Wake-up control methods, devices, equipment and storage media

Abstract

This application provides a wake-up control method, apparatus, device, and storage medium. The wake-up control method includes: responding to a device wake-up request and acquiring a target wake-up voice corresponding to the device wake-up request; receiving a self-organizing network wake-up broadcast corresponding to the target wake-up voice, using the self-organizing network wake-up broadcast to perform wake-up arbitration on each candidate wake-up device, and determining a target wake-up device; and driving the target wake-up device to execute a target operation command corresponding to the target wake-up voice. The technical solution of this application can achieve efficient communication and collaborative operation between devices, optimize wake-up resource allocation, improve wake-up accuracy among multiple devices, and effectively reduce dependence on network resources.

Description

Technical Field

[0001] This application relates to the field of device control technology, specifically to a wake-up control method, apparatus, device, and storage medium. Background Technology

[0002] Currently, with the rapid development of the smart home industry, more and more smart home devices support voice control. Voice control of smart homes has become an important way to improve user experience. However, in scenarios where multiple smart home devices coexist (e.g., multiple air conditioners), the use of the same wake word by each device may cause users to rely on cloud network resources for coordination when making voice control. This can result in the inability to accurately wake up the specified smart home device or ineffective voice wake-up in environments with poor network signal, leading to low wake-up accuracy and impacting the user experience. Summary of the Invention

[0003] This application provides a wake-up control method, apparatus, device, and storage medium, aiming to solve the technical problem in the prior art where multiple devices cannot wake up the smart home that the user needs to control via voice, resulting in low wake-up accuracy.

[0004] On one hand, embodiments of this application provide a wake-up control method, which includes the following steps:

[0005] In response to a device wake-up request, obtain the target wake-up voice corresponding to the device wake-up request;

[0006] Generate the self-organizing network wake-up broadcast corresponding to the target wake-up voice, perform wake-up arbitration on each candidate wake-up device according to the wake-up arbitration information, and determine the target wake-up device among the candidate wake-up devices;

[0007] Drive the target wake-up device to execute the target operation command corresponding to the target wake-up voice.

[0008] In one possible implementation of this application, generating the ad hoc network wake-up broadcast corresponding to the target wake-up voice, performing wake-up arbitration on each candidate wake-up device according to the wake-up arbitration information, and determining the target wake-up device among the candidate wake-up devices includes:

[0009] Obtain the target wake-up word from the target wake-up voice, and determine the candidate wake-up devices corresponding to the target wake-up word for each networking device in the target ad hoc network;

[0010] Receive the self-organizing network wake-up broadcast corresponding to each of the candidate wake-up devices, and determine the wake-up arbitration information corresponding to the candidate wake-up devices;

[0011] Based on the wake-up arbitration information, wake-up arbitration is performed on each candidate wake-up device to determine the target wake-up device among the candidate wake-up devices.

[0012] In one possible implementation of this application, the step of performing wake-up arbitration on each candidate wake-up device based on the wake-up arbitration information to determine the target wake-up device among the candidate wake-up devices includes:

[0013] Obtain the wake-up sound pressure level and / or the wake-up signal-to-noise ratio in the wake-up arbitration information;

[0014] Obtain the target wake-up sound pressure value from the wake-up sound pressure values, and determine the candidate wake-up device corresponding to the target wake-up sound pressure value as the target wake-up device;

[0015] Alternatively, if there are multiple target wake-up sound pressure levels, then the target wake-up signal-to-noise ratio is obtained from the wake-up signal-to-noise ratio, and the candidate wake-up device corresponding to the target wake-up signal-to-noise ratio is determined as the target wake-up device.

[0016] In one possible implementation of this application, the step of performing wake-up arbitration on each candidate wake-up device based on the wake-up arbitration information to determine the target wake-up device among the candidate wake-up devices includes:

[0017] Obtain the sound source localization angle and / or the device address identifier of each candidate wake-up device from the wake-up arbitration information;

[0018] Obtain the target positioning angle from the sound source positioning angle, and determine the candidate wake-up device corresponding to the target positioning angle as the target wake-up device;

[0019] Alternatively, obtain the target address identifier from the device address identifier, and determine the candidate wake-up device corresponding to the target address identifier as the target wake-up device.

[0020] In one possible implementation of this application, before obtaining the target wake-up word in the target wake-up voice and determining the candidate wake-up devices corresponding to the target wake-up word for each networking device in the target ad hoc network, the method further includes:

[0021] In response to a wake word self-learning request, obtain the initial wake word speech corresponding to the wake word self-learning request;

[0022] Extract the initial wake-up features of the initial wake-up word speech, and use a preset speech recognition model to recognize the initial wake-up features to obtain the initial wake-up word;

[0023] In the target ad hoc network, the initial wake-up word of each network device is synchronized, and the device wake-up word of each network device is determined based on the initial wake-up word and the candidate wake-up word.

[0024] In one possible implementation of this application, the step of synchronizing the initial wake-up word of each network device in the target ad hoc network, and determining the device wake-up word of each network device based on the initial wake-up word and candidate wake-up words, includes:

[0025] Obtain the network device prefix of each network device in the target ad hoc network, and determine the target network device combination with the same network device prefix;

[0026] If there are target networking devices with the same initial wake-up word in the target networking device combination, then the initial wake-up word is updated using the candidate wake-up word of the target networking device to obtain the device wake-up word of the target networking device.

[0027] In one possible implementation of this application, driving the target wake-up device to execute the target operation instruction corresponding to the target wake-up voice includes:

[0028] Obtain the target operation command from the target wake-up voice;

[0029] Pre-communication is performed with the target wake-up device according to the preset routing identifier information and the preset self-organizing network protocol to determine the device wake-up status of the target wake-up device;

[0030] If the device is in a woke-up state, then the target wake-up device is driven to execute the target operation instruction.

