CN111696562A - Voice wake-up method, device and storage medium - Google Patents

Abstract

Embodiments of the present application provide a voice wake-up method, a device, and a storage medium, which are used to solve the problem of a low wake-up rate of a master device in a current device group. The method includes: the master device introduces the wake-up identification results of other slave devices in the device group into the wake-up decision of the master device, and determines the master device according to the wake-up identification results of the master device itself and the wake-up identification results of other slave devices in the device group. Whether to wake up, this method improves the accuracy of the master device in the device group being woken up. Further, the master device can also dynamically adjust the preset wake-up threshold of the master device in the device group in a continuous or discrete manner, and determine whether the master device is woken up according to the adjusted wake-up threshold. The dynamically adjusted wake-up threshold is more suitable for the entire device. The actual state of group wake-up, the master device makes wake-up decisions based on the dynamically adjusted wake-up threshold, and the accuracy of wake-up devices is higher.

Description

Voice wake-up method, device and storage medium

technical field

The present application relates to the technical field of terminals, and in particular, to a voice wake-up method, device and storage medium.

Background technique

With the rise of intelligent voice interaction, more and more devices support the function of voice interaction. Among them, voice wake-up, as the beginning of voice interaction, is widely used in different devices, such as smart speakers and smart TVs. When there are devices that support voice wake-up in the space where the user is located, after the user sends a wake-up voice, these wake-up devices will respond to the speaker's request and interact with the user at the same time. At this time, it will cause confusion to the user, not knowing which device to perform voice interaction with.

At present, for the above scenarios, these devices that can be awakened by the user at the same time can be grouped into a device group. In the device group, only the master device responds to the user's wake-up, and other slave devices cooperate with the master device to process the user's command intent, thereby avoiding multiple devices at the same time. be awakened.

However, there are multiple wake-up devices in the device group, and only the wake-up result of the master device is used as the wake-up result of the entire device group, and the wake-up accuracy is not high, for example, the user is far away from the master device, or there are External interference, etc., will reduce the wake-up rate of the master device.

SUMMARY OF THE INVENTION

Embodiments of the present application provide a voice wake-up method, a device, and a storage medium, which improve the wake-up rate of a master device in a device group.

In a first aspect, an embodiment of the present application provides a voice wake-up method. The method is applied to a first electronic device. The first electronic device and at least one second electronic device belong to the same device group. The method includes: acquiring a first electronic device of audio data. Wake-up confidence, the first wake-up confidence is used to indicate the acoustic feature similarity between the wake-up word and the preset wake-up word in the audio data determined by the first electronic device; receive the wake-up recognition result sent by at least one second electronic device, and the wake-up recognition result It is used to indicate that at least one second electronic device is allowed or prohibited to be woken up; according to the first wake-up confidence level and the wake-up identification result, it is determined to allow or prohibit the first electronic device from being woken up.

The first electronic device is a master device in the device group, and the second electronic device is a slave device in the device group.

In the above scheme, the master device determines the master device according to the wake-up recognition results of the master device itself and the wake-up recognition results of other slave devices in the device group by introducing the wake-up recognition results of other slave devices in the device group into the master device's wake-up decision. whether to wake up. This solution can quickly determine whether the master device in the device group performs a wake-up response, makes full use of the wake-up recognition results of other devices in the device group, and improves the accuracy of the master device in the device group wake-up.

Optionally, the wake-up identification result includes at least one of a wake-up identification of at least one second electronic device and a second wake-up confidence; wherein the wake-up identification includes an identification that allows or prohibits the second electronic device from being woken up, and the second wake-up confidence It is used to indicate the similarity of the acoustic features between the wake-up word and the preset wake-up word in the audio data determined by the second electronic device.

If the wake-up identification result only includes the wake-up identifier of at least one second electronic device, the first electronic device may, according to the wake-up identifier, count the number of devices of the second electronic device that are allowed to be woken up, and the number of the devices in the total number of devices in the device group. Proportion.

If the wake-up identification result only includes the second wake-up confidence level of at least one second electronic device, the first electronic device first needs to determine whether the second electronic device is allowed to wake up. Then, count the number of devices of the second electronic device that are allowed to be woken up, and the proportion of the number of devices to the total number of devices in the device group.

If the wake-up identification result includes both the wake-up identification and the second wake-up confidence level of at least one second electronic device, the first electronic device may determine the number of devices of the second electronic device that are allowed to be woken up through any one of the above statistical methods, and The proportion of this number of devices to the total number of devices in the device group.

In a possible design, according to the first wake-up confidence level and the wake-up identification result, determining to allow or prohibit the first electronic device from being awakened includes: if the first wake-up confidence level is greater than or equal to a first threshold, determining to allow the first electronic device to be awakened The device is woken up; or if the first wake-up confidence is less than the first threshold and the first wake-up confidence is greater than the second threshold, according to the first wake-up confidence and the wake-up identification result, it is determined to allow or prohibit the first electronic device from being woken up; or If the first wake-up confidence level is less than or equal to the second threshold, it is determined that the first electronic device is prohibited from being woken up.

The above solution provides a method for determining whether to allow or prohibit the master device from being woken up by comparing the relationship between the first wake-up confidence calculated by the master device and the first threshold and the second threshold based on a preset master device wake-up condition. If the first wake-up confidence level is between the first threshold and the second threshold, it may be determined to allow or prohibit the master device from being woken up in combination with the wake-up identification results sent by other slave devices in the device group. This solution can quickly determine whether the master device in the device group performs a wake-up response, makes full use of the wake-up recognition results of other devices in the device group, and improves the accuracy of the master device in the device group wake-up.

In a possible design, determining whether to allow or prohibit the first electronic device from being woken up according to the first wake-up confidence and the wake-up recognition result includes: counting the wake-up conditions of at least one second electronic device according to the wake-up recognition result; If the preset wake-up condition of the first electronic device is satisfied, it is determined that the first electronic device is allowed to wake up.

Optionally, the wake-up condition includes any one of the following: the second wake-up confidence of all second electronic devices except the first electronic device in the device group is greater than or equal to the third threshold; the device group is allowed to be woken up The proportion of the number of devices of the second electronic device to the total number of devices in the device group is greater than or equal to the first proportion; the proportion of the number of devices of the second electronic device allowed to be awakened in the device group to the total number of devices in the device group is less than the first proportion and greater than The second ratio, and the second wake-up confidence of the second electronic device that is allowed to wake up is greater than or equal to the third threshold; wherein, the third threshold is a preset threshold in the first electronic device that allows the second electronic device to be woken up .

It should be noted that, in order to improve the accuracy of the judgment, a third threshold can be set to check whether there is an electronic device whose second wake-up confidence is lower than the third threshold in the second electronic device that is allowed to be woken up. The first electronic device is prohibited from being woken up. The above-mentioned third wake-up condition can effectively prevent the master device from being woken up by mistake because the preset wake-up thresholds set by some slave devices are too low, thereby improving the accuracy of the master device's wake-up.

In the above solution, the first electronic device counts the wake-up situation of the slave devices in the device group through the wake-up recognition result sent by at least one second electronic device, and judges whether the wake-up situation of the slave device satisfies the preset wake-up condition. If the preset wake-up condition is not satisfied, it is determined to prohibit the first electronic device from being woken up. The scheme fully considers the wake-up recognition results of other slave devices in the device group, optimizes the wake-up conditions of the master device in the device group, and improves the accuracy of the master device's wake-up in the device group.

Optionally, the wake-up identification result further includes a device identifier of at least one second electronic device, where the device identifier is used to indicate the device type of the second electronic device and determine the weight value of the second electronic device.

In a possible design, determining whether to allow or prohibit the first electronic device from being woken up according to the first wake-up confidence level and the wake-up identification result includes: according to the wake-up identification result, counting the number of the second electronic devices in the device group that are allowed to be woken up. a weight value; count the second weight values of all the second electronic devices in the device group; adjust the first threshold based on the first weight value and the second weight value; determine the allowable threshold according to the first wake-up confidence and the adjusted first threshold Or prohibit the first electronic device from being woken up; wherein, the first threshold is a threshold value that allows the first electronic device to be woken up.

The above solution introduces the weight value of the electronic device in the device group, and the weight value is used to indicate the confidence level of the wake-up recognition result of the electronic device. The weight value may be a value related to the device type of the electronic device, and may also be a value related to the software/hardware performance of the electronic device. For example, a smart TV has a weight value of 0.3, a smart speaker has a weight value of 0.6, and a smart light has a weight value of 0.1. For another example, the weight value of the speaker Pro is 0.5, the weight value of the conventional speaker is 0.3, and the weight value of the speaker mini is 0.2.

From the perspective of the wake-up threshold, the above scheme comprehensively analyzes the confidence level of the wake-up recognition results of the slave devices that are allowed to be woken up, combined with the actual wake-up situation of the slave devices in the device group, and then dynamically adjusts the master device in the device group based on a continuous method. Wake-up threshold, and determine whether the current master device needs to make a wake-up response according to the comparison result between the adjusted wake-up threshold and the wake-up confidence level calculated by the current master device. Since the adjusted wake-up threshold of the master device is more in line with the actual wake-up state of the overall device group, the accuracy of the wake-up of the master device in the device group is improved.

Optionally, the first weight value is determined according to the number of devices corresponding to the device type of the second electronic device that is allowed to be awakened in the device group, and the weight value corresponding to each device type of the second electronic device that is allowed to be awakened; The second weight value is determined according to the number of devices corresponding to the device types of all the second electronic devices in the device group, and the weight value corresponding to each device type.

In a possible design, adjusting the first threshold based on the first weight value and the second weight value includes: using the product of the ratio of the first weight value and the second weight value and the maximum threshold adjustment parameter as the threshold adjustment parameter, according to The threshold adjustment parameter adjusts the first threshold.

In a possible design, determining whether to allow or prohibit the first electronic device from being woken up according to the first wake-up confidence level and the wake-up identification result includes: counting the devices of the second electronic device in the device group that are allowed to be woken up according to the wake-up identification result Quantity; determine the threshold adjustment parameter corresponding to the number of devices according to the wake-up threshold adjustment table, the wake-up threshold adjustment table includes the corresponding relationship between the number of devices of the second electronic device that is allowed to be woken up and the threshold adjustment parameter; adjust the first threshold based on the threshold adjustment parameter; The first wake-up confidence and the adjusted first threshold determine whether to allow or prohibit the first electronic device from being woken up.

The above solution also starts from the wake-up threshold, combined with the actual wake-up situation of the slave devices in the device group, counts the number or proportion of the slave devices that are allowed to be woken up, and then dynamically adjusts the wake-up threshold of the master device in the device group based on a discrete method. , and according to the comparison result between the adjusted wake-up threshold and the wake-up confidence level calculated by the current master device, it is determined whether the current master device needs to make a wake-up response. Since the adjusted wake-up threshold of the master device is more in line with the actual wake-up state of the overall device group, the accuracy of the wake-up of the master device in the device group is improved.

In a possible design, determining whether to allow or prohibit the first electronic device from being woken up according to the first wake-up confidence and the adjusted first threshold includes: if the first wake-up confidence is greater than or equal to the adjusted first threshold , it is determined that the first electronic device is allowed to be woken up; or if the first wake-up confidence level is less than the adjusted first threshold, it is determined that the first electronic device is not allowed to be woken up.

In a second aspect, an embodiment of the present application provides a voice wake-up method. The method is applied to a first electronic device. The first electronic device and at least one second electronic device belong to the same device group. The method includes: acquiring a first electronic device of audio data. Wake-up confidence, the first wake-up confidence is used to indicate the acoustic feature similarity between the wake-up word and the preset wake-up word in the audio data determined by the first electronic device; receive the wake-up recognition result sent by at least one second electronic device, and the wake-up recognition result It is used to indicate that the second electronic device is allowed or prohibited to be woken up; the first threshold is adjusted according to the wake-up recognition result, and the first threshold is the threshold that allows the first electronic device to be woken up; according to the adjusted first threshold and the first wake-up confidence, It is determined whether the first electronic device is allowed or prohibited from being woken up.

In a possible design, adjusting the first threshold value according to the wake-up recognition result includes: according to the wake-up recognition result, counting the first weights of the second electronic devices allowed to be woken up in the device group; counting all the second electronic devices in the device group The second weight value of ; adjust the first threshold value based on the first weight value and the second weight value.

Optionally, the first weight value is determined according to the number of devices corresponding to the device type of the second electronic device that is allowed to be awakened in the device group, and the weight value corresponding to each device type of the second electronic device that is allowed to be awakened; The second weight value is determined according to the number of devices corresponding to the device types of all the second electronic devices in the device group, and the weight value corresponding to each device type.

In a possible design, adjusting the first threshold based on the first weight value and the second weight value includes: using the product of the ratio of the first weight value and the second weight value and the maximum threshold adjustment parameter as the threshold adjustment parameter, according to The threshold adjustment parameter adjusts the first threshold.

In a possible design, adjusting the first threshold according to the wake-up recognition result includes: counting the number of devices of the second electronic device allowed to be woken up in the device group according to the wake-up recognition result; determining the threshold corresponding to the number of devices according to the wake-up threshold adjustment table Adjusting parameters, the wake-up threshold adjustment table includes the correspondence between the number of devices of the second electronic device that are allowed to be woken up and the threshold adjustment parameters; the first threshold is adjusted based on the threshold adjustment parameters.

Optionally, the wake-up identification result includes at least one of a wake-up identification of at least one second electronic device and a second wake-up confidence; wherein the wake-up identification includes an identification that allows or prohibits the second electronic device from being woken up, and the second wake-up confidence It is used to indicate the similarity of the acoustic features between the wake-up word and the preset wake-up word in the audio data determined by the second electronic device.

