Scene generation method and device, storage medium and electronic device

Abstract

The invention discloses a scene generation method and device, a storage medium and an electronic device, relates to the technical field of smart home / smart home, and aims to solve the problem that smart home scene generation operation is tedious. In order to achieve the purpose, the scene generation method comprises the steps of obtaining a voice instruction of a user, converting the voice instruction into model input information to recognize the intention of the user, responding to the intention of creating a new scene by the user, generating a corresponding scene according to the model input information, a trained scene extraction model and a preset scene rule, the scene extraction model is used for extracting an equipment action and an action condition from the model input information, and the corresponding scene comprises a scene triggered by the action condition, a scene triggered by a user and a recommended scene. According to the method, the user does not need to execute complex operation, the corresponding scene can be quickly generated only by issuing the voice instruction by the user, the problem that the scene generation operation of the smart home is tedious is solved, and the effect of improving the user experience is achieved.

Description

Technical Field

[0001] The present application relates to the field of smart home / intelligent family technology, and more specifically, to a scene generation method, device, storage medium and electronic device. Background Art

[0002] At present, the smart home scene generation and control technologies mainly include manual settings, automatic triggering based on sensors, and remote control through mobile phone applications. Manual settings require users to memorize a large number of specific instructions and then adjust the parameters of each device one by one. The operation is cumbersome and prone to errors. Although automatic triggering based on sensors has achieved automation to a certain extent, the arrangement and debugging of sensors are relatively complicated, and the flexibility of the scene is limited. Remote control through mobile phone applications requires users to open the mobile phone application, find the device and operate it. The operation is cumbersome and the configuration is not flexible enough, which makes it difficult to meet the personalized needs of users.

[0003] Accordingly, the art needs a new scene generation solution to solve the above problems. Summary of the invention

[0004] In order to overcome the above-mentioned defects, the present application is proposed to solve or at least partially solve the problem of cumbersome operation of smart home scene generation.

[0005] In a first aspect, a scene generation method is provided, the method comprising: obtaining a user's voice command; converting the voice command into model input information to identify the user's intention; in response to the user's intention to create a new scene, generating a corresponding scene according to the model input information, a trained scene extraction model and preset scene rules, the scene extraction model being used to extract device actions and action conditions from the model input information, the corresponding scenes comprising scenes triggered by the action conditions, scenes triggered by the user and recommended scenes.

[0006] In a technical solution of the above-mentioned scene generation method, the corresponding scene is generated according to the model input information, the trained scene extraction model and the preset scene rules, including: in response to the scene extraction model extracting the device action and the corresponding action conditions from the model input information, generating a scene in which the device action is automatically triggered by the action condition.

[0007] In a technical solution of the above-mentioned scene generation method, the corresponding scene is generated according to the model input information, the trained scene extraction model and the preset scene rules, and also includes: in response to the scene extraction model extracting a device action from the model input information and not including the corresponding action condition, generating a scene triggered by the user according to the device action.

[0008] In a technical solution of the above-mentioned scene generation method, the generating of the corresponding scene according to the model input information, the trained scene extraction model and the preset scene rules also includes: in response to the scene extraction model failing to extract the device action from the model input information, generating a recommended scene according to the model input information, the device and the usage frequency of the device action; pushing the recommended scene to the user and obtaining user feedback; executing the recommended scene according to the user feedback or re-executing the user's voice command, converting the voice command into model input information to identify the user's intention and, in response to the user's intention to create a new scene, generating the corresponding scene according to the model input information, the trained scene extraction model and the preset scene rules.

[0009] In a technical solution of the above-mentioned scenario generation method, the generation of recommended scenarios based on the model input information, devices, and the frequency of use of device actions includes: screening devices that match the model input information and / or devices with high usage frequency; obtaining historical device actions of the screened devices; screening device actions that match the model input information and / or are with high usage frequency from the historical device actions; and generating recommended scenarios based on the screened device actions.

[0010] In a technical solution of the above-mentioned scene generation method, the method also includes: obtaining model input information used by the user and the corresponding generated scene; training the scene extraction model based on the model input information used by the user and the corresponding generated scene.

[0011] In a technical solution of the above-mentioned scene generation method, the method also includes: adjusting the usage frequency of the device and the device action according to the user feedback.

