Off-line voice control method, device, equipment, medium and product

Abstract

The invention relates to the technical field of voice control, in particular to an offline voice control method and device, equipment, a medium and a product, and the method comprises the steps: obtaining target intelligent equipment of an area where a user is located; inputting the target intelligent equipment into a pre-constructed user habit model to obtain a control equipment list; performing off-line identification on a voice control instruction sent by a user, and controlling the control equipment list based on an identification result; offline voice control can be efficiently carried out based on low-power-consumption equipment.

Description

Technical Field

[0001] The present invention relates to the field of voice control technology, and in particular to an offline voice control method, device, equipment, medium and product. Background Art

[0002] With the rapid development of Internet of Things technology, voice control has become an important human-computer interaction method in smart homes, wearable devices and other fields.

[0003] Although online algorithms that rely on network connections for processing can provide rich functions and highly accurate recognition, they are often limited by network stability, latency, and user privacy protection issues. In contrast, offline voice control can provide faster response speeds and better privacy protection, but traditionally, offline voice control requires devices to have high computing power and energy efficiency to handle complex voice recognition and command execution.

[0004] There is a technical problem in this field that it is difficult to efficiently perform offline voice control based on low-power devices. Summary of the invention

[0005] The present invention provides an offline voice control method, device, equipment, medium and product, which solve the technical problem that it is difficult to perform offline voice control based on low-power consumption equipment.

[0006] In a first aspect, the present invention provides an offline voice control method, the method comprising: obtaining target smart devices in the user's area; inputting the target smart devices into a pre-built user habit model to obtain a control device list; performing offline recognition of the voice control instructions issued by the user, and controlling the control device list based on the recognition results.

[0007] In some embodiments, the step of obtaining the target smart device in the user's area includes: obtaining the distance between the user terminal and each smart device, the distance between each smart device and / or the distance between the user and the voice control device; determining the user's area based on the distance between the user terminal and each smart device, the distance between each smart device and / or the distance between the user and the voice control device; and using the smart devices in the user's area as target smart devices.

[0008] In some embodiments, the pre-built user habit model includes: building a user habit model based on the user's voice control history, wherein the voice control history includes: the frequency of use of the smart device, the time period of use of the smart device, the user's location when the smart device is used, and / or scene mode.

[0009] In some embodiments, after the step of obtaining a control device list, the method further includes: sending the control device list to a user terminal; receiving feedback from the user terminal on the control device list, wherein the feedback includes adjusting the order of devices, adding devices and / or deleting devices; and using the control device list fed back by the user terminal as the final control device list.

[0010] In some embodiments, the voice control instructions issued by the user are recognized offline, and the step of controlling the control device list based on the recognition result includes: performing synonym conversion on the voice control instructions issued by the user; and performing offline recognition based on the voice control instructions after the synonym conversion.

[0011] In some embodiments, the voice control instructions issued by the user are recognized offline, and the step of controlling the control device list based on the recognition result also includes: using the control instruction of the first device in the control device list as the control instruction of the control device list; and / or setting the smart devices outside the control device list to standby mode.

[0012] In a second aspect, the present invention provides an offline voice control device, which includes: an acquisition module for acquiring target smart devices in the user's area; a processing module for inputting the target smart devices into a pre-built user habit model to obtain a control device list; and a recognition module for offline recognition of voice control instructions issued by the user, and controlling the control device list based on the recognition results.

[0013] In a third aspect, the present invention provides a computer device, comprising a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of any of the offline voice control methods in the above aspects.

[0014] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the offline voice control methods in the above aspects.

[0015] In a fifth aspect, the present invention provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of any of the offline voice control methods in the above aspects.

[0016] The present invention provides an offline voice control method, device, equipment, medium and product, wherein the method comprises: obtaining target smart devices in the area where the user is located; inputting the target smart devices into a pre-built user habit model to obtain a list of control devices; performing offline recognition on the voice control instructions issued by the user, and controlling the list of control devices based on the recognition results; and being able to efficiently perform offline voice control based on low-power devices. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The present invention will be described in more detail below based on embodiments and with reference to the accompanying drawings:

[0018] Figure 1 A flowchart of an offline voice control method provided by an embodiment of the present invention;

[0019] Figure 2 A schematic diagram of the structure of an offline voice control device provided by an embodiment of the present invention;

[0020] Figure 3 A flowchart of an offline voice control method provided for an application example of the present invention.

[0021] In the drawings, the same reference numerals are used for the same components, and the drawings are not drawn to scale. DETAILED DESCRIPTION

[0022] In order to enable those skilled in the art to better understand the scheme of the present invention, and to fully understand and implement how the present invention applies technical means to solve technical problems and achieve the corresponding technical effects, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only an embodiment of a part of the present invention, not all of the embodiments. The embodiments of the present invention and the various features in the embodiments can be combined with each other without conflict, and the technical schemes formed are all within the scope of protection of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work should fall within the scope of protection of the present invention.

[0023] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention 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 invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof 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 that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0024] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0025] With the rapid development of Internet of Things technology, voice control has become an important human-computer interaction method in the fields of smart homes, wearable devices, etc. Although online algorithms that rely on network connections for processing can provide rich functions and highly accurate recognition, they are often limited by network stability, latency, and user privacy protection issues. In contrast, offline voice control can provide faster response speeds and better privacy protection, but traditionally, offline voice control requires devices to have high computing power and energy efficiency to handle complex voice recognition and command execution. There are technical problems in this field that make it difficult to perform offline voice control based on low-power devices.

[0026] In order to solve the above-mentioned technical problem that it is difficult to perform offline voice control based on low-power devices, the present invention proposes an offline voice control method, device, equipment, medium and product. The implementation details of the present invention are described in detail below. The following content is only the implementation details provided for the convenience of understanding and is not necessary for the implementation of this solution.

[0027] Embodiment 1

[0028] Figure 1 is a flowchart of an offline voice control method provided by an embodiment of the present application, such as Figure 1 As shown, in the technical solution of this embodiment, an offline voice control method is provided, which includes: obtaining the target smart device in the user's area; inputting the target smart device into a pre-built user habit model to obtain a control device list; offline recognizing the voice control command issued by the user, and controlling the control device list based on the recognition result.

