Household appliance control method and apparatus, and electronic device

Abstract

The application provides a home appliance control method and device and an electronic device, and belongs to the technical field of smart home. The user instruction is determined through the user voice. When the user instruction is an instruction for recording habits, the voice is processed through a large model to obtain first behavior habit data of the user, and the first behavior habit data is stored in a behavior habit database. When the user instruction is an instruction for controlling the home appliance, second behavior habit data related to the voice is obtained in the behavior habit database. The voice, the current environmental parameter and the second behavior habit data are processed through the large model to obtain control parameters of at least one home appliance. The corresponding home appliance is controlled according to the control parameters. The method identifies the user intention through the voice instruction, processes the voice information through the large model, combines the user behavior habit with the current environmental parameter, intelligently controls the home appliance, and realizes personalized and automated home management.

Description

Technical Field

[0001] This application relates to the field of smart home technology. More specifically, it relates to a method, apparatus, and electronic device for controlling home appliances. Background Technology

[0002] With the increasing popularity of smart homes, users' personalized needs for appliance control are becoming more and more prominent. Traditional appliance control modes often use fixed parameter settings, making it difficult to meet users' diverse needs for appliance functions in different seasons and at different times. This demand is driving the development of the smart appliance control field towards personalization and scenario-based solutions, requiring appliance control systems to understand user habits and provide adaptive appliance adjustment solutions.

[0003] Current methods for controlling smart home appliances require a device that supports voice recognition and connects it to the smart appliances in the home. Users can then control all the smart appliances in the house via simple voice commands.

[0004] Current home appliance control systems primarily rely on large models to analyze and control commands. However, most of these models depend on temporary context windows for prediction, leading to fragmented preference analysis. Simply relying on model analysis capabilities results in low accuracy in controlling home appliances, making it difficult to achieve truly personalized intelligent services. Summary of the Invention

[0005] This application provides a method, apparatus, and electronic device for controlling home appliances. By leveraging the capabilities of external databases and large models, it effectively solves the problem of existing large models lacking long-term memory. Simultaneously, by fully utilizing the powerful analytical capabilities of large models, it achieves accurate analysis of users' home appliance usage preferences, thereby providing users with more personalized and intelligent home appliance control services.

[0006] In a first aspect, embodiments of this application provide a method for controlling household appliances, including:

[0007] In response to the user's first voice, determine the user command corresponding to the first voice;

[0008] When the user instruction is an instruction for recording habits, the first voice is processed through a large model to obtain the user's first behavioral habit data, and the first behavioral habit data is stored in the behavioral habit database.

[0009] When the user instruction is an instruction that requires control of home appliances, the user's second behavioral habit data related to the first voice is obtained from the behavioral habit database.

[0010] By processing the first voice, the current first environmental parameters, and the second behavioral habit data using a large model, at least one first control parameter for a home appliance is obtained.

[0011] The at least one household appliance is controlled according to the first control parameter.

[0012] In the above technical solution, user intent is recognized through voice commands. For inputted habitual commands, a large model is used to process and store user behavior data. For commands to control home appliances, the control parameters of the home appliances are intelligently calculated through the large model, combining voice, current environmental parameters, and user behavior data, and then the control is implemented. The technical effect is to improve the intelligence and personalization of home management, allowing users to enjoy a more convenient and comfortable home life experience.

[0013] In some embodiments, the large model includes a general large model;

[0014] The process of processing the first speech using a large model to obtain the user's first behavioral habit data includes:

[0015] The first speech is recognized to obtain the first text content corresponding to the first speech.

[0016] Obtain the preset prompt words corresponding to the first text content;

[0017] The first text content and the preset prompt words are processed using the general model to obtain the user's first behavioral habit data.

[0018] In the above technical solution, text content is obtained through speech recognition and combined with preset prompts to guide a general large model for accurate analysis, thereby effectively extracting and generating the user's first behavioral habit data, which significantly improves the accuracy and efficiency of data processing.

[0019] In some embodiments, the large model includes a large model of the home appliance control vertical domain, and the second behavioral habit data includes user preference information and data of home appliances;

[0020] The process of processing the first voice, the current first environmental parameters, and the second behavioral habit data using a large model to obtain at least one first control parameter for a home appliance includes:

[0021] The first speech is recognized to obtain the second text content corresponding to the first speech.

[0022] By using the large-scale home appliance control model, the second text content, the first environmental parameters, the user preference information, and the data of the home appliances are processed to obtain at least one first control parameter for a home appliance.

[0023] In the above technical solution, the user's voice, environmental parameters and user behavior data are comprehensively processed by the large-scale model of home appliance control, and control parameters of at least one home appliance are intelligently generated, realizing personalized and intelligent control of home appliances.

[0024] In some embodiments, the method further includes:

[0025] The second text content, the first environmental parameters, the second behavioral habit data, and the first control parameters are processed using a general large model to generate the third behavioral habit data.

[0026] The third behavioral habit data is stored in the behavioral habit database.

[0027] In the above technical solution, by analyzing and processing the current environmental parameters and current control parameters to generate new behavioral habit data and storing the behavioral habit data, the accuracy of home appliance control under the same scenario can be improved in subsequent home appliance control.

[0028] In some embodiments, the method further includes:

[0029] In response to the user's second voice, based on the third text content corresponding to the second voice, the user's fourth behavioral habit data related to the third text content is obtained from the behavioral habit database; the second voice is the voice following the first voice, and the second voice is used to instruct the control of the first home appliance among the at least one home appliance.

[0030] The third text content and the fourth behavioral habit data are processed by the large-scale model of home appliance control to obtain the second control parameters of the first home appliance.

[0031] The first home appliance is controlled according to the second control parameter.

[0032] The aforementioned technical solution effectively integrates the user's immediate intent (expressed via second voice) with long-term behavioral habits by leveraging the deep learning capabilities of a large-scale model for home appliance control, ensuring a high degree of personalization and precision in home appliance control. This not only enhances the convenience of using home appliances but also further strengthens the intelligence and comfort of the home environment by continuously learning and adapting to changes in user habits.