[0012] In a second aspect, a scene generation device is provided, the device comprising: an instruction acquisition module for acquiring a user's voice instructions; an instruction conversion module for converting the voice instructions into model input information to identify the user's intention; a scene generation module, in response to the user's intention to create a new scene, generates a corresponding scene according to the model input information, a trained scene extraction model and preset scene rules, the scene extraction model is used to extract device actions and action conditions from the model input information, the corresponding scenes include scenes triggered by the action conditions, scenes triggered by the user and recommended scenes.

[0013] In a technical solution of the above-mentioned scene generation device, the scene generation module includes: a first generation unit, which extracts device actions and corresponding action conditions from the model input information in response to the scene extraction model, and generates a scene in which the device action is automatically triggered by the action conditions.

[0014] In a technical solution of the above-mentioned scene generation device, the scene generation module also includes: a second generation unit, in response to the scene extraction model extracting a device action from the model input information without including a corresponding action condition, generating a scene triggered by a user according to the device action.

[0015] In a technical solution of the above-mentioned scene generation device, the scene generation module also includes: a third generation unit, in response to the scene extraction model failing to extract the device action from the model input information, generating a recommended scene based on the model input information, the device and the usage frequency of the device action; a push unit, used to push the recommended scene to the user and obtain user feedback; an execution unit, used to execute the recommended scene based on the user feedback or re-execute the user's voice command, convert the voice command into model input information to identify the user's intention, and in response to the user's intention to create a new scene, generate a corresponding scene based on the model input information, the trained scene extraction model and the preset scene rules.

[0016] In a technical solution of the above-mentioned scene generation device, the third generation unit includes: a first screening subunit, used to screen devices that match the model input information and / or devices that are used frequently; an acquisition subunit, used to acquire historical device actions of the screened devices; a second screening subunit, used to screen device actions that match the model input information and / or are used frequently from the historical device actions; and a generation subunit, used to generate recommended scenes based on the screened device actions.

[0017] In a technical solution of the above-mentioned scene generation device, the device also includes: a historical data acquisition module, used to obtain the model input information used by the user and the corresponding generated scenes; a model training module, used to train the scene extraction model based on the model input information used by the user and the corresponding generated scenes.

[0018] In a technical solution of the above-mentioned scene generation device, the device also includes: a frequency adjustment module, which is used to adjust the usage frequency of the device and the device action according to the user feedback.

[0019] In a third aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium includes a stored program, wherein when the program is run, the scene generation method in the above-mentioned first aspect or any corresponding technical solution thereof is executed.

[0020] In a fourth aspect, an electronic device is provided, comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the scene generation method in the above-mentioned first aspect or any corresponding technical solution thereof through the computer program.

[0021] The above one or more technical solutions of the present application have at least one or more of the following beneficial effects:

[0022] In the technical solution of the present application, the user's voice command is obtained, converted into model input information, and the user's intention is identified therefrom. When the user's intention to create a new scene is obtained, the device action and action conditions are extracted from the converted model input information according to the trained scene extraction model, and the corresponding scene is generated according to the preset scene rules. This method does not require the user to perform complex operations, but only requires the user to issue a voice command to quickly generate the corresponding scene, which solves the problem of cumbersome operation of smart home scene generation and achieves the effect of improving the user experience.

[0023] In the technical solution for implementing the present application, a scene extraction model is trained based on the model input information used by the user and the corresponding generated scenes, thereby achieving the purpose of considering the user's historical behavior and usage preferences in the process of generating scenes, thereby achieving personalized scene customization based on user needs and preferences, which is conducive to providing users with more intelligent and personalized services.

[0024] In implementing the technical solution of the present application, by receiving user feedback and adjusting the frequency of use of devices and device actions according to user feedback, dynamic adjustment and optimization of scene rules are achieved, thereby achieving the purpose of further improving the accuracy of scene generation and personalized needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0026] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0027] Figure 1 is a schematic diagram of a hardware environment of an interaction method of a smart device according to an embodiment of the present application;

[0028] Figure 2 is a schematic flow chart of the main steps of a scene generation method according to an embodiment of the present application;

[0029] Figure 3 is a schematic diagram of the overall framework of a scene generation method according to an embodiment of the present application;

[0030] Figure 4is a schematic diagram of the main structural block diagram of a scene generation device according to an embodiment of the present application;

[0031] Figure 5 It is a schematic diagram of the connection relationship between a processor and a memory of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0032] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0033] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices. The terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected, it can be indirectly connected through an intermediate medium, it can also be the internal communication of two elements, it can be a wireless connection, or it can be a wired connection.

