Wake-up Method, Device, System, Medium, and Equipment Based on Collaborative Error Correction

Abstract

An embodiment of this specification provides a wake-up method, device, system, medium, and equipment based on collaborative error correction. The method includes: when receiving a wake-up request sent by a voice device, determining whether there are other voice devices in the space group where the voice device is located; if there are other voice devices, starting to time, and after the timing duration reaches a preset duration, determining whether a wake-up request sent by the other voice device is received during the timing process; if a wake-up request sent by the other voice device is received during the timing process, determining a response voice device from the voice devices that sent the wake-up request, and sending a wake-up response instruction to the response voice device, so that the response voice device enters the wake-up state from the wake-up waiting state. The present invention can reduce the overall probability of false wake-up of all voice devices in the home.

Description

Technical Field

[0001] One or more embodiments of this specification relate to the technical field of voice devices, and in particular, to a wake-up method, device, system, medium, and device based on collaborative error correction. Background Art

[0002] Voice recognition technology is the most widely used human-computer interaction technology at present. Users issue voice commands to control voice devices to perform corresponding operations. For example, controlling an air conditioner to turn on and off. The entire process of voice interaction control includes several key links: voice wake-up, sending voice commands, device execution of actions, and device voice feedback. Voice wake-up means presetting one or more proprietary words for a voice device in advance. Through this proprietary word, the voice device in the wake-up waiting state can be activated and enter the voice command recognition waiting state. Users can further issue various voice commands. After receiving the commands, the voice device executes the corresponding command actions and feeds back the result of the command execution through the built-in playback component of the device to prompt the user.

[0003] When the user says a proprietary word, the voice device can be activated. When the user says a non-proprietary word, the voice device remains in the wake-up waiting state to avoid accidental triggering of the voice by the user's normal interpersonal voice communication. However, in actual scenarios, there are still cases where the voice device is awakened and activated by non-proprietary words or even environmental noise. This type of wake-up is called mis-wake-up in the industry, and the probability index generally ranges from 1 to 3 times per 72 hours. Mis-triggering will disturb the user and reduce the user experience.

[0004] With the wide application of voice recognition, more and more household appliances are equipped with voice recognition systems, providing voice human-computer interaction functions independently. When there are multiple voice devices in the same user's home, especially when there are several individual voice devices in the same room, considering the independent mis-wake-up probability of a single voice device, the overall mis-wake-up probability of the voice devices in this room will increase exponentially. Summary of the Invention

[0005] One or more embodiments of this specification describe a wake-up method, device, system, medium, and device based on collaborative error correction.

[0006] In a first aspect, this specification provides a wake-up method based on collaborative error correction. Voice devices are distributed in the entire space of a household. The voice devices include smart home appliances, and the voice modules in the smart home appliances have voice pickup and voice feedback functions; each of the voice devices is communicatively connected to a cloud platform; the method is executed by a decision-making device, and the decision-making device is a smart home appliance pre-selected from the voice devices; the method includes:

[0007] When receiving a wake-up request sent by a voice device, determine whether there are other voice devices in the space group where the voice device is located; among them, a voice device in the wake-up waiting state sends a wake-up request to the decision-making device after hearing the wake-up word, and each voice device in the entire space of a family has the same wake-up word;

[0008] If there are other voice devices, start timing, and after the timing duration reaches the preset duration, determine whether a wake-up request sent by the other voice device is received during the timing process Wake-up request ;

[0009] If a wake-up request sent by another voice device is received during the timing process, determine a response voice device from the voice devices that sent the wake-up request, and send a wake-up response instruction to the response voice device, so that the response voice device enters the wake-up state from the wake-up waiting state.

[0010] In a second aspect, this specification provides a wake-up device based on collaborative error correction. Voice devices are distributed in the entire space of a family, and the voice devices include smart home appliances. The voice module in the smart home appliances has voice pickup and voice feedback functions; each of the voice devices is communicatively connected to a cloud platform;

[0011] The device is installed on a decision-making device, and the decision-making device is a smart home appliance pre-selected from the voice devices; the device includes:

[0012] A first judgment module, configured to determine whether there are other voice devices in the space group where the voice device is located when receiving a wake-up request sent by a voice device; among them, a voice device in the wake-up waiting state sends a wake-up request to the decision-making device after hearing the wake-up word, and each voice device in the entire space of a family has the same wake-up word;

[0013] A second judgment module, configured to start timing if there are other voice devices, and determine whether a wake-up request sent by other voice devices is received during the timing process after the timing duration reaches the preset duration;

[0014] A target determination module, configured to determine a response voice device from the voice devices that sent the wake-up request and send a wake-up response instruction to the response voice device if a wake-up request sent by other voice devices is received during the timing process, so that the response voice device enters the wake-up state from the wake-up waiting state.

[0015] In a third aspect, an embodiment of the present invention provides a wake-up system based on collaborative error correction, including voice devices distributed within a household and a cloud platform communicatively connected to each voice device; the voice devices include smart home appliances, and the voice module in the smart home appliances has a voice pickup and voice feedback function; the cloud platform is used to control the voice devices in the entire space; the decision-making device is a smart home appliance pre-selected from the voice devices, and the decision-making device has the wake-up device based on collaborative error correction provided in the second aspect.

[0016] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method provided in the first aspect are implemented.

[0017] In a fifth aspect, an embodiment of the present invention provides a voice device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the steps of the method provided in the first aspect are implemented.

[0018] In a sixth aspect, an embodiment of the present invention provides a complete set of home appliance control systems, including an Internet of Things platform, a content platform, a semantic platform, a cloud platform, a voice acquisition module, and a smart terminal; among them, the Internet of Things platform is the Zhihuijia AIOT platform, which has a configuration module and a smart control scenario module. The configuration module and the smart control scenario module transmit information to and control the operation of the smart terminal through a communication network; the cloud platform transmits information to and from the semantic platform and the Internet of Things platform through a communication network respectively; the semantic platform includes a voice parsing module, a function feedback module, and an interaction feedback module. The semantic platform can parse the voice signal acquired by the voice acquisition module and output a control instruction according to the parsed voice signal. The control instruction communicates with the cloud platform and the Internet of Things platform respectively through wireless communication to control or drive the operation of the smart terminal.

[0019] The wake-up method, device, system, medium, and equipment based on collaborative error correction provided by the embodiments of this specification have the following beneficial effects: In the embodiments of the present invention, when the decision-making device receives a wake-up request sent by a voice device, the decision-making device will determine whether there are other voice devices in the space group where the voice device sending the wake-up request is located. If there are other voice devices, it will start timing. After the timing ends, it will determine whether a wake-up request from other voice devices is received during the timing process. If a wake-up request from other voice devices is received, it will determine the answering voice device and then send a wake-up response instruction to the answering voice device. In this way, the answering voice device will enter the wake-up state from the wake-up waiting state. This process is imperceptible to the user and will not affect the original working mode of the voice device. Moreover, when there are multiple voice devices in the space group, the method provided by the embodiments of the present invention is also used to eliminate the abnormal state of a single voice device being accidentally triggered, reducing the overall probability of accidental wake-up of all voice devices in the home. For a single wake-up, when the decision-making device receives wake-up requests sent by multiple voice devices, the decision-making device will select the most suitable voice device as the answering voice device, which can bring a high-quality user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of this specification or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are some embodiments of this specification. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0021] Figure 1 is a flowchart of the wake-up method based on collaborative error correction in an embodiment of this specification;

[0022] Figure 2 is a schematic diagram of four voice devices accessing the same network in an embodiment of this specification;

[0023] Figure 3 is a schematic diagram of the room allocation of four voice devices in an embodiment of this specification;

[0024] Figure 4 is a flowchart of the voice wake-up processing method in an embodiment of this specification;

[0025] Figure 5 is a schematic diagram of the distribution of multiple smart home appliances in a home in an embodiment of this specification;

[0026] Figure 6a is a schematic diagram of a voice terminal in an embodiment of this specification;

[0027] Figure 6b It is a schematic diagram of the distribution of multiple voice terminals within a household in an embodiment of this specification;

[0028] Figure 7 It is a structural block diagram of a wake-up device based on collaborative error correction in an embodiment of this specification;

[0029] Figure 8 It is a structural block diagram of a wake-up system based on collaborative error correction in an embodiment of this specification;

[0030] Figure 9 It is an architecture diagram of a complete set of household appliance systems in an embodiment of this specification;

[0031] Figure 10 It is a schematic diagram of the control principle of a voice command system in an embodiment of this specification. Specific embodiments

[0032] The following describes the solutions provided in this specification in conjunction with the accompanying drawings.