[0033] In some embodiments, the method further includes:

[0034] In response to the user's third voice, the fourth text content corresponding to the third voice is obtained, wherein the third voice is the information fed back by the user after controlling the home appliance based on the first control parameters;

[0035] The fourth text content and the first control parameters are processed using a general large model to obtain the fifth behavior habit data;

[0036] The third behavioral habit data in the behavioral habit database is updated based on the fifth behavioral habit data.

[0037] In the above technical solution, by analyzing user feedback information and updating the third behavioral habit data in the behavioral habit database based on the feedback information, it is possible to accurately update and correct the data in the existing behavioral habit database, and provide users with more personalized services that better suit their preferences and needs for future home appliance control.

[0038] In some embodiments, the method further includes:

[0039] In response to the user's input command for the second home appliance, the second control parameters of the second home appliance are obtained;

[0040] Using the second control parameters, control the second home appliance and obtain the current second environmental parameters;

[0041] The instructions, the second control parameters, and the second environmental parameters are processed using a general large model to generate sixth-behavior habit data.

[0042] The sixth behavioral habit data is stored in the behavioral habit database.

[0043] In the above technical solution, the system monitors in real time the control commands issued by users to home appliances via physical means (such as remote controls, touch screens, etc.) and accurately executes the corresponding operations. Simultaneously, the method conducts in-depth analysis of users' physical operation habits and saves the analysis results to a behavior habit database to ensure the accuracy and completeness of the database content, providing strong support for subsequent intelligent control.

[0044] Secondly, this application provides a home appliance control device, comprising:

[0045] The processing module is used to respond to the user's first voice and determine the user instruction corresponding to the first voice; when the user instruction is an instruction for recording habits, the first voice is processed through a large model to obtain the user's first behavioral habit data, and the first behavioral habit data is stored in the behavioral habit database.

[0046] The acquisition module is used to acquire, in the behavior habit database, second behavior habit data of the user related to the first voice when the user instruction is an instruction that requires control of home appliances;

[0047] The processing module is also used to process the first voice, the current first environmental parameters, and the second behavioral habit data through a large model to obtain at least one first control parameter of a home appliance.

[0048] The control module is used to control the at least one household appliance according to the first control parameters.

[0049] The home appliance control method, device, and electronic device provided in this application determine the content of user commands by recognizing user voice. If the user command involves input habits, a large model is used to process the voice information, extract the user's first behavioral habit data, and store it in a behavioral habit database. When the user command is to control a home appliance, the system retrieves second behavioral habit data related to the current voice from the behavioral habit database. Subsequently, the large model comprehensively considers the voice content, current environmental parameters, and second behavioral habit data to generate control parameters for at least one home appliance. Finally, based on these control parameters, the corresponding home appliance is intelligently controlled, thereby achieving personalized and automated home management based on user behavioral habits and the current environment. Attached Figure Description

[0050] To more clearly illustrate the implementation methods in the embodiments of this application or related technologies, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings.

[0051] Figure 1 A schematic diagram illustrating a scenario for a home appliance control method provided in this application;

[0052] Figure 2 A flowchart illustrating a method for controlling home appliances provided in this application. Figure 1 ;

[0053] Figure 3 A flowchart illustrating a method for controlling home appliances provided in this application. Figure 2 ;

[0054] Figure 4 A flowchart illustrating a method for controlling home appliances provided in this application. Figure 3 ;

[0055] Figure 5 A flowchart illustrating a method for controlling home appliances provided in this application. Figure 4 ;

[0056] Figure 6 A flowchart illustrating a method for controlling home appliances provided in this application. Figure 5 ;

[0057] Figure 7 A flowchart illustrating a method for controlling home appliances provided in this application. Figure 6 ;

[0058] Figure 8 A schematic diagram of the user preference update process in a home appliance control method provided in this application;

[0059] Figure 9 A schematic diagram of the structure of a home appliance control device 60 provided in this application;

[0060] Figure 10 This is a structural schematic diagram of a home appliance control device 70 provided in this application. Detailed Implementation

[0061] To make the objectives, implementation methods and advantages of this application clearer, the exemplary implementation methods of this application will be clearly and completely described below with reference to the accompanying drawings of the exemplary embodiments of this application. Obviously, the described exemplary embodiments are only some embodiments of this application, and not all embodiments.

[0062] It should be noted that the brief descriptions of terms in this application are only for the convenience of understanding the embodiments described below, and are not intended to limit the embodiments of this application. Unless otherwise stated, these terms should be understood in their ordinary and common meaning.

[0063] Furthermore, the terms “comprising” and “having”, and any variations thereof, are intended to cover but not exclusively include, for example, a product or device that includes a series of components is not necessarily limited to those that are explicitly listed, but may include other components that are not explicitly listed or that are inherent to such product or device.

[0064] With the continuous advancement of smart home technology, users' demands for personalized appliance control are increasing. Traditional appliance control methods, due to fixed parameter settings, are no longer able to adapt to the diverse needs of users for appliance functions in different seasons and at different times. This trend is driving the field of smart appliance control towards greater personalization and scenario-based solutions, aiming to provide users with adaptive appliance adjustment solutions by deeply understanding their habits.

[0065] In today's smart home environment, whole-house smart appliance control is primarily achieved through the use of smart devices that support voice recognition. Users simply connect these devices to their smart appliances and can easily control all the appliances in the house using simple voice commands. This control method is not only convenient and efficient but also greatly enhances the level of intelligence in home life, bringing users a more comfortable and convenient living experience.

[0066] Despite significant advancements in whole-house smart appliance control, existing technologies still face several core bottlenecks. The limitations of large-scale model architectures are particularly prominent. Most large models rely on temporary context windows for prediction and lack built-in long-term memory mechanisms, making it difficult for appliance control systems to effectively store and analyze user behavior records across time periods and seasons. This results in fragmented preference analysis, hindering the creation of comprehensive user profiles. Furthermore, relying solely on model analysis capabilities without sufficient integration of external data sources limits the depth and accuracy of preference inference, making it difficult to achieve truly personalized smart services and hindering further improvements in user experience.