[0034] In addition, "module" and "processor" may include hardware, software or a combination of the two. A module may include hardware circuits, various suitable sensors, communication ports, and memories, and may also include software parts, such as program codes, or a combination of software and hardware. The processor may be a central processing unit, a microprocessor, an image processor, a digital signal processor, or any other suitable processor. The processor has data and / or signal processing functions. The processor may be implemented in software, hardware, or a combination of the two. Computer-readable storage media include any suitable media that can store program codes, such as magnetic disks, hard disks, optical disks, flash memory, read-only memory, random access memory, etc.

[0035] In addition, if the meaning of "and / or" appears in this application, it includes three parallel solutions. Taking "A and / or B" as an example, it includes solution A, solution B, or a solution that satisfies both A and B. In addition, the technical solutions between the various embodiments can be combined with each other, but it must be based on the ability of ordinary technicians in this field to implement. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by this application. The term "at least one A or B" or "at least one of A and B" has a similar meaning to "A and / or B" and can include only A, only B, or A and B. The singular terms "one" and "this" can also include plural forms.

[0036] The relevant user personal information that may be involved in the various embodiments of this application is strictly in accordance with the requirements of laws and regulations, following the principles of legality, legitimacy and necessity, based on the reasonable purposes of business scenarios, to process the personal information that users actively provide during the use of products / services or generated due to the use of products / services, as well as the personal information obtained with the user's authorization.

[0037] This application attaches great importance to the security of user personal information and has taken reasonable and feasible security protection measures that meet industry standards to protect user information and prevent personal information from being accessed, disclosed, used, modified, damaged or lost without authorization.

[0038] According to one aspect of an embodiment of the present application, a method for interacting with a smart home device is provided. The method for interacting with a smart home device is widely used in smart home, smart home, smart home device ecosystem, smart home ecosystem and other whole-house intelligent digital control application scenarios. Optionally, in this embodiment, the method for interacting with the smart home device can be applied to Figure 1 In the hardware environment composed of the terminal device 102 and the server 104 shown in FIG. Figure 1 As shown, the server 104 is connected to the terminal device 102 via a network, and can be used to provide services (such as application services, etc.) for the terminal or a client installed on the terminal. A database can be set on the server or independently of the server to provide data storage services for the server 104. Cloud computing and / or edge computing services can be configured on the server or independently of the server to provide data computing services for the server 104.

[0039] The network may include but is not limited to at least one of the following: wired network, wireless network. The wired network may include but is not limited to at least one of the following: wide area network, metropolitan area network, local area network, and the wireless network may include but is not limited to at least one of the following: WIFI (Wireless Fidelity), Bluetooth. The terminal device 102 may be but is not limited to a PC, a mobile phone, a tablet computer, a smart air conditioner, a smart range hood, a smart refrigerator, a smart oven, a smart stove, a smart washing machine, a smart water heater, a smart washing device, a smart dishwasher, a smart projection device, a smart TV, a smart clothes drying rack, a smart curtain, a smart audio and video, a smart socket, a smart speaker, a smart fresh air device, a smart kitchen and bathroom device, a smart bathroom device, a smart sweeping robot, a smart window cleaning robot, a smart mopping robot, a smart air purification device, a smart steamer, a smart microwave oven, a smart kitchen treasure, a smart purifier, a smart water dispenser, a smart door lock, etc.

[0040] According to one aspect of the embodiments of the present application, a scene generation method is provided. Figure 2 , Figure 2 FIG. 1 is a flow chart of the main steps of a scene generation method according to an embodiment of the present application. Figure 2 As shown, the scene generation method mainly includes the following steps S201 to S203:

[0041] Step S201, obtaining the user's voice command.

[0042] In this embodiment, the user's voice command (also called natural language command) refers to a command or instruction issued by the user in the form of natural language, which is used to instruct a computer or other intelligent system to perform a specific task. For example, the user issues a command to the system by speaking, such as asking it to perform operations such as playing music, querying information, setting reminders, etc.