In a possible design, determining whether to allow or prohibit the first electronic device from being woken up according to the adjusted first threshold and the first wake-up confidence includes: if the first wake-up confidence is greater than or equal to the adjusted first threshold , it is determined that the first electronic device is allowed to be woken up; or if the first wake-up confidence level is less than the adjusted first threshold, it is determined that the first electronic device is not allowed to be woken up.

In a third aspect, an embodiment of the present application provides a voice wake-up device, the voice wake-up device is a first electronic device, the first electronic device and at least one second electronic device belong to the same device group, the voice wake-up device includes: an acquisition module, For obtaining the first wake-up confidence of the audio data, the first wake-up confidence is used to indicate the acoustic feature similarity between the wake-up word and the preset wake-up word in the audio data determined by the first electronic device; the receiving module is used to receive at least one A wake-up identification result sent by the second electronic device, where the wake-up identification result is used to indicate that at least one second electronic device is allowed to wake up; the processing module is used to determine whether to allow or prohibit the first wake-up confidence level and the wake-up identification result according to the first wake-up confidence Electronic device wakes up.

Optionally, the wake-up identification result includes at least one of a wake-up identification of at least one second electronic device and a second wake-up confidence; wherein the wake-up identification includes an identification that allows or prohibits the second electronic device from being woken up, and the second wake-up confidence It is used to indicate the similarity of the acoustic features between the wake-up word and the preset wake-up word in the audio data determined by the second electronic device.

Optionally, the processing module is specifically configured to: if the first wake-up confidence is greater than or equal to the first threshold, determine to allow the first electronic device to be awakened; or if the first wake-up confidence is less than the first threshold, and the first wake-up confidence is greater than the second threshold, according to the first wake-up confidence and the wake-up identification result, it is determined to allow or prohibit the first electronic device from being woken up; or if the first wake-up confidence is less than or equal to the second threshold, it is determined to prohibit the first electronic device from being woken up .

Optionally, the processing module is specifically configured to: count the wake-up situation of at least one second electronic device according to the wake-up identification result; if the wake-up situation meets the preset wake-up condition of the first electronic device, determine to allow the first electronic device to be woken up.

Optionally, the wake-up condition includes any one of the following: the second wake-up confidence of all second electronic devices except the first electronic device in the device group is greater than or equal to the third threshold; the device group is allowed to be woken up The proportion of the number of devices of the second electronic device to the total number of devices in the device group is greater than or equal to the first proportion; the proportion of the number of devices of the second electronic device allowed to be awakened in the device group to the total number of devices in the device group is less than the first proportion and greater than The second ratio, and the second wake-up confidence of the second electronic device that is allowed to wake up is greater than or equal to the third threshold; wherein, the third threshold is a preset threshold in the first electronic device that allows the second electronic device to be woken up .

Optionally, the wake-up identification result further includes a device identifier of at least one second electronic device, where the device identifier is used to indicate the device type of the second electronic device and determine the weight value of the second electronic device.

Optionally, the processing module is specifically configured to: count the first weight values of the second electronic devices allowed to be woken up in the device group according to the wake-up identification result; count the second weight values of all the second electronic devices in the device group; A weight value and a second weight value adjust the first threshold; according to the first wake-up confidence and the adjusted first threshold, determine whether to allow or prohibit the first electronic device from being woken up; wherein, the first threshold is to allow the first electronic device to be woken up Wake-up threshold.

Optionally, the first weight value is determined according to the number of devices corresponding to the device type of the second electronic device that is allowed to be awakened in the device group, and the weight value corresponding to each device type of the second electronic device that is allowed to be awakened; The second weight value is determined according to the number of devices corresponding to the device types of all the second electronic devices in the device group, and the weight value corresponding to each device type.

Optionally, the processing module is specifically configured to: use the product of the ratio of the first weight value to the second weight value and the maximum threshold adjustment parameter as the threshold adjustment parameter, and adjust the first threshold according to the threshold adjustment parameter.

Optionally, the processing module is specifically configured to: count the number of devices of the second electronic device allowed to be awakened in the device group according to the wake-up identification result; determine the threshold adjustment parameter corresponding to the number of devices according to the wake-up threshold adjustment table, and the wake-up threshold adjustment table includes: The correspondence between the number of devices of the second electronic device that is allowed to be awakened and the threshold adjustment parameter; the first threshold is adjusted based on the threshold adjustment parameter; according to the first wakeup confidence and the adjusted first threshold, it is determined to allow or prohibit the first electronic device be awakened.

Optionally, the processing module is specifically configured to: if the first wake-up confidence is greater than or equal to the adjusted first threshold, determine to allow the first electronic device to be awakened; or if the first wake-up confidence is less than the adjusted first threshold , it is determined that the first electronic device is prohibited from being woken up.

In a fourth aspect, an embodiment of the present application provides a voice wake-up device, the voice wake-up device is a first electronic device, the first electronic device and at least one second electronic device belong to the same device group, the voice wake-up device includes: an acquisition module, For obtaining the first wake-up confidence of the audio data, the first wake-up confidence is used to indicate the acoustic feature similarity between the wake-up word and the preset wake-up word in the audio data determined by the first electronic device; the receiving module is used to receive at least one The wake-up identification result sent by the second electronic device, where the wake-up identification result is used to indicate that the second electronic device is allowed or prohibited to be woken up; the processing module is configured to adjust the first threshold according to the wake-up identification result, where the first threshold is to allow the first electronic device to be woken up. A wake-up threshold; according to the adjusted first threshold and the first wake-up confidence, determine whether to allow or prohibit the first electronic device from being woken up.

Optionally, the processing module is specifically configured to: count the first weight values of the second electronic devices allowed to be woken up in the device group according to the wake-up identification result; count the second weight values of all the second electronic devices in the device group; A weight value and a second weight value adjust the first threshold.

Optionally, the first weight value is determined according to the number of devices corresponding to the device type of the second electronic device that is allowed to be awakened in the device group, and the weight value corresponding to each device type of the second electronic device that is allowed to be awakened; The second weight value is determined according to the number of devices corresponding to the device types of all the second electronic devices in the device group, and the weight value corresponding to each device type.

Optionally, the processing module is specifically configured to: use the product of the ratio of the first weight value to the second weight value and the maximum threshold adjustment parameter as the threshold adjustment parameter, and adjust the first threshold according to the threshold adjustment parameter.

Optionally, the processing module is specifically configured to: count the number of devices of the second electronic device allowed to be awakened in the device group according to the wake-up identification result; determine the threshold adjustment parameter corresponding to the number of devices according to the wake-up threshold adjustment table, and the wake-up threshold adjustment table includes: The correspondence between the number of devices of the second electronic device that is allowed to be awakened and the threshold adjustment parameter; the first threshold is adjusted based on the threshold adjustment parameter.

Optionally, the wake-up identification result includes at least one of a wake-up identification of at least one second electronic device and a second wake-up confidence; wherein the wake-up identification includes an identification that allows or prohibits the second electronic device from being woken up, and the second wake-up confidence It is used to indicate the similarity of the acoustic features between the wake-up word and the preset wake-up word in the audio data determined by the second electronic device.

Optionally, the processing module is specifically configured to: if the first wake-up confidence is greater than or equal to the adjusted first threshold, determine to allow the first electronic device to be awakened; or if the first wake-up confidence is less than the adjusted first threshold , it is determined that the first electronic device is prohibited from being woken up.

In a fifth aspect, an embodiment of the present application provides a voice wake-up device, including: a memory and a processor, where the memory is used to store a computer program, and the processor is used to call and run the computer program from the memory, so that the processor runs the computer program to execute as described in Section 1. The method of any one of the aspect, or the method of any one of the second aspect.

In a sixth aspect, an embodiment of the present application provides a storage medium, where the storage medium includes a computer program, and the computer program is used to implement the method of any one of the first aspect, or the method of any one of the second aspect.

Embodiments of the present application provide a voice wake-up method, device, and storage medium. The method includes: the master device introduces the wake-up recognition results of other slave devices in the device group into the wake-up decision of the master device, according to the wake-up recognition result of the master device itself. The result and the wake-up identification results of other slave devices in the device group determine whether the master device is woken up, and the method improves the accuracy of the wake-up of the master device in the device group. Further, the master device can also dynamically adjust the preset wake-up threshold of the master device in the device group in a continuous or discrete manner, and determine whether the master device is woken up according to the adjusted wake-up threshold. The dynamically adjusted wake-up threshold is more suitable for the entire device. The actual state of group wake-up, the master device makes wake-up decisions based on the dynamically adjusted wake-up threshold, and the accuracy of wake-up devices is higher.

Description of drawings

1 is a schematic diagram of a scenario of a voice wake-up method provided by an embodiment of the present application;

2 is a schematic diagram of a hardware structure of an electronic device provided by an embodiment of the present application;

3 is a schematic diagram of a software architecture of an electronic device according to an embodiment of the present application;

4a to 4c are schematic diagrams of scenarios of a voice wake-up method provided by an embodiment of the present application;

FIG. 5 is a schematic diagram of a user interface interaction provided by an embodiment of the present application;

FIG. 6 is an interactive schematic diagram of a voice wake-up method provided by an embodiment of the present application;

FIG. 7 is a schematic flowchart of a voice wake-up method provided by an embodiment of the present application;

Fig. 8a is a judgment flowchart of a voice wake-up method provided by an embodiment of the present application;

FIG. 8b is a judgment flowchart of a voice wake-up method provided by an embodiment of the present application;

FIG. 9 is a judgment flowchart of a voice wake-up method provided by an embodiment of the present application;

10 is a schematic flowchart of another voice wake-up method provided by an embodiment of the present application;

11 is a schematic flowchart of another voice wake-up method provided by an embodiment of the present application;

12 is a schematic structural diagram of a voice wake-up device provided by an embodiment of the application;

13 is a schematic structural diagram of a voice wake-up device provided by an embodiment of the application;

FIG. 14 is a schematic diagram of a hardware structure of a voice wake-up device according to an embodiment of the present application.

Detailed ways

The electronic device provided by the embodiment of the present application is an electronic device with a voice wake-up function, that is, a user can wake up the electronic device by voice. Specifically, the user wakes up the electronic device by speaking a wake-up word. The wake-up word may be preset by the user according to his own needs, or may be preset by the electronic device before leaving the factory, which is not limited in this embodiment.

The electronic device acquires the audio data, and detects whether the audio data contains the wake-up word. If the audio data contains the wake-up word, the electronic device is awakened, otherwise the electronic device is not awakened. After the electronic device is awakened, the user can interact with the electronic device through voice. For example, the preset wake-up word is "Xiaoyi Xiaoyi". When the electronic device detects that the audio data contains "Xiaoyi Xiaoyi", the electronic device is woken up. Exemplarily, FIG. 1 shows a schematic diagram of a voice wake-up scenario. As shown in FIG. 1 , the scenario includes an electronic device 10 , an electronic device 20 , an electronic device 30 , and an electronic device 40 . The electronic device 10 , the electronic device 30 , and the electronic device 40 have the same preset wake-up word, such as wake-up word 1 , and the preset wake-up word in the electronic device 20 is wake-up word 2 . When the wake-up word spoken by the user is wake-up word 1, and each electronic device in the scene can receive or collect the wake-up word spoken by the user, the electronic device 10, the electronic device 30, and the electronic device 40 can be woken up. Since the wake-up word spoken by the user is different from the wake-up word preset by the electronic device 20, the electronic device 20 is not woken up.

It can be seen from the above example that when there are multiple electronic devices that support voice wake-up in the space where the user is located, the wake-up word spoken by the user may be received or collected by multiple electronic devices. Similarly, it may cause multiple electronic devices to be awakened, and the multiple electronic devices that are awakened will respond to the user's request at the same time, causing the user to be confused about which device to perform voice interaction with, and reducing the user experience.

In order to avoid the confusion of the user's voice interaction, the voice wake-up method provided by the embodiment of the present application forms a device group of electronic devices that can be woken up by the user at the same time, and only one electronic device in the device group responds to the user's wake-up and will respond The responding electronic device is regarded as the master device of the device group, other electronic devices except the master device are regarded as the slave devices of the device group, and the slave devices cooperate with the master device to process the user's instruction intention. Taking FIG. 1 as an example, the electronic device 10, the electronic device 30 and the electronic device 40 with the same preset wake-up word can be formed into a device group, and the electronic device 10 can be set as the master device of the device group, and the electronic device 30 and the electronic device. 40 is a slave device of the device group, and the slave device cooperates with the master device to process the user's instruction intention. It should be noted that the master device of the device group can be set through preset rules, for example, the electronic device with the strongest processing performance in the device group can be set as the master device, or any electronic device in the device group can be set as the master device by user-defined rules. The master device, which is not limited in this embodiment.

Based on the voice wake-up scenario provided in Figure 1, after receiving or collecting the audio data, the master device of the device group preprocesses the audio data, extracts the wake-up words in the audio data, and uses the voice wake-up model to calculate the difference between the wake-up words in the audio data and the audio data. The similarity value of the acoustic features of the preset wake-up word, if the similarity value is greater than or equal to the preset similarity threshold (also referred to as the preset wake-up threshold), it is considered that a wake-up request is detected, and the master device of the device group is woken up, Otherwise the master device is not woken up. It can be seen from this that whether the master device in the device group wakes up is determined by the wake-up recognition result of the master device itself, and this implementation scheme has the following defects:

First, the device group includes a plurality of electronic devices that can be woken up, and only the wake-up identification result of the master device itself is used as the wake-up identification result of the entire device group, and the accuracy of the result is not high. For example, if there is an interference source on the voice transmission path, or when the user is far away from the main device, only relying on the wake-up recognition result of the main device itself, it is very likely that when the main device needs to be woken up, due to the influence of external factors, the main device will No wake-up voice is detected, or a wake-up voice is detected but the preset wake-up threshold is not reached, causing the entire device group to become unresponsive.