[0067] To address the aforementioned issues, this application proposes a home appliance control method. By utilizing the analytical capabilities of a large model, the method analyzes the user's voice commands to obtain the user's home appliance control behavior habits and generates a behavior habit database. When executing home appliance control, the method combines the user's voice commands with external data stored in the behavior habit database to generate and execute personalized home appliance control data that better matches the user's habits. Furthermore, this method also has the ability to control multiple home appliances simultaneously, effectively promoting the intelligence of home life and significantly improving the overall user experience.

[0068] This solution applies to home appliance control devices, which are flexible and can be embedded and installed as virtual devices in any smart home device, such as televisions and speakers, without being limited to specific hardware. This solution does not restrict the specific installation medium for the home appliance control device, aiming to provide a more convenient and flexible solution for home automation.

[0069] Figure 1 A schematic diagram of a scenario for a home appliance control method provided in this application, such as... Figure 1 As shown, home appliance control devices can intelligently capture users' spoken commands. After analyzing these commands, the control devices store user-habit-related commands in a behavioral habit database. For home appliance control commands, the control devices can perform in-depth analysis by combining multi-dimensional data from the behavioral habit database, thereby generating personalized control command sets. This allows for the automatic execution of corresponding control operations on at least one home appliance associated with each command, achieving an upgrade from simple responses to intelligent services.

[0070] The technical solutions of this application will be described in detail below with reference to specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0071] Figure 2 A flowchart illustrating a method for controlling home appliances provided in this application. Figure 1 ,like Figure 2 As shown, the method includes:

[0072] S101: In response to the user's first voice, determine the user command corresponding to the first voice.

[0073] Understandably, when a user issues a command via voice, the home appliance control device needs to accurately capture and understand the user's voice content and the user's true intention. For example, if a user says "turn on the living room light," the home appliance control device can recognize that this is a command to control the living room light, thereby determining the user's specific needs and providing a basis for subsequent operations.

[0074] S102: When the user's instruction is a habit input instruction, the first voice is processed through the large model to obtain the user's first behavioral habit data, and the first behavioral habit data is stored in the behavioral habit database.

[0075] Understandably, if the user's instruction pertains to recording personal habits, such as "From now on, when I turn on the air conditioner at night, I will also turn on the dehumidifier," the home appliance control device will use a large-scale model to deeply analyze this voice message. The large-scale model will identify the user's behavioral habit (turning on the air conditioner at night and simultaneously turning on the dehumidifier) ​​and combine it with current environmental information (such as season, time, temperature, and humidity) to comprehensively understand this habit. Then, the home appliance control device will organize and store this behavioral habit data in a behavioral habit database so that it can be automatically executed in similar situations in the future.

[0076] S103: When the user's instruction is an instruction that requires control of home appliances, retrieve the user's second behavioral habit data related to the first voice from the behavioral habit database.

[0077] Understandably, when a user issues a command to control home appliances, such as "turn on the air conditioner," the appliance control device immediately searches its behavioral habit database for related historical user behavior. The device searches how the user typically sets the air conditioner parameters (such as temperature and fan speed) in the current season and time, and what other appliances the user usually turns on simultaneously to assist in adjusting the air conditioner (such as humidifiers and fans). This retrieved historical behavioral habit data constitutes secondary behavioral habit data, providing the appliance control device with a reference for executing user commands.

[0078] S104: Process the first voice, the current first environmental parameters, and the second behavioral habit data through a large model to obtain the first control parameters of at least one home appliance.

[0079] Understandably, after acquiring the user's current commands, current environmental parameters (such as outdoor temperature and indoor humidity), and historical behavioral data, home appliance control devices can use a large-scale model to comprehensively analyze this information. This large-scale model can intelligently integrate this information, considering the user's immediate needs, current environmental conditions, and historical habits, thereby generating specific control parameters for at least one home appliance. For example, based on the user's command "cool down the room," combined with the current hot weather and the user's habit of running the fan for a while before turning on the air conditioner, the large-scale model might decide to turn on the fan first, and then automatically adjust the air conditioner to a suitable temperature and fan speed later.

[0080] S105: Control at least one household appliance according to the first control parameter.

[0081] Understandably, after obtaining the initial control parameters generated from the large model, the home appliance control device will immediately control the corresponding home appliances based on these parameters. For example, if the control parameters are a temperature setting of 26 degrees Celsius and a fan speed of medium, the control device will send instructions to the air conditioner to operate according to these parameters. In this way, users can enjoy customized home appliance control services based on their current needs, environmental conditions, and historical habits, improving the convenience and comfort of their lives.

[0082] The home appliance control method provided in this embodiment determines the user command corresponding to the first voice command in response to the user's first voice command. When the user command is a command for recording habits, the first voice command is processed using a large model to obtain the user's first behavioral habit data, which is then stored in a behavioral habit database. When the user command is a command to control home appliances, the user's second behavioral habit data related to the first voice command is retrieved from the behavioral habit database. The first voice command, current first environmental parameters, and second behavioral habit data are processed using a large model to obtain first control parameters for at least one home appliance. Based on the first control parameters, at least one home appliance is controlled. This method identifies user intent through voice commands, processes voice information using a large model, and combines user behavioral habits with current environmental parameters to intelligently control home appliances, achieving personalized and automated home management.

[0083] In some embodiments, the large model includes a general large model. Figure 3 A flowchart illustrating a method for controlling home appliances provided in this application. Figure 2 ,like Figure 3 As shown, in this embodiment... Figure 2 Based on the embodiments, the process of generating the first behavioral habit data is described in detail. The method includes:

[0084] S201: Recognize the first speech and obtain the first text content corresponding to the first speech.

[0085] Understandably, when a user issues a voice command, the home appliance control device first uses advanced speech recognition technology to meticulously analyze and convert the speech. This process is similar to a "translation" step; the control device needs to accurately capture every syllable and intonation in the speech and convert it into readable text content. For example, when a user says, "I turn on the air conditioner at 8 pm every night and set the temperature to 26 degrees Celsius," the control device can accurately convert this speech into corresponding text content using text recognition technology, providing a foundation for subsequent processing and understanding.

