Device awakening method and voice control system

Abstract

The invention is suitable for the technical field of intelligent terminals, and provides an equipment wakeup method and a voice control system, and the method comprises the steps: obtaining a first data set corresponding to a tth wakeup process in a preset wakeup range; wherein the first data set comprises respective first energy values of the plurality of devices in the preset wake-up range; determining respective energy thresholds of the plurality of devices according to the first data set; and sending respective energy thresholds to the plurality of devices to indicate the plurality of devices to determine whether to execute a wake-up task according to the respective energy thresholds in the (t + 1) th wake-up process. Through the method, the voice wake-up efficiency can be improved, so that the experience degree of a user on the voice control equipment is improved.

Description

Technical Field

[0001] This application belongs to the field of smart terminal technology, and in particular relates to a device wake-up method and a voice control system. Background Technology

[0002] As voice recognition technology matures, its application in home appliances has become a trend. More and more households have multiple smart devices that support voice control, such as air conditioners, washing machines, water heaters, and refrigerators, all of which support voice wake-up functionality. In some scenarios, multiple smart devices with the same wake-up word may exist within the same wake-up range. In this case, when a user speaks the wake-up word, multiple devices may be activated, leading to accidental control and negatively impacting the user experience. Summary of the Invention

[0003] This application provides a device wake-up method and a voice control system, which can improve the efficiency of voice wake-up and thus enhance the user's experience with voice-controlled devices.

[0004] In a first aspect, embodiments of this application provide a device wake-up method, including:

[0005] Obtain the first dataset corresponding to the t-th wake-up process within the preset wake-up range; wherein, the first dataset includes the first energy value of each of the multiple devices within the preset wake-up range, and the first energy value is used to characterize the magnitude of the sound energy of the wake word detected by the device in the t-th wake-up process; The energy thresholds of the plurality of devices are determined based on the first dataset; Each of the multiple devices is sent its own energy threshold to instruct the multiple devices to determine whether to perform a wake-up task based on their respective energy thresholds during the (t+1)th wake-up process.

[0006] In this embodiment, the server determines the energy threshold of each of the multiple devices within a preset wake-up range based on their respective first energy values, and sends these energy thresholds to the devices. This allows the devices to determine whether to execute a wake-up task based on their respective energy thresholds during the wake-up process. By learning the energy thresholds of each device on the server side and determining whether to execute a wake-up task based on these thresholds during the wake-up process, the server can reduce interaction between the devices and the server, effectively reducing wake-up latency and improving wake-up efficiency. Furthermore, even if a communication failure occurs between the device and the server, as long as the device has acquired its energy threshold, it can still determine whether to execute a wake-up task based on these thresholds during the wake-up process, thereby increasing the wake-up success rate.

[0007] In one possible implementation of the first aspect, determining the energy threshold of each of the plurality of devices based on the first dataset includes: If there is only one target device among the multiple devices, the energy threshold of the target device is determined according to the first energy value of the target device in the first dataset; wherein, the target device is the device that is woken up during the t-th wake-up process.

[0008] In one possible implementation of the first aspect, determining the energy threshold of the target device based on a first energy value of the target device in the first dataset includes: If the target device is a theoretical device, then the energy threshold of the target device is determined based on the first energy value of the target device in the first dataset; wherein, the theoretical device is the device corresponding to the largest first energy value in the first dataset.

[0009] In one possible implementation of the first aspect, determining the energy threshold of the target device based on a first energy value of the target device in the first dataset includes: If the first dataset includes the first energy values ​​of all devices within the preset wake-up range, then the energy threshold of the target device is determined based on the first energy value of the target device in the first dataset.

[0010] In one possible implementation of the first aspect, determining the energy threshold of the target device based on a first energy value of the target device in the first dataset includes: If the number of second energy values ​​of the target device reaches a preset number, and the target device has an energy threshold, then the energy threshold of the target device is updated according to the energy threshold of the target device, the first energy value, and the preset number of second energy values; If the number of second energy values ​​of the target device reaches a preset number, and the target device does not have an energy threshold, then the energy threshold of the target device is determined based on the first energy value of the target device and the preset number of second energy values. The second energy value of the target device is used to characterize the magnitude of the sound energy of the wake word detected by the target device during the wake-up process prior to the t-th wake-up process.

[0011] In one possible implementation of the first aspect, after obtaining the first dataset corresponding to the t-th wake-up process within a preset wake-up range, the method further includes: If there are multiple target devices among the multiple devices, or if there is only one target device among the multiple devices and the target device is not a theoretical device, then delete the historical data of each of the multiple devices. Wherein, the target device is the device that is woken up during the t-th wake-up process, the theoretical device is the device corresponding to the largest first energy value in the first dataset; the historical data of the device includes the energy value detected by the device in each historical wake-up process and the energy threshold of the device; the historical wake-up process is the wake-up process before the t-th wake-up process.

[0012] In one possible implementation of the first aspect, determining the energy threshold of each of the plurality of devices based on the first dataset includes: If the target device is not among the plurality of devices, a wake-up command is sent to the theoretical device to instruct the theoretical device to perform a wake-up task; wherein, the theoretical device is the device corresponding to the largest first energy value in the first dataset; and the target device is the device that is woken up during the t-th wake-up process. The energy thresholds of the plurality of devices are determined based on the first dataset.

[0013] In one possible implementation of the first aspect, obtaining the first dataset corresponding to the t-th wake-up process within a preset wake-up range includes: When the first energy value reported by the first device within the preset wake-up range is received, the timer is started; If the timer has not reached the set time, the system continues to receive the first energy value reported by devices within the preset wake-up range. If the timer reaches its set time, the receiving of the first energy value reported by devices within the preset wake-up range will stop, and the first dataset will be generated based on the first energy value received during the timer's timeout period.

[0014] Secondly, embodiments of this application provide a voice control system, including: a server and multiple devices; wherein the server is configured to perform the device wake-up method as described in any one of the first aspects.