Second, the preset wake-up threshold of the master device in the device group is a fixed value, but the wake-up threshold may be different in different voice wake-up scenarios. Therefore, the fixed preset wake-up threshold may affect the wake-up recognition result of the master device.

In order to improve the wake-up rate of the master device, other solutions may be used to assist the master device in judging wake-up identification. For example, by re-checking the server model, the master device of the device group obtains the wake-up recognition result from the server by interacting with the server. However, the implementation cost of this method is high, and depending on the network environment, there may be a high delay problem. For another example, the accuracy of wake-up recognition is improved by expanding the voice wake-up model on the device side. The disadvantage of this method is that it occupies more computing and storage resources on the device side, resulting in a higher overall cost of the device.

To sum up, the master device of the device group only relies on its own wake-up recognition result, which has a high misjudgment rate, which is likely to cause the entire device group to not respond after the user sends a wake-up voice, resulting in poor user experience. Although the wake-up rate of the master device can be improved by re-checking on the server side or expanding the device-side voice wake-up model, the implementation cost is high and there is a high delay. In this regard, an embodiment of the present application provides a voice wake-up method. The master device makes full use of the wake-up recognition results on different electronic devices in the device group by introducing the wake-up recognition results of other slave devices in the device group into the wake-up decision of the master device. , according to the wake-up recognition result of the master device itself and the wake-up recognition results of other slave devices in the device group, to determine whether the master device is woken up, because the master device considers the wake-up recognition results of other slave devices in the device group in the wake-up decision. The accuracy with which the master device in the device group is woken up.

In different voice wake-up scenarios, the master device of the device group uses a fixed wake-up threshold to make wake-up decisions, which has a high false positive rate. In this regard, an embodiment of the present application provides a voice wake-up method. The master device also introduces the wake-up recognition results of other slave devices in the device group into the wake-up decision of the master device. The wake-up recognition results of other slave devices use continuous or discrete methods to dynamically adjust the preset wake-up threshold of the master device in the device group, and determine whether the master device is woken up according to the adjusted wake-up threshold. The dynamically adjusted wake-up threshold is more in line with the actual wake-up state of the entire device group. The master device makes wake-up decisions based on the dynamically adjusted wake-up threshold, which improves the accuracy of the master device in the device group being woken up.

The electronic devices in the device group in the embodiments of the present application are described in detail below.

The electronic device in this embodiment of the present application may be a portable electronic device, such as a mobile phone, a tablet computer, an artificial intelligence (artificial intelligence, AI) intelligent voice terminal, a wearable device, an augmented reality (AR)/virtual reality (virtual reality, VR) equipment, etc. Exemplary embodiments of portable electronic devices include, but are not limited to, portable electronic devices powered by or other operating systems. The above-mentioned portable electronic device may also be an in-vehicle terminal, a laptop computer (Laptop), or the like. It should also be understood that the electronic device in the embodiment of the present application may also be a desktop computer, a smart home device (eg, a smart TV, a smart speaker), etc., which is not limited thereto.

By way of example, FIG. 2 shows a schematic diagram of a hardware structure of an electronic device according to an embodiment of the present application. Specifically, as shown in FIG. 2 , the electronic device includes a processor 110, an internal memory 121, an external memory interface 122, a camera 131, a display screen 132, a sensor module 140, a subscriber identification module (SIM) card interface 151, Button 152, audio module 160, speaker 161, receiver 162, microphone 163, headphone jack 164, universal serial bus (USB) interface 170, charging management module 180, power management module 181, battery 182, mobile communication module 191 and wireless communication module 192. In other embodiments, the electronic device may further include motors, indicators, buttons, and the like.

It should be understood that the hardware structure shown in FIG. 2 is only an example. The electronic device of the embodiments of the present application may have more or less components than the electronic device shown in the figures, may combine two or more components, or may have different component configurations. The various components shown in the figures may be implemented in hardware, software, or a combination of hardware and software, including one or more signal processing and/or application specific integrated circuits.

The processor 110 may include one or more processing units. For example, the processor 110 may include an application processor (AP), a modem, a graphics processing unit (GPU), an image signal processor (ISP), a controller, a video codec, a digital A signal processor (digital signal processor, DSP), a baseband processor, and/or a neural-network processing unit (neural-network processing unit, NPU), etc. Wherein, different processing units may be independent devices, or may be integrated in one or more processors.

In some embodiments, a buffer may also be provided in the processor 110 for storing instructions and/or data. For example, the buffer in the processor 110 may be a cache memory. The cache may be used to hold instructions and/or data that have just been used, generated, or recycled by the processor 110 . If the processor 110 needs to use the instruction or data, it can be called directly from the buffer. This helps to reduce the time for the processor 110 to obtain instructions or data, thereby helping to improve the efficiency of the system.

Internal memory 121 may be used to store programs and/or data. In some embodiments, the internal memory 121 includes a stored program area and a stored data area. The storage program area may be used to store an operating system (such as Android, IOS, etc.), a computer program required for at least one function (such as a voice wake-up function, a sound playback function), and the like. The storage data area may be used to store data (such as audio data) created and/or collected during the use of the electronic device. For example, the processor 110 may cause the electronic device to execute a corresponding method by calling programs and/or data stored in the internal memory 121, thereby implementing one or more functions. For example, the processor 110 invokes certain programs and/or data in the internal memory, so that the electronic device executes the voice wake-up method provided in the embodiments of the present application, thereby implementing the voice wake-up function. Wherein, the internal memory 121 may adopt a high-speed random access memory, and/or a non-volatile memory, or the like. For example, the nonvolatile memory may include at least one of one or more magnetic disk storage devices, flash memory devices, and/or universal flash storage (UFS), among others.

The external memory interface 122 can be used to connect an external memory card (eg, a Micro SD card) to expand the storage capacity of the electronic device. The external memory card communicates with the processor 110 through the external memory interface 122 to realize the data storage function. For example, the electronic device can save images, music, videos and other files in the external memory card through the external memory interface 122 .

The camera 131 may be used to capture moving, still images, and the like. Typically, the camera 131 includes a lens and an image sensor. Among them, the optical image generated by the object through the lens is projected onto the image sensor, and then converted into an electrical signal for subsequent processing. Illustratively, the image sensor may be a charge coupled device (CCD) or a complementary metal-oxide-semiconductor (CMOS) phototransistor. The image sensor converts the optical signal into an electrical signal, and then passes the electrical signal to the ISP to convert it into a digital image signal. It should be noted that the electronic device may include 1 or N cameras 131 , where N is a positive integer greater than 1.

Display screen 132 may include a display panel for displaying a user interface. The display panel can be a liquid crystal display (LCD), an organic light-emitting diode (OLED), an active-matrix organic light-emitting diode or an active-matrix organic light-emitting diode (active-matrix organic light-emitting diode). , AMOLED), flexible light-emitting diode (flex light-emitting diode, FLED), Miniled, MicroLed, Micro-oLed, quantum dot light-emitting diode (quantum dot light emittingdiodes, QLED) and so on. It should be noted that the electronic device may include one or M display screens 132 , where M is a positive integer greater than one. For example, the electronic device may implement a display function through a GPU, a display screen 132, an application processor, and the like.

Sensor module 140 may include one or more sensors. For example, a touch sensor 140A, a gyroscope 140B, an acceleration sensor 140C, a fingerprint sensor 140D, a pressure sensor 140E, and the like. In some embodiments, the sensor module 140 may also include an ambient light sensor, a distance sensor, a proximity light sensor, a bone conduction sensor, a temperature sensor, and the like.

The touch sensor 140A may also be referred to as a "touch panel". The touch sensor 140A may be disposed on the display screen 132 , and the touch sensor 140A and the display screen 132 form a touch screen, also referred to as a “touch screen”. The touch sensor 140A is used to detect a touch operation on or near it. The touch sensor 140A may communicate the detected touch operation to the application processor to determine the type of touch event. The electronic device may provide visual output and the like related to touch operations through the display screen 132 . In other embodiments, the touch sensor 140A may also be disposed on the surface of the electronic device, which is different from the location where the display screen 132 is located.

The gyroscope 140B can be used to determine the motion attitude of the electronic device. In some embodiments, the angular velocity of the electronic device about three axes (ie, the x, y, and z axes) may be determined by the gyroscope 140B. The gyroscope 140B can be used for image stabilization. Exemplarily, when the shutter is pressed, the gyroscope 140B detects the shaking angle of the electronic device, calculates the distance to be compensated by the lens module according to the angle, and allows the lens to counteract the shaking of the electronic device through reverse motion, thereby realizing anti-shake. The gyro sensor 140B can also be used for navigation and somatosensory game scenarios.

The acceleration sensor 140C can detect the magnitude of the acceleration of the electronic device in various directions (generally three axes). The magnitude and direction of gravity can be detected when the electronic device is stationary. The acceleration sensor 140C can also be used to identify the posture of the electronic device, and is applied to applications such as switching between horizontal and vertical screens, and pedometers.

The fingerprint sensor 140D is used to collect fingerprints. Electronic devices can use the collected fingerprint characteristics to unlock fingerprints, access application locks, take photos with fingerprints, and answer incoming calls with fingerprints.

The pressure sensor 140E is used to sense pressure signals, and can convert the pressure signals into electrical signals. For example, the pressure sensor 140E may be disposed on the display screen 132 . Among them, touch operations acting on the same touch position but with different touch operation intensities may correspond to different operation instructions.

The SIM card interface 151 is used for connecting a SIM card. The SIM card can be inserted into the SIM card interface 151 or pulled out from the SIM card interface 151 to achieve contact and separation with the electronic device. The electronic device may support 1 or K SIM card interfaces 151 , where K is a positive integer greater than 1. The SIM card interface 151 may support Nano SIM cards, Micro SIM cards, and/or SIM cards, and the like. Multiple cards can be inserted into the same SIM card interface 151 at the same time. The types of the plurality of cards may be the same or different. The SIM card interface 151 can also be compatible with different types of SIM cards. The SIM card interface 151 is also compatible with external memory cards. The electronic device interacts with the network through the SIM card to realize functions such as call and data communication. In some embodiments, the electronic device may also employ an eSIM, that is, an embedded SIM card. The eSIM card can be embedded in the electronic device and cannot be separated from the electronic device.

The keys 152 may include a power key, a volume key, and the like. The keys 152 may be mechanical keys or touch keys. The electronic device may receive key input and generate key signal input related to user settings and function control of the electronic device.

The electronic device may implement audio functions through an audio module 160, a speaker 161, a receiver 162, a microphone 163, an earphone interface 164, an application processor, and the like. For example, audio playback function, recording function, voice wake-up function, etc.

The audio module 160 may be used to perform digital-to-analog conversion and/or analog-to-digital conversion on audio data, and may also be used to encode and/or decode audio data. For example, the audio module 160 may be provided independently of the processor, or may be provided in the processor 110 , or some functional modules of the audio module 160 may be provided in the processor 110 .

The speaker 161, also called "speaker", is used to convert audio data into sound and play the sound. For example, the electronic device 100 may listen to music, answer a speakerphone, or issue a voice prompt through the speaker 161 .

The receiver 162, also referred to as an "earpiece", is used to convert audio data into sound and to play back the sound. For example, when the electronic device 100 answers a call, it can be answered by placing the receiver 162 close to the human ear.

The microphone 163, also called "microphone" or "microphone", is used to collect sound (eg, ambient sound, including sounds made by people, sounds made by equipment, etc.), and convert the sound into audio electrical data. When making a call or sending a voice, the user can make a sound through the human mouth close to the microphone 163, and the microphone 163 collects the sound made by the user. When the voice wake-up function of the electronic device is enabled, the microphone 163 can collect ambient sound in real time and obtain audio data. The situation in which the microphone 163 collects sound is related to the environment in which it is located. For example, when the surrounding environment is relatively noisy and the user speaks the wake-up word, the sound collected by the microphone 163 includes the surrounding environment noise and the sound of the user uttering the wake-up word. For another example, when the surrounding environment is relatively quiet and the user speaks the wake-up word, the sound collected by the microphone 163 is the sound of the user issuing the wake-up word. For another example, when the surrounding environment is relatively noisy, the voice wake-up function of the electronic device is enabled, but the user does not speak a wake-up word to wake up the electronic device, and the sound collected by the microphone 163 is only ambient noise.

It should be noted that the electronic device may be provided with at least one microphone 163 . For example, two microphones 163 are provided in the electronic device, which can implement a noise reduction function in addition to collecting sound. For another example, three, four or more microphones 163 may also be set in the electronic device, so that on the basis of sound collection and noise reduction, sound source identification or directional recording functions can also be implemented.

The earphone jack 164 is used to connect wired earphones. The earphone interface 164 may be a USB interface 170, or a 3.5mm open mobile terminal platform (OMTP) standard interface, a cellular telecommunications industry association of the USA (CTIA) standard interface, etc. .

The USB interface 170 is an interface conforming to the USB standard specification, and may specifically be a Mini USB interface, a Micro USB interface, a USB Type C interface, and the like. The USB interface 170 can be used to connect a charger to charge the electronic device, and can also be used to transmit data between the electronic device and peripheral devices. It can also be used to connect headphones to play audio through the headphones. For example, the USB interface 170 can be used to connect other electronic devices, such as AR devices, computers, and the like, in addition to the headphone interface 164 .

The charging management module 180 is used to receive charging input from the charger. The charger may be a wireless charger or a wired charger. In some wired charging embodiments, the charging management module 180 may receive charging input from the wired charger through the USB interface 170 . In some wireless charging embodiments, the charging management module 180 may receive wireless charging input through a wireless charging coil of the electronic device. While the charging management module 180 charges the battery 182 , the power management module 180 can also supply power to the electronic device.