[0029] The technical problem to be solved in this embodiment is how to perform offline voice control based on low-power devices. In the technical solution of this embodiment, the presence of surrounding smart devices is first detected through device communication methods (such as Bluetooth, Wi-Fi, zigbee, etc.), and the relative distance between devices and the location of the user are estimated by, for example, signal strength, so as to obtain the target smart devices in the area where the user is located. Then these target smart devices are input into the user habit model pre-built based on the user's voice control history, which covers the frequency of use of the smart device, the time period of use, the location of the user when using it, and / or the scene mode, so as to obtain a list of control devices. Finally, the voice control command issued by the user is offline recognized, and the list of control devices is controlled based on the recognition result. For example, in a smart home environment, the user issues a voice command of "I'm going to sleep" in the bedroom. The system determines the smart devices in the bedroom area, such as lights, curtains, air conditioners, etc., as target smart devices in the above manner, and then combines the user habit model to determine that the user usually turns off the lights, draws the curtains, and adjusts the air conditioner to sleep mode in this scene, and then generates a list of control devices and performs corresponding operations. The whole process does not need to rely on network connection, and offline voice control is realized on low-power devices.

[0030] The technical solution of this embodiment, through the above steps, can realize efficient offline voice control on low-power devices. It effectively reduces the processing burden of the device, converts multiple voice recognition problems into one voice recognition problem, significantly reduces the dependence of offline voice control on the computing power of the device, and enables low-performance devices to smoothly run the voice control function. At the same time, due to the streamlining of control tasks, the power consumption of the device in operation and standby states is directly reduced, the battery life is extended, and the user's battery life experience is improved. In addition, intelligent screening ensures that control instructions are directly targeted at the devices that users need most, reduces misoperation and redundant control, improves the accuracy and response speed of control, and provides users with a more convenient and accurate voice control experience, meeting the user's needs for low-power offline voice control in areas such as smart homes.

[0031] Embodiment 2

[0032] Based on the above embodiment, the step of obtaining the target smart device in the user's area includes: obtaining the distance between the user terminal and each smart device, the distance between each smart device and / or the distance between the user and the voice control device; determining the user's area based on the distance between the user terminal and each smart device, the distance between each smart device and / or the distance between the user and the voice control device; and using the smart devices in the user's area as target smart devices.

[0033] The technical problem solved by this embodiment is how to obtain the target smart device in the area where the user is located. In the technical solution of this embodiment, by obtaining the distance between the user terminal and each smart device, the distance between each smart device and / or the distance between the user and the voice control device, these distance data are used to determine the area where the user is located. For example, in a home environment, the smart speaker, as a voice control device, obtains the distance information between each device and the smart speaker and the devices by interacting with surrounding smart devices (such as lights, TVs, air conditioners in the living room, curtains, bedside lamps in the bedroom, etc.). When the user issues a voice command in the living room, the system determines that the user is in the living room area based on these distance data, and then determines the smart devices in the living room area as target smart devices. This distance-based positioning method can accurately screen out smart devices related to the user's current area, provide an accurate target device range for subsequent voice control operations, and improve the pertinence and effectiveness of voice control.

[0034] The technical solution of this embodiment provides a more accurate control object for offline voice control by accurately acquiring the target smart device in the area where the user is located. This helps to reduce unnecessary device search and identification processes, reduce the amount of computing and power consumption of the device, and improve the efficiency and response speed of voice control. At the same time, the precise positioning of the target smart device enables the control instructions to be more accurately conveyed to the device that the user needs to control, avoiding misoperation of devices in other areas and improving the user experience. For example, in actual use, when the user issues voice commands in different rooms, the system can quickly and accurately identify the smart devices in the room and control them, without the need to conduct a comprehensive search and processing of the smart devices in the entire home, which greatly saves resources and time, enables low-power devices to run offline voice control functions more efficiently, and enhances the practicality and stability of the offline voice control system.

[0035] Embodiment 3

[0036] Based on the above embodiments, the pre-constructed user habit model includes: constructing a user habit model based on the user's voice control history, wherein the voice control history includes: the frequency of use of the smart device, the time period of use of the smart device, the user's location when the smart device is used, and / or scene mode.

[0037] The technical problem solved by this embodiment is how to build a user habit model. In the technical solution of this embodiment, a user habit model is built based on the user's voice control history. Specifically, the frequency of use of smart devices, the time period of use, the location of the user during use, and / or the scene mode, etc., are collected, and these data are analyzed and processed by machine learning algorithms to build a personalized user habit model. For example, the system records that the user often issues voice commands such as "turn on the TV", "brighten the lights", and "turn on the air conditioning and cooling" in the living room during the period from 7 to 10 pm, and the light brightness in the living room is usually adjusted to 80%, and the air conditioning temperature is set to 26°C. After a period of data accumulation and analysis, the system will incorporate these usage habits into the user habit model. When the user enters the living room again between 7 and 10 pm and issues a voice command, the system will predict the devices and operations that the user may need to control based on the model, such as automatically turning on the TV and switching to the channel that the user often watches, brightening the lights to 80%, and setting the air conditioner to 26°C cooling mode, thereby providing users with a more personalized and convenient voice control service.

[0038] The technical solution of this embodiment realizes the personalization and intelligence of offline voice control by constructing a user habit model. It can predict the user's needs in advance according to the user's daily usage habits, accurately screen out the smart devices that the user is most likely to use, and recommend and control them, greatly improving the accuracy and efficiency of voice control. At the same time, personalized control also reduces the user's manual operation of the device and improves the user experience. For example, in a smart home scenario, the user does not need to issue detailed control instructions for each device every time. The system can automatically complete a series of operations that conform to the user's habits based on the habit model, which not only saves the user's time and energy, but also reduces the energy consumption of the device, so that low-power devices can run more efficiently while meeting the user's personalized needs, further expanding the application prospects of offline voice control technology in smart homes and other fields.

[0039] Embodiment 4

[0040] Based on the above embodiment, after the step of obtaining the control device list, the method also includes: sending the control device list to the user terminal; receiving feedback from the user terminal on the control device list, wherein the feedback includes adjusting the device order, adding devices and / or deleting devices; and using the control device list fed back by the user terminal as the final control device list.

[0041] The technical problem solved by this embodiment is how to optimize the control device list. In the technical solution of this embodiment, after obtaining the control device list, it is sent to the user terminal, such as a mobile phone APP or a smart control panel. The user can view the recommended control device list on the terminal and provide feedback according to his actual needs, including adjusting the order of devices, adding devices and / or deleting devices. For example, in a family gathering scenario, the user's original control device list may only include the lights and audio in the living room, but the user finds through the terminal that the fan and curtains in the living room need to be controlled, so these two devices can be added directly to the control list on the terminal; or the user thinks that the control order of the lights should be ranked first, so as to quickly adjust the lighting atmosphere, and the order of the devices can be adjusted. After the system receives feedback from the user terminal, the user's adjusted control device list is used as the final control device list for subsequent voice control operations. This way of user participation in optimizing the control device list fully considers the personalized needs of users in different scenarios, making the offline voice control system more flexible and easy to use.