[0031] On the other hand, this application provides a wake-up control device, the wake-up control device comprising:

[0032] The voice acquisition module is configured to respond to a device wake-up request and acquire the target wake-up voice corresponding to the device wake-up request;

[0033] The wake-up arbitration module is configured to generate an ad hoc network wake-up broadcast corresponding to the target wake-up voice, perform wake-up arbitration on each candidate wake-up device according to the wake-up arbitration information, and determine the target wake-up device among the candidate wake-up devices.

[0034] The voice control module is configured to drive the target wake-up device to execute the target operation command corresponding to the target wake-up voice.

[0035] On the other hand, this application also provides a wake-up control device, the wake-up control device comprising:

[0036] One or more processors;

[0037] Memory; and

[0038] One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor to implement the steps of the wake-up control method.

[0039] On the other hand, this application also provides a computer-readable storage medium having a computer program stored thereon, the computer program being loaded by a processor to execute the steps in the wake-up control method.

[0040] This application achieves efficient communication and collaborative operation between devices by responding to a device wake-up request, obtaining the target wake-up voice corresponding to the device wake-up request, generating a self-organizing network wake-up broadcast corresponding to the target wake-up voice, arbitrating wake-up devices among the candidate wake-up devices based on the wake-up arbitration information, and determining the target wake-up device among the candidate wake-up devices. It then drives the target wake-up device to execute the target operation command corresponding to the target wake-up voice. In control scenarios with multiple smart home devices, after detecting a target wake-up voice, a self-organizing network wake-up broadcast corresponding to the target wake-up voice is generated. The target smart home device to be woken up is determined through this self-organizing network wake-up broadcast, and the target smart home device is driven to execute the operation corresponding to the wake-up voice. This achieves efficient communication and collaborative operation between devices, optimizes wake-up resource allocation, improves wake-up accuracy between multiple devices, and effectively reduces dependence on network resources. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 This is a schematic diagram of a scenario for the wake-up control method according to an embodiment of this application;

[0043] Figure 2 This is a flowchart illustrating one embodiment of the wake-up control method in this application.

[0044] Figure 3 A flowchart illustrating an embodiment of the wake-up arbitration in the wake-up control method provided in this application;

[0045] Figure 4 A flowchart illustrating one embodiment of the wake-up word update method in the wake-up control method provided in this application;

[0046] Figure 5 A schematic diagram of one embodiment of the wake-up control device provided in this application;

[0047] Figure 6 This is a schematic diagram of one embodiment of the wake-up control device provided in this application. Detailed Implementation

[0048] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0049] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0050] In this application, the term "exemplary" is used to mean "serving as an example, illustration, or description." Any embodiment described as "exemplary" in this application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed in this application.

[0051] Currently, with the rapid development of the smart home industry, more and more smart home devices support voice control. Voice control of smart homes has become an important way to improve user experience. However, in scenarios where multiple smart home devices coexist (e.g., multiple air conditioners), the use of the same wake word by each device may cause users to rely on cloud network resources for coordination when making voice control. This can result in the inability to accurately wake up the specified smart home device or ineffective voice wake-up in environments with poor network signal, leading to low wake-up accuracy and impacting the user experience.

[0052] Based on this, this application proposes a wake-up control method, apparatus, device, and computer-readable storage medium to solve the technical problem in the prior art where multiple devices cannot wake up the smart home that the user needs to control by voice, resulting in low wake-up accuracy.

[0053] The wake-up control method in this embodiment of the invention is applied to a wake-up control device, which is configured in a wake-up control equipment. The wake-up control equipment includes one or more processors, a memory, and one or more application programs. The one or more application programs are stored in the memory and configured to be executed by the processor to implement the wake-up control method. The wake-up control equipment can be a smart terminal, such as a mobile phone, tablet computer, network device, and smart computer. Optionally, the wake-up control equipment can also be a server or a service cluster composed of multiple servers.

[0054] like Figure 1 As shown, Figure 1 This is a schematic diagram of a wake-up control method according to an embodiment of this application. In this embodiment, the wake-up control scenario includes a wake-up control device 100 (which integrates a wake-up control unit) and a target wake-up device 200. The wake-up control device 100 has a computer-readable storage medium corresponding to the wake-up control method running to execute the steps of the wake-up control method. The target wake-up device 200 is a smart home device in a target ad hoc network that can be woken up by a specific wake word.

[0055] Understandable, Figure 1 The wake-up control device in the wake-up control method scenario shown, or the devices included in the wake-up control device, does not constitute a limitation on the embodiments of the present invention. That is, the number or type of wake-up control device included in the wake-up control method scenario, or the number or type of devices included in each device, does not affect the overall implementation of the technical solution in the embodiments of the present invention, and can all be considered as equivalent substitutions or derivatives of the technical solutions claimed in the embodiments of the present invention.

[0056] In this embodiment of the invention, the wake-up control device 100 is mainly used for: responding to a device wake-up request, obtaining the target wake-up voice corresponding to the device wake-up request; generating a self-organizing network wake-up broadcast corresponding to the target wake-up voice, performing wake-up arbitration on each candidate wake-up device according to the wake-up arbitration information, determining the target wake-up device among the candidate wake-up devices; and driving the target wake-up device to execute the target operation instruction corresponding to the target wake-up voice.

[0057] The wake-up control device 100 in this embodiment of the invention can be an independent wake-up control device, such as a smart terminal like a mobile phone, tablet computer, network device, server, and smart computer, or a smart home device that can be a candidate wake-up device.

[0058] This application provides a wake-up control method, apparatus, device, and computer-readable storage medium, which will be described in detail below.

[0059] It will be understood by those skilled in the art that Figure 1 The application environment shown is only one application scenario related to the solution of this application and does not constitute a limitation on the application scenario of this application. Other application environments may include more than one application scenario. Figure 1 The number of more or fewer wake-up control devices shown, or the wake-up control network connectivity relationships, for example Figure 1 Only one wake-up control device is shown in the diagram. It is understood that the wake-up control method may also include one or more wake-up control devices, which are not specifically limited here. The wake-up control device 100 may also include a memory for storing wake words and other data.