The power management module 181 is used for connecting the battery 182 , the charging management module 180 and the processor 110 . The power management module 181 receives input from the battery 182 and/or the charge management module 180, and supplies power to the processor 110, the internal memory 121, the display screen 132, the camera 131, and the like. The power management module 181 can also be used to monitor parameters such as battery capacity, battery cycle times, battery health status (leakage, impedance). In some other embodiments, the power management module 181 may also be provided in the processor 110 . In other embodiments, the power management module 181 and the charging management module 180 may also be provided in the same device.

The mobile communication module 191 can provide a wireless communication solution including 2G/3G/4G/5G etc. applied on the electronic device. The mobile communication module 191 may include a filter, a switch, a power amplifier, a low noise amplifier (LNA), and the like.

The wireless communication module 192 can provide applications on electronic devices including WLAN (such as Wi-Fi network), Bluetooth (Bluetooth, BT), global navigation satellite system (GNSS), frequency modulation (frequency modulation, FM), Solutions for wireless communication such as near field communication (NFC), infrared technology (infrared, IR). The wireless communication module 192 may be one or more devices integrating at least one communication processing module.

In some embodiments, the antenna 1 of the electronic device is coupled with the mobile communication module 191, and the antenna 2 is coupled with the wireless communication module 192, so that the electronic device can communicate with other devices. Specifically, the mobile communication module 191 can communicate with other devices through the antenna 1 , and the wireless communication module 193 can communicate with other devices through the antenna 2 .

FIG. 3 shows a schematic diagram of a software architecture of an electronic device according to an embodiment of the present application. As shown in FIG. 3 , the electronic device includes an audio collector 401 , an audio processor 402 , an audio recognizer 403 , and an interactor 404 .

The audio collection module 401 is used to store audio data converted from sounds collected by a sound collection device (such as the microphone 163 shown in FIG. 2 or other sensors for collecting sounds), and forward the audio data to audio processing Module 402. For example, the audio collection module 401 can be used to store the audio data obtained from the audio module 160 in a memory (eg, the internal memory 121, or the memory in the processor 110, etc.), and forward the audio data stored in the memory to the audio processing module 402 for processing. It should be noted that, in the embodiment of the present application, the audio collection module 401 may actively acquire the audio data from the audio module 160 after receiving the notification that the audio data is obtained from the audio module 160, or may be obtained after the audio module 160 collects the audio data. , the audio data is sent to the audio collection module 401 , and the embodiment of the present application does not limit the manner in which the audio collection module 401 acquires the audio data from the audio module 160 .

The audio processing module 402 is used to preprocess the audio data, such as channel conversion, smoothing, noise reduction, etc., and send the preprocessed audio data to the audio recognition module 403, so that the subsequent audio recognition module 403 can wake up word detection.

The audio recognition module 403 is configured to perform wake-up word detection on the audio data, and can determine whether there is a preset wake-up word in the audio data through a voice wake-up model. Calculate the similarity value between the wake-up word in the audio data and the acoustic features of the preset wake-up word through the voice wake-up model. If the similarity value is greater than or equal to the preset similarity threshold, send a wake-up call to the microphone 163, the receiver 162, and the earphone. command, after waking up, the pre-stored response voice data is obtained from the audio module 160 through the speaker 161, the response voice data is converted into sound, and a voice response is performed. For example, the audio recognition module 403 can perform wake-up word detection according to the audio data at Q sampling moments in the first time period, which can also be called the wake-up word time window, etc., and is usually set to a value not smaller than the wake-up word issued by the user. The time required for the sound of the word. Wherein, the duration of the interval between two adjacent sampling intervals in the Q sampling moments is the first sampling interval, that is, the audio processing module 402 may send the preprocessed audio data to the audio recognition module 403 at every first sampling interval, The audio recognition module 403 performs a wake-up word detection every first sampling interval according to the audio data of the Q sampling moments recently received. When the audio recognition module 403 detects the wake-up word, it calculates the similarity value between the wake-up word and the preset wake-up word through the voice wake-up model. If the similarity value is greater than or equal to the preset similarity threshold, it is determined to wake up the electronic device, and the audio recognition module 403 A wake-up command is sent to the speaker 161 . It should be noted that, in this embodiment of the present application, the value of the first sampling interval may be 0.1 ms, 0.2 ms, etc., which may be preset or determined according to a preset algorithm, which is not limited in this embodiment. . In other embodiments, the audio recognition module 403 may further perform speech data recognition on the audio data, and recognize semantics in the speech data, and the like.

The interaction module 404 is used for information interaction with other devices, for example, information interaction with other slave devices in the device group. The interaction module 404 is used to send the wake-up recognition results of other slave devices to the audio recognition module 403 when receiving the wake-up recognition results sent by other slave devices in the device group, so that the audio recognition module 403 can combine the wake-up recognition results of other slave devices. Determines whether to wake up the electronic device. The interaction module 404 is further configured to send control instructions to other slave devices in the device group, so that the other slave devices cooperate with the master device to process the user's instruction intention.

Additionally, in some embodiments, the electronic device may further include an audio synthesizer 405 . The audio synthesis module 405 is used for synthesizing corresponding response voice data, and converting the response voice data into sound for playback. For example, the electronic device may play the sound of "What help is needed" through the speaker 161 in response to collecting the sound of "Xiaoyi Xiaoyi" uttered by the user. In this case, the audio synthesis module 405 is used to synthesize corresponding response voice data in response to the collected voice of "Xiaoyi Xiaoyi" from the user, and convert the synthesized response voice data into "What help do I need?" sound and play.

It should be understood that the software structure shown in FIG. 3 is only an example. The electronic device of the embodiments of the present application may have more or less modules than the electronic device shown in the figure, two or more modules may be combined, and so on. The various modules shown in the figures may be implemented in hardware, software, or a combination of hardware and software, including one or more signal processing and/or application specific integrated circuits.

It should be noted that the audio collection module 401, audio identification module 403, audio processing module 402, interaction module 404, and audio synthesis module 405 shown in FIG. 3 may be integrated into one or more of the processor 110 shown in FIG. 2 . Among the processing units, for example, some or all of the audio collection module 401, audio identification module 403, audio processing module 402, interaction module 404, and audio synthesis module 405 may be integrated in one or more of application processors, special-purpose processors, etc. in the processor. It should be noted that, the dedicated processor in the embodiments of the present application may be a DSP, an application specific integrated circuit (application specific integrated circuit, ASIC) chip, or the like.

The following embodiments can all be implemented in an electronic device having the above-mentioned hardware structure and/or software structure.

The usage scenarios of the voice wake-up method provided by the present application will be described in detail below with reference to the accompanying drawings.

Figures 4a to 4c take a smart speaker as an example of an electronic device in the device group. As shown in Figure 4a, the scene includes 6 speakers, which form a device group and are distributed at different locations in area 1 (eg, living room). Among them, the speaker 11 is the master device of the device group, the speakers 12 to 16 are the slave devices of the device group, and each speaker in the device group is connected to each other (FIG. 4a only shows the connection between the master device and each slave device) relation). Each speaker in the device group has the same device type and the same wake word. Each speaker in the device group can receive or collect the wake-up voice sent by the user at the time of wake-up. In the device group, only the main device speaker 11 responds to the wake-up. Whether the speaker 11 is awakened, after the speaker 11 is awakened, the device speaker 12 to the speaker 16 can cooperate with the speaker 11 to process the user's instruction intention. The difference between Fig. 4b and Fig. 4a is that the 6 speakers in the device group are distributed in different regions. As an example, as shown in Fig. 4b, the master device speaker 11 and the slave device speaker 12 are located in region 1, and the slave device speakers 13 to The loudspeaker 16 is located in area 2 (eg bedroom). In this scenario, the user can send a wake-up voice in any of the above two areas. The distance between the user and the main device speaker 11 in the device group is not fixed, and the distance may be short or far away, and the transmission of the wake-up voice is in progress. There may also be noise sources on the path. The difference between Figure 4c and Figure 4a is that the hardware specifications of the speakers in the device group are different, but the wake-up words are the same. According to the hardware specifications of the speakers, the speakers can be subdivided into speaker Pro, conventional speakers and speaker mini. The size relationship of the hardware specifications of each speaker is: speaker Pro > conventional speakers > speaker mini. Among them, the hardware specifications include but are not limited to the processing performance and size of the built-in chip of the speaker. As shown in Figure 4c, the speaker Pro11 with the best hardware specifications is used as the main device of the device group.

The above usage scenarios are only examples. In the scenario, each electronic device in the same device group may be the same type of device. For example, the device group is all smart speakers, as shown in Figure 4a to Figure 4c, or different types of devices, such as devices There are smart speakers, smart phones, smart TVs, etc. in the group, as shown in Figure 1. This embodiment of the present application does not impose any restrictions on the device types of the electronic devices in the device group.

As an example, the user can manage or set the electronic devices in the device group through any electronic device having a display function, such as a smart phone, a tablet computer, and the like. FIG. 5 shows a user interface interaction diagram according to an embodiment of the present application. As shown in FIG. 5, user interface 501 includes a status bar, icon controls for multiple applications (eg, a "smart home" icon), time and weather widgets, and the like. After the electronic device detects a touch operation of a user's finger (or stylus) on an application icon, in response to the touch operation, the electronic device starts the application and displays the user interface of the application on the display screen. Exemplarily, the electronic device detects a touch operation on the "smart home" icon, and in response to the touch operation, displays the user's device interface 502 on the display screen, and the device interface 502 includes a list of smart devices that the user has added (as shown in FIG. 5 ). Shown devices a, b, c, d), first control 503, second control 504, third control 505. The user can add a new smart device to the smart device list by clicking the first control 503, and the user can also add multiple smart devices in the smart device list to the same device group by clicking the second control 504 and the third control 505. . On the device group interface 506, the user can click the fourth control 507 to set the main device in the device group (device a is set as the main device in FIG. 5), and the user can also click the device control 508 to enter the device group Any smart device interface 509 , to query device information of a certain smart device (eg, device model, data record, wake-up sensitivity, etc.). Exemplarily, for the wake-up sensitivity of the smart device, the user can set through the fifth control 510 to enable or disable the wake-up enhancement function of the smart device.

Based on the above introduction of electronic devices and scenarios, the voice wake-up method provided by the present application will be described in detail below with reference to several specific embodiments.

FIG. 6 shows an interactive schematic diagram of the voice wake-up method provided by the embodiment of the present application. Exemplarily, taking a device group composed of three electronic devices as an example, it is assumed that the first electronic device is the master device of the device group, and the second electronic device a and the second electronic device b are slave devices of the device group. As shown in FIG. 6 , when a user sends a wake-up voice, each electronic device in the device group can receive or collect the wake-up voice, and each electronic device in the device group performs wake-up word detection on the wake-up voice to obtain a wake-up recognition result. Specifically, the first electronic device calculates the first wake-up confidence of the wake-up voice; the second electronic device a calculates the second wake-up confidence of the wake-up voice, and determines whether to allow or prohibit the second electronic device a from being woken up; the second electronic device b calculates The third wake-up confidence level of the wake-up voice determines whether the second electronic device b is allowed or prohibited to be woken up. After determining the wake-up recognition result, the second electronic device a and the second electronic device b send the respectively determined wake-up recognition results to the first electronic device, so that the first electronic device can determine the first wake-up confidence level and the device group according to the first electronic device itself. The wake-up identification results sent by other slave devices in the device determine whether to allow or prohibit the first electronic device from being woken up. It should be understood that due to the difference in the distance between each electronic device and the user, the difference in the interference factors on the transmission path, the difference in the processing performance of each electronic device, etc., the wake-up confidence calculated by each electronic device may be the same or different. Results may vary.

As an example, FIG. 7 shows a schematic flowchart of a voice wake-up method. The method takes the first electronic device as the execution body. As shown in FIG. 7 , the method specifically includes the following steps:

Step 101: Acquire audio data.

The first electronic device may collect the sound of the surrounding environment through a sound collecting device, such as a microphone, other sensors for collecting sound, and the like. After collecting ambient sound, the sound collection device converts the ambient sound into audio electrical data, and outputs the audio electrical data to the audio module 160, which performs encoding and/or analog-to-digital conversion to obtain audio data in a corresponding format. After the audio module 160 obtains the audio data in the corresponding format, the audio data in the corresponding format can be sent to the audio collection module 401 in the processor 110 . The audio collection module 401 stores the audio data in the corresponding format in a memory (eg, the internal memory 121 or the memory in the processor 110 , etc.), and sends the audio data in the corresponding format to the audio processing module 402 . The audio processing module 402 preprocesses the audio data in the corresponding format to obtain processed audio data. The preprocessing of audio data includes channel conversion, smoothing, noise reduction, and the like. Optionally, in some embodiments, the audio module 160 acquires audio data in a corresponding format, and may also send a notification to the audio collection module 401 in the processor 110. After receiving the notification, the audio collection module 401 acquires the audio data from the audio module 401. Audio data of the corresponding format, then store the audio data of the corresponding format in the memory, and send the audio data of the corresponding format to the audio processing module 402, and the audio processing module 402 preprocesses the audio data of the corresponding format to obtain the processed audio. data.

Step 102: Calculate the first wake-up confidence level of the audio data. Wherein, the first wake-up confidence level is used to indicate the similarity of acoustic features between the wake-up word in the audio data and the preset wake-up word.

The first electronic device may perform wake word detection on the preprocessed audio data through the processor 110 . Exemplarily, the first electronic device performs wake word detection on the preprocessed audio data through the audio recognition module 403 in the processor 110 . The audio identification module 403 is the preprocessed audio data obtained from the audio processing module 402 .