[0033] In a first aspect, an embodiment of the present invention provides a wake-up method based on collaborative error correction.

[0034] The applicable scenarios of the solutions provided in the embodiments of the present invention may include but are not limited to the following scenarios:

[0035] There are multiple voice devices distributed within a household, and these voice devices include various smart home appliances. For example, refer to Figure 5 , there is a washing machine, a refrigerator, two wall-mounted air conditioners, a cabinet air conditioner, and two televisions installed in the entire space of a household. These smart home appliances all have voice modules, and the voice modules in the smart home appliances can pick up voice commands issued by users and send the voice commands to the cloud platform. The voice module can also perform voice feedback under the control of the cloud platform, that is, inform users of the execution status of some devices by means of voice broadcast.

[0036] However, since there are no smart home appliances installed in areas such as the balcony, the entrance area, and the bathroom, the voice pickup function in these areas will be relatively poor. Therefore, voice terminals can be installed in these areas. Refer to Figure 6a and 6b , voice terminals are installed in positions such as the entrance area, the balcony, the dining room, the study, the secondary bathroom, the master bathroom, the head of the bed in the master bedroom, and the head of the bed in the children's room. The voice terminal can pick up the voice issued by the user and can also access the home local area network through WIFI. Since the volume of the voice terminal is very small and it is inconvenient to install an amplification module, the voice terminal does not have the function of voice broadcast.

[0037] It is understandable that since it is difficult for the optimal sound pickup ranges of various smart home appliances to cover all corners of a household, there may sometimes be a situation where the voice commands of users cannot be responded to. Therefore, in the embodiments of the present invention, voice terminals are arranged in areas within a household that are not covered by smart home appliances. In this way, the optimal sound pickup ranges of various smart home appliances and various voice terminals can cover the entire space within a household.

[0038] Of course, the voice terminal can also have a function of light prompting. For example, when a voice terminal is awakened as a responding voice device, it can be prompted by means of a breathing light. When a voice terminal is used as a responding voice device and the cloud platform successfully controls the corresponding target smart home appliance to execute an instruction, it can control the voice terminal used as the responding voice device to flash a green light three times for prompting. When the cloud platform fails to control the corresponding target smart home appliance to execute an instruction, it can control the voice terminal used as the responding voice device to flash a yellow light three times for prompting.

[0039] Among them, all voice devices within a household (including various voice terminals and various smart home appliances) are communicatively connected to the cloud platform, so that the cloud platform can control various voice devices within the household. Of course, an application program can be installed on the user's mobile terminal, and through this application program, the control, configuration, etc. of various voice terminals within the household can also be performed.

[0040] Example 1: The user issues the wake-up word "Changhong Xiaobai" on the balcony. The voice terminal on the balcony, the TV in the living room, and the floor-standing air conditioner in the living room will all pick up the wake-up word. Then, which voice device should be awakened? This is the problem to be solved in the embodiments of the present invention. If it is determined that the TV in the living room is awakened, the TV in the living room will issue a response "Yes". Then, the user will issue the voice command "Turn on the air conditioner in the living room". At this time, the TV in the living room will pick up this segment of voice and send this segment of voice to the cloud platform. The cloud platform analyzes this segment of voice and then controls the floor-standing air conditioner in the living room to turn on.

[0041] It can be seen that in the embodiments of the present invention, the awakened voice device is called a responding voice device. The function of this responding voice device is to pick up the voice commands issued by the user and send the voice commands to the cloud platform. In this way, the cloud platform will analyze the voice commands and control the execution device to perform corresponding operations according to the analysis results.

[0042] In summary, in a scenario, voice devices are distributed throughout the entire space of a household. The voice devices include smart home appliances and voice terminals. The union of the optimal sound pickup ranges of each smart home appliance and each voice terminal can cover the entire space. The voice module in the smart home appliance has the functions of voice pickup and voice feedback, and the voice module of the voice terminal has the function of voice pickup. All the voice devices are communicatively connected to a cloud platform for controlling voice devices throughout the space.

[0043] An embodiment of the present invention provides a wake-up method based on collaborative error correction. This method can be executed by a decision-making device, which is a smart home appliance pre-selected from the voice devices. In addition to the above scenario, the scenario applicable to this method can also be a scenario where only smart home appliances are included in a household, and the voice module in the smart home appliance has the functions of voice pickup and voice feedback.

[0044] It can be understood that if a voice terminal is involved in some optional embodiments of the present invention, the scenario applicable to such a specific embodiment is: voice devices are distributed throughout the entire space of a household. The voice devices include smart home appliances and voice terminals. The union of the optimal sound pickup ranges of each smart home appliance and each voice terminal can cover the entire space. The voice module in the smart home appliance has the functions of voice pickup and voice feedback, and the voice module of the voice terminal has the function of voice pickup. All the voice devices are communicatively connected to a cloud platform for controlling voice devices throughout the space. For example, there are at least two smart home appliances and at least two voice terminals in a household.

[0045] It can be understood that if a voice terminal is not involved in some optional embodiments of the present invention, the scenario applicable to such a specific embodiment can be the scenario described in the previous paragraph, or of course, it can also be a scenario where only smart home appliances are included in a household, and the voice module in the smart home appliance has the functions of voice pickup and voice feedback.

[0046] See Figure 1 , the method includes the following steps S10 to S30:

[0047] S10. When receiving a wake-up request sent by a voice device, determine whether there are other voice devices in the space group where the voice device is located;

[0048] Among them, a voice device in a wake-up waiting state sends a wake-up request to the decision-making device after detecting a wake-up word, and each voice device in the entire space of a household has the same wake-up word.

[0049] It is understandable that the method provided by the embodiments of the present invention is executed by a decision-making device, and the decision-making device is an intelligent household appliance among various voice devices in a household. Therefore, before executing the method provided by the embodiments of the present invention, it is necessary to first determine the execution device.

[0050] There are multiple voice devices in a household. For example, intelligent voice refrigerators, voice speakers, voice air conditioners, voice TVs, multiple voice terminals, etc. The wake-up words of each voice device in a household are the same, that is to say, each voice device in a household is trained based on the same wake-up word. Moreover, on a mobile terminal, each voice device in this household can be controlled through an application program, and this application program is the preset application program mentioned below. The determination of the decision-making device can be achieved by this preset application program.

[0051] That is to say, each voice device is controlled by the same preset application program on the user's mobile terminal of the household where it is located; the decision-making device is pre-determined by the preset application program; wherein, the process of the preset application program determining the decision-making device includes the following steps S01 to S04:

[0052] S01. Obtain each voice device accessing the home network;

[0053] It is understandable that when each voice device accesses the home network, relevant information of this voice device, such as device identification, device type, device name, etc., will be seen on the preset application program. Therefore, the preset application program can obtain a device list formed by the relevant information of each voice device accessing the home network.

