multiple first keywords correspond one-to-one with multiple spaces in the target area where the target object is located or multiple IoT devices in the multiple spaces;

[0089] Step S706: If the plurality of first keywords successfully match the target keywords in the database of the cloud server, construct family metadata based on the plurality of first keywords, wherein the family metadata is used at least to characterize the location relationship between the plurality of IoT devices and the plurality of spaces;

[0090] Step S708: Generate a target control corresponding to the family metadata through the front-end page generation rules, and generate a first control page based on the target control.

[0091] Through the above steps, firstly, the voice command issued by the target object is obtained, and the voice command is converted into first text using speech recognition technology, or second text sent by the target object is obtained; secondly, the first text or second text is parsed using a large language model to obtain multiple first keywords in the first text or second text, wherein each of the multiple first keywords corresponds one-to-one with multiple spaces in the target area where the target object is located or multiple IoT devices in the multiple spaces; if the multiple first keywords successfully match target keywords in the database of the cloud server, family metadata is constructed based on the multiple first keywords, wherein the family metadata is at least used to characterize the positional relationship between the multiple IoT devices and the multiple spaces; finally, a target control corresponding to the family metadata is generated according to the front-end page generation rules, and a first control page is generated based on the target control. This method solves the problem in related technologies where manually generating the control page by the target object increases the complexity of manual operation and results in low efficiency in control page generation.

[0092] Figure 8 This is a structural block diagram of a control page generation apparatus according to an embodiment of this application, such as... Figure 8 As shown, it includes:

[0093] The acquisition module 82 is used to acquire voice commands issued by the target object and convert the voice commands into first text through speech recognition technology, or acquire second text sent by the target object;

[0094] The parsing module 84 is used to parse the first text or the second text through a large language model to obtain multiple first keywords in the first text or the second text, wherein the multiple first keywords correspond one-to-one with multiple spaces in the target area where the target object is located or multiple IoT devices in the multiple spaces;

[0095] The construction module 86 is used to construct family metadata based on the multiple first keywords when the multiple first keywords are successfully matched with target keywords in the database of the cloud server, wherein the family metadata is used to characterize the location relationship between the multiple IoT devices and the multiple spaces.

[0096] The generation module 88 is used to generate a target control corresponding to the family metadata according to the front-end page generation rules, and generate a first control page according to the target control.

[0097] The above-described device first acquires voice commands issued by the target object and converts these commands into first text using speech recognition technology, or acquires second text sent by the target object. Second, it parses the first or second text using a large language model to obtain multiple first keywords, where each first keyword corresponds one-to-one with multiple spaces or multiple IoT devices within the target area where the target object is located. If the multiple first keywords successfully match target keywords in a cloud server database, it constructs family metadata based on these first keywords, where the family metadata at least characterizes the positional relationships between the multiple IoT devices and the multiple spaces. Finally, it generates target controls corresponding to the family metadata using front-end page generation rules, and generates a first control page based on the target controls. This method solves the problems in related technologies where manually generating control pages by the target object increases the complexity of manual operations and results in low efficiency in control page generation.

[0098] Embodiments of this application also provide a storage medium including a stored program, wherein the program executes any of the methods described above when it is run.

[0099] Optionally, in this embodiment, the storage medium may be configured to store program code for performing the following steps:

[0100] S11, acquire the voice command issued by the target object, and convert the voice command into first text through voice recognition technology, or acquire the second text sent by the target object;

[0101] S12, the first text or the second text is parsed using a large language model to obtain multiple first keywords in the first text or the second text, wherein the multiple first keywords correspond one-to-one with multiple spaces in the target area where the target object is located or multiple IoT devices in the multiple spaces;

[0102] S13, if the plurality of first keywords successfully match the target keywords in the database of the cloud server, construct family metadata based on the plurality of first keywords, wherein the family metadata is at least used to characterize the location relationship between the plurality of IoT devices and the plurality of spaces.