Specifically, the audio recognition module 403 performs wake-up word detection on the preprocessed audio data, and can determine whether there is a preset wake-up word in the audio data through a voice wake-up model. The voice wake-up model is pre-trained with a large amount of audio data, and the voice wake-up model is used to calculate the similarity between the wake-up words in the audio data and the acoustic features of the preset wake-up words. The input of the voice wake-up model is preprocessed audio data, the output of the voice wake-up model may include a wake-up identifier, and the wake-up identifier includes an identifier that allows or prohibits the electronic device from being woken up. Exemplarily, the wake-up flag is 0, which is used to indicate that the electronic device is prohibited from being woken up; the wake-up flag is 1, which is used to indicate that the electronic device is allowed to be woken up. In some embodiments, the output of the voice wake-up model may include a wake-up confidence level, where the wake-up confidence level is used to indicate the similarity between the acoustic features of the wake-up word in the audio data and the preset wake-up word. In some embodiments, the output of the voice wake-up model may include the aforementioned wake-up identification and wake-up confidence at the same time. The audio recognition module 403 obtains the wake-up identification and/or the wake-up confidence level through the voice wake-up model.

Step 103: Receive a wake-up identification result sent by at least one second electronic device. Wherein, the wake-up identification result is used to indicate that at least one second electronic device is allowed or prohibited to be woken up.

In this embodiment, each slave device in the device group has a wake-up recognition function. Similar to the above steps 101 and 102, each second electronic device in the device group can also preprocess the received audio data, and then perform wake-up word detection on the pre-processed audio data, which can be determined by the voice wake-up model. Whether there is a preset wake-up word in the audio data, the similarity of the acoustic features of the wake-up word in the audio data and the preset wake-up word is calculated by the voice wake-up model, and finally the wake-up identification and/or the wake-up confidence level are obtained. Correspondingly, the wake-up identification result sent by each second electronic device may include a wake-up identification and/or a wake-up confidence level. The first electronic device can receive the wake-up recognition result sent by at least one second electronic device in the device group through the interaction module 404. When the interaction module 404 receives the wake-up recognition result sent by the at least one second electronic device, it sends to the audio recognition module 403. The wake-up identification result of the at least one second electronic device, so that the audio identification module 403 determines whether to wake up the first electronic device according to the first wake-up confidence level and the wake-up identification result sent by the at least one second electronic device.

Step 104: Determine whether to allow or prohibit the first electronic device from being woken up according to the first wake-up confidence level and the wake-up identification result sent by the at least one second electronic device.

In this step, the first electronic device may determine whether to allow or prohibit the first electronic device from being woken up through the audio identification module 403 in the processor 110 . Specifically, the audio recognition module 403 firstly judges the relationship between the first wake-up confidence and the preset wake-up threshold, and determines whether to allow or prohibit the first electronic device from being woken up according to the judgment result. The preset wake-up threshold may include one threshold or two thresholds.

In a possible implementation manner, the preset wake-up threshold includes a threshold, which is a first threshold. As shown in FIG. 8a, the audio identification module 403 may determine whether to allow or prohibit the first electronic device from being woken up according to the magnitude relationship between the first wake-up confidence level and the first threshold. Specifically, if the first wake-up confidence level is greater than or equal to the first threshold, the audio recognition module 403 determines that the first electronic device is allowed to be awakened; if the first wake-up confidence level is less than the first threshold, the audio recognition module 403 determines according to the first wake-up confidence level and the wake-up identification result sent by the at least one second electronic device to determine whether to allow or prohibit the first electronic device from being woken up. In this implementation manner, if the first wake-up confidence level calculated by the first electronic device is high, the first electronic device may directly determine that the first electronic device is allowed to be woken up. If the first wake-up confidence calculated by the first electronic device is not very high, the first electronic device needs to combine the wake-up identification results sent by other slave devices in the device group to comprehensively determine whether the first electronic device is allowed to be woken up, and improve the device group. The accuracy of the wake-up of the master device.

In another possible implementation manner, the preset wake-up threshold includes two thresholds, which are a first threshold and a second threshold, wherein the first threshold is greater than the second threshold. As shown in FIG. 8b , the audio recognition module 403 may determine whether to allow or prohibit the first electronic device from being woken up according to the relationship between the first wake-up confidence level and the first threshold and the second threshold. Specifically, if the first wake-up confidence is greater than or equal to the first threshold, the audio recognition module 403 determines to allow the first electronic device to be awakened; if the first wake-up confidence is less than or equal to the second threshold, the audio recognition module 403 determines to prohibit the first electronic device The electronic device is woken up; if the first wake-up confidence level is less than the first threshold and greater than the second threshold, the audio recognition module 403 determines to allow or prohibit the first wake-up confidence level and the wake-up recognition result sent by at least one second electronic device Electronic device wakes up. Compared with the first implementation, this implementation adds a second threshold, which is a relatively low judgment threshold, if the first wake-up confidence calculated by the first electronic device is less than or equal to the second threshold , it can be considered that the first wake-up confidence is indeed very low. The first electronic device does not need to make a comprehensive judgment in combination with the wake-up recognition results sent by other slave devices in the device group, and can directly determine that the first electronic device is prohibited from being woken up. This method can avoid unnecessary The judgment process of the first electronic device improves the processing rate of the first electronic device, and the accuracy of the wake-up of the master device in the device group can also be improved by this method.

It should be noted that the preset wake-up thresholds of each electronic device in the device group in this embodiment of the present application may be the same threshold. For example, the preset wake-up thresholds in the first electronic device and any second electronic device include a threshold, which The threshold is the first threshold, and the first electronic device makes a preliminary determination according to the first threshold and the currently calculated first wake-up confidence. Similarly, the second electronic device makes a preliminary determination according to the first threshold and the currently calculated second wake-up confidence. The wake-up identification is determined, and the wake-up identification result is sent to the first electronic device. In some embodiments, the preset wake-up thresholds of each electronic device in the device group may also be different thresholds. For example, the preset wake-up threshold in the first electronic device includes a threshold, the threshold is the first threshold, and the second electronic device The preset wake-up threshold in the device also includes a threshold, and the threshold can be a value greater than or less than the first threshold. It can be seen that each electronic device in the device group can perform wake-up identification determination based on the respective preset wake-up thresholds.

To sum up, no matter which of the above implementation manners, the following steps are included: according to the first wake-up confidence level and the wake-up identification result sent by at least one second electronic device, it is determined whether to allow or prohibit the first electronic device from being woken up. This step includes the following three possible implementations:

In a first possible implementation manner, the first electronic device counts the wake-up status of the slave devices in the device group through the wake-up identification result sent by at least one second electronic device, and determines whether the wake-up status of the slave device satisfies a preset wake-up condition, If the preset wake-up condition is satisfied, it is determined to allow the first electronic device to be woken up, and if the preset wake-up condition is not satisfied, it is determined to prohibit the first electronic device from being woken up.

In a second possible implementation manner, the first electronic device dynamically adjusts the preset wake-up threshold of the first electronic device in the device group based on a continuous method, and calculates the preset wake-up threshold according to the adjusted preset wake-up threshold and the current first electronic device. The first wake-up confidence is compared to determine whether to allow or prohibit the first electronic device from being woken up.

In a third possible implementation manner, the first electronic device dynamically adjusts the preset wake-up threshold of the first electronic device in the device group based on a discrete method, and calculates the preset wake-up threshold according to the adjusted preset wake-up threshold and the current first electronic device. The first wake-up confidence is compared to determine whether to allow or prohibit the first electronic device from being woken up.

Among them, the first implementation method is based on a preset rule, which fully considers the wake-up recognition results of other slave devices in the device group, optimizes the wake-up conditions of the master device in the device group, and improves the wake-up of the master device in the device group. accuracy. The latter two implementations both optimize the preset wakeup threshold of the master device from the perspective of the wakeup threshold combined with the wakeup identification results of other slave devices in the device group, the purpose of which is the same as that of the first implementation.

The following describes in detail how the first electronic device makes a wake-up decision in combination with the wake-up identification results sent by other slave devices in the device group with reference to FIGS. 9 to 11 . It should be noted that, the judgment processes of the following several embodiments can be executed by the audio recognition module 403 in the processor 110 of the first electronic device.

FIG. 9 shows a judgment flowchart of a voice wake-up method. As shown in FIG. 9 , if the first wake-up confidence level calculated by the first electronic device is less than the first threshold, the method includes the following steps:

Step 201 , according to the wake-up identification result sent by the at least one second electronic device, count the wake-up situation of the at least one second electronic device.

Specifically, the wake-up identification result may include at least one of a wake-up identifier of at least one second electronic device and a second wake-up confidence level. The wake-up identifier includes an identifier that allows or prohibits the second electronic device from being woken up, and the second wake-up confidence level is used to indicate the acoustic feature similarity between the wake-up word and the preset wake-up word in the audio data determined by the second electronic device.

In a possible situation, if the wake-up identification result sent by at least one second electronic device only includes a wake-up identifier, the first electronic device can count the number of devices of the second electronic device that are allowed to be woken up according to the wake-up identifier, and the The number of devices as a percentage of the total number of devices in the device group.

In another possible situation, if the wake-up identification result sent by at least one second electronic device only includes the second wake-up confidence level, the first electronic device first needs to preset the second wake-up confidence level and each second electronic device according to the second wake-up confidence level. The wake-up threshold determines whether the second electronic device is allowed to be woken up. Then, count the number of devices of the second electronic device that are allowed to be woken up, and the proportion of the number of devices to the total number of devices in the device group. The preset wake-up threshold of each second electronic device may be the same value as the preset wake-up threshold of the first electronic device, eg, the first threshold, or may be a different value from the preset wake-up threshold of the first electronic device. The preset wake-up thresholds of the second electronic devices may be the same or different, which is not limited in this embodiment.

In yet another possible situation, if the wake-up identification result sent by at least one second electronic device includes both the wake-up identification and the second wake-up confidence, the first electronic device may determine whether to allow the The number of devices of the second electronic device that are awakened, and the proportion of the number of devices to the total number of devices in the device group.

Step 202: Determine whether the wake-up condition of at least one second electronic device satisfies the preset wake-up condition of the first electronic device, if it meets the preset wake-up condition of the first electronic device, perform step 203; If the wake-up condition of the first electronic device is satisfied, step 204 is executed.

In this embodiment, the preset wake-up condition of the first electronic device includes any one of the following:

(1) The second wake-up confidence levels of all the second electronic devices in the device group except the first electronic device are greater than or equal to the third threshold. The third threshold is a preset threshold in the first electronic device that allows the second electronic device to be woken up, and the threshold may be greater than the judgment threshold of some second electronic devices.

(2) The ratio of the number of devices of the second electronic device allowed to be woken up in the device group to the total number of devices in the device group is greater than or equal to the first ratio. Exemplarily, the first ratio may be set to 80%, and the meaning of this condition is: if 80% of the slave devices in the device group are allowed to be woken up, it can be determined that the first electronic device of the master device is allowed to be woken up.

(3) The ratio of the number of devices of the second electronic device allowed to be awakened in the device group to the total number of devices in the device group is smaller than the first proportion and greater than the second proportion, and the second wake-up confidence level of the second electronic device allowed to be awakened are greater than or equal to the third threshold. Wherein, the first ratio is greater than the second ratio.

It should be noted that the original intention of the above-mentioned third condition is as follows: the preset wake-up thresholds of the second electronic devices may be different, that is, the judgment thresholds of the second electronic devices that are allowed to be woken up are different, for example, the second electronic device a The preset wake-up threshold of the second electronic device b is 0.7, the preset wake-up threshold of the second electronic device b is 0.8, and the preset wake-up threshold of the second electronic device c is 0.5. If the second wake-up confidence levels of the audio data calculated by the second electronic devices are the same, based on different preset wake-up thresholds, the wake-up recognition results determined by the second electronic devices will be different. Based on this, the reference of the number of devices of the second electronic device that is allowed to be woken up obtained by the first electronic device is not high. In order to improve the accuracy of the judgment, a third threshold can be set to check whether there is an electronic device whose second wake-up confidence is lower than the third threshold in the second electronic device that is allowed to be woken up, and if so, it is determined to prohibit the first electronic device be awakened. The above conditions can effectively prevent the master device from being woken up by mistake due to the low preset wake-up threshold set by some slave devices, and improve the accuracy of the master device's wake-up.

Exemplarily, the first ratio is set to 80%, the second ratio is set to 50%, the third threshold is set to 0.7, the slave devices include second electronic devices a, b, c, and d, and the preset values of each second electronic device are The wake-up threshold and the sent wake-up identification results are shown in Table 1. From Table 1, it can be seen that the number of devices of the second electronic device that is allowed to be woken up accounts for 60% (3/5) of the total number of devices in the device group, which is between the first ratio However, among the second electronic devices that are allowed to be woken up, there is an electronic device whose second wake-up confidence is less than the third threshold 0.7, that is, the second electronic device c, and therefore does not meet the third condition above, then determine The first electronic device is not awakened.

Table 1

second electronic device Preset wake-up threshold Second wake-up confidence wake up sign a 0.7 0.8 1 b 0.7 0.5 0 c 0.5 0.5 1 d 0.8 0.8 1

Step 203: Determine to allow the first electronic device to be woken up.

Step 204 , determining that the first electronic device is prohibited from being woken up.

The above solution provides a wake-up condition based on the preset master device, when it is determined that the first wake-up confidence calculated by the master device is less than the first threshold, according to the wake-up recognition results sent by other slave devices in the device group, count the device group. The wake-up condition of the slave device determines whether the wake-up condition of the slave device satisfies the preset master device wake-up condition, and if so, it is determined to allow the master device in the device group to be woken up. This solution can quickly determine whether the master device in the device group performs a wake-up response, makes full use of the wake-up recognition results of other devices in the device group, and improves the accuracy of the master device in the device group wake-up.

It should be understood that if most of the slave devices in the device group except the master device are determined to be allowed to wake up, and the preset wake-up threshold of the master device is too high, the master device may misjudgment, so the following two implementations can be implemented: The method provided by the example, combined with the wake-up conditions of other slave devices in the device group, determines whether the wake-up threshold of the current master device needs to be dynamically adjusted, thereby improving the wake-up accuracy of the master device in the device group.