[0043] Step S202: convert the voice command into model input information to identify the user's intention.

[0044] In this embodiment, if the machine wants to process the user's voice command, it first needs to convert it into model input information, such as text, vector, matrix and other input information that the model can recognize and process, and then identify the user's intention through natural language processing technology.

[0045] In one implementation, assuming that the model input information is text information, that is, converting the voice command into corresponding text information, advanced voice recognition algorithms and models can be used to improve the accuracy and speed of voice recognition, ensuring that the user's voice command can be accurately converted into text information. After obtaining the text information, methods such as rule template parsing and deep learning models can be used to identify the user's intention from the text information. Among them, the methods in the prior art can be referred to for voice recognition and user intention recognition, which will not be repeated here.

[0046] Step S203, in response to the user's intention to create a new scene, generate a corresponding scene based on the model input information, the trained scene extraction model and the preset scene rules. The scene extraction model is used to extract device actions and action conditions from the model input information. The corresponding scenes include scenes triggered by action conditions, scenes triggered by users and recommended scenes.

[0047] In this embodiment, when it is recognized that the instruction issued by the user is intended to create a new scene, the trained scene extraction model is used to extract the device action and action conditions from the converted model input information, and the scene corresponding to the user's instruction is generated based on the device action, action conditions and pre-defined scene generation rules (preset scene rules). Among them, the scene refers to the sum of a series of device states, which can include one or more device actions, and any device action can also have corresponding action conditions. Among them, the generated scenes include but are not limited to scenes triggered by action conditions (automatically), scenes triggered by users (manually), and recommended scenes generated based on model input information (such as recommended device actions).

[0048] In one implementation, a device action is an operation performed by a device, and an action condition is a condition for performing the device action. For example, "Turn on the air conditioner when someone is at home", the device action is "Turn on the air conditioner", and the corresponding action condition is "Someone is at home". That is, the device action will be performed only when the action condition is met. It is understandable that an action condition can correspond to one or more device actions, and a device action can also have multiple action conditions.

[0049] In one embodiment, the device action also includes action parameters. For example, the device action of "turn on the air conditioner" may also include the operating mode and temperature of the air conditioner, wherein the operating mode and temperature are the action parameters.

[0050] In one implementation, a scene may be generated by a terminal and sent to other smart home devices communicating with the terminal in the form of a scene configuration file.

[0051] This embodiment obtains the user's voice command, converts it into model input information, and identifies the user's intention from it. When the user's intention to create a new scene is obtained, the device action and action conditions are extracted from the converted model input information according to the trained scene extraction model, and the corresponding scene is generated according to the preset scene rules. This method does not require the user to perform complex operations, but only requires the user to issue a voice command to quickly generate the corresponding scene, which solves the problem of cumbersome operations in generating smart home scenes and achieves the effect of improving the user experience.

[0052] In one implementation of the embodiment of the present application, the above step S203 may further include the following steps S2031 to S2035:

[0053] Step S2031, in response to the scene extraction model extracting the device action and the corresponding action condition from the model input information, a scene is generated in which the device action is automatically triggered by the action condition.

[0054] In this embodiment, when the scene extraction model extracts the device action and the corresponding action condition from the model input information, an automatic triggering scene consisting of the "action condition" and the "device action" can be generated. That is, when the "action condition" is met, the corresponding "device action" is automatically triggered.

[0055] Step S2032, in response to the scene extraction model extracting the device action from the model input information and not including the corresponding action condition, generating a scene triggered by the user according to the device action.

[0056] In this embodiment, when the scene extraction model only extracts the device action from the model input information but does not extract the corresponding action condition, a scene triggered by the user can be generated based on the extracted "device action". For example, when the user says "play a song", the scene extraction model extracts the device action "play music" based on the model input information, but it does not contain the action condition, which means that the next time "play music" is executed, it still needs to rely on the user's instruction to trigger.

[0057] In one implementation, the device may feed back the generated scene to the user through methods such as screen display or voice announcement, such as displaying the song being played on the screen or prompting the user "playing xxx song for you now".

