Home appliance wake-up method, home appliance, and storage medium

Abstract

This application provides a method for waking up home appliances, a home appliance, and a storage medium, relating to the field of home appliance technology. When the main home appliance detects a voice signal carrying a wake-up keyword, it extracts the signal features of the voice signal; based on the signal features, it determines a first wake-up distance between the sound source emitting the voice signal and the main home appliance, and determines the angle between the propagation direction of the voice signal and the orientation of the home appliance; based on the first wake-up distance and the angle, it determines a wake-up evaluation value for the main home appliance, and receives wake-up evaluation values ​​from at least one slave home appliance; it identifies the home appliance with the highest wake-up evaluation value and controls the home appliance with the highest wake-up evaluation value to perform the wake-up operation, enabling the home appliance to be woken up accurately and uniquely, and meeting the user's wake-up needs.

Description

Technical Field

[0001] This application relates to the field of home appliance technology, and in particular to a method for waking up a home appliance, the home appliance itself, and a storage medium. Background Technology

[0002] With the development of technology, home appliances are becoming increasingly intelligent. For example, home appliances can now be woken up by voice, and after being woken up, voice control can be used to perform certain operations (such as turning off the appliance, connecting to the network, etc.).

[0003] Typically, a household includes multiple home appliances. When a user tries to wake up an appliance using voice commands, multiple appliances may simultaneously respond with the wake-up phrase, which does not meet the user's needs. For example, a user standing in the hallway might try to wake up the living room air conditioner using voice commands, but if the air conditioners in the bedroom and living room respond with the wake-up phrases simultaneously, the wake-up reliability is low, and it negatively impacts the user's experience with the appliances. Summary of the Invention

[0004] This application provides a method for waking up home appliances, a home appliance, and a storage medium to solve the problem that the reliability of waking up home appliances in the prior art is low and affects the user's experience of using the home appliances.

[0005] In a first aspect, this application provides a method for waking up a home appliance, applied to a master home appliance, which is communicatively connected to at least one slave home appliance. The method provided by this application includes:

[0006] When the main home appliance detects a voice signal carrying a wake-up keyword, it extracts the signal features of the voice signal.

[0007] Based on the signal characteristics of the voice signal, determine the first wake-up distance of the sound source emitting the voice signal relative to the main home appliance, and determine the angle between the propagation direction of the voice signal and the orientation of the home appliance.

[0008] The wake-up evaluation value of the main home appliance is determined based on the first wake-up distance and the angle between the two surfaces. The wake-up evaluation value is negatively correlated with the first wake-up distance and the angle between the two surfaces.

[0009] Receive wake-up evaluation values ​​from at least one home appliance;

[0010] Identify the home appliance with the highest wake-up evaluation value and control the home appliance with the highest wake-up evaluation value to perform the wake-up operation, wherein the home appliance with the highest wake-up evaluation value is the master home appliance or one of the slave home appliances.

[0011] In some implementations, the signal characteristics are spectral characteristics. Based on the signal characteristics of the speech signal, determining the angle between the propagation direction of the speech signal and the orientation of the household appliance includes:

[0012] Based on the spectral characteristics of the speech signal, the direct sound signal and the reverberant sound signal in the speech signal are separated.

[0013] Determine the energy of the direct sound signal and the high-frequency attenuation value of the direct sound signal energy, respectively;

[0014] Based on the energy of the direct sound signal and its high-frequency attenuation value, the angle between the direction of the speech signal propagation and the orientation of the household appliance is determined. The energy of the direct sound signal is negatively correlated with the angle between the direct sound signal and the orientation, while the high-frequency attenuation value of the direct sound signal is positively correlated with the angle between the direct sound signal and the orientation.

[0015] In some implementations, the signal characteristics are spectral characteristics. Determining the first wake-up distance of the sound source emitting the voice signal relative to the main home appliance based on the signal characteristics of the voice signal includes:

[0016] Based on the spectral characteristics of the speech signal, the direct sound signal and the reverberant sound signal in the speech signal are separated.