FIG. 10 shows a judgment flowchart of another voice wake-up method. As shown in FIG. 10 , if the first wake-up confidence level calculated by the first electronic device is less than the first threshold, the method includes the following steps:

Step 301 , according to the wake-up identification result sent by at least one second electronic device, count the first weight values of the second electronic devices allowed to be woken up in the device group.

In this embodiment, the weight value of the electronic device in the device group is introduced, and the weight value is used to indicate the confidence level of the wake-up recognition result of the electronic device. The weight value may be a value related to the device type of the electronic device, and may also be a value related to the software/hardware performance of the electronic device. As an example, a smart home includes smart devices of various device types, all of which have a voice wake-up function, including smart TVs, smart speakers, smart lights, smart air conditioners, smart refrigerators, etc. Devices with the same wake word in the device form a device group. Assuming that the device group includes 1 smart TV, 2 smart speakers, and 2 smart lights, the weight values of different device types in the device group can be preset. For example, the weight value of a smart TV is 0.3, and the weight value of a smart speaker is 0.6 , the weight of the smart light is 0.1. As another example, each electronic device in the device group may be the same type of device, such as a smart speaker. As shown in Figure 4c, the smart speaker can be subdivided into speaker Pro, conventional speaker, and speaker mini, and the size relationship of processing performance It is: Speaker Pro > Conventional Speaker > Speaker Mini, the weight value of various speakers can be preset according to the processing performance of the speaker, for example, the weight value of the speaker Pro is 0.5, the weight value of the conventional speaker is 0.3, and the weight value of the speaker mini is 0.2 .

It should be noted that, the weight value of each electronic device in the device group may be preset or recommended by a third-party device (eg, a service platform provided by an electronic device manufacturer). As an example, when an electronic device is newly added to a device group, the third-party device may deliver the weight value of the electronic device to the electronic device or the master device of the device group to which the electronic device is added. As an example, if the electronic device is a slave device in a device group, when the electronic device sends the wake-up recognition result to the master device in the device group, the weight value of the electronic device can be carried in the wake-up recognition result so that the master device can The device acquires the weight value of the newly added electronic device in time. In some embodiments, the weight value of each electronic device may be uniformly maintained by the main device or the third-party device.

In this embodiment, the wake-up identification result sent by the at least one second electronic device may include at least one of a wake-up identification of the at least one second electronic device, a second wake-up confidence level, and a device identification of the at least one second electronic device. The device identifier is used to indicate the device type of the second electronic device. Specifically, the first electronic device first counts the second electronic devices that are allowed to be woken up in the device group according to the wake-up identification result sent by at least one second electronic device. For details, please refer to step 201 in the above embodiment, which is not repeated here. Repeat. After determining the second electronic device that can be woken up, count the device types of the second electronic device that can be woken up and the device types of the second electronic device that can be woken up in the device group according to the device identifier of the second electronic device The corresponding number of devices. Based on the preset weight value of each electronic device, the first weight value of the second electronic device that is allowed to be woken up in the device group is finally calculated. For details, please refer to the following formula:

α＝γ ₁ ×n ₁ +γ ₂ ×n ₂ +…+γ _x ×n _x Formula 1

m=n ₁ +n ₂ +…+n _x Formula 2

In the formula, α represents the first weight value of the second electronic device that is allowed to be woken up in the device group;

γ _i represents the weight value of the i-th second electronic device that is allowed to be woken up;

n _i represents the number of devices of the i-th second electronic device that is allowed to be woken up, where i=1, 2, . . . , x, where x is a positive integer greater than or equal to 1;

x represents the number of device types of the second electronic device allowed to be woken up within the device group;

m represents the total number of devices of the second electronic device allowed to be woken up in the device group.

Exemplarily, Table 2 shows the parameter statistics table of all second electronic devices except the first electronic device in the device group, and Table 2 includes the weight value of each second electronic device, the total number of devices, and the devices that are allowed to be woken up. quantity.

Table 2

second electronic device Weights Total number of devices The number of devices allowed to be woken up a 0.5 2 1 b 0.3 1 0 c 0.2 3 2

It can be seen from Table 2 that the total number of second electronic devices allowed to be awakened in the device group is 3, including 1 second electronic device a and 2 second electronic devices c, where the weight of the second electronic device a is 0.5, the weight value of the second electronic device c is 0.2, according to the above formula 1, the first weight value α=0.5×1+0.2×2=0.9 of the second electronic device allowed to be woken up in the device group.

Step 302: Count the second weight values of all the second electronic devices in the device group.

As an example, the first electronic device may count the device types of all second electronic devices in the device group according to the device identifier in the wake-up identification result sent by at least one second electronic device, and the second electronic device corresponding to each device type. The total number of electronic devices. As another example, the first electronic device may directly determine the device types of all the second electronic devices in the device group and the total number of second electronic devices corresponding to each device type according to the interconnection situation in the current device group. It should be noted that, when the second electronic device added to the device group is interconnected with the first electronic device for the first time, the device information of the second electronic device can be sent to the first electronic device, so that the first electronic device can be pre-stored in the device group. Device information of all the second electronic devices, wherein the device information includes a device identifier and/or a weight value.

Specifically, the second weight value of all the second electronic devices in the device group can be determined by referring to the following formula:

β＝γ ₁ ×n ₁ +γ ₂ ×n ₂ +…+γ _z ×n _z Formula 3

o=n ₁ +n ₂ +…+n _x Formula 4

In the formula, β represents the second weight value of all second electronic devices in the device group;

γ _j represents the weight value of the jth second electronic device;

n _j represents the total number of devices of the jth second electronic device; wherein, j=1,2,...,z, z is a positive integer greater than or equal to 1;

z represents the number of device types of the second electronic device in the device group;

o represents the total number of devices of all second electronic devices in the device group.

Exemplarily, still taking Table 2 as an example, it can be known that the total number of devices of all second electronic devices in the device group is 6, including 2 second electronic devices a, 1 second electronic device b and 3 second electronic devices c, where the weight value of the second electronic device a is 0.5, the weight value of the second electronic device b is 0.3, and the weight value of the second electronic device c is 0.2. According to the above formula 3, all the second electronic devices in the device group The second weight value β=0.5×2+0.3×1+0.2×3=1.9.

Step 303: Adjust the first threshold based on the first weight value and the second weight value.

The first threshold is a threshold that allows the first electronic device to be woken up (ie, a preset wake-up threshold of the first electronic device).

In a specific implementation manner, the first electronic device determines the adjustment parameter of the wake-up threshold of the first electronic device according to the first weight value, the second weight value and the maximum threshold adjustment parameter, and the adjustment parameter can be determined by the following formula:

θ=α/β Formula 5

Δ′=θ×Δ Formula 6

In the formula, θ represents the weight ratio of the second electronic devices allowed to be woken up in the device group to all the second electronic devices;

Δ represents the maximum threshold adjustment parameter, which is a preset value, usually set between 0 and 1, for example, Δ=0.1.

Δ' represents an adjustment parameter of the wake-up threshold of the first electronic device.

The first electronic device adjusts the preset wake-up threshold of the first electronic device based on the above adjustment parameters, and the adjusted wake-up threshold of the first electronic device can be determined by the following formula:

Thredhold _current =Thredhold-Δ′ Equation 7

In the formula, Thredhold _current represents the adjusted wake-up threshold of the first electronic device (that is, the adjusted first threshold);

Thredhold represents a preset wake-up threshold (ie, a first threshold) of the first electronic device.

It can be seen that in this embodiment, the adjustment parameter of the wake-up threshold is determined by calculating the weight value of the device, the adjustment parameter has high accuracy, and the adjustment parameter at adjacent moments is usually continuously changed.

Step 304: Determine whether to allow or prohibit the first electronic device from being woken up according to the first wake-up confidence level and the adjusted first threshold.

Specifically, by comparing the magnitude relationship between the first wake-up confidence level and the adjusted first threshold, it is determined whether to allow or prohibit the first electronic device from being woken up. If the first wake-up confidence calculated by the first electronic device is greater than or equal to the adjusted first threshold, it is determined that the first electronic device is allowed to wake up; if the first wake-up confidence is less than the adjusted first threshold, it is determined to prohibit The first electronic device is awakened.

From the perspective of the wake-up threshold, the above scheme comprehensively analyzes the confidence level of the wake-up recognition results of the slave devices that are allowed to be woken up, combined with the actual wake-up situation of the slave devices in the device group, and then dynamically adjusts the master device in the device group based on a continuous method. Wake-up threshold, and determine whether the current master device needs to make a wake-up response according to the comparison result between the adjusted wake-up threshold and the wake-up confidence level calculated by the current master device. Since the adjusted wake-up threshold of the master device is more in line with the actual wake-up state of the overall device group, the accuracy of the wake-up of the master device in the device group is improved.

FIG. 11 shows a judgment flowchart of another voice wake-up method. As shown in FIG. 11 , if the first wake-up confidence level calculated by the first electronic device is less than the first threshold, the method includes the following steps:

Step 401 , according to the wake-up identification result sent by at least one second electronic device, count the number of devices of the second electronic device allowed to be woken up in the device group.

In this embodiment, the wake-up identification result sent by the at least one second electronic device includes the wake-up identification and/or the second wake-up confidence level of the at least one second electronic device. The first electronic device firstly counts which second electronic devices are allowed to be woken up in the device group according to the wake-up identification result sent by at least one second electronic device, thereby determining the number of devices of the second electronic device allowed to be woken up in the device group, For details, reference may be made to step 201 in the foregoing embodiment, which is not repeated here.

Step 402: Determine threshold adjustment parameters corresponding to the number of devices according to the wake-up threshold adjustment table.

In this embodiment, a wake-up threshold adjustment table is pre-stored in the memory of the first electronic device (for example, the internal memory 121, or the memory in the processor 110, etc.).

In a possible implementation manner, the wake-up threshold adjustment table may include a correspondence between the number of devices of the second electronic device that are allowed to be woken up and the threshold adjustment parameters. As an example, the wake-up threshold adjustment table may include a corresponding relationship between a numerical range of the number of devices of the second electronic device that is allowed to be woken up and a threshold adjustment parameter. Exemplarily, assuming that the total number of second electronic devices other than the first electronic device in the device group is 7, the number of devices of the second electronic device that is allowed to be woken up obtained from the statistics of the first electronic device may fall within the values shown in Table 3. Within any value range of , the first electronic device may determine the current threshold adjustment parameter according to the wake-up threshold adjustment table shown in Table 3.

table 3

It can be seen from Table 3 that the more the number of devices of the second electronic device that is allowed to be woken up, the larger the threshold adjustment parameter.

In another possible implementation manner, the wake-up threshold adjustment table may include a ratio of the number of devices of the second electronic device allowed to be woken up to the total number of devices of all second electronic devices, and a corresponding relationship with the threshold adjustment parameter. It should be understood that the larger the ratio, the larger the threshold adjustment parameter.

It can be seen that, in this embodiment, the threshold adjustment parameters are determined through a preset wake-up threshold adjustment table, and each threshold adjustment parameter is a discrete value.

Step 403: Adjust the first threshold based on the threshold adjustment parameter.

The method for adjusting the preset wake-up threshold of the first electronic device in this embodiment is the same as step 303 in the above embodiment. After determining the threshold adjustment parameters, formula 7 can be used to reduce the preset wake-up threshold of the first electronic device.

Step 404: Determine whether to allow or prohibit the first electronic device from being woken up according to the first wake-up confidence and the adjusted first threshold.

Step 404 in this embodiment is the same as step 304 in the foregoing embodiment. For details, reference may be made to the foregoing embodiment, and details are not repeated here.

The above solution also starts from the wake-up threshold, combined with the actual wake-up situation of the slave devices in the device group, counts the number or proportion of the slave devices that are allowed to be woken up, and then dynamically adjusts the wake-up threshold of the master device in the device group based on a discrete method. , and according to the comparison result between the adjusted wake-up threshold and the wake-up confidence level calculated by the current master device, it is determined whether the current master device needs to make a wake-up response. Since the adjusted wake-up threshold of the master device is more in line with the actual state of wake-up of the overall device group, the accuracy of the wake-up of the master device in the device group is improved.

The voice wake-up method provided by the embodiments of the present application is described in detail above, and the voice wake-up device provided by the embodiments of the present application will be described below. FIG. 12 is a schematic structural diagram of a voice wake-up device provided by an embodiment of the present application. As shown in FIG. 12 , the voice wake-up device 1200 provided by this embodiment of the present application includes:

The obtaining module 1201 is configured to obtain a first wake-up confidence level of the audio data, where the first wake-up confidence level is used to indicate that the wake-up word in the audio data determined by the first electronic device is similar to the acoustic feature of the preset wake-up word Spend;

a receiving module 1202, configured to receive a wake-up identification result sent by the at least one second electronic device, where the wake-up identification result is used to indicate allowing or prohibiting the at least one second electronic device from being woken up;

The processing module 1203 is configured to determine whether to allow or prohibit the first electronic device from being woken up according to the first wake-up confidence level and the wake-up identification result.

Optionally, the wake-up identification result includes at least one of a wake-up identification and a second wake-up confidence of the at least one second electronic device; wherein the wake-up identification includes an identification that allows or prohibits the second electronic device from being woken up. , the second wake-up confidence level is used to indicate the acoustic feature similarity between the wake-up word and the preset wake-up word in the audio data determined by the second electronic device.

Optionally, the processing module 1203 is specifically used for:

If the first wake-up confidence is greater than or equal to a first threshold, determining that the first electronic device is allowed to wake up; or

If the first wake-up confidence level is less than the first threshold, and the first wake-up confidence level is greater than the second threshold, determine whether to allow or prohibit the first wake-up confidence level according to the first wake-up confidence level and the wake-up identification result. an electronic device is awakened; or

If the first wake-up confidence level is less than or equal to the second threshold, it is determined that the first electronic device is prohibited from being woken up.