[0086] S202: Obtain the preset prompt word corresponding to the first text content.

[0087] Understandably, after receiving the initial text content, the appliance control device will further acquire preset prompts corresponding to that text content. These prompts are pre-defined and represent various intentions and instructions that the user might express. For a general-purpose large-scale model, these prompts act like a "learning guide," helping the model better understand and analyze the user's intentions within the text content. For example, for the text "I turn on the air conditioner every night at 8 PM," the preset prompts might include "time" and "turn on the air conditioner," allowing the large-scale model to accurately identify the user's intention to turn on the air conditioner every night at 8 PM.

[0088] S203: Using a general large model, process the first text content and preset prompt words to obtain the user's first behavioral habit data.

[0089] Understandably, after acquiring the initial text content and preset prompts, the home appliance control device inputs this information into a general-purpose model. This model then utilizes its powerful learning capabilities and algorithms to conduct in-depth analysis and processing of the text content and prompts. Through this process, the model can uncover the user's underlying behavioral habits and preferences, such as the user's preference for turning on the air conditioner at 8 PM and setting the temperature to 26 degrees Celsius. This behavioral data is saved by the home appliance control device, forming a user behavior database that provides strong support for future intelligent services and personalized recommendations. Thus, when the user uses the air conditioner again at 8 PM, they don't need to specify the control mode; simply saying "turn on the air conditioner" will allow the control device to provide more precise and personalized services based on the user's historical behavioral habits.

[0090] The home appliance control method provided in this embodiment obtains the first text content corresponding to the first speech by recognizing the first speech. Preset prompt words corresponding to the first text content are then acquired. A general-purpose model is used to process the first text content and the preset prompt words to obtain the user's first behavioral habit data. This method obtains text content through speech recognition and combines it with preset prompt words to guide a general-purpose model for accurate analysis, thereby effectively extracting and generating the user's first behavioral habit data, significantly improving the accuracy and efficiency of data processing.

[0091] In some embodiments, the large model further includes a large model of the home appliance control vertical domain, the second behavioral habits include user preference information and data of home appliances, and the step of generating a first control parameter for at least one home appliance includes:

[0092] The first speech is recognized to obtain the second text content corresponding to the first speech.

[0093] By using a large-scale model of home appliance control, the second text content, the first environmental parameters, user preference information, and data of home appliances are processed to obtain the first control parameters of at least one home appliance.

[0094] Understandably, if the user's first voice instruction is a command to control home appliances, then the first voice is converted into text. This text is then combined with current environmental parameters, historical user preference information extracted from the behavioral habit database, and data from all home appliances in the current household, and input into the large-scale home appliance control model so that the large-scale home appliance control model can generate accurate home appliance control parameters.

[0095] Among them, the home appliance vertical domain model is a model specifically optimized for the home appliance control field. It can infer and generate optimal home appliance control parameters by combining user historical preferences, current voice commands, and environmental parameters. Compared with the general large model, the home appliance vertical domain model has higher professionalism and accuracy, and can adapt to the different models and functional characteristics of different home appliances in a user's home. The home appliance vertical domain model achieves customized development on the basis of the general large model through prompt word engineering and external knowledge base.

[0096] In the process of generating a large-scale model for the home appliance vertical domain, the design of prompt words is crucial. Prompt words not only reflect the training objectives of the model, but also guide the inference path of the model when generating home appliance control parameters. The external knowledge base includes, but is not limited to, data on home appliances, user behavior databases, and detailed information on external environmental conditions. By combining prompt words with the external knowledge base, the large-scale model for home appliance control can generate more accurate and personalized control parameters.

[0097] In some embodiments, after control of at least one home appliance is completed, user behavior habits can also be generated and saved based on the control parameters of the home appliance.

[0098] For example, by processing the second text content, the first environmental parameters, the second behavioral habit data, and the first control parameters through a general large model, the third behavioral habit data can be generated.

[0099] Store the third behavioral habit data in the behavioral habit database.

[0100] Understandably, after transmitting the first control parameters to the corresponding appliance, the appliance control device can comprehensively process the user's voice commands (second text content), current environmental parameters (first environmental parameters), the user's past behavioral habit data (second behavioral habit data, i.e., the user's past operating habits in similar situations), and the specific parameters required to control the appliance this time (first control parameters, such as the air conditioner's set temperature and the fan's speed). The result of this processing is the generation of new behavioral habit data (third behavioral habit data), which reflects the user's latest control preferences in the current situation. This newly generated behavioral habit data is then stored in a dedicated behavioral habit database. The purpose of this is so that when the user wants to control the appliance again in the future, the appliance control device can extract from the database the control method that the user has chosen most frequently in the same or similar situations, i.e., the control method that best matches the user's recent habits, thus more thoughtfully meeting the user's needs.

[0101] In some embodiments, Figure 4 A flowchart illustrating a method for controlling home appliances provided in this application. Figure 3 ,like Figure 4 As shown, in this embodiment... Figure 2 Based on the embodiments, the process of users controlling home appliances through multi-turn dialogue is described in detail. The method includes:

[0102] S301: In response to the user's second voice, based on the third text content corresponding to the second voice, retrieve the user's fourth behavioral habit data related to the third text content from the behavioral habit database.

[0103] The second voice is the voice following the first voice, and the second voice is used to instruct the control of the first home appliance in at least one home appliance.

[0104] Understandably, after a user issues the first voice command and successfully controls some home appliances, the user may issue a second voice command based on changes in the environment or personal needs. This second voice command, following the first, is used to further adjust or control the first of at least one previously controlled home appliance. In response to this second voice command, the home appliance control device first converts it into third text content for further processing. Then, the control device searches its behavioral habit database for user behavioral habit data related to this third text content—a fourth set of behavioral habit data. This data may include adjustments the user made to the first home appliance in similar situations, and the specific parameters of these adjustments. In this way, the control device can better understand the user's current needs and prepare to perform the corresponding control operation.