[0015] In one possible implementation of the second aspect, the device is configured as follows: If the energy threshold of the device has been received from the server, when a wake word is detected, it is determined whether the third energy value of the detected wake word is greater than the energy threshold. If the third energy value is greater than the energy threshold, then the wake-up task is executed; The third energy value is reported to the server.

[0016] In one possible implementation of the second aspect, the device is configured as follows: If the energy threshold of the device is not received from the server, when a wake word is detected, the third energy value of the detected wake word is reported to the server. If a wake-up command is received from the server, the wake-up task is executed.

[0017] Thirdly, embodiments of this application provide a device wake-up device, including: An acquisition unit is used to acquire a first dataset corresponding to the t-th wake-up process within a preset wake-up range; wherein, the first dataset includes first energy values ​​of multiple devices within the preset wake-up range, and the first energy value is used to characterize the magnitude of the sound energy of the wake word detected by the device in the t-th wake-up process; A determining unit is configured to determine the energy threshold of each of the plurality of devices based on the first dataset; The sending unit is used to send their respective energy thresholds to the plurality of devices, so as to instruct the plurality of devices to determine whether to perform a wake-up task based on their respective energy thresholds during the (t+1)th wake-up process.

[0018] Fourthly, embodiments of this application provide a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the device wake-up method as described in any one of the first aspects above.

[0019] Fifthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the device wake-up method as described in any one of the first aspects above.

[0020] Sixthly, embodiments of this application provide a computer program product that, when run on a terminal device, causes the terminal device to execute the device wake-up method described in any of the first aspects above.

[0021] It is understood that the beneficial effects of the second to sixth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1This is a schematic diagram illustrating the application scenario provided in the embodiments of this application; Figure 2 This is a schematic diagram of another application scenario provided by an embodiment of this application; Figure 3 This is a schematic diagram of the voice control system provided in an embodiment of this application; Figure 4 This is a schematic diagram of the processing flow of device 32 provided in the embodiments of this application; Figure 5 This is a flowchart illustrating the device wake-up method provided in an embodiment of this application; Figure 6 This is a schematic diagram of the server processing flow provided in the embodiments of this application; Figure 7 This is a structural block diagram of the device wake-up device provided in the embodiments of this application; Figure 8 This is a schematic diagram of the structure of the terminal device provided in the embodiments of this application. Detailed Implementation

[0024] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0025] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0026] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0027] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0028] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0029] References to "one embodiment" or "some embodiments" in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized.

[0030] As voice recognition technology matures, its application in home appliances has become a trend. More and more households have multiple smart devices that support voice control, such as air conditioners, washing machines, water heaters, and refrigerators, all of which support voice wake-up functionality.

[0031] In some application scenarios, multiple smart devices with the same wake word may exist within the same wake-up range. For example, see [link to example]. Figure 1 This is a schematic diagram illustrating an application scenario provided in an embodiment of this application. For example... Figure 1 As shown, when a user speaks a wake word, devices 11, 12, and 13 can all receive the wake word, meaning they are within the same wake-up range. If the wake words for devices 11, 12, and 13 are the same (e.g., "XXX"), in one scenario, the user expects to wake up device 11, but when the user speaks "XXX", all three devices are woken up, leading to accidental activation and negatively impacting the user experience.

[0032] To address the aforementioned issues, some embodiments propose a unique wake-up method. For example, see [link to example]. Figure 2 This is a schematic diagram of another application scenario provided in the embodiments of this application. Figure 2 In the application scenario shown, it is in Figure 1 A server is added to the application scenario shown. When the user speaks the wake word "XXX", devices 11, 12, and 13 all receive the wake word "XXX" and send their respective energy values ​​for that wake word to the server. Correspondingly, the server determines the uniquely awakened device based on the received energy values ​​from each device and sends a wake-up command to the determined device, instructing it to perform the wake-up task. For example, the server determines the device with the highest energy value among the received energy values ​​as the uniquely awakened device.

[0033] Using the above method, only one smart device among multiple smart devices with the same wake-up word within the same wake-up range is uniquely woken up. In this unique wake-up method, each device needs to send its energy value to the server during each wake-up process. The server then decides which device is uniquely woken up before issuing the wake-up command. In this approach, each device needs to wait for the server's decision, and the server needs to obtain the energy values ​​sent by all devices within the same wake-up range before making a decision. This increases wake-up latency, resulting in low wake-up efficiency. Furthermore, if a communication failure occurs between a device and the server, the server may be unable to make a decision, leading to wake-up failure.

[0034] Based on this, this application provides a device wake-up method. In this embodiment, the server determines the energy threshold of each of the multiple devices within a preset wake-up range based on their respective first energy values, and sends the respective energy thresholds to the multiple devices. This allows the multiple devices to determine whether to execute a wake-up task based on their respective energy thresholds during the wake-up process. Through this method, the server learns the energy thresholds of each device, and the devices determine whether to execute a wake-up task based on these thresholds during the wake-up process. This eliminates the need to wait for the server's decision, reducing interaction between the device and the server, effectively reducing wake-up latency, and improving wake-up efficiency. Furthermore, even if a communication failure occurs between the device and the server, as long as the device has acquired the energy threshold, it can determine whether to execute a wake-up task based on the energy threshold during the wake-up process, thereby improving the wake-up success rate.

[0035] First, let's introduce the voice control system. See [link / reference] Figure 3 This is a schematic diagram of the voice control system provided in an embodiment of this application. Figure 3 As shown, the voice control system includes a server 31 and multiple devices 32.

[0036] The server 31 is configured to execute the device wake-up method described in the following embodiments. It is understood that by executing the device wake-up method, the server 31 can determine the energy threshold of each of the multiple devices 32, and send the respective energy thresholds to the multiple devices 32 to instruct them to determine whether to perform a wake-up task based on their respective energy thresholds.