[0042] The technical solution of this embodiment greatly improves the flexibility and adaptability of the offline voice control system through the user's optimization of the control device list. The user can adjust the control device list at any time according to the changes in the actual scene to ensure that the voice control instructions can accurately act on the required equipment, avoid unnecessary equipment control and operation, and improve the accuracy and efficiency of control. At the same time, the user's participation also enhances their sense of control and satisfaction with the system and improves the user experience. For example, in different activity scenarios (such as watching movies, gatherings, rest, etc.), users can quickly personalize the control device list on the terminal, so that the system can better meet the voice control needs in various scenarios, and further expand the practicality and effectiveness of offline voice control technology in complex and changeable actual application scenarios.

[0043] Embodiment 5

[0044] Based on the above embodiment, the voice control instructions issued by the user are recognized offline, and the step of controlling the control device list based on the recognition result includes: performing synonym conversion on the voice control instructions issued by the user; and performing offline recognition based on the voice control instructions after synonym conversion.

[0045] The technical problem solved by this embodiment is how to perform offline recognition of the voice control instructions issued by the user. In the technical solution of this embodiment, the voice control instructions issued by the user are first converted to synonyms. For example, due to accents, habits, etc., users may issue different expressions such as "turn on the light" and "turn on the light", but these expressions actually have the same meaning as the standard instruction "turn on the light". The system converts these non-standard voice instructions into standard instruction forms through a pre-set synonym library, and then performs offline recognition based on the voice control instructions after synonym conversion. During the offline recognition process, the system uses the locally stored voice recognition model and algorithm to analyze and recognize the converted instructions to determine the corresponding control intention and target device. For example, in a low-power intelligent lighting system, the user issues various "turn on the light" expressions. After synonym conversion and offline recognition, the system can accurately recognize that the user's intention is to control the lighting device to turn on, thereby achieving precise control of the light without relying on network connection for voice instruction recognition and processing, ensuring that voice control operations can be completed efficiently and accurately on low-power devices.

[0046] The technical solution of this embodiment improves the adaptability and recognition accuracy of the offline voice control system to different users' voice habits through synonym conversion and offline recognition of user voice control commands. Even on low-power devices, it can quickly and accurately understand the user's command intentions, reduce control errors caused by differences in voice command expressions, and improve user experience. At the same time, the offline recognition method avoids the impact of network delays and instability on voice control, and ensures the timeliness and reliability of control. For example, in actual use, different users may have a variety of different voice expressions for the same operation. The system's synonym conversion and offline recognition functions can effectively handle these differences, allowing low-power devices to accurately respond to various user commands, broadening the scope of application of offline voice control technology, and providing strong support for its application in more low-power IoT devices.

[0047] Embodiment 6

[0048] Based on the above embodiment, the voice control instructions issued by the user are recognized offline, and the step of controlling the control device list based on the recognition result also includes: using the control instruction of the first device in the control device list as the control instruction of the control device list; and / or setting the smart devices outside the control device list to standby mode.

[0049] The technical problem solved by this embodiment is how to control the control device list based on the recognition results. In the technical solution of this embodiment, the control instruction of the first device in the control device list is used as the control instruction of the entire control device list, which can simplify the control process and improve the control efficiency. For example, in a smart home scenario, the control device list includes smart lights, smart curtains and smart air conditioners. When the user issues a voice command of "going to bed", the system determines that the operation corresponding to the command is to turn off the lights, draw the curtains and adjust the air conditioner temperature according to the user habit model. At this time, the turning off instruction of the smart light is used as the starting control instruction of the entire list, and the operations of turning off the lights, drawing the curtains and adjusting the air conditioner temperature are executed in sequence, avoiding the complex instruction scheduling and sorting process, making the control more efficient and orderly. At the same time, the smart devices outside the control device list are set to standby mode. For example, in the above scenario, the TV, speakers and other devices in the living room that are not in the control device list will be set to standby mode, reducing unnecessary device operation and power consumption.

[0050] The technical solution of this embodiment, through the above control method, further optimizes the control process and reduces power consumption when controlling the control device list based on the recognition results. The simplified control instruction execution method improves the speed and accuracy of control, reduces the response time of the device, and improves the user experience. Setting non-control devices to standby mode effectively reduces the overall power consumption of the device, extends the battery life of low-power devices, and enhances the practicality and stability of the system. For example, in a smart home environment, when users use voice control in different scenarios, the system can quickly and accurately perform control operations, while reducing the energy consumption of irrelevant devices, so that low-power devices can better meet users' needs for offline voice control.

[0051] Embodiment 7

[0052] Figure 2 is a schematic diagram of the structure of an offline voice control device provided in an embodiment of the present application, such as Figure 2 As shown, in the technical solution of this embodiment, an offline voice control device is provided, which includes: an acquisition module, used to acquire the target smart device in the user's area; a processing module, used to input the target smart device into a pre-built user habit model to obtain a control device list; a recognition module, used to perform offline recognition of the voice control instructions issued by the user, and control the control device list based on the recognition result.

[0053] The technical problem to be solved in this embodiment is how to perform offline voice control based on low-power devices. In the technical solution of this embodiment, the presence of surrounding smart devices is first detected through device communication methods (such as Bluetooth, Wi-Fi, zigbee, etc.), and the relative distance between devices and the location of the user are estimated by, for example, signal strength, so as to obtain the target smart devices in the area where the user is located. Then these target smart devices are input into the user habit model pre-built based on the user's voice control history, which covers the frequency of use of the smart device, the time period of use, the location of the user when using it, and / or the scene mode, so as to obtain a list of control devices. Finally, the voice control command issued by the user is offline recognized, and the list of control devices is controlled based on the recognition result. For example, in a smart home environment, the user issues a voice command of "I'm going to sleep" in the bedroom. The system determines the smart devices in the bedroom area, such as lights, curtains, air conditioners, etc., as target smart devices in the above manner, and then combines the user habit model to determine that the user usually turns off the lights, draws the curtains, and adjusts the air conditioner to sleep mode in this scene, and then generates a list of control devices and performs corresponding operations. The whole process does not need to rely on network connection, and offline voice control is realized on low-power devices.