[0058] Step S2033, in response to the scenario extraction model failing to extract the device action from the model input information, generating a recommended scenario based on the model input information, the device, and the usage frequency of the device action.

[0059] In this embodiment, when there is no obvious device action in the model input information, for example, when the user says "It's so hot", although the instruction does not contain an obvious action, common sense shows that the device to be controlled is the air conditioner or fan, and the device action to be executed is "turn on the air conditioner" or "turn on the fan". Furthermore, the device action to be executed can be determined based on the frequency of use of the air conditioner and fan by the user.

[0060] In one implementation, a common sense base or knowledge graph may be integrated into the scene extraction model to achieve the purpose of matching device actions based on model input information when the instructions do not contain obvious device actions.

[0061] In one embodiment, the recommended scenario may include multiple device actions. For example, if the user's instruction received is "It's so hot", the generated recommended scenario already includes the device action of "Turn on the air conditioner". One or more other device actions may be added to the scenario based on the frequency of use of the device and the device action.

[0062] In an optional implementation, the above step S2033 may include the following steps S20331 to S20334:

[0063] Step S20331, screening devices that match the model input information and / or devices with high usage frequency.

[0064] In this embodiment, still taking the instruction "It's too hot" as an example, the device that matches the model input information is an air conditioner. Assuming that based on the user's historical usage of the device, it is known that speakers and smart curtains are household devices that the user uses more frequently, the air conditioner, speakers and smart curtains can be used together to generate recommended scenes for the user.

[0065] Step S20332, obtaining the historical device actions of the filtered devices.

[0066] In this embodiment, based on the device screened in step S20331, the device actions during its historical use are obtained.

[0067] Step S20333: Filter out device actions that match the model input information and / or are frequently used from historical device actions.

[0068] In this embodiment, frequently used device actions are screened from historical device actions. For example, the most frequently used device action for air conditioners is "turn on the air conditioner to cooling mode 26 degrees"; the most frequently used device action for speakers is "play music by xx band"; and the most frequently used device action for smart curtains is "close the curtains."

[0069] Step S20334: Generate recommended scenarios based on the filtered device actions.

[0070] In this embodiment, based on the device actions that match the model input information and / or are frequently used in step S20333, the following recommended scenarios can be generated for the user: "Turn on the air conditioner to cooling mode 26 degrees", "Play music by xx band", "Close the curtains".

[0071] Step S2034: Push the recommended scenario to the user and obtain user feedback.

[0072] In this embodiment, the generated recommended scenarios are pushed to the user and user feedback is obtained. For example, the user is asked whether he needs help generating scenarios such as "turn on the air conditioner to cooling mode 26 degrees", "play music from xx band", and "close the curtains". The user feedback can be an affirmative or negative answer to the recommended scenario.

[0073] Step S2035, execute the recommended scenario based on user feedback or re-execute to obtain the user's voice command, convert the voice command into model input information to identify the user's intention and respond to the user's intention to create a new scenario, and generate the corresponding scenario based on the model input information, the trained scenario extraction model and the preset scenario rules.

[0074] In this embodiment, if the recommended scenario is pushed to the user and the user gives an affirmative answer, the recommended scenario can be executed; if the recommended scenario is pushed to the user and the user gives a negative answer, it is necessary to re-acquire the user's voice command and generate a corresponding scenario for it, that is, execute steps S201 to S203 again.

[0075] In one embodiment, the usage frequency of the device and the device action can be adjusted according to user feedback. Specifically, if the recommended scenario is rejected by the user, the usage frequency of the device and the device action included in the recommended scenario can be reduced, or a recommendation coefficient can be set for the device and the device action, and the usage frequency of the device and the device action in the recommended scenario can be adjusted using the recommendation coefficient.

[0076] In one implementation of the embodiment of the present application, after the above step S203, step S204 is further included, which specifically includes the following steps S2041 and S2042:

[0077] Step S2041, obtaining the model input information used by the user and the corresponding generated scene.

[0078] Step S2042: training a scene extraction model based on the model input information used by the user and the corresponding generated scenes.

[0079] In this embodiment, a scene extraction model is trained based on the model input information used by the user and the corresponding generated scenes, so that the user's historical behavior and usage preferences are taken into consideration in the process of generating scenes, thereby realizing personalized scene customization according to user needs and preferences, which is conducive to providing users with more intelligent and personalized services.