[0054] S02. Obtain the power-on state data, computing power data, and user usage habit data of each voice device joining the home network;

[0055] It is understandable that after a voice device runs for a period of time, a lot of data will be generated. For example, user usage habit data of the voice device and power-on state data of the voice device, and these data will be recorded in the memory inside the voice device. The preset application program will obtain this data from each voice device. At the same time, the preset application program will also obtain the computing power data of each voice device.

[0056] Among them, the usage habit data of the voice device, for example, the time period of the day when the user is accustomed to watching TV, the season when the user is accustomed to turning on the air conditioner, etc. The power-on state data of the voice device refers to whether the voice device is always powered on, or powered on during a certain time period of the day, or rarely powered on, etc. The computing power data of the voice device refers to the data that can reflect the computing speed and computing volume of the voice device. Since the decision-making device needs to perform wake-up decision-making processing while realizing its own functions, the computing power of the voice device needs to be considered here.

[0057] S03. Determine the decision-making ability score of each voice device according to the power-on state data, computing power data, and user usage habit data of each voice device;

[0058] It can be understood that when calculating the decision-making ability score of each voice device, not only the user's usage habits and power-on state need to be considered, but also the computing power of the voice device needs to be considered, that is, not only the external factors of the voice device are considered, but also the internal factors of the voice device are considered. Finally, a score reflecting the comprehensive decision-making ability of the voice device can be calculated.

[0059] In specific implementation, in S03, the first calculation formula can be specifically used to calculate the decision-making ability score of each voice device, and the first calculation formula includes:

[0060] P1 = u * (d 2 - 1 / y)

[0061] In the formula, P1 is the decision-making ability score; d is the average daily power-on duration of the voice device in the past month; y is the CPU computing power of the voice device; u is the usage mark of the voice device by the household user in the current season; if the household user uses the voice device in the current season, the usage mark corresponding to the voice device is 1; if the household user does not use the voice device in the current season, the usage mark corresponding to the voice device is 0.

[0062] In the above first calculation formula, u is the user's usage habit data. If the household user does not use the voice device in the current season, the usage mark corresponding to the voice device is 0, and at this time P1 is 0. If the household user uses the voice device in the current season, the usage mark corresponding to the voice device is 1, and at this time P1 = d 2 - 1 / y. For example, for the voice air conditioner as a voice device, some families do not use it in winter and only use it in summer.

[0063] Among them, d is the average daily power-on duration of the voice device in the past month. This parameter takes into account the usage of the user in the recent period and also reflects the average power-on situation every day. For example, if the user only turns on the voice TV at night in the past month, the average daily power-on duration of the voice TV in the past month is only a few hours. While the voice refrigerator is always powered on, and at this time, the average daily power-on duration of the voice refrigerator in the past month is 24 hours. The parameter d is a key parameter. The longer the power-on duration of the voice device, the more time the voice device can perform wake-up decision processing, which can reduce the occurrence of missed wake-up requests. The larger d is, the larger P1 is.

[0064] Among them, y reflects the computing power of the voice device, for example, the computing power of the CPU. The computing power of different CPUs is calculated in different ways. For example, some CPUs measure the computing power by word length, and some CPUs measure the computing power by double-precision floating-point computing power, which can be specifically determined according to the actual situation of the CPU. The larger y is, the larger P1 is, but the relationship between y and P1 is not a proportional relationship.

[0065] It can be seen that the above first calculation formula can very reasonably reflect the decision-making ability of a voice device.

[0066] S04. Use the voice device with the highest decision-making ability score as the decision-making device.

[0067] That is to say, in S03, calculate the decision-making ability scores of each voice device accessing the home network, and then select the voice device with the highest decision-making ability score as the decision-making device.

[0068] In specific implementation, after determining a decision-making device in the above manner, the method provided in the embodiment of the present invention can be executed by using this decision-making device. However, if the decision-making device loses power, other voice devices need to be replaced as the decision-making device. Of course, there may also be a situation where the previously determined decision-making device is not powered on, and at this time, other voice devices also need to be replaced as the decision-making device.

[0069] Since the above situations may occur in the actual scenario, the process of the preset application program determining the decision-making device may further include: the preset application program detects whether the current decision-making device is in a powered-on state at a preset time interval through a heartbeat instruction; if it is not in a powered-on state, select the voice device with the highest decision-making ability score among the voice devices in a powered-on state as the current decision-making device.

[0070] That is to say, the preset application on the mobile terminal detects whether the current decision-making device is powered on at time intervals, specifically by means of heartbeat instructions. For example, the preset application on the mobile terminal sends a heartbeat instruction to the decision-making device at regular intervals. If the mobile terminal can receive the feedback information from the decision-making device, it means that the decision-making device is powered on at this time. If the feedback information from the decision-making device cannot be received, it means that the decision-making device is not powered on at this time.

[0071] If the preset application on the mobile terminal detects that the current decision-making device is not powered on and needs to replace other voice devices, it will select the voice device with the highest decision-making ability score among all the powered-on voice devices as the current decision-making device at this time, which can avoid affecting the wake-up process due to the decision-making device not being powered on.

[0072] Of course, in addition to automatically determining the decision-making device, the user can also mark the room where each voice device is located on the preset application, that is, set a corresponding room identifier for each voice device.

[0073] In specific implementation, the above space grouping can be a physical space grouping or a virtual space grouping. When the above space grouping is a physical space grouping, one room corresponds to one space grouping, and one space grouping includes at least one voice device. That is to say, the space grouping is multiple groups obtained by dividing the voice devices in each room in the way of physical space grouping.

[0074] Specifically, each device can be grouped on the preset application. Specifically, the preset application can be used to provide a configuration interface for the user to configure a corresponding room identifier for each voice device accessing the home network. Correspondingly, in S10, determining whether there are other voice devices in the space grouping where the voice device is located includes: obtaining the room identifier corresponding to the voice device from the preset application, and using the room identifier as the first room identifier; determining whether there is a room identifier among the room identifiers of other voice devices that is the same as the first room identifier; if there is, there are other voice devices in the room where the voice device is located; otherwise, there are no other voice devices in the room where the voice device is located.

[0075] It is understandable that the user can set the corresponding room identifier for each voice device on the configuration interface of the preset application, so that the preset application can know which voice devices are in each room. Further, when the decision-making device determines whether there are other voice devices in the space group where the voice device is located, it will obtain the room identifier corresponding to the voice device that sends the wake-up request, that is, the first room identifier, from the preset application of the mobile terminal, and then determine whether there is a room identifier in the room identifiers of other voice devices that is the same as the first room identifier, so as to determine whether there are other voice devices in the room where the voice device that sends the wake-up request is located.

[0076] In an actual scenario, if the user utters a wake-up word, the voice device in a room that is in the wake-up waiting state and monitors this wake-up word will send a wake-up request to the decision-making device. Then, the decision-making device determines whether there are other voice devices in the room where the voice device is located, and then executes subsequent steps.

[0077] It is understandable that if the voice device that sends the wake-up request is the decision-making device itself, the processing method is similar.

[0078] S20. If there are other voice devices, start timing, and after the timing duration reaches the preset duration, determine whether a wake-up request sent by other voice devices is received during the timing process;

[0079] It is understandable that since the wake-up words of each voice device are the same, and if there are other voice devices in a room in addition to the preset device that sends the wake-up request, at this time the decision-making device will start timing, stop timing when the timing duration reaches the preset duration, and it is possible to receive a wake-up request sent by other voice devices during the timing process, or it is also possible not to receive a wake-up request sent by other voice devices.