[0017] Determine the energy of the direct sound signal and the energy of the reverberant sound signal respectively;

[0018] Based on the ratio of the energy of the direct sound signal to the energy of the reverberant sound signal and a preset mapping relationship, the first wake-up distance of the sound source emitting the voice signal relative to the main home appliance is determined, wherein the ratio of the energy of the direct sound signal to the energy of the reverberant sound signal is positively correlated with the first wake-up distance.

[0019] In some embodiments, the method provided in this application further includes:

[0020] When the main home appliance detects a voice signal carrying a wake-up keyword, it collects CSI data of the Wi-Fi signal.

[0021] CSI data is input into a pre-trained human position awareness model to determine the second wake-up distance of the sound source emitting the voice signal relative to the main home appliance. The human position awareness model is trained based on multiple training samples input into the neural network. Each training sample includes historical CSI data and the corresponding historical wake-up distance.

[0022] According to the formula D3=K1D1+K2D2, the first wake-up distance of the sound source emitting the voice signal relative to the main home appliance is corrected. Where D1 is the first wake-up distance before correction, D2 is the second wake-up distance, D3 is the first wake-up distance after correction, K1 is the first weighting coefficient, and K2 is the second weighting coefficient.

[0023] In some implementations, the main home appliance collects CSI data of the Wi-Fi signal when it detects a voice signal carrying a wake-up keyword.

[0024] CSI data is input into a pre-trained human position awareness model to determine the direction and distance of the sound source emitting the voice signal relative to the main home appliance. The human position awareness model is trained based on multiple training samples input into a neural network. Each training sample includes historical CSI data and the corresponding historical wake-up distance and direction.

[0025] Based on the direction and distance of the sound source emitting the voice signal relative to the main home appliance, the preset three-dimensional space model of the house, and the position of the main home appliance in the three-dimensional space model of the house, the system identifies whether the sound source emitting the voice signal and the main home appliance are located in the same room and obtains the identification result.

[0026] Based on the first wake-up distance and the angle between the two surfaces, determine the wake-up evaluation value of the main home appliance, including:

[0027] Based on the first wake-up distance, the angle between the faces, and the recognition result, the wake-up evaluation value of the main home appliance is determined. When the recognition result indicates that the sound source emitting the voice signal is located in the same room as the main home appliance, the recognition result is positively correlated with the wake-up evaluation value. When the recognition result indicates that the sound source emitting the voice signal is located in the same room as the main home appliance, the recognition result is negatively correlated with the wake-up evaluation value.

[0028] In some embodiments, before the main home appliance extracts the signal features of the voice signal upon detecting a voice signal carrying a wake-up keyword, the method provided in this application further includes:

[0029] The target home appliance receives IP addresses from other home appliances;

[0030] The target home appliance sorts its own IP address according to the size relationship between its own IP address and the IP addresses of other home appliances.

[0031] If the target home appliance has the largest or smallest IP address, the target home appliance will identify itself as the master home appliance and other home appliances as slave home appliances.

[0032] Secondly, this application also provides a method for waking up a home appliance, applied to a slave home appliance, wherein the slave home appliance is communicatively connected to a master home appliance. The method provided in this application further includes:

[0033] When a home appliance detects a voice signal carrying a wake-up keyword, the signal features of the voice signal are extracted.

[0034] Based on the signal characteristics of the speech signal, determine the first wake-up distance of the sound source emitting the speech signal relative to the home appliance, and determine the angle between the propagation direction of the speech signal and the orientation of the home appliance.

[0035] The wake-up evaluation value of the home appliance is determined based on the first wake-up distance and the angle between the two surfaces, wherein the wake-up evaluation value is negatively correlated with the first wake-up distance and the angle between the two surfaces.

[0036] Send wake-up assessment values ​​from the home appliances to the main home appliance;

[0037] Receive a wake-up notification sent by the master appliance when the wake-up evaluation value of the slave appliance is the highest, and perform a wake-up operation in response to the wake-up notification.