[0105] For example, in the first voice command, the user turns on the fan and air conditioner to quickly cool down. After a while, feeling less hot, the user issues a second voice command: "It's not hot anymore, I can turn off the fan and turn up the temperature a bit." The appliance control unit responds to this voice command, converts it into text, and searches its behavioral database for the user's past control habits regarding the fan and air conditioner in similar situations (such as when the weather turns cooler or the indoor temperature drops). The appliance control unit might discover that the user typically turns off the fan and raises the air conditioner temperature to a comfortable level in this situation.

[0106] S302: The third text content and the fourth behavioral habit data are processed by the large-scale model of home appliance control to obtain the second control parameters of the first home appliance.

[0107] Understandably, after acquiring the fourth behavioral habit data related to the third text content, the home appliance control device will utilize a comprehensive home appliance control vertical model to process both the third text content and the fourth behavioral habit data. This model intelligently infers the second control parameters that the first home appliance should be adjusted to based on the user's current commands, past control habits, and possible environmental factors. These control parameters may include the device's on / off status, operating mode, temperature setting, fan speed adjustment, etc. In this way, the home appliance control device can more accurately understand the user's intentions and generate control commands that conform to the user's habits.

[0108] S303: Control the first household appliance according to the second control parameter.

[0109] Understandably, after obtaining the second control parameters, the home appliance control device can control the first home appliance based on these parameters. This means that the control device can send corresponding control commands to the first home appliance, causing it to adjust or operate according to the user's expectations. In this way, users can easily achieve intelligent control of home appliances without manual operation or memorizing complex control steps. Furthermore, because the control device generates control commands based on the user's habits and current needs, the control results are usually more in line with the user's expectations and preferences.

[0110] The home appliance control method provided in this embodiment responds to a user's second voice prompt and, based on the corresponding third text content, retrieves fourth behavioral habit data related to the user's third text content from a behavioral habit database. The second voice prompt follows the first voice prompt and is used to instruct control of the first home appliance among at least one home appliance. The third text content and fourth behavioral habit data are processed using a large-scale home appliance control model to obtain second control parameters for the first home appliance. Based on these second control parameters, the first home appliance is controlled. This method enables intelligent adjustment and optimization of the home appliance control parameters after a single home appliance control operation by capturing and analyzing the text content converted from the user's subsequent voice prompt (i.e., the second voice prompt) and combining it with the user's previously established behavioral habit database. It effectively integrates the user's immediate intent (expressed through the second voice prompt) with long-term behavioral habits, ensuring a high degree of personalization and accuracy in home appliance control.

[0111] In some embodiments, Figure 5 A flowchart illustrating a method for controlling home appliances provided in this application. Figure 4 ,like Figure 5 As shown, in this embodiment... Figure 2 Based on the embodiments, the process of updating the behavior habit database according to user feedback information is described in detail. The method includes:

[0112] S401: In response to the user's third voice, obtain the fourth text content corresponding to the third voice. The third voice is the information fed back by the user after controlling the home appliance based on the first control parameters.

[0113] Understandably, after a user operates a home appliance based on the initial control parameters set by the appliance control device (such as the temperature setting for a smart air conditioner or the brightness adjustment for smart lights), the control device continues to listen for and respond to the user's subsequent voice feedback—the third voice. This third voice is a direct reaction from the user to the appliance's control effect, unlike immediate control commands; it's feedback or a suggestion from the user regarding the device's control performance. Users may use the third voice to express dissatisfaction with the control result, offer suggestions for improvement, or indicate a preference for different control strategies in certain situations. The appliance control device uses voice recognition technology to convert the user's third voice into fourth text content for further analysis and processing. For example, if a user says, "The air conditioner is a bit cold; I'll turn it up a bit next time," the appliance control device will treat this statement as fourth text content, preparing for the next step of processing.

[0114] It's important to note the difference between the second and third voice commands: the second voice command is based on the user's current experience, requesting immediate adjustments to the home appliance to suit their needs; while the third voice command is feedback or a complaint from the user after controlling the device. Although immediate adjustments may not be necessary, the user hopes that the appliance control device can adjust its parameters based on this feedback in similar situations in the future to better suit their personalized habits.

[0115] S402: The fourth text content and the first control parameter are processed by a general large model to obtain the fifth behavior habit data.

[0116] Understandably, home appliance control devices utilize a general large model to comprehensively analyze the user's fourth textual content (i.e., feedback information) and the previously set first control parameters. This large model can deeply understand the user's feedback intent, including dissatisfaction with the control results, specific descriptions of suggestions for improvement, and the control strategies the user hopes the appliance control device will adopt in similar future situations. Simultaneously, the large model also combines the specific values ​​of the first control parameters to infer more realistic and detailed user behavioral habits or preferences. The processed result, namely the fifth behavioral habit data, will contain the user's expectations and preferences for future control strategies, providing strong support for optimizing the user behavioral habit database.

[0117] S403: Update the third behavioral habit data in the behavioral habit database based on the fifth behavioral habit data.

[0118] Understandably, the home appliance control device will update the third behavioral habit data previously stored in the behavioral habit database based on the fifth behavioral habit data obtained from the large model processing. This third behavioral habit data is generated and stored by the user during previous interactions and may contain some initial preferences or habits. However, as user feedback accumulates and the home appliance control device continues to learn, this initial data may need adjustment to more accurately reflect the user's actual habits and expectations. By updating the third behavioral habit data, the home appliance control device can ensure that the stored user habit information is always up-to-date and best meets the user's personalized needs. In this way, in future control processes, the home appliance control device can better meet the user's expectations and preferences, providing more intelligent and personalized services.

[0119] The home appliance control method provided in this embodiment obtains a fourth text content corresponding to a third voice command in response to the user's third voice command. The third voice command is the feedback information after the user controls the home appliance based on a first control parameter. The fourth text content and the first control parameter are processed using a general large model to obtain fifth behavioral habit data. Based on the fifth behavioral habit data, the third behavioral habit data in the behavioral habit database is updated. This method can accurately update and correct the data in the existing behavioral habit database, providing users with more personalized services that better suit their preferences and needs for future home appliance control.