[0080] S30. If a wake-up request sent by other voice devices is received during the timing process, determine a response voice device from the voice devices that send the wake-up request, and send a wake-up response instruction to the response voice device, so that the response voice device enters the wake-up state from the wake-up waiting state;

[0081] Further, in S30, if a wake-up request sent by other voice devices is received during the timing process, and a response voice device is determined from the voice devices that send the wake-up request, it includes:

[0082] If a wake-up request sent by other voice devices is received during the timing process, obtain the sound pickup parameters of each voice device that sends the wake-up request;

[0083] Determine the response voice device from the voice devices that send the wake-up request according to the sound pickup parameters of each voice device that sends the wake-up request.

[0084] It can be understood that when a wake-up request sent by other voice devices is received, it means that more than one voice device has responded to the voice uttered by the user. At this time, the possibility of false wake-up is relatively small. Therefore, at this time, it is necessary to select a wake-up object with the highest probability among the voice devices that send the wake-up request. If no wake-up request from other voice devices is received during the timing process, the possibility that the voice device that sends the wake-up request is falsely awakened is very high.

[0085] If a wake-up request sent by other voice devices is received during the timing process, at this time, the sound pickup parameters of each voice device that sends the wake-up request can be obtained, and then the next calculation can be carried out using the sound pickup parameters. The so-called sound pickup parameters refer to some parameters when the voice device collects the voice uttered by the user.

[0086] Among them, the sound pickup parameters may include the sound pickup angle and the sound pickup intensity. If the user utters voice towards the voice device to be awakened, the sound pickup angle at this time is 90°. However, if the user does not utter voice towards the voice device to be awakened, the sound pickup angle at this time is not 90°. It can be seen that the sound pickup angle reflects the sound pickup angle of the voice device when the user utters voice. The sound pickup intensity refers to the intensity of the sound picked up by the voice device.

[0087] Based on the above sound pickup parameters, determine the response voice device according to the sound pickup parameters of each voice device that sends the wake-up request. The specific process may include the following steps S41 to S43:

[0088] S41. For each voice device that sends the wake-up request, determine whether the number of voice devices with a sound pickup angle in the range of 60 to 120 degrees is greater than 1;

[0089] Among them, 60 to 120 degrees is a preferred sound pickup angle. When the sound pickup angle is in the range of 60 to 120 degrees, the possibility that the voice device is the response voice device is very high.

[0090] It can be understood that in this step, first determine the number of voice devices with a sound pickup angle in the range of 60 to 120 degrees. If the number is greater than 1, it is necessary to further screen among these voice devices with a sound pickup angle in the range of 60 to 120 degrees.

[0091] S42. If the number of voice devices with a pickup angle in the range of 60 to 120 degrees is greater than 1, then according to the pickup angle and pickup sound intensity of each voice device with a pickup angle in the range of 60 to 120 degrees, calculate the wake-up score corresponding to the voice device; the wake-up score of a voice device is used to represent the probability that the user wakes up the voice device.

[0092] In this step, for each voice device with a pickup angle in the range of 60 to 120 degrees, according to the pickup angle and pickup sound intensity of the voice device, calculate the wake-up score of each voice device, so as to know the probability that the user wants to wake up the voice device.

[0093] Further, in this step S42, the second calculation formula can be used to calculate the wake-up score corresponding to the voice device, and the second calculation formula includes:

[0094] When r is in the first range, P2 = a*s + b / |r - 90|

[0095] When r is in the second range, P2 = b*s - a*|r - 90| + c

[0096] In the formula, P2 is the wake-up score, s is the pickup sound intensity, r is the pickup angle, the first range is: r is greater than or equal to 60 and less than 80, or, r is greater than 100 and less than or equal to 120; the second range is: r is greater than or equal to 80 and less than or equal to 100; a and b are preset weights, and a > b, c is 10*a + b / 10.

[0097] Among them, when r is in the first range, the greater the pickup intensity, the greater P2; the closer the pickup angle is to 90, the greater P2. Therefore, the wake-up score of the voice device with a high pickup intensity and a pickup angle close to 90 is relatively high. Moreover, when r is in the first range, the weight of the pickup sound intensity is a, and a > b, indicating that more attention is paid to the pickup sound intensity at this time. That is to say, in this case, the pickup sound intensity is more important.

[0098] Among them, when r is in the second range, the greater the pickup intensity, the greater P2, and the closer the pickup angle is, the greater P2. Therefore, the wake-up score of the voice device with a high pickup intensity and a pickup angle close to 90 is relatively high. When r is in the second range, the weight of the pickup angle is a, and a > b, indicating that more attention is paid to the pickup angle at this time. That is to say, in this case, the pickup angle is more important.

[0099] Further, when the picked-up sound intensity is the same, the wake-up score when r is in the second range should be greater than the wake-up score when r is in the first range. To ensure this requirement, in the embodiments of the present invention, a parameter c is added to the calculation formula when r is in the second range, and c is set to 10*a + b / 10. This value can ensure that when the picked-up sound intensity is the same, the wake-up score when r is in the second range is greater than the wake-up score when r is in the first range.

[0100] Among them, c = 10*a + b / 10 is calculated to ensure that b*s - a*|r - 90| + c is greater than or equal to a*s + b / |r - 90| when r is 100 and the picked-up sound intensity is 0. Based on this c value, when r is at any value within the second range, it can be ensured that when the picked-up sound intensity is the same, the wake-up score when r is in the second range is greater than the wake-up score when r is in the first range.

[0101] S43. Use the voice device with the highest wake-up score as the answering voice device.

[0102] It can be understood that the higher the wake-up score of a voice device, the greater the probability that the user wants to wake up this voice device. If the wake-up score of a voice device is lower, it means that the probability that the user wants to wake up this voice device is lower. Therefore, here the voice device with the highest wake-up score is used as the answering voice device.

[0103] In specific implementation, according to the respective sound-picking parameters of the voice devices that send wake-up requests, determine the answering voice device. The specific process may further include at least one of the following:

[0104] (1) If the number of voice devices with the sound-picking angle in the range of 60 - 120 degrees is 1, use the voice device with the sound-picking angle in the range of 60 - 120 degrees as the answering voice device;

[0105] It can be understood that if there is only one voice device with the sound-picking angle in the range of 60 - 120 degrees, the probability that this one voice device is the voice device that the user wants to wake up is the greatest. In this case, the picked-up sound intensity of the voice device does not need to be considered. Therefore, at this time, this one voice device is used as the answering voice device.

[0106] (2) If the number of voice devices with the sound-picking angle in the range of 60 - 120 degrees is 0, use the voice device with the highest picked-up sound intensity as the answering voice device.

[0107] It can be understood that if the number of voice devices with the sound-picking angle in the range of 60 - 120 degrees is 0, it means that the user does not send voice to any voice device. At this time, the sound-picking angle does not need to be considered. At this time, the voice device with the highest picked-up sound intensity can be selected from the voice devices that send wake-up requests as the answering voice device.

[0108] It can be seen that the answering voice device can be determined according to the above process in different situations.

[0109] It is understandable that the above are all processing methods when a wake-up request sent by other voice devices is received during the timing process. In practice, if no wake-up request sent by other voice devices is received during the timing process, it can be considered that the voice device sending the wake-up request is mis-triggered. At this time, the decision device can send a termination wake-up response instruction to the voice device that sent the wake-up request. The voice device that receives the termination wake-up response instruction will not enter the wake-up state from the wake-up waiting state, but will continue to maintain the wake-up waiting state.