[0080] In an example of an application scenario of the present application, Figure 3 FIG. 1 is a schematic diagram of the overall framework of a scene generation method according to an embodiment of the present application. Figure 3 As shown, it is assumed that multiple smart home devices communicate with a terminal, and the terminal generates corresponding scenes according to the user's voice command. The specific process is as follows:

[0081] First, when the terminal detects that the user initiates a wake-up command, it uses automatic speech recognition (ASR) to convert the audio of the user interaction into text, that is, to convert the voice command into text information (text); then it uses natural language processing to perform intent analysis and slot extraction, that is, to identify the user intent, device action and action conditions from the text information; then it determines whether it is the intention to create a new scene. If not, it exits the scene creation process; if it is the intention to create a new scene, it further determines whether the user instruction (text information) contains an action (device action) and whether the user instruction contains a condition (action condition), which specifically includes the following situations:

[0082] The user instruction contains an action and a condition. At this time, an automatic triggering scenario can be generated, that is, a scenario in which the action condition automatically triggers the device action. For details, please refer to step S2031 of the above embodiment, which will not be repeated here.

[0083] The user instruction contains an action but no condition. In this case, a manually triggered scene, i.e., a scene triggered by the user, can be generated and previewed. The preview is to display the generated scene to the user. For details, please refer to step S2032 of the above embodiment, which will not be repeated here.

[0084] If the user instruction does not contain an action, an action can be recommended to the user based on the user's instruction and the user's commonly used devices, such as asking the user "Do you need help generating a scene of <device action 1><device action 2><device action 3>?" For details, please refer to step S2033 of the above embodiment, which will not be repeated here. It should be noted that the number of actions in the recommended scene can be set according to the specific situation, and this embodiment does not impose specific restrictions on this.

[0085] After generating the recommended scenario, it is also necessary to obtain user feedback and determine whether the user's answer is a positive intention. If it is a positive intention, generate the scene and perform a preview; if it is a negative intention, exit the (current) scene creation process, obtain the user's voice command again and re-execute the scene creation process, which is the above-mentioned scene generation method.

[0086] In this embodiment, speech recognition technology is used to process voice commands, thereby achieving the purpose of improving the accuracy and speed of speech recognition; natural language processing technology is used to parse and understand the text information output by speech recognition, identify user needs and intentions, and achieve the effect of improving the accuracy and flexibility of scene generation; through the preset scene algorithm, the purpose of quickly generating corresponding scenes according to user needs and intentions is achieved, thereby improving the efficiency and quality of scene generation; the user's needs and intentions are combined with the user's historical behavior data and preference settings to meet the user's needs for personalized scenes; according to user feedback and adjustment of scene generation rules, the scene extraction model is trained using historical data, thereby achieving the effect of improving the adaptability and personalization level of the generated scenes.

[0087] It should be pointed out that although the various steps in the above embodiments are described in a specific order, those skilled in the art can understand that in order to achieve the effect of the present application, different steps do not have to be executed in such an order. They can be executed simultaneously (in parallel) or in other orders. These adjusted schemes are equivalent to the technical schemes described in this application, and therefore will also fall within the scope of protection of this application.

[0088] Another aspect of the present application also provides a scene generation device, such as Figure 4 As shown, the device includes: an instruction acquisition module 401, used to acquire the user's voice instructions; an instruction conversion module 402, used to convert the voice instructions into model input information to identify the user's intention; a scene generation module 403, in response to the user's intention to create a new scene, generates a corresponding scene according to the model input information, a trained scene extraction model and preset scene rules, the scene extraction model is used to extract device actions and action conditions from the model input information, and the corresponding scenes include scenes triggered by action conditions, scenes triggered by users and recommended scenes.

[0089] In a technical solution of the above-mentioned scene generation device, the scene generation module includes: a first generation unit, in response to the scene extraction model, extracts the device action and the corresponding action condition from the model input information, and generates a scene in which the device action is automatically triggered by the action condition.

[0090] In a technical solution of the above-mentioned scene generation device, the scene generation module also includes: a second generation unit, in response to the scene extraction model extracting the device action from the model input information without including the corresponding action condition, generating a scene triggered by the user according to the device action.