Optionally, the processing module 1203 is specifically used for:

Statistics on the wake-up situation of the at least one second electronic device according to the wake-up identification result;

If the wake-up condition satisfies a preset wake-up condition of the first electronic device, it is determined that the first electronic device is allowed to wake up.

Optionally, the wake-up condition includes any of the following:

The second wake-up confidences of all second electronic devices in the device group except the first electronic device are greater than or equal to a third threshold;

The ratio of the number of devices of the second electronic device allowed to be woken up in the device group to the total number of devices in the device group is greater than or equal to the first ratio;

The ratio of the number of devices of the second electronic device that can be woken up in the device group to the total number of devices in the device group is smaller than the first ratio and greater than the second ratio, and the second wake-up of the second electronic device that is allowed to wake up is allowed The confidence levels are all greater than or equal to the third threshold;

The third threshold is a preset threshold in the first electronic device that allows the second electronic device to be woken up.

Optionally, the wake-up identification result further includes a device identifier of the at least one second electronic device, where the device identifier is used to indicate the device type of the second electronic device, and determine the weight value of the second electronic device. .

Optionally, the processing module 1203 is specifically used for:

Calculate the first weight value of the second electronic device that is allowed to be woken up in the device group according to the wake-up identification result;

Counting second weight values of all second electronic devices in the device group;

Adjusting a first threshold based on the first weight value and the second weight value;

According to the first wake-up confidence level and the adjusted first threshold, determine whether to allow or prohibit the first electronic device from being woken up;

Wherein, the first threshold is a threshold for allowing the first electronic device to be woken up.

Optionally, the first weight value is based on the number of devices corresponding to the device type of the second electronic device that is allowed to be woken up in the device group, and the weight value corresponding to each device type of the second electronic device that is allowed to be woken up. The second weight value is determined according to the number of devices corresponding to the device types of all the second electronic devices in the device group, and the weight value corresponding to each device type.

Optionally, the processing module 1203 is specifically used for:

The product of the ratio of the first weight value to the second weight value and the maximum threshold adjustment parameter is used as a threshold adjustment parameter, and the first threshold is adjusted according to the threshold adjustment parameter.

Optionally, the processing module 1203 is specifically used for:

Counting the number of devices of the second electronic device allowed to be woken up in the device group according to the wake-up identification result;

Determine threshold adjustment parameters corresponding to the number of devices according to a wake-up threshold adjustment table, where the wake-up threshold adjustment table includes a correspondence between the number of devices of the second electronic device that is allowed to be woken up and the threshold adjustment parameters;

Adjusting the first threshold based on the threshold adjustment parameter;

and determining whether to allow or prohibit the first electronic device from being woken up according to the first wake-up confidence and the adjusted first threshold.

Optionally, the processing module 1203 is specifically used for:

If the first wake-up confidence level is greater than or equal to the adjusted first threshold, determining that the first electronic device is allowed to wake up; or

If the first wake-up confidence level is less than the adjusted first threshold, it is determined that the first electronic device is prohibited from being woken up.

The voice wake-up device provided in this embodiment of the present application is used to execute the technical solution executed by the first electronic device in the method embodiments shown in FIG. 7 , FIG. 8 a , 8 b , and FIG. 9 . No longer.

FIG. 13 is a schematic structural diagram of a voice wake-up device provided by an embodiment of the present application. As shown in FIG. 13 , the voice wake-up device 1300 provided by this embodiment of the present application includes:

The acquiring module 1301 is configured to acquire the first wake-up confidence level of the audio data, where the first wake-up confidence level is used to indicate that the wake-up word in the audio data determined by the first electronic device is similar to the acoustic feature of the preset wake-up word Spend;

a receiving module 1302, configured to receive a wake-up identification result sent by the at least one second electronic device, where the wake-up identification result is used to indicate allowing or prohibiting the second electronic device from being woken up;

The processing module 1303 is configured to adjust a first threshold according to the wake-up identification result, where the first threshold is a threshold for allowing the first electronic device to be woken up; according to the adjusted first threshold and the first wake-up confidence , and determine whether to allow or prohibit the first electronic device from being woken up.

Optionally, the processing module 1303 is specifically used for:

Calculate the first weight value of the second electronic device that is allowed to be woken up in the device group according to the wake-up identification result;

Counting second weight values of all second electronic devices in the device group;

The first threshold is adjusted based on the first weight value and the second weight value.

Optionally, the first weight value is based on the number of devices corresponding to the device type of the second electronic device that is allowed to be woken up in the device group, and the weight value corresponding to each device type of the second electronic device that is allowed to be woken up. The second weight value is determined according to the number of devices corresponding to the device types of all the second electronic devices in the device group, and the weight value corresponding to each device type.

Optionally, the processing module 1303 is specifically used for:

The product of the ratio of the first weight value to the second weight value and the maximum threshold adjustment parameter is used as a threshold adjustment parameter, and the first threshold is adjusted according to the threshold adjustment parameter.

Optionally, the processing module 1303 is specifically used for:

Counting the number of devices of the second electronic device allowed to be woken up in the device group according to the wake-up identification result;

Determine threshold adjustment parameters corresponding to the number of devices according to a wake-up threshold adjustment table, where the wake-up threshold adjustment table includes a correspondence between the number of devices of the second electronic device that is allowed to be woken up and the threshold adjustment parameters;

The first threshold is adjusted based on the threshold adjustment parameter.

Optionally, the wake-up identification result includes at least one of a wake-up identification and a second wake-up confidence of the at least one second electronic device; wherein the wake-up identification includes an identification that allows or prohibits the second electronic device from being woken up. , the second wake-up confidence level is used to indicate the acoustic feature similarity between the wake-up word and the preset wake-up word in the audio data determined by the second electronic device.

Optionally, the processing module 1303 is specifically used for:

If the first wake-up confidence level is greater than or equal to the adjusted first threshold, determining that the first electronic device is allowed to wake up; or

If the first wake-up confidence level is less than the adjusted first threshold, it is determined that the first electronic device is prohibited from being woken up.

The voice wake-up device provided in this embodiment of the present application is used to execute the technical solution executed by the first electronic device in the method embodiments shown in FIG. 10 and FIG. 11 .

It should be noted that it should be understood that the above division of the modules of the voice wake-up device is only a division of logical functions, and may be fully or partially integrated into a physical entity in actual implementation, or may be physically separated. And these modules can all be implemented in the form of software calling through processing elements; they can also all be implemented in hardware; some modules can also be implemented in the form of calling software through processing elements, and some modules can be implemented in hardware. For example, the processing module may be a separately established processing element, or it may be integrated into a certain chip of the above-mentioned device, and it may also be stored in the memory of the above-mentioned device in the form of program code, and a certain processing element of the above-mentioned device Call and execute the function of the above processing module. The implementation of other modules is similar. In addition, all or part of these modules can be integrated together, and can also be implemented independently. The processing element described here may be an integrated circuit with signal processing capability. In the implementation process, each step of the above-mentioned method or each of the above-mentioned modules can be completed by an integrated logic circuit of hardware in the processor element or an instruction in the form of software.

For example, the above modules may be one or more integrated circuits configured to implement the above methods, such as: one or more application specific integrated circuits (ASIC), or one or more digital microprocessors (digital) signal processor, DSP), or, one or more field programmable gate array (field programmable gate array, FPGA) and so on. For another example, when one of the above modules is implemented in the form of a processing element scheduling program code, the processing element may be a general-purpose processor, such as a central processing unit (CPU) or other processors that can invoke program codes. For another example, these modules can be integrated together and implemented in the form of a system-on-a-chip (SOC).

FIG. 14 is a schematic diagram of a hardware structure of a voice wake-up device according to an embodiment of the present application. As shown in FIG. 14 , the voice wake-up device 1400 provided by this embodiment of the present application may include:

Processor 1401 , memory 1402 and communication interface 1403 . The memory 1402 is used to store computer programs; the processor 1401 is used to execute the computer programs stored in the memory 1402 to implement the method executed by the first electronic device in any of the above method embodiments. The communication interface 1403 is used for data communication or signal communication with at least one second electronic device or server.

Optionally, the memory 1402 may be independent or integrated with the processor 1401 . When the memory 1402 is a device independent of the processor 1401 , the voice wake-up device 1400 may further include: a bus 1404 for connecting the memory 1402 and the processor 1401 .

In a possible implementation manner, the processing module 1203 in FIG. 12 may be integrated in the processor 1401 and implemented, and the receiving module 1202 may be integrated in the communication interface 1403 and implemented. The processing module 1303 in FIG. 13 may be integrated in the processor 1401 and implemented, and the receiving module 1302 may be integrated in the communication interface 1403 and implemented.

In a possible implementation manner, the processor 1401 can be used to implement the information processing operation of the first electronic device in the above method embodiment, and the communication interface 1403 can be used to implement the signal transceiving operation of the first electronic device in the above method embodiment.

The voice wake-up device provided in this embodiment can be used to execute the method executed by the first electronic device in any of the above method embodiments, and its implementation principle and technical effect are similar, and details are not repeated here.

Embodiments of the present application further provide a computer-readable storage medium, where computer-executable instructions are stored in the computer-readable storage medium, and when the computer-executable instructions are executed by a processor, are used to implement the first method in any of the foregoing method embodiments. A technical solution for electronic equipment.

The embodiments of the present application further provide a program, which, when the program is executed by the processor, is used to execute the technical solution of the first electronic device in any of the foregoing method embodiments.

Embodiments of the present application further provide a computer program product, including program instructions, where the program instructions are used to implement the technical solution of the first electronic device in any of the foregoing method embodiments.

Embodiments of the present application further provide a chip, including: a processing module and a communication interface, where the processing module can execute the technical solutions of the first electronic device in the foregoing method embodiments.

Further, the chip also includes a storage module (eg, memory), the storage module is used for storing instructions, the processing module is used for executing the instructions stored in the storage module, and the execution of the instructions stored in the storage module causes the processing module to execute the first electronic The technical solution of the equipment.

In the above-mentioned embodiments, it may be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented in software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present application are generated. The computer may be a general purpose computer, special purpose computer, computer network, or other programmable device. The computer instructions may be stored in or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be downloaded from a website site, computer, server, or data center Transmission to another website site, computer, server, or data center is by wire (eg, coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (eg, infrared, wireless, microwave, etc.). The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that includes an integration of one or more available media. The usable media may be magnetic media (eg, floppy disks, hard disks, magnetic tapes), optical media (eg, DVDs), or semiconductor media (eg, solid state disks (SSDs)), and the like.

In this application, "at least one" means one or more, and "plurality" means two or more. "And/or", which describes the association relationship of the associated objects, indicates that there can be three kinds of relationships, for example, A and/or B, which can indicate: the existence of A alone, the existence of A and B at the same time, and the existence of B alone, where A, B can be singular or plural. The character "/" generally indicates that the related objects before and after are an "or" relationship; in the formula, the character "/" indicates that the related objects are a "division" relationship. "At least one item(s) below" or similar expressions thereof refer to any combination of these items, including any combination of single item(s) or plural items(s). For example, at least one item (a) of a, b, or c can represent: a, b, c, a-b, a-c, b-c, or a-b-c, where a, b, c can be single or multiple indivual.

It can be understood that, the various numbers and numbers involved in the embodiments of the present application are only for the convenience of description, and are not used to limit the scope of the embodiments of the present application.

It can be understood that, in the embodiments of the present application, the size of the sequence numbers of the above-mentioned processes does not imply the order of execution, and the execution order of each process should be determined by its functions and internal logic, rather than the implementation of the present application. The implementation of the examples constitutes no limitation.

Claims (38)

1. A voice wake-up method is applied to a first electronic device, wherein the first electronic device and at least one second electronic device belong to the same device group, and the method comprises the following steps:

acquiring a first awakening confidence level of audio data, wherein the first awakening confidence level is used for indicating the acoustic feature similarity of an awakening word in the audio data determined by the first electronic device and a preset awakening word;

receiving a wake-up identification result sent by the at least one second electronic device, wherein the wake-up identification result is used for indicating that the at least one second electronic device is allowed or prohibited to be woken up;

and determining to allow or prohibit the first electronic equipment to be awakened according to the first awakening confidence level and the awakening identification result.

2. The method of claim 1, wherein the wake up recognition result comprises at least one of a wake up identity of the at least one second electronic device, a second wake up confidence;

the awakening identifier comprises an identifier which allows or prohibits the second electronic device to be awakened, and the second awakening confidence coefficient is used for indicating the acoustic feature similarity between the awakening word in the audio data determined by the second electronic device and a preset awakening word.

3. The method according to claim 1 or 2, wherein the determining to allow or prohibit the first electronic device from being woken up according to the first wake-up confidence level and the wake-up recognition result comprises:

if the first awakening confidence coefficient is larger than or equal to a first threshold value, determining that the first electronic equipment is allowed to be awakened; or

If the first awakening confidence degree is smaller than the first threshold value and the first awakening confidence degree is larger than a second threshold value, determining to allow or prohibit the first electronic equipment to be awakened according to the first awakening confidence degree and the awakening identification result; or

And if the first awakening confidence coefficient is smaller than or equal to the second threshold value, determining to prohibit the first electronic equipment from being awakened.

4. The method of claim 3, wherein the determining to allow or prohibit the first electronic device from being woken up according to the first wake-up confidence level and the wake-up recognition result comprises:

counting the awakening condition of the at least one second electronic device according to the awakening identification result;

and if the awakening condition meets the preset awakening condition of the first electronic equipment, determining that the first electronic equipment is allowed to be awakened.