[0120] In some embodiments, Figure 6 A flowchart illustrating a method for controlling home appliances provided in this application. Figure 5 ,like Figure 6 As shown, in this embodiment... Figure 2 Based on the embodiments, the process of updating the behavior habit database according to the user's physical operation records is described in detail. The method includes:

[0121] S501: In response to a user input command for a second home appliance, obtain the second control parameters of the second home appliance.

[0122] Understandably, users operate and input commands into a second home appliance using physical means, such as a remote control or touchscreen. These physical input commands are captured and collected by the appliance's control device. The control device then parses these commands to extract the specific parameters used to control the second appliance, i.e., the second control parameters. These parameters may include the appliance's on / off status, operating mode, temperature setting, fan speed adjustment, etc., which directly reflect the user's intention to operate the appliance through physical means.

[0123] S502: Use the second control parameters to control the second household appliance and obtain the current second environmental parameters.

[0124] Understandably, upon acquiring the second control parameters, the appliance control device will immediately use these parameters to control the second appliance. This means the control device will send corresponding control commands to the second appliance, causing it to adjust or operate according to the user's expectations. Simultaneously, the control device will also acquire current environmental parameters, such as indoor temperature, humidity, and light intensity. These environmental parameters are crucial for understanding the user's control behavior and preferences, as they provide background information about the user's operation of the appliance.

[0125] S503: Through a general large model, it processes instructions, second control parameters, and second environmental parameters to generate sixth-order habit data.

[0126] Understandably, home appliance control devices can utilize a general large-scale model to comprehensively analyze the user's physical input commands, secondary control parameters, and current secondary environmental parameters. This large-scale model can deeply understand the user's control intentions, operational behaviors, and the surrounding environmental context, thereby inferring the user's behavioral habits or preferences in this specific situation. The processed results, namely the sixth behavioral habit data, will include the user's control habits and preferences for the second home appliance, as well as their correlation with environmental parameters. This data is crucial for the system's intelligent control and personalized services.

[0127] S504: Store the sixth behavioral habit data in the behavioral habit database.

[0128] Understandably, home appliance control devices will store the generated sixth behavioral habit data in a behavioral habit database. This way, in similar future scenarios, the control device can automatically adjust the control parameters of a second home appliance based on the stored behavioral habit data to meet the user's expectations and preferences. The continuous updating and improvement of the behavioral habit database will enable home appliance control devices to more accurately understand user control behaviors and preferences, providing more intelligent and personalized services.

[0129] The home appliance control method provided in this embodiment responds to user input commands for a second home appliance, obtains the second control parameters of the second home appliance, uses the second control parameters to control the second home appliance, and obtains the current second environmental parameters. Through a general large model, the commands, second control parameters, and second environmental parameters are processed to generate sixth behavioral habit data. This sixth behavioral habit data is stored in a behavioral habit database. This method can monitor and implement user commands for controlling home appliances using physical means such as remote controls, while simultaneously analyzing and saving the user's physical operation behavioral habits, ensuring the accuracy of the behavioral habit database content.

[0130] For example, the following provides an illustration of methods for controlling home appliances. Figure 7 A flowchart illustrating a method for controlling home appliances provided in this application. Figure 6 ,like Figure 7 As shown, the implementation steps of the method include:

[0131] 1. Speech understanding and intent recognition module;

[0132] This application utilizes a prompt word project to implement natural language processing functionality based on a large model. The large model possesses the ability to handle the diversity and ambiguity of natural language, accurately understanding user needs and even interpreting colloquial commands precisely. During user interaction with the system (i.e., the home appliance control device, hereinafter referred to as the system), voice input is first converted into text data by the Automatic Speech Recognition (ASR) module. The ASR module employs a mature third-party solution, which will not be elaborated upon in this paper. After text conversion, the system uses a Large Language Model (LLM) to perform semantic analysis (NLU) on the user's statements, thereby determining the intent type of the statements (i.e., the user commands described above). The intent types can be categorized into the following three types:

[0133] User Preference Habits Intent: If the system recognizes that the user's statement is intended to express preference information, it will transmit the voice command to the user habit analysis module;

[0134] Home appliance control intent: If the intent is identified as related to home appliance adjustment, the system will transmit the user's voice command to the home appliance vertical domain big model for further processing;

[0135] Other intentions: If the statement is unrelated to home appliance control, the system will invoke the dialogue capabilities of the general large model to engage in natural casual conversation with the user.

[0136] 2. Generate control parameters for a large-scale model of home appliances in the vertical domain;

[0137] The Home Appliance Vertical Domain Model is a model specifically optimized for the home appliance control field. It combines user historical preferences, current voice commands, and environmental parameters to infer and generate optimal home appliance control parameters. Compared to general-purpose models, the Home Appliance Vertical Domain Model offers higher professionalism and accuracy, adapting to the different models and functionalities of various home appliances in a user's home. The Home Appliance Vertical Domain Model achieves customized development based on general-purpose models through prompt word engineering and a Retrieval-Augmented Generation (RAG) knowledge base. The module's input parameters are as follows:

[0138] User voice commands: Specific operational requirements converted into text by the voice understanding module;

[0139] User history preference information: User habit data retrieved from external servers, such as commonly used air conditioning temperature, light brightness, etc.;

[0140] Real-time environmental parameters: such as current indoor temperature, humidity, light intensity and other sensor data.

[0141] The module infers from the input information. When the user says, "Turn the air conditioner to a more comfortable temperature," the model combines the current indoor temperature with the user's historical preferences to adjust the air conditioner temperature to the user's preferred range (e.g., 23°C). When the user says, "I want to watch TV," the model refers to the user's past preference records and the current time to automatically open TV programs or recommended programs that the user frequently watches during the current time period.