[0060] It should be noted that, Figure 1 The scenario diagram of the wake-up control method shown is merely an example. The scenario of the wake-up control method described in the embodiments of the present invention is for the purpose of more clearly illustrating the technical solution of the embodiments of the present invention, and does not constitute a limitation on the technical solution provided by the embodiments of the present invention.

[0061] Based on the scenarios described above for the wake-up control method, various embodiments of the wake-up control method disclosed in this invention are proposed.

[0062] like Figure 2 As shown, Figure 2 This is a flowchart illustrating one embodiment of the wake-up control method in this application. The wake-up control method includes the following steps 201 to 203:

[0063] 201. Respond to the device wake-up request and obtain the target wake-up voice corresponding to the device wake-up request;

[0064] The wake-up control method in this embodiment is applied to a wake-up control device. The type and number of wake-up control devices are not specifically limited; that is, the wake-up control device can be one or more smart terminals or smart home devices. For example, the wake-up control device can be a smart device specifically used to determine the target wake-up device in the target ad hoc network, such as a mobile phone, tablet computer, network device, or smart computer. Optionally, the wake-up control device can also be a smart home device to be woken up. For example, the wake-up control device can also be any voice-wakeable smart home device such as a smart air conditioner, smart light fixture, or smart TV. In a specific embodiment, the wake-up control scenario can be an air conditioner control scenario, in which there are multiple smart air conditioner devices capable of responding to user voice control. Any smart air conditioner device in this air conditioner control scenario can act as a wake-up control device to execute the wake-up control method in this embodiment to achieve the unique wake-up of the nearest designated air conditioner to perform the corresponding operation.

[0065] Specifically, the application scenario in this embodiment is the coexistence of smart home devices with specific wake words that can be woken up and executed by voice. To avoid the technical problem of multiple smart home devices being woken up simultaneously by a single wake word in this application scenario, resulting in resource waste and a poor user experience, each smart home device and wake-up control device in this embodiment can perform self-organizing network operation. This enables each smart home device and wake-up control device to communicate through a specific network protocol, and in subsequent steps, wake-up arbitration can be performed using self-organizing network wake-up broadcast in the target self-organizing network to achieve efficient communication and collaboration between devices, thereby waking up the specific target device and improving the wake-up success rate.

[0066] Specifically, during operation, the wake-up control device also constructs a target ad hoc network for wake-up control. It monitors ambient sound in real time through a microphone array pre-set by the network devices within the target ad hoc network. When a target wake-up word is detected, the network device closest to the user within the target ad hoc network is selected as the wake-up control device to respond to the device wake-up request. This retrieves the target wake-up voice corresponding to the device request. The device wake-up request is an operation instruction that drives the wake-up control device to wake up the nearest smart home device in the target ad hoc network based on the target wake-up voice. The target wake-up voice is a voice control command carrying the target wake-up word and the target operation instruction. For example, in one specific embodiment, when a user needs to control a designated smart home device in the target ad hoc network, they speak the target wake-up voice containing the target wake-up word and the corresponding target operation instruction to wake up the smart home device and control it to execute the target operation instruction. For example, the target wake-up voice could be something like "Hey A, Hey A, turn on cooling mode." Optionally, in other embodiments, a specific smart home device or mobile terminal designated in the target ad hoc network can also be used as the wake-up control device for wake-up control. Optionally, in another specific embodiment, the target wake-up voice may consist only of the target wake-up word.

[0067] Specifically, before responding to a device wake-up request, the wake-up control device and other smart home devices pre-construct a target self-organizing network based on preset routing identification information and a preset self-organizing network protocol. That is, upon startup, the wake-up control device and other smart home devices automatically connect to a preset local area network (LAN) via a preset router to construct the target self-organizing network. In one specific embodiment, this target self-organizing network can be a wireless network LAN encrypted using WPA2 (Wi-Fi Protected Access 2) or a higher level of encryption protocol. The preset routing identification information is the SSID (Service Set Identifier) ​​of the router corresponding to the target self-organizing network. The wake-up control device and other smart home devices within this target self-organizing network can communicate with each other using the IEEE 802.11s standard or similar self-organizing network protocols. That is, after joining the target self-organizing network, the networking devices in the target self-organizing network can automatically exchange information such as IP addresses and subnet masks according to the self-organizing network protocol of the target sub-network, and recognize each other through service discovery protocols such as Bonjour or mDNS to realize direct communication between devices.

[0068] Optionally, during operation, the networking devices in the target ad hoc network can also utilize a preset synchronization service for time synchronization, thereby ensuring accurate time alignment among the networking devices and providing a unified time reference for subsequent wake-up arbitration. In one specific embodiment, the preset synchronization service can be an NTP (Network Time Protocol) synchronization service.

[0069] 202. Receive the ad hoc network wake-up broadcast corresponding to the target wake-up voice, and use the ad hoc network wake-up broadcast to perform wake-up arbitration on each candidate wake-up device to determine the target wake-up device;

[0070] Specifically, after acquiring the target wake-up voice, the wake-up control device also receives a corresponding ad hoc network wake-up broadcast generated by the networking device based on the target wake-up voice. This ad hoc network wake-up broadcast is used to arbitrate wake-up candidate devices in the target ad hoc network associated with the target wake-up word, thereby determining the target wake-up device. The candidate wake-up device is a networking device in the target ad hoc network whose wake-up word is the same as the target wake-up word.

[0071] Specifically, after acquiring the target wake-up voice, the wake-up control device performs speech recognition on the target wake-up voice to determine the target wake-up word. It then accesses the wake-up word database corresponding to the target ad hoc network, comparing the device wake-up words in the database with the target wake-up word to filter out candidate wake-up devices whose device wake-up words match the target wake-up word. These candidate wake-up devices are smart home devices within the target ad hoc network that may be woken up by voice.