[0054] The technical solution of this embodiment, through the above steps, can realize efficient offline voice control on low-power devices. It effectively reduces the processing burden of the device, converts multiple voice recognition problems into one voice recognition problem, significantly reduces the dependence of offline voice control on the computing power of the device, and enables low-performance devices to smoothly run the voice control function. At the same time, due to the streamlining of control tasks, the power consumption of the device in operation and standby states is directly reduced, the battery life is extended, and the user's battery life experience is improved. In addition, intelligent screening ensures that control instructions are directly targeted at the devices that users need most, reduces misoperation and redundant control, improves the accuracy and response speed of control, and provides users with a more convenient and accurate voice control experience, meeting the user's needs for low-power offline voice control in areas such as smart homes.

[0055] Based on the above embodiment, the step of obtaining the target smart device in the user's area includes: obtaining the distance between the user terminal and each smart device, the distance between each smart device and / or the distance between the user and the voice control device; determining the user's area based on the distance between the user terminal and each smart device, the distance between each smart device and / or the distance between the user and the voice control device; and using the smart devices in the user's area as target smart devices.

[0056] The technical problem solved by this embodiment is how to obtain the target smart device in the area where the user is located. In the technical solution of this embodiment, by obtaining the distance between the user terminal and each smart device, the distance between each smart device and / or the distance between the user and the voice control device, these distance data are used to determine the area where the user is located. For example, in a home environment, the smart speaker, as a voice control device, obtains the distance information between each device and the smart speaker and the devices by interacting with surrounding smart devices (such as lights, TVs, air conditioners in the living room, curtains, bedside lamps in the bedroom, etc.). When the user issues a voice command in the living room, the system determines that the user is in the living room area based on these distance data, and then determines the smart devices in the living room area as target smart devices. This distance-based positioning method can accurately screen out smart devices related to the user's current area, provide an accurate target device range for subsequent voice control operations, and improve the pertinence and effectiveness of voice control.

[0057] The technical solution of this embodiment provides a more accurate control object for offline voice control by accurately acquiring the target smart device in the area where the user is located. This helps to reduce unnecessary device search and identification processes, reduce the amount of computing and power consumption of the device, and improve the efficiency and response speed of voice control. At the same time, the precise positioning of the target smart device enables the control instructions to be more accurately conveyed to the device that the user needs to control, avoiding misoperation of devices in other areas and improving the user experience. For example, in actual use, when the user issues voice commands in different rooms, the system can quickly and accurately identify the smart devices in the room and control them, without the need to conduct a comprehensive search and processing of the smart devices in the entire home, which greatly saves resources and time, enables low-power devices to run offline voice control functions more efficiently, and enhances the practicality and stability of the offline voice control system.

[0058] Based on the above embodiments, the pre-constructed user habit model includes: constructing a user habit model based on the user's voice control history, wherein the voice control history includes: the frequency of use of the smart device, the time period of use of the smart device, the user's location when the smart device is used, and / or scene mode.

[0059] The technical problem solved by this embodiment is how to build a user habit model. In the technical solution of this embodiment, a user habit model is built based on the user's voice control history. Specifically, the frequency of use of smart devices, the time period of use, the location of the user during use, and / or the scene mode, etc., are collected, and these data are analyzed and processed by machine learning algorithms to build a personalized user habit model. For example, the system records that the user often issues voice commands such as "turn on the TV", "brighten the lights", and "turn on the air conditioning and cooling" in the living room during the period from 7 to 10 pm, and the light brightness in the living room is usually adjusted to 80%, and the air conditioning temperature is set to 26°C. After a period of data accumulation and analysis, the system will incorporate these usage habits into the user habit model. When the user enters the living room again between 7 and 10 pm and issues a voice command, the system will predict the devices and operations that the user may need to control based on the model, such as automatically turning on the TV and switching to the channel that the user often watches, brightening the lights to 80%, and setting the air conditioner to 26°C cooling mode, thereby providing users with a more personalized and convenient voice control service.

[0060] The technical solution of this embodiment realizes the personalization and intelligence of offline voice control by constructing a user habit model. It can predict the user's needs in advance according to the user's daily usage habits, accurately screen out the smart devices that the user is most likely to use, and recommend and control them, greatly improving the accuracy and efficiency of voice control. At the same time, personalized control also reduces the user's manual operation of the device and improves the user experience. For example, in a smart home scenario, the user does not need to issue detailed control instructions for each device every time. The system can automatically complete a series of operations that conform to the user's habits based on the habit model, which not only saves the user's time and energy, but also reduces the energy consumption of the device, so that low-power devices can run more efficiently while meeting the user's personalized needs, further expanding the application prospects of offline voice control technology in smart homes and other fields.

[0061] Based on the above embodiment, after the step of obtaining the control device list, the device also includes: sending the control device list to the user terminal; receiving feedback from the user terminal on the control device list, wherein the feedback includes adjusting the device order, adding devices and / or deleting devices; and using the control device list fed back by the user terminal as the final control device list.

[0062] The technical problem solved by this embodiment is how to optimize the control device list. In the technical solution of this embodiment, after obtaining the control device list, it is sent to the user terminal, such as a mobile phone APP or a smart control panel. The user can view the recommended control device list on the terminal and provide feedback according to his actual needs, including adjusting the device order, adding devices and / or deleting devices. For example, in a family gathering scenario, the user's original control device list may only include the lights and audio in the living room, but the user finds through the terminal that the fan and curtains in the living room need to be controlled, so these two devices can be added directly to the control list on the terminal; or the user thinks that the control order of the lights should be ranked first, so as to quickly adjust the lighting atmosphere, and the order of the devices can be adjusted. After the system receives feedback from the user terminal, the user's adjusted control device list is used as the final control device list for subsequent voice control operations. This way of user participation in optimizing the control device list fully considers the personalized needs of users in different scenarios, making the offline voice control system more flexible and easy to use.

[0063] The technical solution of this embodiment greatly improves the flexibility and adaptability of the offline voice control system through the user's optimization of the control device list. The user can adjust the control device list at any time according to the changes in the actual scene to ensure that the voice control instructions can accurately act on the required equipment, avoid unnecessary equipment control and operation, and improve the accuracy and efficiency of control. At the same time, the user's participation also enhances their sense of control and satisfaction with the system and improves the user experience. For example, in different activity scenarios (such as watching movies, gatherings, rest, etc.), users can quickly personalize the control device list on the terminal, so that the system can better meet the voice control needs in various scenarios, and further expand the practicality and effectiveness of offline voice control technology in complex and changeable actual application scenarios.

[0064] Based on the above embodiment, the voice control instructions issued by the user are recognized offline, and the step of controlling the control device list based on the recognition result includes: performing synonym conversion on the voice control instructions issued by the user; and performing offline recognition based on the voice control instructions after synonym conversion.