[0110] It is understandable that the above are all for the situation where there are other voice devices in the space group where the voice device sending the wake-up request is located. When there are no other voice devices in the space group where the voice device sending the wake-up request is located, the method provided by the embodiment of the present invention may further include the following steps: sending a wake-up response instruction to the voice device that sent the wake-up request, so that the voice device enters the wake-up state.

[0111] That is to say, there is only one voice device in a space group, and this voice device sends a wake-up request to the decision device. At this time, the decision device believes that the user just wants to wake up this voice device. Therefore, the decision device will send a wake-up response instruction to this device, and the voice device that receives this instruction will enter the wake-up state from the wake-up waiting state.

[0112] It is understandable that the decision device in the embodiment of the present invention may be in the same room as the voice device that sent the wake-up request, or may not be in the same space group as the voice device that sent the wake-up request.

[0113] It is understandable that each voice device in the embodiment of the present invention is connected to the same network. For example, it is connected to the same home network through Ethernet, Bluetooth, WIFI, etc. In this way, each voice device can communicate with each other and transfer data.

[0114] For example, see Figure 2 , in a family, there are voice device A, voice device B, voice device C, and voice device D. Voice device A and voice device B use WIFI network technology, voice device C uses wired Ethernet technology, and voice device D uses Bluetooth communication technology. They are all connected to the home network device, and voice device A is selected as the decision device.

[0115] The user marks the room identifier of each voice device in the application program of their mobile phone. For example, see Figure 3, use the APP software to assign voice devices A, voice devices B, and voice devices C to room A, assign voice device D to room B, and send the device list and device ID information of room A and room B to voice device A.

[0116] See also Figure 4 When voice device B is triggered by the user's voice and needs to be awakened, it first initiates a wake-up request to the decision-making device, i.e., voice device A, through the voice device B network. The decision-making device detects that there are also voice devices A and C in the room A group of voice device B, and then enters a timing wait. After the timing ends, it determines whether voice devices A and C have also initiated a wake-up request. After determining that only voice device B submitted a wake-up request this time, it is determined to be a false wake-up, and the decision-making device sends a termination response instruction to voice device B.

[0117] During the above judgment process, users will not be aware of it, which effectively reduces the impact of false wake-ups of multi-voice devices.

[0118] It can be seen that the method provided by the embodiment of the present invention is a voice wake-up solution that supports collaborative error correction, which is aimed at the problem that the probability of false triggering increases exponentially when there are multiple voice devices in a home. In this solution, it is necessary to first connect the device to the network, select the decision-making device, and mark the room, and then execute the above steps S1 to S4 provided by the embodiment of the present invention.

[0119] In a specific implementation, the space grouping can also be a virtual space grouping. The space grouping is a virtual space grouping obtained by dividing the space according to the position and orientation of the user who issued the wake-up word in the whole space, and a virtual space grouping includes at least two voice devices; the space grouping is predetermined by the cloud platform. The virtual space grouping is not set by the user on the application, but is divided by the cloud platform according to the user's position and orientation, and does not require user setting.

[0120] Regarding virtual space grouping, the cloud platform determines the space grouping method including:

[0121] a1. Obtaining the position and orientation of the user who issued the wake-up word in the entire space in the historical wake-up task;

[0122] a2. Obtain the sound intensity of the preset wake-up word picked up by each voice device, and select a first voice device from each voice device according to the sound intensity; wherein the sound intensity picked up by each first voice device is higher than the sound intensity picked up by other voice devices, and the maximum difference between the sound intensities picked up by each first voice device is within a preset range;

[0123] a3. Obtain the pickup angles of each of the first voice devices, and select, from the first voice devices, second voice devices whose pickup angles fall within the optimal pickup angle range according to the pickup angles of each of the first voice devices;

[0124] a4. Group the second voice devices spatially with respect to the position and orientation of the user.

[0125] For example, when the user is in the middle position between the living room and the dining room and facing the balcony and says "Changhong Xiaobai", the sound intensities picked up by the 5 voice devices in the entrance, living room and dining room are slightly higher than those picked up by the voice devices in the laundry room and the kitchen, and the sound intensities picked up by the 5 voice devices in the entrance, living room and dining room are much higher than those picked up by the voice devices in the study, bedroom and bathroom. The first voice devices are selected according to the sound intensities picked up by each voice device. Since the user stands in the middle position between the living room and the dining room, the sound intensities picked up by the 5 voice devices in the entrance, living room and dining room are about the same at this time, and these 5 voice devices are the voice devices with the highest sound intensities among all voice devices. Therefore, these 5 voice devices are used as the first voice devices.

[0126] Furthermore, since the user is facing the balcony, 3 voice devices in the living room and the balcony face the user's sound-emitting direction, while the 2 voice devices in the entrance and the dining room face away from the user's sound-emitting direction. Therefore, the pickup angles of the 3 voice devices in the living room and the balcony are within the optimal pickup angle range, while the pickup angles of the 2 voice devices in the entrance and the dining room are not within the optimal pickup angle range. Therefore, the 3 voice devices in the living room and the balcony are used as the second voice devices, and these 3 second voice devices form a virtual spatial group. Of course, if the user is facing the dining room direction, the 2 voice devices in the dining room and the entrance form a virtual spatial group.

[0127] It can be seen that a virtual spatial group can be corresponding to the position and orientation of the user. Actually, when the user is at a certain position, a virtual spatial group can be corresponding. For example, when the user is sitting on the sofa in the living room watching TV, the corresponding virtual spatial group is the virtual spatial group formed by the 3 voice devices in the living room and the balcony.

[0128] In practice, when the user is at different positions and orientations in the entire space of the home, for multiple wake-up requests, multiple virtual spatial groups can be formed, and the multiple virtual spatial groups are saved and can be directly used later.

[0129] An embodiment of the present invention proposes a virtual space grouping. The virtual space grouping is an automatically divided virtual space grouping based on past experience. The virtual space grouping is a device grouping determined according to the user's position and orientation. Each voice device within a virtual space grouping can be located in the same physical space or in multiple physical spaces. For example, when the user stands in the middle of the living room and the dining room and sends a wake-up request towards the balcony, the cabinet air conditioner and the TV in the living room and the voice terminal located on the balcony will be grouped into a virtual space grouping. The virtual space grouping is a virtual space grouping, and there are at least two voice devices in this virtual space grouping. If there is only one voice device, there is no need to divide it into a group.

[0130] It can be understood that when the decision-making device receives a wake-up request sent by a voice device, it determines whether there are other voice devices within the virtual space grouping where the voice device is located. If there are other voice devices, it determines whether it can receive wake-up requests sent by other voice devices within a certain period of time. If it receives wake-up requests sent by other voice devices, it indicates that the possibility of false wake-up at this time is relatively small, and then a voice device is selected from the voice devices that send the wake-up requests as the response voice device. However, if there are other voice devices within the virtual space grouping, but no wake-up requests sent by other voice devices are received within a certain period of time, it indicates that the possibility of false wake-up is relatively large, and at this time, the response voice device is not determined.

[0131] Of course, since there are at least two voice devices within a virtual space grouping, when a voice device within a virtual space grouping sends a wake-up request to the decision-making device, there will be no situation where there are no other voice devices within the virtual space grouping. Only when physically grouping the space will there be a situation where there is only one voice device within a physical grouping space.

[0132] For the virtual space grouping, there may be a situation where a voice device does not have a corresponding virtual space grouping. For example, there is only one voice device in a room. When the cloud platform determines the virtual space grouping, this voice device is not grouped into any virtual space grouping. For the situation where a voice device does not have a corresponding virtual space grouping, when the decision-making device receives a wake-up request sent by this voice device, it will directly use this voice device as the response voice device.