[0072] Specifically, the wake-up control device also acquires the ad hoc network wake-up broadcasts associated with the target wake-up voice broadcast by each candidate wake-up device, and uses the wake-up arbitration information in the ad hoc network wake-up broadcasts to perform wake-up arbitration on each candidate wake-up device, determining the target wake-up device that is uniquely woken up and executes the corresponding operation command among the candidate wake-up devices. The ad hoc network wake-up broadcast is a broadcast wake-up request packet carrying wake-up arbitration information generated by the wake-up control device or other networking devices when the target wake-up voice is detected.

[0073] The wake-up arbitration information includes any one or more of the following: wake-up sound pressure level, wake-up signal-to-noise ratio, sound source positioning angle, device address identifier, and timestamp, reflecting the information received by each candidate device from the target wake-up voice.

[0074] Among them, the wake-up sound pressure level (WPL) is a parameter characterizing the speech intensity of the target wake-up speech. The wake-up signal-to-noise ratio (SNR) is a parameter evaluating the ratio of the target wake-up speech signal to background noise. The sound source localization parameter is the azimuth angle of the sound source of the target wake-up speech relative to the candidate wake-up device. The device address identifier is a unique device identification information used to mark the device. The timestamp is a representation of the reception time corresponding to the candidate wake-up device receiving the target wake-up speech.

[0075] Specifically, after determining the candidate wake-up devices, the wake-up control device also receives the self-organizing network wake-up broadcast sent by each candidate wake-up device, as well as the wake-up arbitration information in the self-organizing network wake-up broadcast, and uses the wake-up arbitration information to determine the most suitable target wake-up device among the candidate wake-up devices.

[0076] Specifically, the wake-up control device compares the wake-up sound pressure value, wake-up signal-to-noise ratio, and wake-up sound source positioning angle in the wake-up arbitration information to perform wake-up arbitration on each candidate wake-up device, thereby determining the target wake-up device that is uniquely awakened by the target wake-up voice.

[0077] Optionally, in one embodiment, the wake-up control device uses the wake-up sound pressure level and / or wake-up signal-to-noise ratio (SNR) for wake-up arbitration. That is, the wake-up control device acquires the wake-up sound pressure level and / or wake-up SNR from the wake-up arbitration information, compares the wake-up sound pressure levels of each candidate wake-up device, determines the maximum wake-up sound pressure level among all wake-up sound pressure levels, determines the maximum wake-up sound pressure level as the target sound pressure level, and acquires the candidate wake-up device corresponding to the target sound pressure level. If there is only one target sound pressure level among the wake-up sound pressure levels, then the candidate wake-up device corresponding to the target sound pressure level is determined as the target wake-up device.

[0078] Optionally, if there are multiple target sound pressure values ​​with the same value among the wake-up sound pressure values, that is, if there are multiple maximum wake-up sound pressure values ​​with the same value among the wake-up sound pressure values, then the wake-up control device further compares the wake-up signal-to-noise ratio of each maximum wake-up sound pressure value with the corresponding candidate wake-up device, determines the target signal-to-noise ratio with the highest signal-to-noise ratio among the wake-up signal-to-noise ratios, and determines the candidate wake-up device corresponding to the target signal-to-noise ratio as the target wake-up device.

[0079] Optionally, in other embodiments, if there are multiple target signal-to-noise ratios in the wake-up signal-to-noise ratio and a unique target wake-up device cannot be arbitrated, the wake-up control device further uses the sound source positioning angle and / or device address identifier in the wake-up arbitration information to perform wake-up arbitration on each candidate wake-up device, thereby determining the target wake-up device that is uniquely woken up by the target wake-up voice.

[0080] Optionally, in other embodiments, the wake-up control device can also perform a weighted summation based on a preset scoring model and preset weights corresponding to the wake-up arbitration information to determine the wake-up score of each candidate wake-up device, and determine the candidate wake-up device with the highest wake-up score as the target wake-up device.

[0081] Optionally, in other embodiments, the wake-up control device also includes a built-in timer. If no wake-up command or arbitration result is received within the set period corresponding to the timer, the wake-up control device automatically performs a wake-up operation according to the wake-up priority of each candidate wake-up device, and determines the candidate wake-up device with the highest wake-up priority as the target wake-up device.

[0082] 203. Drive the target wake-up device to execute the target operation command corresponding to the target wake-up voice.

[0083] Specifically, after determining the unique target wake-up device corresponding to the target wake-up voice, the wake-up control device conducts pre-communication with the target wake-up device, sends a wake-up command to the target wake-up device through the target ad hoc network, detects the device wake-up status of the target wake-up device, and drives the target wake-up device to execute the target operation command corresponding to the target wake-up voice based on the device wake-up status.

[0084] Specifically, after constructing the target ad hoc network and determining the target wake-up device corresponding to the target wake-up voice, the wake-up control device communicates with the target wake-up device through the preset routing identifier information and preset ad hoc network protocol of the target ad hoc network, and reads the wake-up status identifier of the target wake-up device to determine the device wake-up status of the target wake-up device through the wake-up status identifier.

[0085] Optionally, if the target wake-up device is in an unwake-up state, the wake-up control device will re-wake up the target wake-up device through the target self-organizing network until the target wake-up device changes to a wake-up state.

[0086] Optionally, if the wake-up state of the target wake-up device is already awake, then the target wake-up device is driven to execute the target operation command carried by the target wake-up voice, or the target wake-up device is driven to output a successful wake-up reminder, and obtain the target operation command subsequently issued by the user, and drive the target wake-up device to execute the target operation command.