[0038] Thirdly, this application also provides a household appliance, including:

[0039] System-on-a-chip (SoC);

[0040] Memory used to store executable instructions for system-on-a-chip;

[0041] The system-on-a-chip is configured to execute instructions to implement the home appliance wake-up method provided in the first aspect of this application.

[0042] In some implementations, the system-on-a-chip integrates a WiFi chip, a display and interaction module, a voice recognition module, and a main processing module, wherein...

[0043] The WiFi chip is used to collect CSI data of WiFi signals and input the CSI data into a pre-trained human position awareness model to determine the second wake-up distance of the sound source emitting the voice signal relative to the main home appliance.

[0044] The speech recognition module is used to extract the signal features of the speech signal, and based on the signal features of the speech signal, to determine the first wake-up distance of the sound source emitting the speech signal relative to the main home appliance, and to determine the angle between the propagation direction of the speech signal and the orientation of the home appliance.

[0045] The main processing module determines the wake-up evaluation value of the main home appliance based on the first wake-up distance and the angle between the two sides; and determines the home appliance with the largest wake-up evaluation value based on at least one wake-up evaluation value sent from the home appliance, and controls the home appliance with the largest wake-up evaluation value to perform the wake-up operation.

[0046] The display interaction module is used to process the data to be displayed by home appliances and transmit it to the display panel of the home appliances for display.

[0047] Fourthly, this application also provides a storage medium that, when the instructions in the storage medium are executed by the system-on-a-chip of a home appliance, enables the home appliance to execute the home appliance wake-up method provided in the first aspect of this application.

[0048] This application provides a method for waking up a home appliance, the home appliance itself, and a storage medium. When the main home appliance detects a voice signal carrying a wake-up keyword, it extracts the signal features of the voice signal. Based on these features, it determines a first wake-up distance between the source of the voice signal and the main home appliance, and determines the angle between the propagation direction of the voice signal and the orientation of the home appliance. Based on the first wake-up distance and the angle, it determines a wake-up evaluation value for the main home appliance and receives wake-up evaluation values ​​from at least one slave home appliance. It then identifies the home appliance with the highest wake-up evaluation value and controls it to perform the wake-up operation. Understandably, the closer the user is to the home appliance and the smaller the angle between the propagation direction of the voice signal and the orientation of the home appliance, the higher the probability that the user will wake up the appliance. Since the wake-up evaluation value is negatively correlated with the first wake-up distance and the angle, controlling the home appliance with the highest wake-up evaluation value to perform the wake-up operation ensures that the home appliance can be woken up precisely and uniquely, meeting the user's wake-up needs. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 A schematic diagram illustrating the interaction between the master home appliance and the slave home appliance provided in this application embodiment;

[0051] Figure 2 This is one of the flowcharts for a home appliance wake-up method provided in the embodiments of this application;

[0052] Figure 3 A flowchart for determining the first wake-up distance provided in an embodiment of this application;

[0053] Figure 4 A flowchart for determining the included angle of a plane provided in an embodiment of this application;

[0054] Figure 5 The second flowchart of the home appliance wake-up method provided in the embodiments of this application;

[0055] Figure 6This is one of the functional module block diagrams of the home appliance wake-up device provided in the embodiments of this application;

[0056] Figure 7 This is the second functional module block diagram of the home appliance wake-up device provided in the embodiments of this application;

[0057] Figure 8 A circuit connection block diagram of a home appliance provided in an embodiment of this application. Detailed Implementation

[0058] Embodiments of the present disclosure will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the disclosure. Furthermore, descriptions of well-known structures and technologies are omitted in the following description to avoid unnecessarily obscuring the concepts of the present disclosure.

[0059] The accompanying drawings illustrate various structural schematics according to embodiments of the present disclosure. These drawings are not to scale, and some details have been enlarged for clarity, and some details may have been omitted. The shapes of the various regions and layers shown in the drawings, as well as their relative sizes and positional relationships, are merely exemplary and may deviate from reality due to manufacturing tolerances or technical limitations. Furthermore, those skilled in the art can design regions / layers with different shapes, sizes, and relative positions as needed.