[0133] It is understandable that in order to pick up voice commands issued by users in all corners of a household's entire space, multiple voice terminals are set up so that the union of the optimal sound pickup ranges of the smart home appliances and the voice terminals can cover the entire space. This can avoid or greatly reduce the problem of not responding to user commands due to not picking up user commands. Moreover, each voice terminal is connected to the cloud platform, so that control over each voice terminal in the entire space of a household can be achieved.

[0134] In one embodiment, the method provided by the embodiments of the present invention can be a proximity wake-up method with collaborative error correction. The steps of this proximity wake-up method are the same as the above steps S10 to S30. Further, when determining the answering voice device, specifically, one answering voice device can be determined from the voice devices that send the wake-up request based on the proximity principle. The so-called proximity principle, for example, selecting the voice device closest to the user from the voice devices that send the wake-up request as the answering voice device. Another example is to select the voice device closest to the user from multiple voice devices that meet the requirements of sound intensity and pickup angle range as the answering voice device.

[0135] In one embodiment, when the user issues wake-up words such as "Changhong Xiaobai" or "Haier Xiaohei", the decision-making device can also be directly woken up, and the decision-making device is used to wake up and control other voice devices.

[0136] Of course, each voice device in a household can also have different wake-up words, but the above method is executed for each voice device with the same wake-up word.

[0137] In the above text, the voice terminal does not have a voice feedback function. If a speaker module is integrated into the voice terminal so that the voice terminal has a voice feedback function, at this time, the voice terminal can be regarded as a smart home appliance because generally the voice terminal has the function of a status indicator light, and at this time, the voice terminal can be regarded as a smart home appliance with the function of a status indicator light.

[0138] In the above text, the voice device can be directly communicatively connected to the cloud platform, or can communicate with the cloud platform through a home local area network.

[0139] Of course, in the above text, the pickup angle range of 60 degrees to 120 degrees can also be extended to 50 - 130, 40 - 140, 30 - 150, 20 - 160, 10 - 170, etc.

[0140] The method provided by the embodiments of the present invention. In the embodiments of the present invention, when the decision-making device receives a wake-up request sent by a voice device, the decision-making device will determine whether there are other voice devices in the space group where the voice device sending the wake-up request is located. If there are other voice devices, it will start timing. After the timing ends, it will determine whether a wake-up request from other voice devices is received during the timing process. If a wake-up request from other voice devices is received, it will determine a responding voice device, and then send a wake-up response instruction to the responding voice device. In this way, the responding voice device will enter the wake-up state from the wake-up waiting state. This process is imperceptible to the user and will not affect the original working mode of the voice device. Moreover, when there are multiple voice devices in the space group, the method provided by the embodiments of the present invention is also used to eliminate the abnormal state of a single voice device being accidentally triggered, reducing the overall probability of accidental wake-up of all voice devices in the home. For one wake-up, when the decision-making device receives wake-up requests sent by multiple voice devices, at this time, the decision-making device will select a most suitable voice device as the responding voice device according to the sound pickup parameters, which can bring a high-quality user experience while achieving wake-up nearby.

[0141] In a second aspect, the embodiments of the present invention provide a wake-up device based on collaborative error correction. Voice devices are distributed in the entire space of a home, and the voice devices include smart home appliances. The voice module in the smart home appliances has voice pickup and voice feedback functions; each of the voice devices is communicatively connected to a cloud platform; the device is installed on a decision-making device, and the decision-making device is a smart home appliance pre-selected from the voice devices; see Figure 7 , the device includes:

[0142] A first judgment module, configured to judge whether there are other voice devices in the space group where the voice device is located when receiving a wake-up request sent by a voice device; wherein, a voice device in a wake-up waiting state sends a wake-up request to the decision-making device after detecting a wake-up word, and each voice device in the entire space of a home has the same wake-up word;

[0143] A second judgment module, configured to start timing if there are other voice devices, and judge whether a wake-up request sent by other voice devices is received during the timing process after the timing duration reaches a preset duration;

[0144] A target determination module, configured to determine a responding voice device from the voice devices sending the wake-up request and send a wake-up response instruction to the responding voice device if a wake-up request sent by other voice devices is received during the timing process, so that the responding voice device enters the wake-up state from the wake-up waiting state.

[0145] In one embodiment, the sound pickup parameters include a sound pickup angle and a sound pickup intensity; the target determination module specifically includes:

[0146] A first judgment unit, configured to: for each voice device that sends a wake-up request, judge whether the number of voice devices with a sound pickup angle within the range of 60 to 120 degrees is greater than 1;

[0147] A first calculation unit, configured to: if the number of voice devices with a sound pickup angle within the range of 60 to 120 degrees is greater than 1, calculate the wake-up score corresponding to the voice device according to the sound pickup angle and the sound pickup intensity of each voice device with a sound pickup angle within the range of 60 to 120 degrees; the wake-up score of a voice device is used to represent the probability that the user wakes up the voice device; the voice device with the highest wake-up score is used as the response voice device;

[0148] A first determination unit, configured to: if the number of voice devices with a sound pickup angle within the range of 60 to 120 degrees is 1, use the voice device with a sound pickup angle within the range of 60 to 120 degrees as the response voice device; if the number of voice devices with a sound pickup angle within the range of 60 to 120 degrees is 0, use the voice device with the highest sound pickup intensity as the response voice device.

[0149] In one embodiment, the first calculation unit is specifically configured to: calculate the wake-up score corresponding to the voice device by using a second calculation formula, and the second calculation formula includes:

[0150] When r is within the first range, P2 = a*s + b / |r - 90|

[0151] When r is within the second range, P2 = b*s - a*|r - 90| + c

[0152] In the formula, P2 is the wake-up score; s is the sound pickup intensity; r is the sound pickup angle; the first range is: r is greater than or equal to 60 and less than 80, or r is greater than 100 and less than or equal to 120; the second range is: r is greater than or equal to 80 and less than or equal to 100; a and b are preset weights, and a > b, and c is 10*a + b / 10.

[0153] It can be understood that the device provided in the second aspect corresponds to the method provided in the first aspect. For the explanations, examples, beneficial effects, etc. of the content in this aspect, reference can be made to the relevant content in the first aspect, which will not be elaborated here.

[0154] In a third aspect, an embodiment of the present invention provides a wake-up system based on collaborative error correction. See Figure 8, including voice devices distributed and set within a household and a cloud platform communicatively connected to each voice device; the voice devices include smart home appliances, and the voice module in the smart home appliances has voice pickup and voice feedback functions; the cloud platform is used to control the voice devices in the entire space; the decision-making device is a smart home appliance pre-selected from the voice devices, and the decision-making device has the wake-up device based on collaborative error correction provided in the second aspect.

[0155] It can be understood that this wake-up system includes each voice device and the cloud platform. Of course, it can also include an application program on a mobile terminal, and this application program can control each voice device. Figure 8 In which N is a positive integer greater than 2.

[0156] It can be understood that for the explanations, examples, beneficial effects, etc. of the relevant content in the system provided in the third aspect, reference can be made to the relevant content in the first aspect and the second aspect, and details will not be elaborated here.

[0157] Fourthly, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method provided in the first aspect are implemented.

[0158] Specifically, a system or device equipped with a storage medium can be provided. On this storage medium, software program codes for implementing the functions of any one of the above-mentioned embodiments are stored, and the computer (or CPU or MPU) of this system or device reads and executes the program codes stored in the storage medium.

[0159] In this case, the program code read from the storage medium itself can implement the functions of any one of the above-mentioned embodiments. Therefore, the program code and the storage medium storing the program code constitute a part of the present invention.