[0087] In this application, the wake-up control device responds to a device wake-up request, obtains the target wake-up voice corresponding to the device wake-up request, generates a self-organizing network wake-up broadcast corresponding to the target wake-up voice, performs wake-up arbitration on each candidate wake-up device according to the wake-up arbitration information, determines the target wake-up device among the candidate wake-up devices, and drives the target wake-up device to execute the target operation instruction corresponding to the target wake-up voice. This enables efficient communication and collaborative operation between devices in a control scenario with multiple smart home devices. After detecting a target wake-up voice, a self-organizing network wake-up broadcast corresponding to the target wake-up voice is generated. The target smart home device to be woken up is determined through the self-organizing network wake-up broadcast of this internal self-organizing network, and the target smart home device is driven to execute the operation corresponding to the wake-up voice. This achieves efficient communication and collaborative operation between devices, optimizes wake-up resource allocation, improves wake-up accuracy between multiple devices, and effectively reduces dependence on network resources.

[0088] like Figure 3 As shown, Figure 3 A flowchart illustrating an embodiment of the wake-up arbitration in the wake-up control method provided in this application is shown below. Figure 3 As shown, in this embodiment, the wake-up control method further includes steps 301 to 303:

[0089] 301. Obtain the sound source positioning angle and / or the device address identifier of each candidate wake-up device from the wake-up arbitration information;

[0090] 302. Obtain the target positioning angle from the sound source positioning angle, and determine the candidate wake-up device corresponding to the target positioning angle as the target wake-up device;

[0091] 303. Obtain the target address identifier from the device address identifier, and determine the candidate wake-up device corresponding to the target address identifier as the target wake-up device.

[0092] Based on the above embodiments, in this embodiment, after determining the candidate wake-up devices, the wake-up control device also sends the self-organizing network wake-up broadcast sent by each candidate wake-up device, as well as the wake-up arbitration information in the self-organizing network wake-up broadcast, and determines the target wake-up device most suitable for wake-up among the candidate wake-up devices through the wake-up arbitration information.

[0093] Specifically, when the wake-up control device cannot determine the unique target wake-up device using the wake-up sound pressure value and wake-up signal-to-noise ratio, the wake-up control device also determines the unique target wake-up device of the target wake-up voice by using the sound source positioning angle and / or device address identifier.

[0094] Specifically, the wake-up control device acquires the sound source positioning angle of the wake-up arbitration information corresponding to each candidate wake-up device, calculates the sound source angle difference between the sound source positioning angle and the target sound source, compares each sound source angle difference, determines the target sound source angle difference that is closest to the target sound source, and determines the sound source positioning angle corresponding to the target sound source angle difference as the target positioning angle. If the number of target positioning angles is unique, the candidate wake-up device corresponding to the target positioning angle is determined as the target wake-up device.

[0095] Optionally, in other embodiments, when the wake-up control device cannot determine a unique target wake-up device through other wake-up arbitration information, the wake-up control device further obtains the device address identifier in the wake-up arbitration information, and by comparing the device address identifier, determines the device address identifier with the smallest device address identifier as the target address identifier, and determines the candidate wake-up device corresponding to the target address identifier as the closest target wake-up device, and uses the target wake-up device as the unique wake-up device corresponding to the target wake-up voice.

[0096] Specifically, after determining the unique target wake-up device corresponding to the target wake-up voice, the wake-up control device conducts pre-communication with the target wake-up device, sends a wake-up command to the target wake-up device through the target ad hoc network, detects the device wake-up status of the target wake-up device, and drives the target wake-up device to execute the target operation command corresponding to the target wake-up voice based on the device wake-up status.

[0097] In this embodiment, the wake-up control device obtains the sound source localization angle and / or the device address identifiers of each candidate wake-up device from the wake-up arbitration information; obtains the target localization angle from the sound source localization angle, and determines the candidate wake-up device corresponding to the target localization angle as the target wake-up device; or, obtains the target address identifier from the device address identifiers, and determines the candidate wake-up device corresponding to the target address identifier as the target wake-up device. This achieves wake-up arbitration using multiple wake-up arbitration information sets, thereby accurately arbitrating the target wake-up device that is closest to the target wake-up voice from multiple dimensions, improving the reliability and accuracy of wake-up arbitration.

[0098] like Figure 4 As shown, Figure 4 This is a flowchart illustrating one embodiment of the wake-up control method for updating the wake-up word provided in this application. Specifically, in this embodiment, the wake-up control method further includes steps 401 to 403:

[0099] 401. Respond to the wake word self-learning request and obtain the initial wake word speech corresponding to the wake word self-learning request;

[0100] 402. Extract the initial wake-up features of the initial wake-up word speech, and use a preset speech recognition model to recognize the initial wake-up features to obtain the initial wake-up word;

[0101] 403. Synchronize the initial wake-up word of each network device in the target ad hoc network, and determine the device wake-up word of each network device based on the initial wake-up word and the candidate wake-up word.

[0102] Based on the above embodiments, in this embodiment, before responding to the target wake-up voice and determining the target wake-up device corresponding to the target wake-up voice, the wake-up control device also performs wake-up word self-learning for each network device in the target ad hoc network to generate device wake-up words for each network device.

[0103] Specifically, during operation, the wake-up control device responds to a wake-up word self-learning request and receives an initial wake-up word voice input by the user in a conversational manner, containing at least one custom wake-up word. This wake-up word self-learning request drives the wake-up control device to perform wake-up word self-learning on designated network devices to determine the operational instructions for the device wake-up words of the network devices.

[0104] Specifically, after acquiring the initial wake-up word voice, the wake-up control device detects the initial wake-up word voice to remove unqualified data in the initial wake-up word voice, and uses the wake-up control device or the corresponding self-learning networking device to output a re-input reminder in order to obtain a qualified initial wake-up word voice input by the user.

[0105] Specifically, after acquiring a qualified initial wake-up word speech, the wake-up control device performs preprocessing operations such as noise reduction and normalization on the initial wake-up word speech to obtain preprocessed initial wake-up word speech. It then extracts the acoustic signal features of the preprocessed initial wake-up word speech to obtain the initial wake-up features corresponding to the initial wake-up word speech. These initial wake-up features are acoustic signal features that include any one or more of the following characteristics: silver tone, volume, speech rate, and pitch. Optionally, in a specific embodiment, these features can be MFCC (Melbourne Frequency Cepstrahl Coefficients) features, spectrogram features, and speech-to-text features.