[0091] In a technical solution of the above-mentioned scene generation device, the scene generation module also includes: a third generation unit, in response to the scene extraction model failing to extract the device action from the model input information, generating a recommended scene based on the model input information, the device and the frequency of use of the device action; a push unit, used to push the recommended scene to the user and obtain user feedback; an execution unit, used to execute the recommended scene based on the user feedback or re-execute the user's voice command, convert the voice command into model input information to identify the user's intention, and in response to the user's intention to create a new scene, generate a corresponding scene based on the model input information, the trained scene extraction model and the preset scene rules.

[0092] In a technical solution of the above-mentioned scene generation device, the third generation unit includes: a first screening subunit, used to screen devices that match the model input information and / or devices that are used frequently; an acquisition subunit, used to obtain historical device actions of the screened devices; a second screening subunit, used to screen device actions that match the model input information and / or are used frequently from historical device actions; and a generation subunit, used to generate recommended scenarios based on the screened device actions.

[0093] In a technical solution of the above-mentioned scene generation device, the device also includes: a historical data acquisition module, used to obtain model input information used by the user and the corresponding generated scenes; a model training module, used to train the scene extraction model based on the model input information used by the user and the corresponding generated scenes.

[0094] In a technical solution of the above-mentioned scene generation device, the device also includes: a frequency adjustment module, which is used to adjust the usage frequency of the device and the device action according to user feedback.

[0095] In an example of an application scenario of the present application, the scene generation device is designed to enable users to conveniently and efficiently generate and control various scenes through voice commands. The core logic is to convert the user's complex voice commands into executable scene configurations, and operate the relevant smart devices through the device control module to achieve the scene effect expected by the user. As an example, the instruction acquisition module is used to collect user interaction audio (user voice commands) after the device receives a wake-up event. The instruction conversion module is used to perform preliminary filtering, noise reduction and other processing on the user's voice commands to obtain the processed voice signal, and then perform feature extraction on it, including features such as phonemes, syllables, speaking speed, and intonation, and finally convert it into text (model input information). The scene generation module uses natural language processing technology to semantically understand the text, parse out the user's needs and intentions, and then match the corresponding scene modes and parameters according to the parsed user needs, combined with preset scene rules, to generate a scene configuration file, wherein the scene includes (necessary) actions and (non-necessary) conditions.

[0096] In one embodiment, the apparatus further includes a scene execution module, which is used to read the generated scene configuration file and send the control instructions therein to the multiple smart devices controlled by it.

[0097] In one embodiment, the device further includes a communication module for sending a control signal to a corresponding smart device according to a unified device control protocol after the device control module receives a control instruction, thereby implementing operations such as switching on and off the smart device and adjusting parameters.

[0098] In one embodiment, the device further includes a feedback optimization module for obtaining feedback results and optimizing and adjusting the scene generation rules and algorithms to improve the accuracy and adaptability of subsequent scene generation.

[0099] It is understood by those skilled in the art that all or part of the processes in the method for implementing the above-mentioned embodiment of the present application can also be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable storage medium may include: any entity or device, medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory, random access memory, electrical carrier signal, telecommunication signal and software distribution medium, etc. that can carry the computer program code.

[0100] Another aspect of the present application also provides a computer-readable storage medium.

[0101] In an embodiment of a computer-readable storage medium according to the present application, the computer-readable storage medium may be configured to store a program for executing the scene generation method of the above-mentioned method embodiment, and the program may be loaded and run by a processor to implement the scene generation method in the above-mentioned first aspect or any technical solution corresponding thereto. For ease of explanation, only the parts related to the embodiment of the present application are shown. For specific technical details not disclosed, please refer to the method part of the embodiment of the present application. The computer-readable storage medium may be a storage device formed by various electronic devices. Optionally, the computer-readable storage medium in the embodiment of the present application is a non-temporary computer-readable storage medium.

[0102] Another aspect of the present application provides an electronic device.

[0103] In an embodiment of an electronic device according to the present application, the electronic device may include at least one processor; and a memory connected to the at least one processor; wherein a computer program is stored in the memory, and when the computer program is executed by the at least one processor, the scene generation method in the first aspect or any corresponding technical solution thereof is implemented. The electronic device described in the present application may include a PC, a mobile phone, a tablet computer, a smart speaker, a smart washing machine, a smart projection device, a smart TV, etc. See attached Figure 5 , Figure 5 FIG. 4 exemplarily shows that the memory 11 and the processor 12 are communicatively connected via a bus.