[0142] 3. User habit analysis module;

[0143] Figure 8 This application provides a schematic diagram of the user preference update process in a home appliance control method, as shown below. Figure 8 As shown, the user habit analysis module is the core of the system's intelligence. It is responsible for analyzing users' long-term operational behaviors and historical dialogue records, uncovering potential user preference patterns, and continuously optimizing the user experience through dynamic updates. This module primarily utilizes the natural language processing and data analysis capabilities of large-scale models to provide decision support for subsequent home appliance control. This module mainly includes the following aspects:

[0144] 1) Data recording module:

[0145] User physical operation log: Records every home appliance control command from the user, such as air conditioner temperature setting, light brightness adjustment, TV program selection, etc.

[0146] Dialogue information: including voice dialogue records of single-turn and multi-turn interactions, such as "turn on the air conditioner and set it to 26 degrees", "lower it by two degrees";

[0147] Environmental information: By combining sensor data such as indoor and outdoor temperature, humidity, and weather conditions, we analyze the triggering conditions of user preferences.

[0148] 2) Habit Analysis Module:

[0149] High-frequency habit extraction: By analyzing user operation records, this method identifies user appliance operation preferences over a specific time period and utilizes prompt words from a large model. Below is an example prompt word:

[0150] Based on historical conversation records and the current time, we analyze and generate information on users' air conditioning mode usage preferences in different seasons and at different times of the day. For example, if a user selects the heating mode more than 60% of all operation commands during winter nights, it can be inferred that the user prefers to turn on the air conditioner's heating mode on winter nights.

[0151] A user's operation log contains records of each time they control the air conditioner, including the time, the user's request, and the inference answer from the large model. Examples are as follows: 2025-02-24 18:00 User request: I'm home from get off work, it's so stuffy, can you adjust the air conditioner? Large model's inference answer: You feel it's stuffy, probably because you just came in from outside. I will turn on the fresh air function of the air conditioner to alleviate the stuffiness. 2025-02-24 20:00 User request: Adjust the air conditioner mode to heating mode at 28 degrees Celsius. Large model's inference answer: The air conditioner has been adjusted to heating mode at 28 degrees Celsius as requested.

[0152] Contextual Association Analysis: By analyzing user action records and historical dialogue records, this analysis reveals user operational preferences in specific times, environments, and scenarios, and utilizes prompt words from a large-scale model to achieve this. Below is an example prompt word:

[0153] Based on historical conversation records, combined with the current time and the user's current feelings, user preference information is analyzed and generated. For example, if a user expresses a desire to feel hot when they first return home in the summer, and in this scenario, the number of times they use the "cooling mode + 24 degrees + high fan speed" combination command accounts for 60% of all mode adjustments in that scenario, it can be inferred that the user prefers to turn on the air conditioner in cooling mode, set to 24 degrees and high fan speed, when they return home after exercising in the summer.

[0154] Behavior anomaly correction: Corrects erroneous user input commands through large model reasoning. For example, if a user misspoke, saying "I'm so hot, turn up the temperature" but actually meant "turn down the temperature," the system can understand and correct the error by combining the context.

[0155] 4. Preference update module;

[0156] The system automatically updates preference data based on the user's repeated adjustments. For example, if a user frequently sets the air conditioner to a certain temperature range, the system will set that as the default preference. If the user's habits change, the system will also dynamically adjust to ensure the accuracy of personalized services. All preference analysis results are updated in the database for use by other modules.

[0157] Figure 9 This is a schematic diagram of the structure of a home appliance control device 60 provided in this application. Figure 9 As shown, the device includes:

[0158] The processing module 601 is used to respond to the user's first voice and determine the user instruction corresponding to the first voice; when the user instruction is an instruction to input habits, the first voice is processed through a large model to obtain the user's first behavior habit data, and the first behavior habit data is stored in the behavior habit database.

[0159] The acquisition module 602 is used to acquire, in the behavior habit database, second behavior habit data of the user related to the first voice when the user instruction is an instruction that requires control of home appliances;

[0160] The processing module 601 is further configured to process the first voice, the current first environmental parameters, and the second behavioral habit data through a large model to obtain at least one first control parameter of a home appliance.

[0161] The control module 603 is used to control the at least one household appliance according to the first control parameters.

[0162] In one possible implementation, the large model includes a general large model;

[0163] The processing module 601 is further configured to recognize the first speech and obtain the first text content corresponding to the first speech;

[0164] The acquisition module 602 is further configured to acquire a preset prompt word corresponding to the first text content;

[0165] The processing module 601 is further configured to process the first text content and the preset prompt words through the general large model to obtain the user's first behavioral habit data.

[0166] In one possible implementation, the processing module 601 is further configured to recognize the first speech and obtain the second text content corresponding to the first speech.

[0167] The processing module 601 is further configured to process the second text content, the first environmental parameters, the user preference information, and the data of the home appliance through the home appliance control vertical domain large model to obtain at least one first control parameter of the home appliance.

[0168] In one possible implementation, the device further includes: a storage module 604;

[0169] The processing module 601 is further configured to process the second text content, the first environmental parameters, the second behavioral habit data and the first control parameters through a general large model to generate third behavioral habit data;

[0170] The storage module 604 is used to store the third behavioral habit data into the behavioral habit database.

[0171] In one possible implementation, the acquisition module 602 is further configured to, in response to a user's second voice, acquire, based on the third text content corresponding to the second voice, user's fourth behavioral habit data related to the third text content in the behavioral habit database; the second voice is the voice following the first voice, and the second voice is used to instruct the control of the first home appliance among the at least one home appliance.

[0172] The processing module 601 is further configured to process the third text content and the fourth behavioral habit data through the large model of the home appliance control vertical domain to obtain the second control parameters of the first home appliance.

[0173] The control module 603 is also used to control the first home appliance according to the second control parameters.

[0174] In one possible implementation, the acquisition module 602 is further configured to acquire a fourth text content corresponding to the third voice in response to the user's third voice, wherein the third voice is information fed back by the user after controlling the home appliance based on the first control parameters.

[0175] The processing module 601 is further configured to process the fourth text content and the first control parameters through a general large model to obtain the fifth behavior habit data;

[0176] The processing module 601 is further configured to update the third behavioral habit data in the behavioral habit database based on the fifth behavioral habit data.