[0103] Optionally, in this embodiment, the storage medium may also be configured to store program code for performing the following steps:

[0104] S21, acquire the voice command issued by the target object, and convert the voice command into first text through voice recognition technology, or acquire the second text sent by the target object;

[0105] S22, the first text or the second text is parsed using a large language model to obtain multiple first keywords in the first text or the second text, wherein the multiple first keywords correspond one-to-one with multiple spaces in the target area where the target object is located or multiple IoT devices in the multiple spaces;

[0106] S23, if the plurality of first keywords successfully match the target keywords in the database of the cloud server, construct family metadata based on the plurality of first keywords, wherein the family metadata is at least used to characterize the location relationship between the plurality of IoT devices and the plurality of spaces;

[0107] S24, generate a target control corresponding to the family metadata according to the front-end page generation rules, and generate a first control page according to the target control.

[0108] Embodiments of this application also provide an electronic device including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.

[0109] Optionally, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor and the input / output device is connected to the processor.

[0110] Optionally, in this embodiment, the processor can be configured to perform the following steps via a computer program:

[0111] S11, acquire the voice command issued by the target object, and convert the voice command into first text through voice recognition technology, or acquire the second text sent by the target object;

[0112] S12, the first text or the second text is parsed using a large language model to obtain multiple first keywords in the first text or the second text, wherein the multiple first keywords correspond one-to-one with multiple spaces in the target area where the target object is located or multiple IoT devices in the multiple spaces;

[0113] S13, if the plurality of first keywords successfully match the target keywords in the database of the cloud server, construct family metadata based on the plurality of first keywords, wherein the family metadata is at least used to characterize the location relationship between the plurality of IoT devices and the plurality of spaces.

[0114] Optionally, in this embodiment, the processor may also be configured to perform the following steps via a computer program:

[0115] S21, acquire the voice command issued by the target object, and convert the voice command into first text through voice recognition technology, or acquire the second text sent by the target object;

[0116] S22, the first text or the second text is parsed using a large language model to obtain multiple first keywords in the first text or the second text, wherein the multiple first keywords correspond one-to-one with multiple spaces in the target area where the target object is located or multiple IoT devices in the multiple spaces;

[0117] S23, if the plurality of first keywords successfully match the target keywords in the database of the cloud server, construct family metadata based on the plurality of first keywords, wherein the family metadata is at least used to characterize the location relationship between the plurality of IoT devices and the plurality of spaces;

[0118] S24, generate a target control corresponding to the family metadata according to the front-end page generation rules, and generate a first control page according to the target control.

[0119] Optionally, in this embodiment, the storage medium may include, but is not limited to, various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0120] Optionally, specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementations, and will not be repeated here.

[0121] Obviously, those skilled in the art should understand that the modules or steps of this application described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those presented here, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, this application is not limited to any particular combination of hardware and software.

[0122] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the principles of this application should be included within the protection scope of this application.

Claims

1. A method for constructing family metadata, characterized in that, Applied to terminal devices, including: Acquire voice commands issued by the target object and convert the voice commands into first text using speech recognition technology, or acquire second text sent by the target object; The first text or the second text is parsed using a large language model to obtain multiple first keywords in the first text or the second text, wherein the multiple first keywords correspond one-to-one with multiple spaces in the target area where the target object is located or multiple IoT devices in the multiple spaces; If the plurality of first keywords successfully match the target keywords in the database of the cloud server, family metadata is constructed based on the plurality of first keywords, wherein the family metadata is used at least to characterize the location relationship between the plurality of IoT devices and the plurality of spaces.

2. The method for constructing family metadata according to claim 1, characterized in that, After constructing family metadata based on the multiple first keywords, the method further includes: Determine a plurality of second keywords that correspond to the plurality of spaces, which are included in the plurality of first keywords, and a plurality of third keywords that correspond to the plurality of IoT devices in the plurality of spaces; Multiple first controls corresponding to the multiple second keywords and multiple second controls corresponding to the multiple third keywords are generated according to the front-end page generation rules. The second positional relationship between the plurality of first controls and the plurality of second controls is determined based on the first positional relationship between the plurality of spaces and the plurality of IoT devices; The first control page of the terminal device is generated based on the plurality of first controls, the plurality of second controls, and the second positional relationship.

3. The method for constructing family metadata according to claim 2, characterized in that, After generating the first control page of the terminal device based on the plurality of first controls, the plurality of second controls, and the second positional relationship, the method further includes: The system receives an adjustment instruction sent by the target object and obtains the fourth keyword corresponding to the adjustment instruction through an IoT interface. The adjustment instruction is used to add or delete the fourth keyword in the family metadata. According to the adjustment instruction, the control corresponding to the fourth keyword is adjusted on the first control page through the control interface of the front-end page, wherein the front-end page is used to display the first control page.