[0037] In one embodiment, device 32 is configured to: If the device's energy threshold has been received from the server, when a wake word is detected, it is determined whether the third energy value of the detected wake word is greater than the energy threshold. If the third energy value is greater than the energy threshold, then the wake-up task is executed and the third energy value is reported to the server; If the third energy value is not greater than the energy threshold, then the third energy value is reported to the server.

[0038] In another embodiment, device 21 is configured to: If the device's energy threshold is not received from the server, when a wake word is detected, the third energy value of the detected wake word is reported to the server. If a wake-up command is received from the server, the wake-up task will be executed.

[0039] For example, see Figure 4 This is a schematic diagram of the processing flow of device 32 provided in an embodiment of this application. It is intended as an example and not a limitation. Figure 4 As shown, the processing flow of device 32 may include the following steps: S401, when the device detects a wake word, it acquires the third energy value of the wake word.

[0040] S402, the device determines whether an energy threshold exists locally.

[0041] It's understandable that the server's process of learning the energy thresholds for each device can run in parallel with the device's processing flow. In some cases, the server has already sent the energy threshold to the device before it detects the wake word; in this case, the device has a locally available energy threshold. In other cases, the server has not yet sent the energy threshold to the device before it detects the wake word; in this case, the device does not have a locally available energy threshold.

[0042] S403, if a local energy threshold exists, the device determines whether the third energy value is greater than the energy threshold.

[0043] S404, If the third energy value is greater than the energy threshold, the device will perform a wake-up task.

[0044] S405, the device reports the third energy value to the server.

[0045] S406 If the third energy value is not greater than the energy threshold, or if the energy threshold does not exist locally, the device reports the third energy value to the server.

[0046] Optionally, when the device reports the third energy threshold to the server, it may carry identification information indicating whether a wake-up task has been performed. For example, in step S405, when the device reports the third energy threshold to the server, the identification information indicates that the device has performed a wake-up task; in step S406, when the device reports the third energy threshold to the server, the identification information indicates that the device has not performed a wake-up task.

[0047] In some implementations, the device can enable the unique wake-up feature by default, or the user can choose to enable or disable it. For example, users can configure the device through its application to choose whether to enable or disable the unique wake-up feature.

[0048] Optionally, if the device has enabled the unique wake-up function, the server sends a registration interface to the device. When the device recognizes the wake word, it calls the registration interface to send registration information to the server. The registration information may include the received wake word, the third energy value, the device ID, and identification information indicating whether a wake-up task has been performed, etc.

[0049] As can be seen from the above processing flow, the energy threshold of each device is learned on the server side. During the wake-up process, the device side determines whether to execute the wake-up task based on the energy threshold, without waiting for the server's decision. This reduces the interaction between the device and the server, effectively reduces wake-up latency, and helps improve wake-up efficiency. In addition, even if the device and the server experience a communication failure, as long as the device has acquired the energy threshold, it can determine whether to execute the wake-up task based on the energy threshold during the wake-up process, thereby improving the wake-up success rate.

[0050] See Figure 5 This is a flowchart illustrating the device wake-up method provided in an embodiment of this application. It is understood that the device wake-up method in this embodiment can be executed by the server in the aforementioned voice control system. As an example and not a limitation, the method may include the following steps: S501, obtain the first dataset corresponding to the t-th wake-up process within the preset wake-up range.

[0051] The first dataset includes the first energy values ​​of multiple devices within the preset wake-up range. These first energy values ​​characterize the magnitude of the sound energy of the wake-up word detected by the device during the t-th wake-up process. For example, the energy value can be volume, sound energy, sound intensity, loudness, etc. Factors such as the distance and angle between the user and the device, and the volume of the user's voice, all affect the magnitude of the energy value received by the device when it recognizes the wake-up word.

[0052] Understandably, the first dataset includes the energy values ​​of various devices within the same wake-up range (i.e., the preset wake-up range) receiving the same wake-up word. This first dataset serves as a learning sample for the server. In other words, when learning the energy thresholds of devices, the server learns the energy thresholds of each device within that wake-up range based on the energy values ​​of each device within that wake-up range receiving the same wake-up word.

[0053] For example, Table 1 shows the energy values ​​of the wake-up word "Hello XX" recognized by multiple devices during four wake-up processes when the user says the wake-up word "Hello XX".

[0054] Table 1 equipment Family ID wake word Unique wake-up switch First energy value Second energy value 3rd energy value 4th energy value A 100001 Hello XX open 90 65 52 40 B 100001 Hello XX open 25 35 45 60 C 100001 Hello XX close 10 20 60 55 D 100001 Hello YY open E 100002 Hello XX open 5 10 20 35 As shown in Table 1, the home ID of device E is different from that of devices A, B, C, and D, indicating that devices A, B, C, and D belong to the same wake-up range, while device E belongs to a different wake-up range. Therefore, the energy value of device E cannot be used as the same learning sample as the energy values ​​of devices A, B, C, and D. During the four wake-up processes, the wake-up word of device D is different from that of devices A, B, and C. Therefore, although device D belongs to the same wake-up range as devices A, B, and C, it does not need to respond when the wake-up word "Hello XX" is spoken. Consequently, the energy value of the wake-up word "Hello XX" recognized by device D is 0. Although device C belongs to the same wake-up range as devices A and B, its only wake-up function is in a disabled state, meaning there is no need to learn the energy threshold of device C. Therefore, the energy value of device C is not included in the learning samples. Based on this, only the energy values ​​of device A and device B can be used as learning samples. Specifically, the first energy value (90) of device A and the first energy value (25) of device B are recorded as the first dataset, the second energy value (65) of device A and the second energy value (35) of device B are recorded as the second dataset, the third energy value (52) of device A and the third energy value (45) of device B are recorded as the third dataset, and the fourth energy value (40) of device A and the fourth energy value (60) of device B are recorded as the fourth dataset.