[0106] Specifically, after acquiring the initial wake-up feature, the wake-up control device further fine-tunes the model weights of the preset speech recognition model based on the initial wake-up feature to obtain the fine-tuned speech recognition model. In the subsequent self-learning process, the speech recognition model is used to recognize the initial wake-up feature to obtain the initial wake-up word of the network device.

[0107] Specifically, after setting the initial wake-up word, each network device also synchronizes the initial wake-up word of each network device in the target self-organizing network, detects the network device prefix of each network device in the target self-organizing network, groups the target self-organizing network devices according to the network device prefix, and assigns network devices with the same network device prefix to the corresponding network device combination to obtain the target network device combination with the same network device.

[0108] Specifically, after determining the target network device combination, handshake communication is performed between each network device in the target network device combination to determine if there are target network devices with the same wake-up word. If there are target network devices with the same wake-up word in the target network device combination, the initial wake-up word of any target network device is updated using the candidate wake-up word of the target network device to obtain the device wake-up word of the target network device. That is, if multiple target network devices have the same initial wake-up word / device wake-up word, a candidate wake-up word that is different from other target network devices is obtained for each target network device, and the candidate wake-up word is updated to the device wake-up word of the target network device to reduce the repetition of network device wake-up words and avoid the abnormal wake-up state of waking up multiple network devices at the same time. Here, the candidate wake-up word is a different wake-up word pre-set for each target network device. The candidate wake-up word can be pre-set according to the device type and usage scenario, and users can personalize or adjust the device's wake-up word through the wake-up control application.

[0109] Specifically, after determining the wake-up word of the target network device, the target network device is driven to verbally announce the wake-up word reminder corresponding to that wake-up word. For example, in a specific embodiment, the wake-up word reminder could be "Please wake me up with Xiao B Xiao B".

[0110] In this embodiment, the wake-up control device responds to a wake-up word self-learning request, obtains the initial wake-up word speech corresponding to the request, extracts the initial wake-up features of the initial wake-up word speech, and uses a preset speech recognition model to identify the initial wake-up features to obtain the initial wake-up word. The initial wake-up word of each network device is synchronized in the target ad hoc network, and the device wake-up word of each network device is determined based on the initial wake-up word and candidate wake-up words. This enables self-learning and network updates of the wake-up words of network devices, reducing the probability of repeated wake-up words among network devices, avoiding abnormal wake-up states where multiple network devices are simultaneously woken up, and improving the user's voice control experience.

[0111] To better implement the wake-up control method in the embodiments of this application, a wake-up control device is also provided in the embodiments of this application, such as... Figure 5 As shown, Figure 5 This is a schematic diagram of the structure of the wake-up control device provided in the embodiments of this application. Specifically, the wake-up control device 500 includes:

[0112] The voice acquisition module 501 is configured to respond to a device wake-up request and acquire the target wake-up voice corresponding to the device wake-up request.

[0113] The wake-up arbitration module 502 is configured to generate an ad hoc network wake-up broadcast corresponding to the target wake-up voice, perform wake-up arbitration on each candidate wake-up device according to the wake-up arbitration information, and determine the target wake-up device among the candidate wake-up devices.

[0114] The voice control module 503 is configured to drive the target wake-up device to execute the target operation command corresponding to the target wake-up voice.

[0115] In one possible implementation of this embodiment, the wake-up control device generates a self-organizing network wake-up broadcast corresponding to the target wake-up voice, performs wake-up arbitration on each candidate wake-up device according to the wake-up arbitration information, and determines the target wake-up device among the candidate wake-up devices, including:

[0116] Obtain the target wake-up word from the target wake-up voice, and determine the candidate wake-up devices corresponding to the target wake-up word for each networking device in the target ad hoc network;

[0117] Receive the self-organizing network wake-up broadcast corresponding to each of the candidate wake-up devices, and determine the wake-up arbitration information corresponding to the candidate wake-up devices;

[0118] Based on the wake-up arbitration information, wake-up arbitration is performed on each candidate wake-up device to determine the target wake-up device among the candidate wake-up devices.

[0119] In one possible implementation of this embodiment, the wake-up control device performs wake-up arbitration on each candidate wake-up device according to the wake-up arbitration information to determine the target wake-up device among the candidate wake-up devices, including:

[0120] Obtain the wake-up sound pressure level and / or the wake-up signal-to-noise ratio in the wake-up arbitration information;

[0121] Obtain the target wake-up sound pressure value from the wake-up sound pressure values, and determine the candidate wake-up device corresponding to the target wake-up sound pressure value as the target wake-up device;

[0122] Alternatively, if there are multiple target wake-up sound pressure levels, then the target wake-up signal-to-noise ratio is obtained from the wake-up signal-to-noise ratio, and the candidate wake-up device corresponding to the target wake-up signal-to-noise ratio is determined as the target wake-up device.

[0123] In one possible implementation of this embodiment, the wake-up control device performs wake-up arbitration on each candidate wake-up device according to the wake-up arbitration information to determine the target wake-up device among the candidate wake-up devices, including:

[0124] Obtain the sound source localization angle and / or the device address identifier of each candidate wake-up device from the wake-up arbitration information;

[0125] Obtain the target positioning angle from the sound source positioning angle, and determine the candidate wake-up device corresponding to the target positioning angle as the target wake-up device;

[0126] Alternatively, obtain the target address identifier from the device address identifier, and determine the candidate wake-up device corresponding to the target address identifier as the target wake-up device.

[0127] In one possible implementation of this embodiment, before the wake-up control device acquires the target wake-up word in the target wake-up voice and determines the candidate wake-up devices corresponding to the target wake-up word for each networking device in the target ad hoc network, it further includes:

[0128] In response to a wake word self-learning request, obtain the initial wake word speech corresponding to the wake word self-learning request;

[0129] Extract the initial wake-up features of the initial wake-up word speech, and use a preset speech recognition model to recognize the initial wake-up features to obtain the initial wake-up word;

[0130] In the target ad hoc network, the initial wake-up word of each network device is synchronized, and the device wake-up word of each network device is determined based on the initial wake-up word and the candidate wake-up word.