[0160] Embodiments of the storage medium for providing program codes include floppy disks, hard disks, magneto-optical disks, optical disks (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), magnetic tapes, non-volatile memory cards, and ROMs. Optionally, the program code can be downloaded from a server computer through a communication network.

[0161] In addition, it should be clear that not only can the actual operations be completed in part or in whole by executing the program code read by the computer, but also by means of instructions based on the program code to make the operating system, etc. operating on the computer, so as to implement the functions of any one of the above-mentioned embodiments.

[0162] In addition, it can be understood that the program code read from the storage medium is written into the memory provided in the expansion board inserted into the computer or into the memory provided in the expansion module connected to the computer. Subsequently, based on the instructions of the program code, the CPU or the like installed on the expansion board or the expansion module is made to execute part or all of the actual operations, thereby implementing the functions of any one of the above embodiments.

[0163] It can be understood that for the explanations, examples, beneficial effects, etc. of the content in the medium provided in the fourth aspect, reference can be made to the relevant content in the first aspect and the second aspect, which will not be elaborated here.

[0164] In a fifth aspect, an embodiment of the present invention provides a voice device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the steps of the method provided in the first aspect are implemented.

[0165] It can be understood that this voice device is the decision-making device.

[0166] It can be understood that for the explanations, examples, beneficial effects, etc. of the content in the voice device provided in the fifth aspect, reference can be made to the relevant content in the first aspect and the second aspect, which will not be elaborated here.

[0167] In a sixth aspect, to improve the intelligent linkage operation of smart home appliances, an embodiment of the present invention further provides a complete set of home appliance control systems. This complete set of home appliance control systems includes an Internet of Things platform 1, a content platform 2, a semantic platform 3, a cloud platform 4, a voice acquisition module 7, and a smart terminal 5. The Internet of Things platform 1 can be the Zhihuijia AIoT platform, and this Internet of Things platform has a configuration module 11 and an intelligent control scenario module 12. Among them, the configuration module 11 and the intelligent control scenario module 12 perform information transmission with the smart terminal 5 through a communication network and control the operation of the smart terminal 5; the cloud platform 4 performs information transmission with the semantic platform 3 and the Internet of Things platform 1 respectively through a communication network; the mobile terminal 6 realizes information transmission with the cloud platform 1 wirelessly; the semantic platform 3 includes a voice parsing module 33, a function feedback module 32, and an interaction feedback module 31; the semantic platform 3 can parse the voice signal acquired by the voice acquisition module 7 and output a control instruction according to the parsed voice signal. This control instruction communicates with the cloud platform and the Internet of Things platform through wireless communication to further control or drive the normal operation of the smart terminal.

[0168] Further, to achieve remote control and mobile terminal control, this complete set of home appliance control systems includes a mobile terminal 6 and a voice broadcast module 8. The mobile terminal can communicate with the Internet of Things platform through an APP, or the mobile terminal 6 communicates with the Zhihuijia AIoT platform to control the normal operation of the smart terminal; the voice broadcast module 8 completes voice broadcast.

[0169] Furthermore, the voice command system of the above-mentioned complete set of household appliances system includes a voice pickup module 71, a voice command parsing module 34, a logic judgment module 9, a designated device module 91, a non-designated device module 92, an AI hierarchical management module 311, an AI intelligent grouping module 312, a voice broadcast module 81, an intelligent device instruction execution module 10, and a feedback instruction execution result module 11; when the voice pickup module obtains a user's voice command, it transmits the voice command to the voice command parsing module 33, and then drives the control of the intelligent terminal to run through the judgment result of the logic judgment module 9. The intelligent terminal feeds back and broadcasts the execution result of the voice command; among them, the AI hierarchical management module 311 can perform hierarchical management, and the AI intelligent grouping module 312 can achieve intelligent grouping; among them, the designated device module 91 can perform designated management and control on specific intelligent terminals, and the non-designated device module 92 can perform unified management and control on non-specific intelligent terminals. Among them, the designated devices can be defined by spatial location (such as room location) and device name (such as air conditioner or refrigerator). The logic judgment module can control the intelligent terminal according to the priority; the AI hierarchical management module can classify and manage the functions of air conditioners, refrigerators, washing machines, and TVs according to the functions of the intelligent terminal. This classification can be divided into general skills, exclusive skills, and public skills; the AI intelligent grouping module can group and manage the intelligent terminals (such as air conditioners, refrigerators, washing machines, and TVs) within the spatial area according to the spatial area (living room, bedroom, study, kitchen, balcony).

[0170] It can be understood that the voice pickup module 71 and the voice acquisition module can be one module, and of course, they can also not be the same module. The voice command parsing module 34 and the voice parsing module 33 can be the same module, and of course, they can also not be the same module. The voice playback module 8 and the voice broadcast module 81 can be one module, and of course, they can also not be the same module.

[0171] Each embodiment in this specification is described in a progressive manner. The same or similar parts among the embodiments can be referred to each other, and the key points of each embodiment are the differences from other embodiments. In particular, for the device embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiments.

[0172] Those skilled in the art should be able to realize that in one or more of the above examples, the functions described in the present invention can be implemented by hardware, software, a plug-in, or any combination thereof. When implemented using software, these functions can be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium.

[0173] The specific embodiments described above further elaborate on the object, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only for the specific embodiments of the present invention and is not intended to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made on the basis of the technical solution of the present invention shall be included within the protection scope of the present invention.

Claims

1. A wake-up method based on collaborative error correction, characterized in that Voice devices are distributed throughout the entire space of a household. The voice devices include smart home appliances, and the voice modules in the smart home appliances have voice pickup and voice feedback functions. All the voice devices are communicatively connected to a cloud platform. The method is executed by a decision-making device, which is a smart home appliance pre-selected from the voice devices. The method includes: When receiving a wake-up request sent by a voice device, determining whether there are other voice devices in the space group where the voice device is located. Among them, a voice device in the wake-up waiting state sends a wake-up request to the decision-making device after detecting a wake-up word, and all the voice devices in the entire space of a household have the same wake-up word. If there are other voice devices, start timing, and after the timing duration reaches a preset duration, determine whether a wake-up request sent by the other voice devices is received during the timing process. If a wake-up request sent by other voice devices is received during the timing process, determine a response voice device from the voice devices that sent the wake-up request, and send a wake-up response instruction to the response voice device to enable the response voice device to enter the wake-up state from the wake-up waiting state. Each voice device is controlled by the same preset application on the user's mobile terminal in the household. The decision-making device is pre-determined by the preset application. Among them, the process of the preset application determining the decision-making device includes: Obtain each voice device accessing the home network. Obtain the power-on status data, computing power data, and user usage habit data of each voice device joining the home network. Determine the decision-making ability score of each voice device according to the power-on status data, computing power data, and user usage habit data of each voice device. Use the voice device with the highest decision-making ability score as the decision-making device. The determining the decision-making ability score of each voice device according to the power-on status data, computing power data, and user usage habit data of each voice device includes: calculating the decision-making ability score of each voice device using a first calculation formula, and the first calculation formula includes: P1 = u * (d 2 - 1 / y) In the formula, P1 is the decision-making ability score; d is the average daily power-on duration of the voice device in the past month; y is the CPU computing power of the voice device; u is the usage mark of the voice device by the household user in the current season. If the household user uses the voice device in the current season, the usage mark corresponding to the voice device is 1; if the household user does not use the voice device in the current season, the usage mark corresponding to the voice device is 0.

2. The method according to claim 1, wherein The wake-up method based on collaborative error correction is a collaborative error correction and proximity wake-up method. The determining a response voice device from the voice devices that sent the wake-up request if a wake-up request sent by other voice devices is received during the timing process includes: determining a response voice device from the voice devices that sent the wake-up request based on the proximity principle.