5. The method of claim 4, wherein the wake-up condition comprises any one of:

the second awakening confidence degrees of all second electronic equipment except the first electronic equipment in the equipment group are greater than or equal to a third threshold;

the proportion of the number of the devices of the second electronic device which is allowed to be awakened in the device group to the total number of the devices of the device group is greater than or equal to a first proportion;

the proportion of the number of the devices of the second electronic device allowed to be awakened in the device group to the total number of the devices of the device group is smaller than the first proportion and larger than a second proportion, and second awakening confidences of the second electronic devices allowed to be awakened are all larger than or equal to a third threshold;

the third threshold is a preset threshold allowing the second electronic device to be awakened in the first electronic device.

6. The method of claim 2, wherein the wake up recognition result further comprises a device identifier of the at least one second electronic device, wherein the device identifier is used for indicating a device type of the second electronic device, and determining a weight value of the second electronic device.

7. The method of claim 6, wherein the determining to allow or prohibit the first electronic device from being woken up according to the first wake-up confidence level and the wake-up recognition result comprises:

counting a first weight value of a second electronic device which is allowed to be awakened in the device group according to the awakening identification result;

counting second weighted values of all second electronic equipment in the equipment group;

adjusting a first threshold based on the first weight value and the second weight value;

determining to allow or prohibit the first electronic device to be awakened according to the first awakening confidence level and the adjusted first threshold;

wherein the first threshold is a threshold that allows the first electronic device to be woken up.

8. The method according to claim 7, wherein the first weight value is determined according to the number of devices corresponding to the device type of the second electronic device allowed to be woken in the device group and the weight values corresponding to the device types of the second electronic devices allowed to be woken; the second weight value is determined according to the number of devices corresponding to the device types of all the second electronic devices in the device group and the weight value corresponding to each device type.

9. The method of claim 7, wherein adjusting the first threshold based on the first and second weight values comprises:

and taking the product of the ratio of the first weight value to the second weight value and a maximum threshold value adjusting parameter as a threshold value adjusting parameter, and adjusting the first threshold value according to the threshold value adjusting parameter.

10. The method according to claim 1 or 2, wherein the determining to allow or prohibit the first electronic device from being woken up according to the first wake-up confidence level and the wake-up recognition result comprises:

counting the number of the second electronic equipment allowed to be awakened in the equipment group according to the awakening identification result;

determining a threshold adjustment parameter corresponding to the equipment number according to a wake-up threshold adjustment table, wherein the wake-up threshold adjustment table comprises a corresponding relation between the equipment number of the second electronic equipment allowed to be woken up and the threshold adjustment parameter;

adjusting a first threshold based on the threshold adjustment parameter;

and determining to allow or prohibit the first electronic equipment to be awakened according to the first awakening confidence and the adjusted first threshold.

11. The method according to claim 7 or 10, wherein the determining to allow or prohibit the first electronic device from being woken up according to the first wake-up confidence and the adjusted first threshold comprises:

if the first awakening confidence is greater than or equal to the adjusted first threshold, determining that the first electronic device is allowed to be awakened; or

And if the first awakening confidence coefficient is smaller than the adjusted first threshold, determining to prohibit the first electronic device from being awakened.

12. A voice wake-up method is applied to a first electronic device, wherein the first electronic device and at least one second electronic device belong to the same device group, and the method comprises the following steps:

acquiring a first awakening confidence level of audio data, wherein the first awakening confidence level is used for indicating the acoustic feature similarity of an awakening word in the audio data determined by the first electronic device and a preset awakening word;

receiving a wake-up identification result sent by the at least one second electronic device, wherein the wake-up identification result is used for indicating that the second electronic device is allowed or prohibited to be woken up;

adjusting a first threshold value according to the awakening identification result, wherein the first threshold value is a threshold value for allowing the first electronic device to be awakened;

and determining to allow or prohibit the first electronic equipment to be awakened according to the adjusted first threshold and the first awakening confidence level.

13. The method of claim 12, wherein adjusting the first threshold according to the wake-up recognition result comprises:

counting a first weight value of a second electronic device which is allowed to be awakened in the device group according to the awakening identification result;

counting second weighted values of all second electronic equipment in the equipment group;

adjusting the first threshold based on the first and second weight values.

14. The method according to claim 13, wherein the first weight value is determined according to the number of devices corresponding to the device type of the second electronic device allowed to be woken in the device group, and the weight values corresponding to the device types of the second electronic devices allowed to be woken; the second weight value is determined according to the number of devices corresponding to the device types of all the second electronic devices in the device group and the weight value corresponding to each device type.

15. The method of claim 13, wherein adjusting the first threshold based on the first and second weight values comprises:

and taking the product of the ratio of the first weight value to the second weight value and a maximum threshold value adjusting parameter as a threshold value adjusting parameter, and adjusting the first threshold value according to the threshold value adjusting parameter.

16. The method of claim 12, wherein adjusting the first threshold according to the wake-up recognition result comprises:

counting the number of the second electronic equipment allowed to be awakened in the equipment group according to the awakening identification result;

determining a threshold adjustment parameter corresponding to the equipment number according to a wake-up threshold adjustment table, wherein the wake-up threshold adjustment table comprises a corresponding relation between the equipment number of the second electronic equipment allowed to be woken up and the threshold adjustment parameter;

adjusting the first threshold based on the threshold adjustment parameter.

17. The method according to any of claims 12-16, wherein the wake up recognition result comprises at least one of a wake up identity of the at least one second electronic device, a second wake up confidence;

the awakening identifier comprises an identifier which allows or prohibits the second electronic device to be awakened, and the second awakening confidence coefficient is used for indicating the acoustic feature similarity between the awakening word in the audio data determined by the second electronic device and a preset awakening word.

18. The method according to any of claims 12-16, wherein the determining to allow or prohibit the first electronic device from being woken up according to the adjusted first threshold and the first wake-up confidence level comprises:

if the first awakening confidence is greater than or equal to the adjusted first threshold, determining that the first electronic device is allowed to be awakened; or

And if the first awakening confidence coefficient is smaller than the adjusted first threshold, determining to prohibit the first electronic device from being awakened.

19. The utility model provides a voice wake-up equipment, its characterized in that, voice wake-up equipment is first electronic equipment, first electronic equipment and at least one second electronic equipment belong to same equipment group, voice wake-up equipment includes:

the acquiring module is used for acquiring a first awakening confidence coefficient of the audio data, wherein the first awakening confidence coefficient is used for indicating the acoustic feature similarity of an awakening word in the audio data determined by the first electronic device and a preset awakening word;

a receiving module, configured to receive a wake-up identification result sent by the at least one second electronic device, where the wake-up identification result is used to indicate that the at least one second electronic device is allowed or prohibited to be woken up;

and the processing module is used for determining to allow or prohibit the first electronic equipment to be awakened according to the first awakening confidence coefficient and the awakening identification result.

20. The device of claim 19, wherein the wake up recognition result comprises at least one of a wake up identity of the at least one second electronic device, a second wake up confidence level;

the awakening identifier comprises an identifier which allows or prohibits the second electronic device to be awakened, and the second awakening confidence coefficient is used for indicating the acoustic feature similarity between the awakening word in the audio data determined by the second electronic device and a preset awakening word.

21. The device according to claim 19 or 20, wherein the processing module is specifically configured to:

if the first awakening confidence coefficient is larger than or equal to a first threshold value, determining that the first electronic equipment is allowed to be awakened; or

If the first awakening confidence degree is smaller than the first threshold value and the first awakening confidence degree is larger than a second threshold value, determining to allow or prohibit the first electronic equipment to be awakened according to the first awakening confidence degree and the awakening identification result; or

And if the first awakening confidence coefficient is smaller than or equal to the second threshold value, determining to prohibit the first electronic equipment from being awakened.

22. The device according to claim 21, wherein the processing module is specifically configured to:

counting the awakening condition of the at least one second electronic device according to the awakening identification result;

and if the awakening condition meets the preset awakening condition of the first electronic equipment, determining that the first electronic equipment is allowed to be awakened.

23. The device of claim 22, wherein the wake-up condition comprises any one of:

the second awakening confidence degrees of all second electronic equipment except the first electronic equipment in the equipment group are greater than or equal to a third threshold;

the proportion of the number of the devices of the second electronic device which is allowed to be awakened in the device group to the total number of the devices of the device group is greater than or equal to a first proportion;

the proportion of the number of the devices of the second electronic device allowed to be awakened in the device group to the total number of the devices of the device group is smaller than the first proportion and larger than a second proportion, and second awakening confidences of the second electronic devices allowed to be awakened are all larger than or equal to a third threshold;

the third threshold is a preset threshold allowing the second electronic device to be awakened in the first electronic device.

24. The device of claim 20, wherein the wake up recognition result further comprises a device identifier of the at least one second electronic device, wherein the device identifier is used for indicating a device type of the second electronic device, and determining a weight value of the second electronic device.

25. The device according to claim 24, wherein the processing module is specifically configured to:

counting a first weight value of a second electronic device which is allowed to be awakened in the device group according to the awakening identification result;

counting second weighted values of all second electronic equipment in the equipment group;

adjusting a first threshold based on the first weight value and the second weight value;

determining to allow or prohibit the first electronic device to be awakened according to the first awakening confidence level and the adjusted first threshold;

wherein the first threshold is a threshold that allows the first electronic device to be woken up.

26. The device according to claim 25, wherein the first weight value is determined according to the number of devices corresponding to the device type of the second electronic device allowed to be woken in the device group, and the weight values corresponding to the respective device types of the second electronic device allowed to be woken; the second weight value is determined according to the number of devices corresponding to the device types of all the second electronic devices in the device group and the weight value corresponding to each device type.

27. The device according to claim 25, wherein the processing module is specifically configured to:

and taking the product of the ratio of the first weight value to the second weight value and a maximum threshold value adjusting parameter as a threshold value adjusting parameter, and adjusting the first threshold value according to the threshold value adjusting parameter.

28. The device according to claim 19 or 20, wherein the processing module is specifically configured to:

counting the number of the second electronic equipment allowed to be awakened in the equipment group according to the awakening identification result;

determining a threshold adjustment parameter corresponding to the equipment number according to a wake-up threshold adjustment table, wherein the wake-up threshold adjustment table comprises a corresponding relation between the equipment number of the second electronic equipment allowed to be woken up and the threshold adjustment parameter;

adjusting a first threshold based on the threshold adjustment parameter;

and determining to allow or prohibit the first electronic equipment to be awakened according to the first awakening confidence and the adjusted first threshold.

29. The device according to claim 25 or 28, wherein the processing module is specifically configured to:

if the first awakening confidence is greater than or equal to the adjusted first threshold, determining that the first electronic device is allowed to be awakened; or

And if the first awakening confidence coefficient is smaller than the adjusted first threshold, determining to prohibit the first electronic device from being awakened.

30. The utility model provides a voice wake-up equipment, its characterized in that, voice wake-up equipment is first electronic equipment, first electronic equipment and at least one second electronic equipment belong to same equipment group, voice wake-up equipment includes:

the acquiring module is used for acquiring a first awakening confidence coefficient of the audio data, wherein the first awakening confidence coefficient is used for indicating the acoustic feature similarity of an awakening word in the audio data determined by the first electronic device and a preset awakening word;

a receiving module, configured to receive a wake-up identification result sent by the at least one second electronic device, where the wake-up identification result is used to indicate that the second electronic device is allowed or prohibited to be woken up;

the processing module is used for adjusting a first threshold value according to the awakening identification result, wherein the first threshold value is a threshold value allowing the first electronic device to be awakened; and determining to allow or prohibit the first electronic equipment to be awakened according to the adjusted first threshold and the first awakening confidence level.

31. The device according to claim 30, wherein the processing module is specifically configured to:

counting a first weight value of a second electronic device which is allowed to be awakened in the device group according to the awakening identification result;

counting second weighted values of all second electronic equipment in the equipment group;

adjusting the first threshold based on the first and second weight values.

32. The device according to claim 31, wherein the first weight value is determined according to the number of devices corresponding to the device type of the second electronic device allowed to be woken in the device group, and the weight values corresponding to the respective device types of the second electronic device allowed to be woken; the second weight value is determined according to the number of devices corresponding to the device types of all the second electronic devices in the device group and the weight value corresponding to each device type.

33. The device according to claim 31, wherein the processing module is specifically configured to:

and taking the product of the ratio of the first weight value to the second weight value and a maximum threshold value adjusting parameter as a threshold value adjusting parameter, and adjusting the first threshold value according to the threshold value adjusting parameter.

34. The device according to claim 30, wherein the processing module is specifically configured to:

counting the number of the second electronic equipment allowed to be awakened in the equipment group according to the awakening identification result;

determining a threshold adjustment parameter corresponding to the equipment number according to a wake-up threshold adjustment table, wherein the wake-up threshold adjustment table comprises a corresponding relation between the equipment number of the second electronic equipment allowed to be woken up and the threshold adjustment parameter;

adjusting the first threshold based on the threshold adjustment parameter.

35. The device according to any of claims 30-34, wherein the wake up recognition result comprises at least one of a wake up identity of the at least one second electronic device, a second wake up confidence;

the awakening identifier comprises an identifier which allows or prohibits the second electronic device to be awakened, and the second awakening confidence coefficient is used for indicating the acoustic feature similarity between the awakening word in the audio data determined by the second electronic device and a preset awakening word.

36. The device according to any one of claims 30 to 34, wherein the processing module is specifically configured to:

if the first awakening confidence is greater than or equal to the adjusted first threshold, determining that the first electronic device is allowed to be awakened; or

And if the first awakening confidence coefficient is smaller than the adjusted first threshold, determining to prohibit the first electronic device from being awakened.

37. A voice wake-up device, comprising: a memory for storing a computer program and a processor for calling and executing the computer program from the memory, such that the processor executes the computer program to perform the method of any of claims 1 to 11, or the method of any of claims 12 to 18.

38. A storage medium, characterized in that the storage medium comprises a computer program for implementing the method according to any one of claims 1 to 11 or the method according to any one of claims 12 to 18.

Images