[0131] In one possible implementation of this embodiment, the wake-up control device synchronizes the initial wake-up word of each network device in the target ad hoc network, and determines the device wake-up word of each network device based on the initial wake-up word and candidate wake-up words, including:

[0132] Obtain the network device prefix of each network device in the target ad hoc network, and determine the target network device combination with the same network device prefix;

[0133] If there are target networking devices with the same initial wake-up word in the target networking device combination, then the initial wake-up word is updated using the candidate wake-up word of the target networking device to obtain the device wake-up word of the target networking device.

[0134] In one possible implementation of this embodiment, the wake-up control device drives the target wake-up device to execute the target operation command corresponding to the target wake-up voice, including:

[0135] Obtain the target operation command from the target wake-up voice;

[0136] Pre-communication is performed with the target wake-up device according to the preset routing identifier information and the preset self-organizing network protocol to determine the device wake-up status of the target wake-up device;

[0137] If the device is in a woke-up state, then the target wake-up device is driven to execute the target operation instruction.

[0138] In this embodiment, the wake-up control device responds to a device wake-up request, obtains the target wake-up voice corresponding to the device wake-up request, generates a self-organizing network wake-up broadcast corresponding to the target wake-up voice, performs wake-up arbitration on each candidate wake-up device according to the wake-up arbitration information, determines the target wake-up device among the candidate wake-up devices, and drives the target wake-up device to execute the target operation instruction corresponding to the target wake-up voice. This enables efficient communication and collaborative operation between devices in a control scenario with multiple smart home devices. After detecting a target wake-up voice, a self-organizing network wake-up broadcast corresponding to the target wake-up voice is generated. The target smart home device to be woken up is determined through the self-organizing network wake-up broadcast of this internal self-organizing network, and the target smart home device is driven to execute the operation corresponding to the wake-up voice. This achieves efficient communication and collaborative operation between devices, optimizes wake-up resource allocation, improves wake-up accuracy between multiple devices, and effectively reduces dependence on network resources.

[0139] This invention also provides a wake-up control device, such as... Figure 6 As shown, Figure 6 This is a schematic diagram of one embodiment of the wake-up control device provided in this application.

[0140] The wake-up control device integrates any of the wake-up control devices provided in the embodiments of the present invention, and the wake-up control device includes:

[0141] One or more processors;

[0142] Memory; and

[0143] One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor in the steps of the wake-up control method described in any of the embodiments of the above-described wake-up control method.

[0144] Specifically, the wake-up control device may include components such as a processor 601 with one or more processing cores, a memory 602 with one or more computer-readable storage media, a power supply 603, and an input unit 604. Those skilled in the art will understand that... Figure 6 The wake-up control device structure shown does not constitute a limitation on the wake-up control device and may include more or fewer components than shown, or combine certain components, or have different component arrangements. Wherein:

[0145] The processor 601 is the control center of the wake-up control device. It connects various parts of the wake-up control device via various interfaces and lines. By running or executing software programs and / or modules stored in the memory 602, and by calling data stored in the memory 602, it performs various functions and processes data of the wake-up control device, thereby providing overall monitoring of the wake-up control device. Optionally, the processor 601 may include one or more processing cores; preferably, the processor 601 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and applications, and the modem processor mainly handles wireless communication. It is understood that the modem processor may not be integrated into the processor 601.

[0146] The memory 602 can be used to store software programs and modules. The processor 601 executes various functional applications and data processing by running the software programs and modules stored in the memory 602. The memory 602 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created based on the use of the wake-up control device, etc. In addition, the memory 602 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 602 may also include a memory controller to provide the processor 601 with access to the memory 602.

[0147] The wake-up control device also includes a power supply 603 that supplies power to the various components. Preferably, the power supply 603 can be logically connected to the processor 601 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. The power supply 603 may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0148] The wake-up control device may also include an input unit 604, which can be used to receive input digital or character information, and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

[0149] Although not shown, the wake-up control device may also include a display unit, etc., which will not be described in detail here. Specifically, in this embodiment, the processor 601 in the wake-up control device loads the executable files corresponding to the processes of one or more applications into the memory 602 according to the following instructions, and the processor 601 runs the applications stored in the memory 602 to realize various functions, as follows:

[0150] In response to a device wake-up request, obtain the target wake-up voice corresponding to the device wake-up request;

[0151] Generate the self-organizing network wake-up broadcast corresponding to the target wake-up voice, perform wake-up arbitration on each candidate wake-up device according to the wake-up arbitration information, and determine the target wake-up device among the candidate wake-up devices;

[0152] Drive the target wake-up device to execute the target operation command corresponding to the target wake-up voice.

[0153] Therefore, embodiments of the present invention provide a computer-readable storage medium, which may include: read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, etc. A computer program is stored thereon, which is loaded by a processor to execute the steps in any of the wake-up control methods provided in the embodiments of the present invention. For example, the computer program loaded by the processor can execute the following steps:

[0154] In response to a device wake-up request, obtain the target wake-up voice corresponding to the device wake-up request;

[0155] Generate the self-organizing network wake-up broadcast corresponding to the target wake-up voice, perform wake-up arbitration on each candidate wake-up device according to the wake-up arbitration information, and determine the target wake-up device among the candidate wake-up devices;

[0156] Drive the target wake-up device to execute the target operation command corresponding to the target wake-up voice.

[0157] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the detailed descriptions of other embodiments above, which will not be repeated here.

[0158] In practice, each of the above units or structures can be implemented as an independent entity or can be arbitrarily combined to be implemented as the same or several entities. For the specific implementation of each of the above units or structures, please refer to the previous method embodiments, which will not be repeated here.