[0055] In one embodiment, S501 includes: When the first energy value reported by the first device within the preset wake-up range is received, the timer is started; If the timer has not reached the set time, it will continue to receive the first energy value reported by devices within the preset wake-up range; If the timer reaches its set time, the receiving of the first energy value reported by devices within the preset wake-up range will stop, and the first dataset will be generated based on the first energy value received during the timer's timeout period.

[0056] The method described above effectively sets a time window for each wake-up process, ensuring that data received within that window belongs to the same wake-up process. This reduces data congestion between different wake-up processes, improves the accuracy of the dataset used to learn the energy threshold, and ultimately enhances the learning accuracy of the energy threshold.

[0057] Optionally, when the first energy value reported by device A is received, it is determined whether energy values ​​reported by other devices within the same wake-up range as device A have been received within a preset time period prior to receiving the first energy value reported by device A. If not, it means that device A is the first device to report an energy value within the preset wake-up range, and a timer is started. If so, it means that device A is not the first device to report an energy value within the preset wake-up range, and the first energy value reported by device A is added to the dataset corresponding to the wake-up range to which device A belongs.

[0058] like Figure 4 As described in the embodiments, the device can report registration information such as the received wake-up word, third energy value, and device ID to the server. Optionally, the server can determine the wake-up range to which the device belongs based on the device ID in the device's registration information. For example, devices within the same wake-up range may have the same flag information in their device IDs (such as the family ID shown in Table 1), and the server can determine the wake-up range to which the device belongs based on the flag information in the device ID.

[0059] In one example, when an energy value 'a' is reported by device A, it is determined whether any other devices within the same wake-up range as device A have reported energy values ​​within a preset time period prior to receiving the energy value 'a' reported by device A. If not, it indicates that device A is the first device within the preset wake-up range to report an energy value, and a timer is started. During the timer's duration, each time the server receives registration information reported by device B, it determines whether device B belongs to the same wake-up range as device A based on the device ID in device B's registration information, and whether the wake-up word in device B's registration information is the same as the wake-up word in device A's registration information. If yes, it indicates that the registration information reported by device B and device A belong to the same wake-up process, and the energy value 'b' reported by device B is added to the first dataset. If no, it indicates that the registration information reported by device B and device A do not belong to the same wake-up process, and another timer is started to generate a second dataset, with device B as the first device in the second dataset.

[0060] Understandably, a server can "serve" multiple devices within a single wake-up range, or it can "serve" multiple devices within multiple wake-up ranges. In the case where the server "serves" multiple devices within multiple wake-up ranges, as illustrated in the example above, it may learn the energy thresholds of multiple preset wake-up range devices in parallel.

[0061] In one embodiment, after S501, the method includes: If there are multiple target devices among multiple devices, or if there is only one target device among multiple devices and the target device is not the theoretical device, then delete the historical data of each device among the multiple devices; If only one target device exists among multiple devices, or if no target device exists among multiple devices, then execute S502.

[0062] Wherein, the target device is the device that is woken up during the t-th wake-up process, the theoretical device is the device corresponding to the largest first energy value in the first dataset; the historical data of the device includes the energy value detected by the device in each historical wake-up process and the energy threshold of the device; the historical wake-up process is the wake-up process before the t-th wake-up process.

[0063] For example, during the timer's countdown, the server receives registration information reported by multiple devices. Based on the identifier information in the registration information indicating whether a wake-up task has been executed, it determines whether a target device exists among the multiple devices and the number of target devices. Based on the energy value in the registration information, it determines the theoretical device with the largest energy value among the multiple devices. If multiple target devices exist among the multiple devices, or if only one target device exists among the multiple devices and the target device is not a theoretical device, then the historical data of each device among the multiple devices is deleted. If only one target device exists among the multiple devices, or if no target device exists among the multiple devices, then S502 is executed.

[0064] Continuing with the example in Table 1, for the third wake-up process, assuming the actual result of the third wake-up process is that device A is not woken up and device B is woken up, then the target device is device B, which is different from the theoretical device (device A). Therefore, the first, second, and third energy values ​​of device A, as well as the energy threshold of device A determined based on the first, second, and third energy values ​​of device A, are deleted.

[0065] In the above embodiments, conditions are essentially set for the learning samples (i.e., the dataset used for the energy threshold) for the energy threshold. When the conditions are met, such as when only one target device exists among multiple devices, or when no target device exists among multiple devices, the learning samples are valid, meaning the energy threshold can be learned from the learning samples. When the conditions are not met, such as when multiple target devices exist among multiple devices, or when only one target device exists among multiple devices and the target device is not the theoretical device, the learning samples are invalid. This method achieves the screening of learning samples, improves the accuracy of the learning samples, and helps improve the learning accuracy of the energy threshold.

[0066] S502, determine the energy thresholds of multiple devices based on the first dataset.

[0067] In one embodiment, S502 includes: determining the energy threshold of each of the plurality of devices based on a first dataset.

[0068] In another embodiment, S502 includes: If there is only one target device among multiple devices, the energy threshold of the target device is determined based on the first energy value of the target device in the first dataset; where the target device is the device that is woken up during the t-th wake-up process.

[0069] In this embodiment, a learning condition is set to learn only the energy threshold of the target device among multiple devices, that is, only the energy threshold of the awakened device among multiple devices. Since the awakening of a device indicates that the energy value of the device is valid, while the energy value of an unawakened device may be invalid, learning only the energy threshold of the awakened device can reduce the impact of invalid energy values ​​on the learning accuracy of the energy threshold, thereby improving the learning accuracy of the capability threshold.

[0070] In one implementation, if there is only one target device among multiple devices, and the target device is a theoretical device, then the energy threshold of the target device is determined based on the first energy value of the target device in the first dataset.