[0065] The technical problem solved by this embodiment is how to perform offline recognition of the voice control instructions issued by the user. In the technical solution of this embodiment, the voice control instructions issued by the user are first converted to synonyms. For example, due to accents, habits, etc., users may issue different expressions such as "turn on the light" and "turn on the light", but these expressions actually have the same meaning as the standard instruction "turn on the light". The system converts these non-standard voice instructions into standard instruction forms through a pre-set synonym library, and then performs offline recognition based on the voice control instructions after synonym conversion. During the offline recognition process, the system uses the locally stored voice recognition model and algorithm to analyze and recognize the converted instructions to determine the corresponding control intention and target device. For example, in a low-power intelligent lighting system, the user issues various "turn on the light" expressions. After synonym conversion and offline recognition, the system can accurately recognize that the user's intention is to control the lighting device to turn on, thereby achieving precise control of the light without relying on network connection for voice instruction recognition and processing, ensuring that voice control operations can be completed efficiently and accurately on low-power devices.

[0066] The technical solution of this embodiment improves the adaptability and recognition accuracy of the offline voice control system to different users' voice habits through synonym conversion and offline recognition of user voice control commands. Even on low-power devices, it can quickly and accurately understand the user's command intentions, reduce control errors caused by differences in voice command expressions, and improve user experience. At the same time, the offline recognition method avoids the impact of network delays and instability on voice control, and ensures the timeliness and reliability of control. For example, in actual use, different users may have a variety of different voice expressions for the same operation. The system's synonym conversion and offline recognition functions can effectively handle these differences, allowing low-power devices to accurately respond to various user commands, broadening the scope of application of offline voice control technology, and providing strong support for its application in more low-power IoT devices.

[0067] Based on the above embodiment, the voice control instructions issued by the user are recognized offline, and the step of controlling the control device list based on the recognition result also includes: using the control instruction of the first device in the control device list as the control instruction of the control device list; and / or setting the smart devices outside the control device list to standby mode.

[0068] The technical problem solved by this embodiment is how to control the control device list based on the recognition results. In the technical solution of this embodiment, the control instruction of the first device in the control device list is used as the control instruction of the entire control device list, which can simplify the control process and improve the control efficiency. For example, in a smart home scenario, the control device list includes smart lights, smart curtains and smart air conditioners. When the user issues a voice command of "going to bed", the system determines that the operation corresponding to the command is to turn off the lights, draw the curtains and adjust the air conditioner temperature according to the user habit model. At this time, the turning off instruction of the smart light is used as the starting control instruction of the entire list, and the operations of turning off the lights, drawing the curtains and adjusting the air conditioner temperature are executed in sequence, avoiding the complex instruction scheduling and sorting process, making the control more efficient and orderly. At the same time, the smart devices outside the control device list are set to standby mode. For example, in the above scenario, the TV, speakers and other devices in the living room that are not in the control device list will be set to standby mode, reducing unnecessary device operation and power consumption.

[0069] The technical solution of this embodiment, through the above control method, further optimizes the control process and reduces power consumption when controlling the control device list based on the recognition results. The simplified control instruction execution method improves the speed and accuracy of control, reduces the response time of the device, and improves the user experience. Setting non-control devices to standby mode effectively reduces the overall power consumption of the device, extends the battery life of low-power devices, and enhances the practicality and stability of the system. For example, in a smart home environment, when users use voice control in different scenarios, the system can quickly and accurately perform control operations, while reducing the energy consumption of irrelevant devices, so that low-power devices can better meet users' needs for offline voice control.

[0070] Embodiment 8

[0071] In the technical solution of this embodiment, a computer device is provided, including a memory, a processor and a computer program stored in the memory, and the processor executes the computer program to implement the steps of any offline voice control method of the above embodiments.

[0072] In the technical solution of this embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of any offline voice control method of the above embodiments are implemented.

[0073] In the technical solution of this embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps of any offline voice control method of the above embodiments are implemented.

[0074] The processor may include, but is not limited to, for example, one or more processors or microprocessors. Each processor may be an application specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), a field programmable gate array (FPGA), a controller, a microcontroller, a microprocessor or other electronic components to implement the method in the above embodiment. The computer readable storage medium may be implemented by any type of volatile or non-volatile storage device or a combination thereof, and the computer readable storage medium may include, but is not limited to, for example, a random access memory (RAM), a read-only memory (ROM), a flash memory, an EPROM memory, an EEPROM memory, a register, a computer storage medium (such as a hard disk, a floppy disk, a solid state drive, a removable disk, a CDROM, a DVDROM, a Blu-ray disc, etc.).

[0075] The computer-readable storage medium may also store at least one computer executable program / instruction, which may be, for example, a computer-readable instruction. The computer-readable storage medium includes, but is not limited to, for example, a volatile memory and / or a non-volatile memory. The volatile memory may include, for example, a random access memory (RAM) and / or a cache memory (cache), etc. The computer-readable storage medium may include, for example, a read-only memory (ROM), a hard disk, a flash memory, etc. For example, a non-transitory computer-readable storage medium may be connected to a computing device such as a computer, and then, when the computing device runs the computer-readable instructions stored on the computer-readable storage medium, the various methods described above may be performed.

[0076] In addition, the computer device may also include (but not limited to) a data bus, an input / output (I / O) bus, a display, and an input / output device (e.g., a keyboard, a mouse, a speaker, etc.). The processor may communicate with external devices via the I / O bus via a wired or wireless network. In one embodiment, the at least one computer executable instruction may also be compiled into or constitute a software product / computer program product, wherein one or more computer executable instructions are executed by the processor to perform the various functions and / or steps of the method in the embodiments described in the present technology.

[0077] Embodiment 9

[0078] Based on the above embodiments, this embodiment provides an application example.

[0079] This application example provides a method to improve the offline voice control accuracy rate of low-performance devices through intelligent screening and optimization.

[0080] With the rapid development of IoT technology, voice control has become an important human-computer interaction method in smart homes, wearable devices and other fields. However, although online algorithms that rely on network connections for processing can provide rich functions and highly accurate recognition, they are often limited by network stability, latency and user privacy protection issues. In contrast, offline voice control can provide faster response speed and better privacy protection, but traditionally, offline voice control requires devices to have high computing power and energy efficiency to handle complex voice recognition and command execution.

[0081] There is a technical problem in this field that it is difficult to perform offline voice control efficiently.