[0104] In some embodiments of the present application, the electronic device may further include at least one sensor, and the sensor is used to sense information. The sensor is connected to any type of processor mentioned in the present application. The electronic device described in the present application may be, but is not limited to, a PC, a mobile phone, a tablet computer, a smart air conditioner, a smart range hood, a smart refrigerator, a smart oven, a smart stove, a smart washing machine, a smart water heater, a smart washing equipment, a smart dishwasher, a smart projection equipment, a smart TV, a smart clothes drying rack, a smart curtain, a smart audio and video, a smart socket, a smart speaker, a smart speaker, a smart fresh air equipment, a smart kitchen and bathroom equipment, a smart bathroom equipment, a smart sweeping robot, a smart window cleaning robot, a smart mopping robot, a smart air purification equipment, a smart steamer, a smart microwave oven, a smart kitchen treasure, a smart purifier, a smart water dispenser, a smart door lock, etc.

[0105] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A scene generation method, characterized in that: The method comprises: Get the user's voice command; Converting the voice command into model input information to identify the user's intention; In response to the user's intention to create a new scene, a corresponding scene is generated according to the model input information, a trained scene extraction model and preset scene rules. The scene extraction model is used to extract device actions and action conditions from the model input information. The corresponding scenes include scenes triggered by the action conditions, scenes triggered by the user and recommended scenes.

2. The scene generation method according to claim 1, characterized in that: Generating a corresponding scene according to the model input information, the trained scene extraction model and the preset scene rules includes: In response to the scenario extraction model extracting the device action and the corresponding action condition from the model input information, a scenario is generated in which the device action is automatically triggered by the action condition.

3. The scene generation method according to claim 1, characterized in that: The generating of the corresponding scene according to the model input information, the trained scene extraction model and the preset scene rules also includes: In response to the scenario extraction model extracting a device action from the model input information without including a corresponding action condition, a scenario triggered by a user is generated according to the device action.

4. The scene generation method according to claim 1, characterized in that: The generating of the corresponding scene according to the model input information, the trained scene extraction model and the preset scene rules also includes: In response to the scenario extraction model failing to extract a device action from the model input information, generating a recommended scenario based on the model input information, a device, and a usage frequency of the device action; Pushing the recommended scenario to the user and obtaining user feedback; The steps of executing the recommended scenario or re-executing the user's voice command according to the user feedback, converting the voice command into model input information to identify the user's intention and responding to the user's intention to create a new scenario, and generating the corresponding scenario according to the model input information, the trained scenario extraction model and the preset scenario rules.

5. The scene generation method according to claim 4, characterized in that: Generating the recommended scenario according to the model input information, the device, and the usage frequency of the device action includes: Screening devices that match the model input information and / or devices with high usage frequency; Get the historical device actions of the filtered devices; Filtering, from the historical device actions, device actions that match the model input information and / or are frequently used; Generate recommended scenarios based on the filtered device actions.

6. The scene generation method according to claim 1, characterized in that: The method further comprises: Obtain the model input information used by the user and the corresponding generated scenes; The scene extraction model is trained based on the model input information used by the user and the corresponding generated scenes.

7. The scene generation method according to claim 4, characterized in that: The method further comprises: The usage frequency of the device and the device actions is adjusted according to the user feedback.

8. A scene generation device, characterized in that: The device comprises: A command acquisition module, used to acquire the user's voice command; An instruction conversion module, used to convert the voice instruction into model input information to identify the user's intention; The scenario generation module, in response to the user's intention to create a new scenario, generates a corresponding scenario based on the model input information, the trained scenario extraction model and the preset scenario rules. The scenario extraction model is used to extract device actions and action conditions from the model input information. The corresponding scenarios include scenarios triggered by the action conditions, scenarios triggered by the user and recommended scenarios.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein the program executes the scene generation method according to any one of claims 1 to 7 when executed.

10. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to execute the scene generation method according to any one of claims 1 to 7 through the computer program.

Images