[0071] Understandably, the target device is the device actually woken up during the wake-up process, while the theoretical device is determined by the server based on the energy values ​​reported by each device. In other words, the target device represents the actual execution result on the device side, while the theoretical device represents the decision result on the server side. When the target device is a theoretical device, it means that the actual execution result on the device side is consistent with the decision result on the server side; when the target device is not a theoretical device, it means that the actual execution result on the device side is inconsistent with the decision result on the server side.

[0072] Continuing with the example in Table 1, for the first wake-up process, assuming the actual result is that device A is woken up and device B is not, then the target device is device A. The server determines the theoretical device as device A (with the highest energy value) based on the dataset from the first wake-up process (including the first energy value of device A (90) and device B (25)). Since the target device is the theoretical device, the energy threshold for device A is determined based on its first energy value of 90. There is no need to determine the energy threshold for device B in this case.

[0073] The above implementation effectively validates the learning samples. The energy threshold of the target device is only learned when the target device is a theoretical device—that is, when the actual execution result on the device side matches the decision result on the server side. This approach reduces the impact of invalid learning samples on the learning accuracy of the energy threshold, thereby improving the learning accuracy of the capability threshold.

[0074] In one implementation, if the first dataset includes the first energy values ​​of all devices within a preset wake-up range, then the energy threshold of the target device is determined based on the first energy value of the target device in the first dataset.

[0075] It is understandable that the first dataset includes the first energy values ​​of all devices within the preset wake-up range that have enabled the unique wake-up function and have the same wake-up word. The energy threshold of the target device is then determined based on the first energy value of the target device in the first dataset.

[0076] Continuing with the example in Table 1, if the preset wake-up range is home ID 100001, the devices with the unique wake-up function enabled within this preset wake-up range include device A, device B, and device D. Device D's wake-up word is different from that of devices A and B. Therefore, the first data set for devices A and B needs to include the first energy values ​​of both devices A and B. If the first data set lacks the first energy value of device A or device B, it is considered invalid. If the first data set includes the first energy values ​​of both devices A and B, it is considered valid. Then, the energy threshold of the target device is determined based on the first energy value of the target device in the first data set.

[0077] The above implementation effectively validates the learning samples. The first dataset includes the first energy values ​​of all devices within a preset wake-up range; that is, the energy threshold of the target device is only learned if the learning samples are valid. This approach reduces the impact of invalid learning samples on the learning accuracy of the energy threshold, thereby improving the learning accuracy of the capability threshold.

[0078] In one implementation, the step of determining the energy threshold of the target device based on the first energy value of the target device in the first dataset includes: If the number of second energy values ​​of the target device reaches a preset number, and the target device does not have an energy threshold, then the energy threshold of the target device is determined based on the first energy value of the target device and the preset number of second energy values.

[0079] The second energy value of the target device is used to characterize the magnitude of the sound energy of the wake word detected by the target device during the wake-up process before the tth wake-up process.

[0080] Continuing with the example in Table 1, during the four wake-up cycles, the devices actually woken up are as follows: device A is woken up in the first three cycles, and device B is woken up in the fourth cycle. During the first three wake-up cycles, the energy value of device A is a valid sample, and during the fourth wake-up cycle, the energy value of device B is also a valid sample. During the fourth wake-up cycle, device A has two second energy values ​​(i.e., device A's first energy value of 90 and second energy value of 65), while device B has zero second energy values. Assuming the preset quantity is 2, the number of second energy values ​​for device A reaches the preset quantity. Therefore, the energy threshold for device A is calculated based on its two second energy values ​​(i.e., device A's first energy value of 90 and second energy value of 65) and its first energy value (i.e., device A's third energy value of 52). Since the number of second energy values ​​for device B does not reach the preset quantity, there is no need to calculate the energy threshold for device B.

[0081] Optionally, a statistical value of the second energy value and the first energy value of the target device can be calculated as an energy threshold for the target device. For example, the statistical value can be the average value or the maximum value.

[0082] Understandably, in some cases, the average energy values ​​from multiple devices may be the same, meaning multiple devices may have the same energy threshold, which could lead to false alarms. To address this issue, it is preferable that the statistical value be the maximum value. This reduces the likelihood of multiple devices being woken up simultaneously due to the same average value, thus improving the accuracy of unique wake-up.

[0083] To further improve the reliability of the learning samples, optionally, it can be determined whether the difference between the maximum energy value and the second largest energy value in the first dataset is greater than a preset value; if so, the first dataset is used as a valid learning sample; otherwise, the first dataset is ignored.

[0084] For example, see Table 2, which shows the energy values ​​reported by Device A and Device B during the last 20 wake-up processes.

[0085] Table 2 equipment 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 A 65 61 60 63 67 63 64 66 69 58 62 64 61 60 69 64 65 62 61 63 B 61 59 56 62 66 62 58 63 67 55 59 63 57 58 62 61 64 60 60 62 Difference 4 2 4 1 1 1 6 3 2 3 3 2 4 2 7 3 1 2 1 1 As shown in Table 2, assuming the preset value is 5, the energy difference 6 between device A and device B during the 7th wake-up process is greater than the preset value 5, and the energy difference 7 between device A and device B during the 15th wake-up process is greater than the preset value 5. Therefore, only the datasets from the 7th and 15th wake-up processes can be used as effective learning samples.

[0086] In another implementation, the step of determining the energy threshold of the target device based on the first energy value of the target device in the first dataset includes: If the number of second energy values ​​of the target device reaches a preset number, and the target device has an energy threshold, then the energy threshold of the target device is updated according to the energy threshold of the target device, the first energy value, and the preset number of second energy values.