[0082] In order to overcome this challenge, the present invention proposes a screening optimization method that focuses on optimizing the number of devices that need to be controlled by offline voice devices. By accurately screening and limiting the control scope to the most critical and most commonly used devices, the requirements for device performance and power consumption can be significantly reduced while ensuring control accuracy. The method of the present invention not only makes offline voice control more efficient and accurate, but also supports device operations under more precise control requirements without sacrificing user experience, rather than simply expanding to more types of devices.

[0083] This patent solves the key technical problems faced by offline voice control technology when it is applied to low-power, low-cost devices. Traditionally, offline voice control requires too high a computing power and battery life of the device because it needs to process control instructions of multiple devices at the same time, which limits its application on a wide range of devices. For example, when a user intends to turn off the lights and go to bed, the user needs to issue a voice command "I'm going to bed." However, after the user turns off the lights, he wants to turn on the night light for a while, so he immediately issues a voice command "Turn on the night light and turn off the light after 30 minutes." This patent proposes a method for screening and optimizing the number of devices, which not only effectively reduces the processing burden of the device, but also significantly improves the accuracy and efficiency of voice control.

[0084] The specific method of the present invention is as follows:

[0085] ① Area and distance perception: The technical solution of the present invention uses device communication methods (such as Bluetooth, Wi-Fi, Zigbee, etc.) to detect the presence of surrounding smart devices, and estimates the relative distance and location between devices through, for example, signal strength. Prioritize the selection of smart devices that are closer to the user's current area, because these smart devices are more likely to be the target devices that the user is currently or about to use.

[0086] ② User habit analysis: Combine historical user usage data, such as the frequency of use of smart devices, usage time period, scene mode, etc., to build a personalized habit model for users. Through machine learning algorithms, identify the most commonly used smart device combinations in different regions, voice devices, and scenes, filter and recommend related smart devices, and obtain a list of control devices.

[0087] ③ Intelligent recommendation and user confirmation: Based on the area and distance perception results and the user habit analysis results, the technical solution of the present invention intelligently recommends a list of control devices that are most likely to meet the current needs. The user views the recommended control device list through voice commands or touch interface, and selects and confirms according to user needs. The technical solution of the present invention supports users to customize the order of devices or add / delete devices.

[0088] ④ Optimize control logic: After the user confirms the recommended control device list, the technical solution of the present invention will perform special optimization on these devices. The content of special optimization includes but is not limited to: 1. Streamlining the control instruction set to only retain instructions that are valid for the selected device; 2. Optimizing the voice recognition algorithm to improve the recognition accuracy of control instructions for these devices; 3. Adjusting the communication protocol between devices to reduce unnecessary data transmission and power consumption.

[0089] The technical solution of the present invention can significantly lower the performance threshold: by reducing the number of devices processed simultaneously, the technical solution of the present invention significantly reduces the dependence of offline voice control on the computing power of the device, converts multiple voice recognition problems into one voice recognition problem, and enables lower-performance devices to efficiently run the voice control function.

[0090] The technical solution of the present invention can significantly reduce power consumption: the simplification of control tasks directly reduces the power consumption of the device in operation and standby states, effectively prolongs the battery life, and improves the user's battery life experience.

[0091] The technical solution of the present invention can improve control accuracy and efficiency: intelligent screening ensures that control instructions are directly targeted at the devices most needed by users, reduces misoperation and redundant control, and improves control accuracy and response speed.

[0092] The technical solution of the present invention can enhance user privacy protection: since the control logic is mainly completed in an offline state, the demand for data transmission is reduced, thereby reducing the risk of user data leakage and enhancing privacy protection capabilities.

[0093] The technical solution of the present invention can widely expand the application scenarios: by solving the performance and power consumption limitation problems, this patent opens up a broader application space for offline voice control technology, enabling it to adapt to more types of low-power, low-cost IoT devices, and promote the development of smart homes, wearable devices and other fields.

[0094] This embodiment has an intelligent screening mechanism: The core of this patent is to propose an intelligent screening method that can screen out the most critical and commonly used devices for priority control based on the user's usage habits and scenario requirements. The screening of the present invention not only takes into account the physical location of the device (such as confirming the distance through signal interaction between devices), but also combines the user's historical usage data, thereby achieving a more accurate and personalized control experience.

[0095] This embodiment has a resource optimization strategy: resources are optimized by reducing the number of devices processed simultaneously, including but not limited to: streamlining the control instruction set to retain only instructions that are valid for selected devices; optimizing the voice recognition algorithm to improve the recognition accuracy of control instructions for these devices; adjusting the communication protocol between devices to reduce unnecessary data transmission and power consumption.

[0096] This embodiment has user interaction and confirmation: The technical solution of the present invention also includes user interaction, which enhances the flexibility and user satisfaction of the technical solution of the present invention by providing an interface for users to select and confirm control devices. Users can customize the control device list according to their needs and preferences, and the personalized settings of the present invention further enhance the user experience.

[0097] This embodiment has enhanced privacy protection: since most of the control logic is completed offline, the need for data transmission is reduced, and the technical solution of the present invention also has significant advantages in protecting user privacy. This design reduces the risk of user data leakage and enhances the user's trust in technology.

[0098] The data is recorded on the device in advance. In order to prevent the data from being too concentrated, a common method is to store the data in a dispersed manner on each device. If the user sets the instruction a online, after the instruction a is set, the instruction will be saved on the voice entry device A and each related device. After multiple executions, the instruction a action is optimized according to the user's habits. After intelligent screening, the corresponding actions and devices will also be adjusted.

[0099] If the user issues a voice command to voice portal A, the data stored on voice portal A is that command a involves devices B and C, so A will transmit command a to devices B and C through the local area network. After receiving command a, devices B and C will instruct certain actions b and c. After successful execution, they will return successful execution to device A. Device A will return the voice to the user to indicate successful execution, such as "Master, executed" or "OK, executed successfully" or "Scene executed", etc. Subsequently, command a will be adjusted according to user habits, and data analysis and processing will be performed on the local host / gateway of the home. If device D is added, corresponding to action d, this data will be saved on device D.

[0100] Figure 3 It is a flow chart of an offline voice control method provided by an application example of the present invention.

[0101] like Figure 3 As shown, the method of this application example includes: device discovery and access to the smart home system; optimizing, screening and recommending devices according to user habit models, device areas, distances, etc.; user selection and confirmation, the user views the recommendation list, and selects according to personal needs and preferences; data collection and user habit analysis; control instruction optimization and execution; the user confirms whether the instruction control meets the requirements; if it meets the requirements, the user data is saved locally, and the instruction is optimized according to user feedback and new technical models; if it does not meet the requirements, it goes to the above-mentioned user selection and confirmation, the user views the recommendation list, and selects according to personal needs and preferences.