[0060] In the context of this disclosure, when a layer / element is referred to as being "above" another layer / element, the layer / element may be directly above the other layer / element, or there may be an intermediate layer / element between them. Additionally, if a layer / element is "above" another layer / element in one orientation, then when the orientation is reversed, the layer / element may be "below" the other layer / element.

[0061] Explanation of technical terms in this application:

[0062] Channel state information (CSI) data describes how a signal propagates through a channel from the transmitter to the receiver. It characterizes a combination of factors, such as scattering, fading, and energy attenuation with distance. Therefore, when people move around at home, it affects signal propagation; the signal is scattered by the person, and these physical phenomena are recorded in the CSI data.

[0063] Direct sound: Direct sound is a sound wave that travels directly from its source to the receiver without any reflection. In indoor environments, direct sound is one of the main paths of sound propagation.

[0064] Reverberation: Reverberation is the sound wave that reaches the receiving point after being reflected once or multiple times on various surfaces in a room (such as walls, ceilings, floors, etc.).

[0065] The technical solutions of this application and how they solve the aforementioned technical problems will be described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0066] This application provides a method for waking up a home appliance, applied to the main home appliance 101. For example... Figure 1 As shown, the master appliance 101 is communicatively connected to at least one slave appliance 102. The master appliance 101 and slave appliance 102 can be, but are not limited to, air conditioners, air handling units, refrigerators, etc. It should be noted that the master appliance 101 and slave appliance 102 can be determined as follows: the target appliance receives IP addresses from other appliances; the target appliance sorts its own IP address relative to the IP addresses of the other appliances; if the target appliance's IP address is the largest or smallest, the target appliance determines itself as the master appliance 101 and the other appliances as slave appliances 102. Other appliances determine themselves as slave appliances 102 if they detect that the target appliance's IP address is the largest or smallest. Figure 2 As shown, the method provided in this application embodiment includes S201-S205.

[0067] S201: When the main home appliance 101 detects a voice signal carrying a wake-up keyword, it extracts the signal features of the voice signal.

[0068] The wake-up keyword can be, but is not limited to, "Hello, Xiaomei". Methods for extracting signal features include, but are not limited to, first using filters or spectral subtraction to denoise the monitored speech signal to eliminate background and environmental noise. The denoised speech signal is then segmented into multiple short speech frames, and a Hamming window or rectangular window is applied to each frame to reduce discontinuities at frame edges. Each windowed speech frame undergoes a short-time Fourier transform to extract the spectral features of the speech signal; alternatively, spectral features can be extracted based on Mel-frequency cepstral coefficients.

[0069] S202: Based on the signal characteristics of the voice signal, determine the first wake-up distance of the sound source emitting the voice signal relative to the main home appliance 101, and determine the angle between the propagation direction of the voice signal and the orientation of the home appliance.

[0070] Specifically, such as Figure 3 As shown, determining the first wake-up distance can be specifically implemented as follows:

[0071] S301: Based on the spectral characteristics of the speech signal, separate the direct sound signal and the reverberation signal in the speech signal.

[0072] Specifically, the spectral features of the speech signal can be input into a pre-trained signal separation model to separate the direct sound signal and the reverberant sound signal in the speech signal. The signal separation model is trained by inputting multiple training samples into a neural network or hidden Markov network. Each training sample can include historical spectral features and the corresponding direct sound signal and reverberant sound signal. Alternatively, blind source separation techniques can also be used to separate the direct sound signal and the reverberant sound signal in the speech signal.

[0073] S302: Determine the energy of the direct sound signal and the energy of the reverberant sound signal respectively.

[0074] The energy of the direct sound signal can be determined by summing the squares of the direct sound signal; the energy of the reverberant sound signal can be determined by summing the squares of the direct sound signal.

[0075] S303: Based on the ratio of the energy of the direct sound signal to the energy of the reverberant sound signal and the preset mapping relationship, determine the first wake-up distance of the sound source emitting the voice signal relative to the main home appliance 101.