4. The method for constructing family metadata according to claim 1, characterized in that, Before parsing the first or second text using a large language model to obtain multiple first keywords from the first or second text, the method further includes: The format of the third text is changed to obtain a standard format third text, wherein the third text is either the first text or the second text; The standard format third text is subjected to noise filtering to obtain a clearer third text; Multiple fifth keywords are extracted from the clarified third text, wherein the multiple fifth keywords include at least one of the following: the multiple first keywords and the sixth keyword, wherein the sixth keyword is used to indicate the time or action corresponding to the third text.

5. The method for constructing family metadata according to claim 1, characterized in that, After constructing family metadata based on the multiple first keywords, the method further includes: Obtain the scene control command issued by the target object from the first text or the second text, wherein the scene control command is used to control the target IoT device to perform the target action; The target space is determined from the plurality of spaces according to the scene control instructions, and the second control page corresponding to the target space is determined. The control commands are displayed on the second control page.

6. The method for constructing family metadata according to claim 2, characterized in that, After constructing family metadata based on the multiple first keywords, the method further includes: The status of multiple IoT devices in the multiple spaces is obtained through the IoT interface; If, among the plurality of IoT devices, the first IoT device is in a state of being out of use, the third keyword corresponding to the first IoT device is deleted from the home metadata. When a second IoT device is added to the multiple spaces, the home metadata is updated according to the seventh keyword corresponding to the second IoT device.

7. The method for constructing family metadata according to claim 1, characterized in that, After parsing the first or second text using a large language model to obtain multiple first keywords from the first or second text, the method further includes: If, among the plurality of first keywords, an eighth keyword fails to match with multiple keywords in the database of the cloud server, an error message is received from the cloud server, wherein the error message contains the reason why the eighth keyword fails to match with the plurality of keywords.

8. A device for constructing family metadata, characterized in that, include: The acquisition module is used to acquire voice commands issued by the target object and convert the voice commands into first text through speech recognition technology, or acquire second text sent by the target object; The parsing module is used to parse the first text or the second text through a large language model to obtain multiple first keywords in the first text or the second text, wherein the multiple first keywords correspond one-to-one with multiple spaces in the target area where the target object is located or multiple IoT devices in the multiple spaces; A construction module is used to construct family metadata based on the multiple first keywords when the multiple first keywords are successfully matched with target keywords in the database of the cloud server. The family metadata is used to characterize the location relationship between the multiple IoT devices and the multiple spaces.

9. A method for controlling page generation, characterized in that, Applied to terminal devices, including: Acquire voice commands issued by the target object and convert the voice commands into first text using speech recognition technology, or acquire second text sent by the target object; The first text or the second text is parsed using a large language model to obtain multiple first keywords in the first text or the second text, wherein the multiple first keywords correspond one-to-one with multiple spaces in the target area where the target object is located or multiple IoT devices in the multiple spaces; If the plurality of first keywords successfully match the target keywords in the database of the cloud server, family metadata is constructed based on the plurality of first keywords, wherein the family metadata is used at least to characterize the location relationship between the plurality of IoT devices and the plurality of spaces; The target control corresponding to the family metadata is generated by the front-end page generation rules, and the first control page is generated based on the target control.

10. A device for controlling the generation of pages, characterized in that, include: The acquisition module is used to acquire voice commands issued by the target object and convert the voice commands into first text through speech recognition technology, or acquire second text sent by the target object; The parsing module is used to parse the first text or the second text through a large language model to obtain multiple first keywords in the first text or the second text, wherein the multiple first keywords correspond one-to-one with multiple spaces in the target area where the target object is located or multiple IoT devices in the multiple spaces; A construction module is used to construct family metadata based on the multiple first keywords when the multiple first keywords are successfully matched with target keywords in the database of the cloud server, wherein the family metadata is used to characterize the location relationship between the multiple IoT devices and the multiple spaces; The generation module is used to generate target controls corresponding to the family metadata according to the front-end page generation rules, and generate a first control page based on the target controls.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein the program, when executed, performs the method described in any one of claims 1 to 7 or 9.

12. An electronic device comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to execute the method described in any one of claims 1 to 7 or 9 via the computer program.

Images