[0102] This application example includes the following contents.

[0103] 1. Device Discovery and Access

[0104] After the technical solution of the present invention is started, it automatically scans and discovers available smart devices around through built-in wireless communication technologies such as Wi-Fi and Bluetooth. Among them, voice-triggered scanning, click-triggered network scanning on the APP, and real-time scanning are all supported. Real-time scanning is generally when the network is just configured or a new device is found in the configured network. The voice device will broadcast "Device A is found" through voice whether to configure the network.

[0105] Identify the basic information of each smart device, such as type, model, firmware version, etc., and create a smart device list. The device list may be in the following form, for example: Device 1: device information of device 1; Device 2: device information of device 2; Device 3: device information of device 3.

[0106] Invite the user to confirm the access of the smart device through the application or control panel, and set the necessary initial parameters for the smart device (such as the name and location of the smart device).

[0107] 2. Equipment screening and recommendation

[0108] Combine the user habit model and the signal strength between smart devices (the distance is determined by the signals sent between devices) to screen out the most critical and most frequently used smart devices. Prioritize smart devices that are closer to users and more frequently used and put them in the priority control list.

[0109] User habit model: Build a personalized user habit model based on device usage frequency, time period, scenario mode, etc. Through machine learning algorithms, identify the most commonly used device combinations by users in different regions, voice devices, and scenarios, and filter and recommend related devices.

[0110] The user's personalized habit model includes: what device is controlled by a certain voice control command of the user. That is, according to the user's habits, at a certain time point, using control command A, device B and device C will be changed to a specific state. Using the same command with different users or at different time points will control different results. If command D + device control E is used multiple times, the system will also associate command D with device control E. When command D is used, device E will be controlled.

[0111] Through the application or control panel, the user is presented with a list of recommended control devices and is provided with the option to customize them.

[0112] Step 1: The devices connected to the system include curtain control devices, air conditioners, fans and lighting control devices. Create a smart device list based on the device information of these devices.

[0113] Step 2:

[0114] 2.1 Prioritize the distance between the device and the user and the frequency of device use to filter the devices in the smart device list to obtain a priority control list, for example, the list includes: "lighting control device", "air conditioning", and "curtain control device".

[0115] For example, the user issues control commands in the bedroom multiple times (more than twice) around 10 p.m.: ① "I'm going to bed" (based on the distance between the user and the device, confirming that the user is in the bedroom, the system will start the bedroom sleeping scene: turn off the lights, close the curtains, and adjust the air conditioner to sleep mode); ② "Turn on the night light"; ③ "Time the night light to turn off after 30 minutes". Therefore, the system will adjust the control of the command "I'm going to bed" to: turn on the sleeping scene, turn on the night light and time it to turn off after 30 minutes. After this adjustment, the original three voice control commands are merged into one voice control command, reducing the computing resources required for voice offline control.

[0116] 2.2 Input the device information, current time, and current area category (such as bedroom, living room, dining room, etc.) in the priority control list into the constructed personalized usage model, and output the control device list?

[0117] The obtained device control list is "lighting control device", "air conditioning", and "curtain control device".

[0118] The personalized usage model is trained based on voice commands, time, user area, control area, distance between device and user, device information, control frequency, device weight, etc. Users do not need to actively input data, the system will automatically filter according to user habits. For example, devices closer to the user will be given priority. If the user will adjust and choose a farther device or a specific device, the next time the control is performed, the farther or specific device will be controlled based on the last control situation.

[0119] By further optimizing the priority control list, the entire model will build a more complete user personality and control more accurate devices as users use it more and more frequently.

[0120] If the user uses the system to perform voice control "I'm going to bed" for the first time, the system will default to the sleeping scene, turn off the lights, close the curtains, and set the air conditioner to sleep mode. If the user is used to turning on the night light before going to bed and turning off the light after 30 minutes, subsequent "I'm going to bed" commands will be optimized to add the actions of turning on the night light and turning off the light after a certain time.

[0121] 2.3 The control device list is displayed to the user through an application or control panel to generate a display interface, and the display interface is provided with user-customizable device options so that the user can add, delete devices in the device list, and adjust the order of devices in the control device list.

[0122] 3. User selection and confirmation

[0123] The user reviews the recommended list and selects according to personal needs and preferences. The user is allowed to adjust the order of devices, add or delete devices to meet the control needs in a specific scenario. The user confirms the final list of control devices, and the system records and saves the user's selection.

[0124] The devices in the recommended list are sorted based on the following sorting rules: main voice commands, distance between the device and the user, and frequency of device use. For example, after the user uses the door lock to open the door, he will say the voice command "play a song" to the voice air conditioner when entering the door. After playing the song, the user will frequently adjust the living room lights to soft lights and occasionally close the living room curtains. For the voice command "play a song" of the living room voice cabinet, the order of the devices is: ① air conditioner; ② living room smart light; ③ living room curtains. According to the technical solution of the present invention, subsequent operations can be performed on the air conditioner, living room smart light, and living room curtains based on the "play a song" command.

[0125] In some embodiments, the system will continuously optimize the voice command control device by combining various parameters. For example, if a user sends the same command A to the same device at the same time in the same location, the weight of device B will increase because other devices B are controlled before and after the control command A is issued. When the weight of device B is high enough, when the user executes command A again, device B will also be controlled.

[0126] In some implementations, the frequency of device B is increased, and the weight of action b of device B in instruction A will also be higher. When the weight reaches a certain value, instruction A will be optimized.

[0127] For example, in the scenario: the first time a user uses the system to perform voice control "I'm going to sleep", the system routinely executes the linked sleeping scenario: turn off the lights, close the curtains, and set the air conditioner to sleep mode. If the user executes "turn on the night light, and turn off the light after 30 minutes" after executing the sleeping scenario, the weight of this action of the night light will be higher. After executing it multiple times, the subsequent "I'm going to sleep" command will be optimized. When the user issues the "I'm going to sleep" command again, the command will be adjusted to: turn off the lights, close the curtains, set the air conditioner to sleep mode, turn on the night light, and turn off the night light after 30 minutes.

[0128] 4: Data collection and user habit analysis.

[0129] The technical solution of the present invention collects the interaction data between users and smart devices, including the frequency of device use, time period, scene mode, etc. The data is processed and analyzed using the local model (if the user chooses to synchronize data online, the online model can be used for analysis) to build a personalized usage model for the user. The personalized usage model of the user is updated regularly to adapt to changes in user habits and improve the accuracy of prediction.