[0076] Among them, the ratio of the energy of the direct sound signal to the energy of the reverberant sound signal is positively correlated with the first wake-up distance.

[0077] For example, a preset mapping relationship can be included in a preset mapping table. The ratio of the energy of the direct sound signal to the energy of the reverberant sound signal can be used to find the first wake-up distance of the sound source emitting the speech signal relative to the main home appliance 101 from the preset mapping table. Alternatively, the ratio of the energy of the direct sound signal to the energy of the reverberant sound signal can be input into a pre-trained wake-up distance determination model to obtain the first wake-up distance of the sound source emitting the speech signal relative to the main home appliance 101. The wake-up distance determination model is trained by inputting multiple training samples into a neural network. Each training sample includes the ratio of the energy of the historical direct sound signal to the energy of the reverberant sound signal and the corresponding historical first wake-up distance.

[0078] In some embodiments, the method provided in this application further includes: correcting the first wake-up distance, and the correction may include the following steps 1-3.

[0079] Step 1: When the main home appliance 101 detects a voice signal carrying a wake-up keyword, it collects the CSI data of the Wi-Fi signal.

[0080] Step 2: Input the CSI data into the pre-trained human position awareness model to determine the second wake-up distance of the sound source emitting the voice signal relative to the main home appliance 101.

[0081] The human position awareness model is trained by inputting multiple training samples into a neural network. Each training sample includes historical CSI data and the corresponding historical wake-up distance.

[0082] Step 3: Based on the formula D3 = K1D1 + K2D2, correct the first wake-up distance of the sound source emitting the voice signal relative to the main home appliance 101, where D1 is the first wake-up distance before correction, D2 is the second wake-up distance, D3 is the first wake-up distance after correction, K1 is the first weighting coefficient, and K2 is the second weighting coefficient. For example, K1 = 0.5, K2 = 0.5.

[0083] Understandably, based on the above steps 1-3, the accuracy of the first wake-up distance of the sound source emitting the voice signal relative to the main home appliance 101 can be improved.

[0084] In some implementations, the signal characteristics are spectral characteristics, such as... Figure 4 As shown, the method for determining the included angle of the plane can be specifically implemented as follows:

[0085] S401: Based on the spectral characteristics of the speech signal, separate the direct sound signal and the reverberation signal in the speech signal.

[0086] S402: Determine the energy of the direct sound signal and the high-frequency attenuation value of the direct sound signal energy, respectively.

[0087] S403: Determine the angle between the direction of propagation of the voice signal and the orientation of the household appliance based on the energy of the direct sound signal and the high-frequency attenuation value of the direct sound signal energy.

[0088] Among them, the energy of the direct sound signal is negatively correlated with the angle between the two surfaces, while the high-frequency attenuation value of the direct sound signal energy is positively correlated with the angle between the two surfaces.

[0089] For example, the angle between the propagation direction of the speech signal and the orientation of the home appliance can be determined by looking up a preset mapping table based on the energy of the direct sound signal and its high-frequency attenuation value. Alternatively, the energy of the direct sound signal and its high-frequency attenuation value can be input into a pre-trained angle determination model to determine the angle between the propagation direction of the speech signal and the orientation of the home appliance. The angle determination model is trained by inputting multiple training samples into a neural network. Each training sample includes the energy of a historical direct sound signal, the high-frequency attenuation value of the historical direct sound signal, and the corresponding historical angle.

[0090] S203: Determine the wake-up evaluation value of the main home appliance 101 based on the first wake-up distance and the angle between the two surfaces.

[0091] The wake-up evaluation value is negatively correlated with the first wake-up distance and the angle between the face and the device. For example, the wake-up evaluation value of the main home appliance 101 can be found in a preset mapping table based on the first wake-up distance and the angle between the face and the device. Understandably, a set of first wake-up distances and angles between the face and the device can correspond to a wake-up evaluation value, and the smaller the first wake-up distance and angle between the face and the device, the larger the wake-up evaluation value.