3. The method according to claim 1, characterized in that, The determining a response voice device from the voice devices that sent the wake-up request if a wake-up request sent by other voice devices is received during the timing process includes: If a wake-up request sent by another voice device is received during the timing process, obtain the pickup parameters of each voice device that sends the wake-up request. According to the respective pickup parameters of the voice devices that send the wake-up request, determine the response voice device from the voice devices that send the wake-up request.

4. The method according to claim 1, characterized in that, The process of the preset application determining the decision device further includes: the preset application detects whether the current decision device is in a powered state at a preset time interval by means of a heartbeat instruction; if it is not in a powered state, select a voice device with the highest decision-making ability score among the voice devices in the powered state as the current decision device.

5. The method according to claim 1, wherein The spatial grouping is a plurality of groups obtained by dividing the voice devices in each room in the manner of physical space grouping. One room corresponds to one spatial grouping, and one spatial grouping includes at least one voice device. Correspondingly, the preset application is used to provide a configuration interface for the user to configure the corresponding room identifier for each voice device accessing the home network; correspondingly, determining whether there are other voice devices in the spatial grouping where the voice device is located includes: obtaining the room identifier corresponding to the voice device from the preset application, and using the room identifier as the first room identifier; determining whether there is a room identifier in the room identifiers of other voice devices that is the same as the first room identifier; if so, there are other voice devices in the room where the voice device is located; otherwise, there are no other voice devices in the room where the voice device is located.

6. The method according to claim 3, wherein The pickup parameters include the pickup angle and the pickup sound intensity; correspondingly, determining the response voice device according to the respective pickup parameters of the voice devices that send the wake-up request includes: For each voice device that sends the wake-up request, determine whether the number of voice devices with a pickup angle in the range of 60 to 120 degrees is greater than 1. If the number of voice devices with a pickup angle in the range of 60 to 120 degrees is greater than 1, calculate the wake-up score corresponding to the voice device according to the pickup angle and the pickup sound intensity of each voice device with a pickup angle in the range of 60 to 120 degrees; the wake-up score of a voice device is used to represent the probability that the user wakes up the voice device; select the voice device with the highest wake-up score as the response voice device. If the number of voice devices with a pickup angle in the range of 60 to 120 degrees is 1, use the voice device with a pickup angle in the range of 60 to 120 degrees as the response voice device. If the number of voice devices with a pickup angle in the range of 60 to 120 degrees is 0, use the voice device with the highest pickup sound intensity as the response voice device.

7. The method according to claim 6, wherein Calculating the wake-up score corresponding to the voice device includes: calculating the wake-up score corresponding to the voice device using a second calculation formula, and the second calculation formula includes: When r is in the first range, P2 = a*s + b / |r - 90| When r is in the second range, P2 = b*s - a*|r - 90| + c Wherein, P2 is the wake-up score; s is the picked-up sound intensity; r is the picked-up sound angle; the first range is: r is greater than or equal to 60 and less than 80, or, r is greater than 100 and less than or equal to 120; the second range is: r is greater than or equal to 80 and less than or equal to 100; a and b are preset weights, and a is greater than b, and c is 10*a + b / 10.

8. The method according to any one of claims 1 to 7, characterized in that, The method further includes at least one of the following: If there is no other voice device in the space group where the voice device sending the wake-up request is located, send a wake-up response instruction to the voice device sending the wake-up request, so that the voice device enters the wake-up state; If no wake-up request sent by other voice devices is received during the timing process, determine that the voice device sending the wake-up request is a false trigger, and send a termination wake-up response instruction to the voice device sending the wake-up request, so that the voice device sending the wake-up request continues to remain in the wake-up waiting state.

9. The method according to claim 1, characterized in that, The space group is a virtual space group obtained by dividing according to the position and orientation of the user who utters the wake-up word in the entire space, and at least two voice devices are included in one virtual space group; The space group is pre-determined by the cloud platform; Wherein, the method for the cloud platform to determine the space group includes: the cloud platform obtains the position and orientation of the user who utters the wake-up word in the entire space in the historical wake-up task; obtains the sound intensity of the preset wake-up word picked up by each voice device, and selects the first voice device from each voice device according to the sound intensity; wherein, the sound intensity picked up by each of the first voice devices is higher than the sound intensity picked up by other voice devices, and the maximum difference between the sound intensities picked up by each of the first voice devices is within a preset range; obtains the picked-up sound angle of each of the first voice devices, and selects the second voice device whose picked-up sound angle falls within the optimal picked-up sound angle range from the first voice devices; forms the space group for the position and the orientation of the user by each of the second voice devices.

10. A wake-up device based on collaborative error correction for performing the method according to any one of claims 1 to 9, characterized in that, Voice devices are distributed in the entire space of a family, and the voice devices include smart home appliances, and the voice module in the smart home appliances has voice pickup and voice feedback functions; each of the voice devices is communicatively connected to the cloud platform; The device is installed on a decision-making device, and the decision-making device is a smart home appliance pre-selected from the voice devices; the device includes: A first judgment module, configured to judge whether there is any other voice device in the space group where the voice device is located when receiving a wake-up request sent by a voice device; wherein, a voice device in the wake-up waiting state sends a wake-up request to the decision-making device after detecting the wake-up word, and each voice device in the entire space of a family has the same wake-up word; A second judgment module, configured to start timing if there is any other voice device, and judge whether a wake-up request sent by any other voice device is received during the timing process after the timing duration reaches a preset duration; A target determination module, configured to, if a wake-up request sent by another voice device is received during the timing process, determine a response voice device from the voice devices that send the wake-up request, and send a wake-up response instruction to the response voice device, so that the response voice device enters the wake-up state from the wake-up waiting state.

11. A wake-up system based on collaborative error correction, characterized in that, It includes voice devices distributed in a household and a cloud platform communicatively connected to each voice device; the voice devices include smart home appliances, and the voice modules in the smart home appliances have voice pickup and voice feedback functions; the cloud platform is used to control the voice devices in the entire space; The decision-making device is a smart home appliance pre-selected from the voice devices, and the decision-making device is equipped with the wake-up device based on collaborative error correction described in claim 10.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-9.

13. A voice device, characterized in that, It includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the method according to any one of claims 1-9.

14. A complete household appliance control system for performing the method according to any one of claims 1 to 9, characterized in that, It includes an Internet of Things platform (1), a content platform (2), a semantic platform (3), a cloud platform (4), a voice acquisition module (7), and a smart terminal (5); among them, the Internet of Things platform (1) is the Zhihuijia AIoT platform, and the Internet of Things platform has a configuration module (11) and a smart control scenario module (12). The configuration module (11) and the smart control scenario module (12) transmit information with the smart terminal (5) through a communication network and control the operation of the smart terminal (5); the cloud platform (4) transmits information with the semantic platform (3) and the Internet of Things platform (1) through a communication network respectively; the semantic platform (3) includes a voice parsing module (33), a function feedback module (32), and an interaction feedback module (31). The semantic platform (3) can parse the voice signal acquired by the voice acquisition module (7), and output a control instruction according to the parsed voice signal. The control instruction communicates with the cloud platform (4) and the Internet of Things platform (1) respectively through wireless communication to control or drive the operation of the smart terminal (5).

15. The home appliance control system according to claim 14, characterized in that, It further includes a mobile terminal (6) and a voice broadcast module (8). The mobile terminal (6) communicates with the Internet of Things platform (1) through an APP, or the mobile terminal (6) communicates with the Zhihuijia AIoT platform to control the operation of the smart terminal (5), and the voice broadcast module (8) is used for voice broadcast.

Images