[0130] For example, at 7:30 every morning, the user will control the bedside lamp to turn on, the cloth curtains to open, and the screen curtains to remain on. When the user uses the control command "turn on the lights, open the curtains" at 7 o'clock, the bedside lamp and curtains will be turned on 100%. When the user uses "turn on the lights, open the curtains" normally, all the lights and curtains in this area will be turned on.

[0131] V. Control Optimization and Execution

[0132] The technical solution of the present invention optimizes the list of control devices confirmed by the user, including streamlining the control instruction set, optimizing the voice recognition algorithm, etc. The communication protocol between devices is adjusted to reduce unnecessary data transmission and power consumption. Users control the devices through voice instructions or applications, and the system ensures the accuracy and efficiency of the control instructions.

[0133] For commands that are frequently used by users, we carefully optimize and streamline the recognition algorithm, so that the recognition rate of commands is higher and the controlled devices will maintain a higher sensitivity. However, if some devices are controlled very rarely during the day, these devices will be in standby mode during the day. This method can reduce the power consumption of the devices.

[0134] The simplified command set simplifies the user's commands. For example, the command set related to "getting up" in the bedroom includes: I want to get up, I got up, ahaha I got up, I woke up, wake me up, etc. The "prepare" in the "prepare to get up" command set in the bedroom may be "Xiaoge Xiaoge, wake me up in ten minutes" in the delayed voice commands of the user in the bedroom in the morning. In the "prepare to get up" command set, the "prepare" command is "delay for 10 minutes". Similarly, the "prepare" in "prepare to go to bed" may be "delay for 30 minutes" according to user habits;

[0135] Optimizing voice recognition is to optimize the recognition of commands commonly used by users (such as: come back, get up, sleep, I'm leaving, etc.). The optimization of the algorithm can increase the recognition rate of entries. When "fat" and "fat coming" are recognized, they will be associated with "come back". Only targeted optimization is performed, mainly because this solution is an offline voice solution, and data needs to be stored locally. The optimization of the instruction set and the optimization of the recognition algorithm are essentially based on the user's repeated habits. After the system is integrated, it is simplified, and then the specific integrated actions and algorithms are saved locally to provide control accuracy and recognition rate.

[0136] After removing the devices on the control device list (screening out the devices) from the device list (devices to which the current device is connected), the remaining devices are switched to standby mode.

[0137] In addition, some controls require intermediate devices to transmit commands. Based on the characteristics of mesh and ad hoc networks, for example, a Bluetooth device B in a restaurant is rarely controlled or started, and it is more of a Bluetooth transmission relay point (Bluetooth transmission). Generally, after 8 pm, the user will control the bedroom Bluetooth device C in the living room. The command needs to pass through the living room Bluetooth device A and the restaurant Bluetooth device B before it can be transmitted to device C. Therefore, even if the user is at home, before 8 pm, Bluetooth B can be set to standby. Similarly, if the user is not at home, or the user usually works after 12 pm, the transmission module of the device can be set to standby to reduce unnecessary data transmission and power consumption.

[0138] 6. Continuous optimization and updating

[0139] The technical solution of the present invention continuously monitors the user's usage and saves it locally. Based on user feedback and new technology trends, the algorithm and user experience are continuously optimized. Updates are released regularly to fix potential problems, introduce new features, and improve the overall performance and stability of the system.

[0140] Optimize personalized models and algorithms for high-frequency language commands. Locally, too much data cannot be stored, so there must be a certain priority. When voice control command A reaches a certain weight, the command will be optimized. The optimization method is saved on the server. There are mature algorithms on the market, and each company will continue to update them. After the command is optimized, it will be downloaded locally. Considering privacy, the latest algorithm can also be downloaded locally. After optimization locally, the relevant algorithm can be deleted to release capacity.

[0141] Through the above steps, the best implementation method of this patent can provide an efficient, convenient and safe solution for the field of smart home control, significantly improving the user's voice control experience.

[0142] In the embodiments provided by the present invention, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the above-mentioned module, a program segment or a part of a code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0143] It should be noted that, in the present invention, the terms "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element limited by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0144] Although the embodiments disclosed in the present invention are as above, the above contents are only embodiments adopted for facilitating the understanding of the present invention and are not intended to limit the present invention. Any technician in the technical field to which the present invention belongs can make any modifications and changes in the form and details of the implementation without departing from the spirit and scope disclosed in the present invention, but the patent protection scope of the present invention shall still be subject to the scope defined in the attached claims.

Claims

1. An offline voice control method, characterized in that: The method comprises: Get the target smart devices in the user's area; Input the target smart device into a pre-built user habit model to obtain a control device list; The voice control command issued by the user is recognized offline, and the control device list is controlled based on the recognition result.

2. The offline voice control method according to claim 1, characterized in that: The step of obtaining the target smart device in the area where the user is located includes: Obtaining the distance between the user terminal and each smart device, the distance between each smart device, and / or the distance between the user and the voice control device; Determine the area where the user is located based on the distance between the user terminal and each smart device, the distance between each smart device and / or the distance between the user and the voice control device; The smart devices in the user's area are used as target smart devices.

3. The offline voice control method according to claim 1, characterized in that: The pre-built user habit model includes: A user habit model is constructed based on the user's voice control history, wherein the voice control history includes: the frequency of use of the smart device, the time period of use of the smart device, the location of the user when the smart device is used, and / or scene mode.

4. The offline voice control method according to claim 1, characterized in that: After the step of obtaining the control device list, the method further includes: Sending the control device list to a user terminal; receiving feedback from a user terminal on the control device list, wherein the feedback includes adjusting the device order, adding a device, and / or deleting a device; The control device list fed back by the user terminal is used as the final control device list.

5. The offline voice control method according to claim 1, characterized in that: The step of offline identifying the voice control command issued by the user and controlling the control device list based on the identification result includes: Convert the voice control commands issued by the user into synonyms; Offline recognition of voice control instructions based on synonym conversion.

6. The offline voice control method according to claim 5, characterized in that: The step of offline identifying the voice control command issued by the user and controlling the control device list based on the identification result also includes: The control instruction of the first device in the control device list is used as the control instruction of the control device list; and / or the smart devices outside the control device list are set to standby state.

7. An offline voice control device, characterized in that: The device comprises: An acquisition module, used to acquire target smart devices in the area where the user is located; A processing module, used for inputting the target smart device into a pre-built user habit model to obtain a control device list; The recognition module is used to perform offline recognition of the voice control instructions issued by the user and control the control device list based on the recognition result.

8. A computer device comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the offline voice control method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the offline voice control method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the offline voice control method according to any one of claims 1 to 6 are implemented.

Images