[0159] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0160] The wake-up control method provided by the embodiments of this application has been described in detail above. Specific embodiments have been used to illustrate the principle and implementation of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation and application scope based on the idea of ​​the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A wake-up control method, characterized in that, The wake-up control method includes: In response to a device wake-up request, obtain the target wake-up voice corresponding to the device wake-up request; Obtain the target wake-up word in the target wake-up voice, determine the candidate wake-up devices corresponding to the target wake-up word for each networking device in the target ad hoc network, receive the ad hoc network wake-up broadcast corresponding to each candidate wake-up device, determine the wake-up arbitration information corresponding to the candidate wake-up device, and obtain the wake-up sound pressure value and the wake-up signal-to-noise ratio in the wake-up arbitration information. Obtain the target wake-up sound pressure value from the wake-up sound pressure values. If there are multiple target wake-up sound pressure values, obtain the target wake-up signal-to-noise ratio from the wake-up signal-to-noise ratio. Then, determine the candidate wake-up device corresponding to the target wake-up signal-to-noise ratio as the target wake-up device. Drive the target wake-up device to execute the target operation command corresponding to the target wake-up voice.

2. The wake-up control method according to claim 1, characterized in that, Before obtaining the target wake-up word in the target wake-up voice and determining the candidate wake-up devices corresponding to the target wake-up word for each networking device in the target ad hoc network, the method further includes: In response to a wake word self-learning request, obtain the initial wake word speech corresponding to the wake word self-learning request; Extract the initial wake-up features of the initial wake-up word speech, and use a preset speech recognition model to recognize the initial wake-up features to obtain the initial wake-up word; In the target ad hoc network, the initial wake-up word of each network device is synchronized, and the device wake-up word of each network device is determined based on the initial wake-up word and the candidate wake-up word.

3. The wake-up control method according to claim 2, characterized in that, The process of synchronizing the initial wake-up word for each network device in the target ad hoc network, and determining the device wake-up word for each network device based on the initial wake-up word and candidate wake-up words, includes: Obtain the network device prefix of each network device in the target ad hoc network, and determine the target network device combination with the same network device prefix; If there are target networking devices with the same initial wake-up word in the target networking device combination, then the initial wake-up word is updated using the candidate wake-up word of the target networking device to obtain the device wake-up word of the target networking device.

4. The wake-up control method according to any one of claims 1-3, characterized in that, The step of driving the target wake-up device to execute the target operation instruction corresponding to the target wake-up voice includes: Obtain the target operation command from the target wake-up voice; Pre-communication is performed with the target wake-up device according to the preset routing identifier information and the preset self-organizing network protocol to determine the device wake-up status of the target wake-up device; If the device is in a woke-up state, then the target wake-up device is driven to execute the target operation instruction.

5. A wake-up control method, characterized in that, The wake-up control method includes: In response to a device wake-up request, obtain the target wake-up voice corresponding to the device wake-up request; Obtain the target wake-up word in the target wake-up voice, determine the candidate wake-up devices corresponding to the target wake-up word for each networking device in the target ad hoc network, receive the ad hoc network wake-up broadcast corresponding to each candidate wake-up device, and determine the wake-up arbitration information corresponding to the candidate wake-up device; Obtain the sound source localization angle and the device address identifier of each candidate wake-up device from the wake-up arbitration information; Obtain the target positioning angle from the sound source positioning angles. If there are multiple target positioning angles, obtain the target address identifier from the device address identifiers and determine the candidate wake-up device corresponding to the target address identifier as the target wake-up device. Drive the target wake-up device to execute the target operation command corresponding to the target wake-up voice.

6. The wake-up control method according to claim 5, characterized in that, Before obtaining the target wake-up word in the target wake-up voice and determining the candidate wake-up devices corresponding to the target wake-up word for each networking device in the target ad hoc network, the method further includes: In response to a wake word self-learning request, obtain the initial wake word speech corresponding to the wake word self-learning request; Extract the initial wake-up features of the initial wake-up word speech, and use a preset speech recognition model to recognize the initial wake-up features to obtain the initial wake-up word; In the target ad hoc network, the initial wake-up word of each network device is synchronized, and the device wake-up word of each network device is determined based on the initial wake-up word and the candidate wake-up word.

7. The wake-up control method according to claim 6, characterized in that, The process of synchronizing the initial wake-up word for each network device in the target ad hoc network, and determining the device wake-up word for each network device based on the initial wake-up word and candidate wake-up words, includes: Obtain the network device prefix of each network device in the target ad hoc network, and determine the target network device combination with the same network device prefix; If there are target networking devices with the same initial wake-up word in the target networking device combination, then the initial wake-up word is updated using the candidate wake-up word of the target networking device to obtain the device wake-up word of the target networking device.

8. A wake-up control device, characterized in that, The wake-up control device includes: The voice acquisition module is configured to respond to a device wake-up request and acquire the target wake-up voice corresponding to the device wake-up request; The wake-up arbitration module is configured to acquire the target wake-up word in the target wake-up voice, determine the candidate wake-up devices corresponding to the target wake-up word for each networking device in the target ad hoc network, receive the ad hoc network wake-up broadcast corresponding to each candidate wake-up device, determine the wake-up arbitration information corresponding to the candidate wake-up device, acquire the wake-up sound pressure value and the wake-up signal-to-noise ratio in the wake-up arbitration information, acquire the target wake-up sound pressure value in the wake-up sound pressure value, and if there are multiple target wake-up sound pressure values, acquire the target wake-up signal-to-noise ratio in the wake-up signal-to-noise ratio, and determine the candidate wake-up device corresponding to the target wake-up signal-to-noise ratio as the target wake-up device; The voice control module is configured to drive the target wake-up device to execute the target operation command corresponding to the target wake-up voice.

9. A wake-up control device, characterized in that, The wake-up control device includes: One or more processors; Memory; and One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor to implement the steps of the wake-up control method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It stores a computer program, which is loaded by a processor to execute the steps of the wake-up control method according to any one of claims 1 to 7.

Images