[0092] S204: Receive a wake-up evaluation value from at least one home appliance 102.

[0093] The method by which the wake-up evaluation value is determined from the home appliance 102 is the same as that used by the main home appliance 101, and will not be described in detail here. The home appliance 102 can transmit the wake-up evaluation value to the main home appliance 101 via wireless communication.

[0094] S205: Identify the home appliance with the highest wake-up evaluation value and control the home appliance with the highest wake-up evaluation value to perform a wake-up operation.

[0095] Among them, the home appliance with the highest wake-up evaluation value is the main home appliance 101 or one of the secondary home appliances 102.

[0096] In summary, the home appliance wake-up method provided in this application involves the main home appliance 101 extracting signal features from a voice signal carrying a wake-up keyword upon detecting it. Based on these features, it determines a first wake-up distance between the voice signal source and the main home appliance 101, and the angle between the voice signal propagation direction and the orientation of the home appliance. Based on the first wake-up distance and the angle, it determines a wake-up evaluation value for the main home appliance 101 and receives wake-up evaluation values ​​from at least one secondary home appliance 102. Finally, it identifies the home appliance with the highest wake-up evaluation value and controls it to perform the wake-up operation. Understandably, the closer the user is to the home appliance and the smaller the angle between the voice signal propagation direction and the orientation of the home appliance, the higher the likelihood of the user waking up the appliance. Since the wake-up evaluation value is negatively correlated with the first wake-up distance and the angle, controlling the home appliance with the highest wake-up evaluation value to perform the wake-up operation ensures that the home appliance can be woken up precisely and uniquely, meeting the user's wake-up needs.

[0097] In some implementations, the method provided in this application may further include:

[0098] Step A: When the main home appliance 101 detects a voice signal carrying a wake-up keyword, it collects the CSI data of the Wi-Fi signal.

[0099] Step B: Input the CSI data into the pre-trained human position perception model to determine the direction and distance of the sound source emitting the voice signal relative to the main home appliance 101.

[0100] The human position awareness model is trained by inputting multiple training samples into a neural network. Each training sample includes historical CSI data and the corresponding historical wake-up distance and direction.

[0101] Step C: Based on the direction and distance of the sound source emitting the voice signal relative to the main home appliance 101, the preset three-dimensional space model of the house, and the position of the main home appliance 101 in the three-dimensional space model of the house, identify whether the sound source emitting the voice signal and the main home appliance 101 are located in the same room, and obtain the identification result.

[0102] The above-mentioned S203 can be specifically implemented as follows: based on the first wake-up distance, the angle between the faces, and the recognition result, determine the wake-up evaluation value of the main home appliance 101. When the recognition result is that the sound source emitting the voice signal is located in the same room as the main home appliance 101, the recognition result is positively correlated with the wake-up evaluation value. When the recognition result is that the sound source emitting the voice signal is located in the same room as the main home appliance 101, the recognition result is negatively correlated with the wake-up evaluation value.

[0103] This makes the wake-up evaluation value of the determined main home appliance 101 more reliable.

[0104] In addition, this application embodiment also provides another method for waking up home appliances, applied to a slave home appliance 102, which is communicatively connected to a master home appliance 101. It should be noted that the basic principle and technical effects of the home appliance wake-up method provided in this application embodiment are the same as those in the above embodiments. For the sake of brevity, any parts not mentioned in this application embodiment can be referred to the corresponding content in the above embodiments. Figure 5 As shown, the method provided in this application embodiment includes:

[0105] S501: When the home appliance 102 detects a voice signal carrying a wake-up keyword, it extracts the signal features of the voice signal.

[0106] S502: Based on the signal characteristics of the voice signal, determine the first wake-up distance of the sound source emitting the voice signal relative to the home appliance 102, and determine the angle between the propagation direction of the voice signal and the orientation of the home appliance 102.

[0107] S503: Determine the wake-up evaluation value of the home appliance 102 based on the first wake-up distance and the angle between the two surfaces, wherein the wake-up evaluation value is negatively correlated with the first wake-up distance and the angle between the two surfaces.