[0177] In one possible implementation, the acquisition module 602 is further configured to acquire the second control parameters of the second home appliance in response to a user's input command for the second home appliance.

[0178] The control module 603 is also used to control the second home appliance using the second control parameters and to obtain the current second environmental parameters;

[0179] The processing module 601 is also used to process the instruction, the second control parameter, and the second environmental parameter through a general large model to generate sixth behavior habit data;

[0180] The storage module 604 is also used to store the sixth behavioral habit data into the behavioral habit database.

[0181] The home appliance control device 60 provided in this application embodiment can execute the home appliance control method in the above method embodiment. Its implementation principle and technical effects are similar, and will not be repeated here. It should be noted that the above... Figure 9 The division of modules shown is merely illustrative. This application does not limit the division of modules or the naming of modules.

[0182] Figure 10 This is a structural schematic diagram of a home appliance control device 70 provided in this application. Figure 10 As shown, the electronic device 70 provided in this embodiment includes at least one processor 701 and a memory 702. Optionally, the device 70 further includes a communication component 703. The processor 701, memory 702, and communication component 703 are connected via a bus.

[0183] In a specific implementation, at least one processor 701 executes computer execution instructions stored in memory 702, causing at least one processor 701 to perform the above-described method.

[0184] The specific implementation process of processor 701 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0185] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0186] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0187] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0188] This application also provides a computer-readable storage medium, which may include various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk. Specifically, the computer-readable storage medium stores program instructions, which are used in the methods described in the above embodiments.

[0189] This application also provides a program product including execution instructions stored in a readable storage medium. At least one control module of a display device can read the execution instructions from the readable storage medium, and the at least one control module executes the execution instructions to cause the display device to implement the home appliance control methods provided in the various embodiments described above.

[0190] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0191] For ease of explanation, the above description has been provided in conjunction with specific embodiments. However, the above exemplary discussion is not intended to be exhaustive or to limit the embodiments to the specific forms disclosed above. Various modifications and variations can be obtained based on the above teachings. The selection and description of the above embodiments are for the purpose of better explaining the principles and practical applications, thereby enabling those skilled in the art to better utilize the described embodiments and various different variations of embodiments suitable for specific use considerations.

Claims

1. A method for controlling household appliances, characterized in that, The method includes: In response to the user's first voice, determine the user command corresponding to the first voice; When the user instruction is an instruction for recording habits, the first voice is processed through a large model to obtain the user's first behavioral habit data, and the first behavioral habit data is stored in the behavioral habit database. When the user instruction is an instruction that requires control of home appliances, the user's second behavioral habit data related to the first voice is obtained from the behavioral habit database. By processing the first voice, the current first environmental parameters, and the second behavioral habit data using a large model, at least one first control parameter for a home appliance is obtained. The at least one household appliance is controlled according to the first control parameter.

2. The method according to claim 1, characterized in that, The large model includes a general large model; The process of processing the first speech using a large model to obtain the user's first behavioral habit data includes: The first speech is recognized to obtain the first text content corresponding to the first speech. Obtain the preset prompt words corresponding to the first text content; The first text content and the preset prompt words are processed using the general model to obtain the user's first behavioral habit data.

3. The method according to claim 1, characterized in that, The large model includes a large model for the home appliance control vertical domain, and the second behavioral habit data includes user preference information and data on home appliances. The process of processing the first voice, the current first environmental parameters, and the second behavioral habit data using a large model to obtain at least one first control parameter for a home appliance includes: The first speech is recognized to obtain the second text content corresponding to the first speech; By using the large-scale home appliance control model, the second text content, the first environmental parameters, the user preference information, and the data of the home appliances are processed to obtain at least one first control parameter for a home appliance.

4. The method according to claim 3, characterized in that, The method further includes: The second text content, the first environmental parameters, the second behavioral habit data, and the first control parameters are processed using a general large model to generate the third behavioral habit data. The third behavioral habit data is stored in the behavioral habit database.

5. The method according to any one of claims 1-4, characterized in that, The method further includes: In response to the user's second voice, based on the third text content corresponding to the second voice, the user's fourth behavioral habit data related to the third text content is obtained from the behavioral habit database; the second voice is the voice following the first voice, and the second voice is used to instruct the control of the first home appliance among the at least one home appliance. The third text content and the fourth behavioral habit data are processed by the large-scale model of home appliance control to obtain the second control parameters of the first home appliance. The first home appliance is controlled according to the second control parameter.

6. The method according to claim 4, characterized in that, The method further includes: In response to the user's third voice, the fourth text content corresponding to the third voice is obtained, wherein the third voice is the information fed back by the user after controlling the home appliance based on the first control parameters; The fourth text content and the first control parameters are processed using a general large model to obtain the fifth behavior habit data; The third behavioral habit data in the behavioral habit database is updated based on the fifth behavioral habit data.

7. The method according to claim 6, characterized in that, The method further includes: In response to the user's input command for the second home appliance, the second control parameters of the second home appliance are obtained; Using the second control parameters, control the second home appliance and obtain the current second environmental parameters; The instructions, the second control parameters, and the second environmental parameters are processed using a general large model to generate sixth-behavior habit data. The sixth behavioral habit data is stored in the behavioral habit database.

8. A control device for household appliances, characterized in that, The device includes: The processing module is used to respond to the user's first voice and determine the user instruction corresponding to the first voice; when the user instruction is an instruction for recording habits, the first voice is processed through a large model to obtain the user's first behavioral habit data, and the first behavioral habit data is stored in the behavioral habit database. The acquisition module is used to acquire, in the behavior habit database, second behavior habit data of the user related to the first voice when the user instruction is an instruction that requires control of home appliances; The processing module is also used to process the first voice, the current first environmental parameters, and the second behavioral habit data through a large model to obtain at least one first control parameter of a home appliance. The control module is used to control the at least one household appliance according to the first control parameters.

9. An electronic device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions that, when executed by a processor, implement the method described in any one of claims 1-7.

Images