[0087] Continuing the example above, during the fourth wake-up process, the energy threshold of device A is determined to be 69 (i.e., the average of the first energy value of device A (90), the second energy value (65), and the third energy value). Assuming that during the fifth wake-up process, the fifth energy value of device A is 75, the fifth energy threshold of device B is 60, and device A is actually woken up during the fifth wake-up process, then device A has three second energy values ​​(i.e., the first energy value of device A (90), the second energy value (65), and the third energy value (52), and device B has one second energy value (i.e., the fourth energy value of device B (60)). Since the number of second energy values ​​(3) of device A reaches the preset number (2), the energy threshold of device A is calculated and updated based on the three second energy values ​​of device A (i.e., the first energy value of device A (90), the second energy value (65), and the third energy value (52)) and the first energy value of device A (i.e., the fifth energy value of device A (75)). Since the number of second energy values ​​of device B does not reach the preset number, there is no need to calculate the energy threshold of device B.

[0088] The method described above is equivalent to learning the device's energy threshold based on multiple learning samples. Compared to determining the energy threshold based on a single learning sample, the calculated energy threshold is more reliable. Furthermore, as the number of wake-ups increases, this method can continuously update the device's energy threshold, ensuring that the learned energy threshold reflects real-world application scenarios and thus improving the accuracy of unique wake-up.

[0089] Optionally, for the initial wake-up process, the energy threshold of the target device can be set based on the energy value of the target device during the first wake-up process. Then, during subsequent wake-up processes, when the number of second energy values ​​of the target device reaches a preset number, the energy threshold of the target device can be updated.

[0090] Optionally, for the initial wake-up process, different initial values ​​of energy thresholds are set for each device, and then in subsequent wake-up processes, when the number of second energy values ​​of a certain device reaches a preset number, the energy threshold of that device is updated.

[0091] Optionally, for the initial wake-up process, the energy threshold for each device is empty. In subsequent wake-up processes, the energy threshold for a device is only calculated when the number of second energy values ​​for that device reaches a preset number.

[0092] In another embodiment, S502 includes: If the target device is not among the multiple devices, a wake-up command is sent to the theoretical device to instruct it to perform a wake-up task; the energy thresholds of each of the multiple devices are determined based on the first dataset.

[0093] Continuing with the example in Table 1, during the first wake-up process, neither device A nor device B has an energy threshold. In this case, devices A and B cannot make a decision locally based on the energy threshold and need to report their respective energy values ​​to the server for decision-making. That is, there is no target device among the multiple devices. Accordingly, the server determines the theoretical device based on the first energy values ​​received from device A and device B. For example, the device corresponding to the maximum value between the first energy values ​​of device A and device B is determined as the theoretical device (i.e., device A), and a wake-up command is sent to device A. After device A executes the wake-up task, device A can be used as the target device in the first wake-up process. Since the target device in the first wake-up process is the same as the theoretical device, the energy threshold of device A can be determined based on its first energy value.

[0094] S503 sends its own energy threshold to multiple devices to instruct the multiple devices to determine whether to perform a wake-up task based on their respective energy thresholds during the t+1 wake-up process.

[0095] The processing flow for each device can be found in [reference]. Figure 4 The descriptions in the embodiments will not be repeated here.

[0096] Continuing with the example in Table 1, suppose that during the 4th wake-up process, the energy threshold for device A is determined to be 69, while the energy threshold for device B is not determined. In this case, the server sends the energy threshold for device A to device A. During the 5th wake-up process, suppose device A recognizes a 5th energy value of 75 for the wake word. Device A determines locally that this 5th energy value is greater than the energy threshold, so it executes the wake-up task and reports its registration information to the server. Since device B does not have an energy threshold, device B reports its 5th energy value to the server and waits for instructions. Based on the identifier in the registration information reported by device A indicating that a wake-up task has been executed, the server determines that device A has executed the wake-up task. Therefore, it does not send a wake-up command to device B again, or it can send a command to device B indicating that a wake-up is not required.

[0097] For example, see Figure 6 This is a schematic diagram of the server processing flow provided in an embodiment of this application. For example... Figure 6 As shown, the server's processing flow includes the following steps: S601, after the processing flow begins, the server receives registration information reported by each device during the timer period and generates the first dataset based on the registration information.

[0098] Understandably, upon receiving registration information from the first device within the preset wake-up range, a timer is started, thus initiating the processing flow. Of course, the first dataset also includes the energy value from the registration information reported by the first device.

[0099] S602, the server determines the theoretical device based on the first dataset.

[0100] Optionally, the server may identify the device with the largest energy value in the first dataset as the theoretical device.

[0101] S603, the server determines whether the target device exists during this wake-up process.

[0102] If a target device exists during this wake-up process, then execute S604; if no target device exists during this wake-up process, then execute S610.

[0103] In one implementation, the registration information includes identification information indicating whether the device is performing a wake-up task. Accordingly, the server can determine the target device based on this identification information.

[0104] S604, if the theoretical device is the target device, the server determines whether the target device is the theoretical device.

[0105] If the theoretical device is the target device, then execute S606; if the theoretical device is not the target device, then execute S605.

[0106] S605: If the theoretical device is not the target device, the server deletes the historical data of all devices within the preset wake-up range.

[0107] It is understandable that this refers to all devices within the preset wake-up range that have enabled the unique wake-up function.

[0108] Optionally, if there are multiple target devices, the case can be handled as if the theoretical device is not a target device.

[0109] S606, if the theoretical device is the target device, the server determines whether the first dataset includes the energy values ​​of all devices within the preset wake-up range.

[0110] Understandably, this refers to the energy values ​​of all devices within the preset wake-up range that have enabled the unique wake-up function in the first dataset.

[0111] If the first dataset includes the energy values ​​of all devices within the preset wake-up range, then execute S607; if the first dataset does not include the energy value of any device within the preset wake-up range, then end the current processing flow.

[0112] S607, if the first dataset includes the energy values ​​of all devices within the preset wake-up range, the server will use the energy value of the target device in the first dataset as one learning sample of the target device.

[0113] S608, the server determines whether the number of learning samples for the target device has reached N.

[0114] It is understandable that N is the number of second energy values ​​of the target device + 1, that is, the learning samples of the target device at this time include a preset number (N-1) of second energy values ​​(historical data) and first energy values ​​(data of this wake-up process).