[0108] S504: Send the wake-up evaluation value from home appliance 102 to the main home appliance 101.

[0109] S505: Receive a wake-up notification sent by the main home appliance 101 when the wake-up evaluation value of the secondary home appliance 102 is the maximum, and perform a wake-up operation in response to the wake-up notification.

[0110] Please see Figure 6 This application provides a home appliance wake-up device 600, applied to a main home appliance 101, which is communicatively connected to at least one slave home appliance 102. It should be noted that the basic principle and technical effects of the home appliance wake-up device 600 provided in this application are the same as those in the above embodiments. For the sake of brevity, any parts not mentioned in this application can be referred to the corresponding content in the above embodiments. The device 600 provided in this application includes a feature extraction unit 601, a location determination unit 602, an evaluation value determination unit 603, a data receiving unit 604, and a device wake-up unit 605.

[0111] The feature extraction unit 601 is used to extract the signal features of the speech signal when a speech signal carrying a wake-up keyword is detected.

[0112] The orientation determination unit 602 is used to determine, based on the signal characteristics of the voice signal, the first wake-up distance of the sound source emitting the voice signal relative to the main home appliance 101, and to determine the angle between the propagation direction of the voice signal and the orientation of the home appliance.

[0113] The evaluation value determination unit 603 is used to determine the wake-up evaluation value of the main home appliance 101 based on the first wake-up distance and the angle between the two surfaces, wherein the wake-up evaluation value is negatively correlated with the first wake-up distance and the angle between the two surfaces.

[0114] The data receiving unit 604 is used to receive a wake-up evaluation value from at least one home appliance 102;

[0115] The device wake-up unit 605 is used to determine the home appliance with the largest wake-up evaluation value and control the home appliance with the largest wake-up evaluation value to perform a wake-up operation, wherein the home appliance with the largest wake-up evaluation value is the master home appliance 101 or one of the slave home appliances 102.

[0116] In some embodiments, the signal characteristics are spectral characteristics. The orientation determination unit 602 is specifically used to separate the direct sound signal and the reverberant sound signal in the speech signal based on the spectral characteristics of the speech signal; determine the energy of the direct sound signal and the high-frequency attenuation value of the direct sound signal energy respectively; and determine the angle between the propagation direction of the speech signal and the orientation of the household appliance based on the energy of the direct sound signal and the high-frequency attenuation value of the direct sound signal energy, wherein the energy of the direct sound signal is negatively correlated with the angle between the direct sound signal and the orientation of the household appliance, and the high-frequency attenuation value of the direct sound signal energy is positively correlated with the angle between the direct sound signal and the orientation.

[0117] In some embodiments, the orientation determination unit 602 is specifically used to separate the direct sound signal and the reverberant sound signal in the speech signal according to the spectral characteristics of the speech signal; determine the energy of the direct sound signal and the energy of the reverberant sound signal respectively; and determine the first wake-up distance of the sound source emitting the speech signal relative to the main home appliance 101 according to the ratio of the energy of the direct sound signal and the energy of the reverberant sound signal and a preset mapping relationship, wherein the ratio of the energy of the direct sound signal and the energy of the reverberant sound signal is positively correlated with the first wake-up distance.

[0118] In some embodiments, the apparatus 600 provided in this application further includes: a data acquisition unit, used to acquire CSI data of Wi-Fi signals when a voice signal carrying a wake-up keyword is detected;

[0119] The distance determination unit is used to input CSI data into a pre-trained human position awareness model to determine the second wake-up distance of the sound source emitting the speech signal relative to the main home appliance 101. The human position awareness model is trained on a neural network based on multiple training samples input into the neural network; each training sample includes historical CSI data and the corresponding historical wake-up distance.

[0120] The distance correction unit is used to correct the first wake-up distance of the sound source emitting the voice signal relative to the main home appliance 101 according to the formula D3=K1D1+K2D2, where D1 is the first wake-up distance before correction, D2 is the second wake-up distance, D3 is the first wake-up distance after correction, K1 is the first weighting coefficient, and K2 is the second weighting coefficient.