[0115] If the number of learning samples for the target device reaches N, then execute S609; if the number of learning samples for the target device does not reach N, then end the current processing flow.

[0116] S609, if the number of learning samples of the target device reaches N, the server determines the energy threshold of the target device based on the latest N learning samples of the target device.

[0117] Steps S606-S609 can be found in the description of the relevant embodiments of step S502, and will not be repeated here.

[0118] It should be noted that in practical applications, the decision steps of S603, S604, S606, and S608 can be executed one by one or all of them. In addition, the decision steps of S604, S606, and S608 can be executed in parallel or sequentially. Figure 6 This is merely an example of serial execution, and the specific order of the judgment steps in S604, S606, and S608 is not specified.

[0119] S610, if the target device does not exist in this wake-up process, the server sends a wake-up command to the theoretical device to instruct the theoretical device to perform the wake-up task.

[0120] Understandably, after the theoretical device performs the wake-up task, the theoretical device is recorded as the target device in this wake-up process, and S606 continues to be executed.

[0121] In some implementations, if the number of devices with the unique wake-up function enabled within the preset wake-up range changes, or the wake word of any device changes, the server deletes the historical data of all devices with the unique wake-up function enabled within the preset wake-up range and relearns the function.

[0122] In this embodiment, the server determines the energy threshold of each of the multiple devices within a preset wake-up range based on their respective first energy values, and sends these energy thresholds to the devices. This allows the devices to determine whether to execute a wake-up task based on their respective energy thresholds during the wake-up process. By learning the energy thresholds of each device on the server side and determining whether to execute a wake-up task based on these thresholds during the wake-up process, the server can reduce interaction between the devices and the server, effectively reducing wake-up latency and improving wake-up efficiency. Furthermore, even if a communication failure occurs between the device and the server, as long as the device has acquired its energy threshold, it can still determine whether to execute a wake-up task based on these thresholds during the wake-up process, thereby increasing the wake-up success rate.

[0123] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0124] Corresponding to the device wake-up method described in the above embodiments, Figure 7 This is a structural block diagram of the device wake-up device provided in the embodiments of this application. For ease of explanation, only the parts related to the embodiments of this application are shown.

[0125] Reference Figure 7 The device wake-up device 7 includes: The acquisition unit 71 is used to acquire a first dataset corresponding to the t-th wake-up process within a preset wake-up range; wherein, the first dataset includes the first energy values ​​of each of the multiple devices within the preset wake-up range, and the first energy value is used to characterize the magnitude of the sound energy of the wake word detected by the device in the t-th wake-up process.

[0126] The determining unit 72 is used to determine the energy threshold of each of the plurality of devices based on the first dataset.

[0127] The sending unit 73 is used to send its respective energy threshold to the plurality of devices, so as to instruct the plurality of devices to determine whether to perform a wake-up task based on their respective energy thresholds during the t+1 wake-up process.

[0128] Optionally, the determining unit 72 is also used for: If there is only one target device among the multiple devices, the energy threshold of the target device is determined according to the first energy value of the target device in the first dataset; wherein, the target device is the device that is woken up during the t-th wake-up process.

[0129] Optionally, the determining unit 72 is also used for: If the target device is a theoretical device, then the energy threshold of the target device is determined based on the first energy value of the target device in the first dataset; wherein, the theoretical device is the device corresponding to the largest first energy value in the first dataset.

[0130] Optionally, the determining unit 72 is also used for: If the first dataset includes the first energy values ​​of all devices within the preset wake-up range, then the energy threshold of the target device is determined based on the first energy value of the target device in the first dataset.

[0131] Optionally, the determining unit 72 is also used for: If the number of second energy values ​​of the target device reaches a preset number, and the target device has an energy threshold, then the energy threshold of the target device is updated according to the energy threshold of the target device, the first energy value, and the preset number of second energy values; If the number of second energy values ​​of the target device reaches a preset number, and the target device does not have an energy threshold, then the energy threshold of the target device is determined based on the first energy value of the target device and the preset number of second energy values. The second energy value of the target device is used to characterize the magnitude of the sound energy of the wake word detected by the target device during the wake-up process prior to the t-th wake-up process.

[0132] Optionally, the determining unit 72 is also used for: If there are multiple target devices among the multiple devices, or if there is only one target device among the multiple devices and the target device is not a theoretical device, then delete the historical data of each of the multiple devices. Wherein, the target device is the device that is woken up during the t-th wake-up process, the theoretical device is the device corresponding to the largest first energy value in the first dataset; the historical data of the device includes the energy value detected by the device in each historical wake-up process and the energy threshold of the device; the historical wake-up process is the wake-up process before the t-th wake-up process.

[0133] Optionally, the determining unit 72 is also used for: If the target device is not among the plurality of devices, a wake-up command is sent to the theoretical device to instruct the theoretical device to perform a wake-up task; wherein, the theoretical device is the device corresponding to the largest first energy value in the first dataset; the target device is the device that is woken up during the t-th wake-up process; and the energy threshold of each of the plurality of devices is determined according to the first dataset.

[0134] Optionally, the acquisition unit 71 is also used for: When the first energy value reported by the first device within the preset wake-up range is received, the timer is started; If the timer has not reached the set time, the system continues to receive the first energy value reported by devices within the preset wake-up range. If the timer reaches its set time, the receiving of the first energy value reported by devices within the preset wake-up range will stop, and the first dataset will be generated based on the first energy value received during the timer's timeout period.

[0135] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0136] in addition, Figure 7 The device wake-up device shown can be a software unit, a hardware unit, or a combination of software and hardware built into an existing terminal device. It can also be integrated into the terminal device as an independent component, or it can exist as an independent terminal device.

[0137] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0138] Figure 8 This is a schematic diagram of the structure of the terminal device provided in the embodiments of this application. For example... Figure 8 As shown, the terminal device 8 of this embodiment includes: at least one processor 80 ( Figure 8 (Only one is shown in the diagram) a processor, a memory 81, and a computer program 82 stored in the memory 81 and executable on the at least one processor 80, wherein the processor 80 executes the computer program 82 to implement the steps in any of the above-described device wake-up method embodiments.

[0139] The terminal device may be a desktop computer, laptop, handheld computer, or cloud server, etc. This terminal device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that... Figure 8 This is merely an example of terminal device 8 and does not constitute a limitation on terminal device 8. It may include more or fewer components than shown in the figure, or combine certain components, or different components, such as input / output devices, network access devices, etc.

[0140] The processor 80 can be a Central Processing Unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0141] In some embodiments, the memory 81 may be an internal storage unit of the terminal device 8, such as a hard disk or memory of the terminal device 8. In other embodiments, the memory 81 may be an external storage device of the terminal device 8, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the terminal device 8. Furthermore, the memory 81 may include both internal and external storage units of the terminal device 8. The memory 81 is used to store the operating system, applications, boot loader, data, and other programs, such as the program code of the computer program. The memory 81 can also be used to temporarily store data that has been output or will be output.

[0142] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the steps in the above-described method embodiments.

[0143] This application provides a computer program product that, when run on a terminal device, enables the terminal device to implement the steps described in the various method embodiments.

[0144] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a device / terminal equipment, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0145] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0146] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0147] In the embodiments provided in this application, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling or direct coupling or communication connection may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0148] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0149] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A device wake-up method, characterized in that, include: Obtain the first dataset corresponding to the t-th wake-up process within the preset wake-up range; wherein, the first dataset includes the first energy value of each of the multiple devices within the preset wake-up range, and the first energy value is used to characterize the magnitude of the sound energy of the wake word detected by the device in the t-th wake-up process; The energy thresholds of the plurality of devices are determined based on the first dataset; Each of the multiple devices is sent its own energy threshold to instruct the multiple devices to determine whether to perform a wake-up task based on their respective energy thresholds during the (t+1)th wake-up process.

2. The device wake-up method as described in claim 1, characterized in that, Determining the energy threshold of each of the plurality of devices based on the first dataset includes: If there is only one target device among the multiple devices, the energy threshold of the target device is determined according to the first energy value of the target device in the first dataset; wherein, the target device is the device that is woken up during the t-th wake-up process.

3. The device wake-up method as described in claim 2, characterized in that, Determining the energy threshold of the target device based on the first energy value of the target device in the first dataset includes: If the target device is a theoretical device, then the energy threshold of the target device is determined based on the first energy value of the target device in the first dataset; wherein, the theoretical device is the device corresponding to the largest first energy value in the first dataset.

4. The device wake-up method as described in claim 2, characterized in that, Determining the energy threshold of the target device based on the first energy value of the target device in the first dataset includes: If the first dataset includes the first energy values ​​of all devices within the preset wake-up range, then the energy threshold of the target device is determined based on the first energy value of the target device in the first dataset.

5. The device wake-up method according to any one of claims 2 to 4, characterized in that, Determining the energy threshold of the target device based on the first energy value of the target device in the first dataset includes: If the number of second energy values ​​of the target device reaches a preset number, and the target device has an energy threshold, then the energy threshold of the target device is updated according to the energy threshold of the target device, the first energy value, and the preset number of second energy values; If the number of second energy values ​​of the target device reaches a preset number, and the target device does not have an energy threshold, then the energy threshold of the target device is determined based on the first energy value of the target device and the preset number of second energy values. The second energy value of the target device is used to characterize the magnitude of the sound energy of the wake word detected by the target device during the wake-up process prior to the t-th wake-up process.

6. The device wake-up method as described in claim 2, characterized in that, After obtaining the first dataset corresponding to the t-th wake-up process within the preset wake-up range, the method further includes: If there are multiple target devices among the multiple devices, or if there is only one target device among the multiple devices and the target device is not a theoretical device, then delete the historical data of each of the multiple devices. Wherein, the target device is the device that is woken up during the t-th wake-up process, the theoretical device is the device corresponding to the largest first energy value in the first dataset; the historical data of the device includes the energy value detected by the device in each historical wake-up process and the energy threshold of the device; the historical wake-up process is the wake-up process before the t-th wake-up process.

7. The device wake-up method as described in claim 1, characterized in that, Determining the energy threshold of each of the plurality of devices based on the first dataset includes: If the target device is not among the plurality of devices, a wake-up command is sent to the theoretical device to instruct the theoretical device to perform a wake-up task; wherein, the theoretical device is the device corresponding to the largest first energy value in the first dataset; and the target device is the device that is woken up during the t-th wake-up process. The energy thresholds of the plurality of devices are determined based on the first dataset.

8. The device wake-up method as described in claim 1, characterized in that, The step of obtaining the first dataset corresponding to the t-th wake-up process within the preset wake-up range includes: When the first energy value reported by the first device within the preset wake-up range is received, the timer is started; If the timer has not reached the set time, the system continues to receive the first energy value reported by devices within the preset wake-up range. If the timer reaches its set time, the receiving of the first energy value reported by devices within the preset wake-up range will stop, and the first dataset will be generated based on the first energy value received during the timer's timeout period.

9. A voice control system, characterized in that, include: Servers and multiple devices; The server is configured to perform the device wake-up method as described in any one of claims 1 to 8.

10. The voice control system as described in claim 9, characterized in that, The device is configured to: If the energy threshold of the device has been received from the server, when a wake word is detected, it is determined whether the third energy value of the detected wake word is greater than the energy threshold; if the third energy value is greater than the energy threshold, the wake-up task is executed. The third energy value is reported to the server; If the energy threshold of the device is not received from the server, when a wake word is detected, the third energy value of the detected wake word is reported to the server; if a wake-up command is received from the server, the wake-up task is executed.

Images