[0121] In some implementations, the data acquisition unit is also used to acquire CSI data of the Wi-Fi signal when a voice signal carrying a wake-up keyword is detected.

[0122] The orientation determination unit 602 is also used to input CSI data into a pre-trained human position perception model to determine the direction and distance of the sound source emitting the voice signal relative to the main home appliance 101.

[0123] The human position awareness model is trained by inputting multiple training samples into a neural network. Each training sample includes historical CSI data and the corresponding historical wake-up distance and direction.

[0124] The device 600 provided in this application embodiment further includes: a room determination unit, used to identify whether the source of the voice signal and the main home appliance 101 are located in the same room based on the direction and distance of the source of the voice signal relative to the main home appliance 101, a preset three-dimensional space model of the house, and the position of the main home appliance 101 in the three-dimensional space model of the house, and to obtain the identification result.

[0125] The evaluation value determination unit 603 is specifically used to determine the wake-up evaluation value of the main home appliance 101 based on the first wake-up distance, the angle between the faces, and the recognition result. When the recognition result indicates that the sound source emitting the voice signal is located in the same room as the main home appliance 101, the recognition result is positively correlated with the wake-up evaluation value. When the recognition result indicates that the sound source emitting the voice signal is located in the same room as the main home appliance 101, the recognition result is negatively correlated with the wake-up evaluation value.

[0126] In some embodiments, the apparatus 600 provided in this application further includes:

[0127] Data receiving unit 604 is used to receive IP addresses from other home appliances;

[0128] The sorting unit is used to sort the IP addresses of itself and other home appliances according to their size relationship.

[0129] The master-slave device determination unit is used to determine itself as the master home appliance 101 and other home appliances as slave home appliances 102 when the IP address of the target home appliance is the largest or smallest.

[0130] Additionally, please see Figure 7 This application also provides a home appliance wake-up device 700, applied to a slave home appliance 102, which is communicatively connected to a master home appliance 101. It should be noted that the basic principle and technical effects of the home appliance wake-up device 700 provided in this application are the same as those in the above embodiments. For the sake of brevity, any parts not mentioned in this application can be referred to the corresponding content in the above embodiments. The device 700 provided in this application includes:

[0131] The feature extraction unit 701 is used to extract the signal features of the speech signal when a speech signal carrying a wake-up keyword is detected.

[0132] The orientation determination unit 702 is used to determine, based on the signal characteristics of the voice signal, the first wake-up distance of the sound source emitting the voice signal relative to the orientation of the home appliance 102, and to determine the angle between the propagation direction of the voice signal and the orientation of the home appliance 102.

[0133] The evaluation value determination unit 703 is used to determine the wake-up evaluation value of the home appliance 102 based on the first wake-up distance and the angle between the two surfaces, wherein the wake-up evaluation value is negatively correlated with the first wake-up distance and the angle between the two surfaces.

[0134] The data transmission unit 704 is used to send the wake-up evaluation value of the home appliance 102 to the main home appliance 101;

[0135] The device wake-up unit 705 is used to receive a wake-up notification sent by the main home appliance 101 when the wake-up evaluation value of the secondary home appliance 102 is the maximum, and to perform a wake-up operation in response to the wake-up notification.

[0136] Additionally, please see Figure 8 This application also provides a home appliance device 800, including a system-on-a-chip 802; a memory 804 for storing executable instructions of the system-on-a-chip 802; wherein the system-on-a-chip 802 is configured to execute instructions to implement the home appliance wake-up method provided in the above embodiments of this application.

[0137] In some implementations, the system-on-a-chip 802 integrates a WiFi chip 820, a display interaction module 822, a voice recognition module 816, and a main processing module 818. The WiFi chip 820 collects CSI data of the WiFi signal and inputs the CSI data into a pre-trained human position perception model to determine a second wake-up distance between the sound source emitting the voice signal and the main home appliance. The voice recognition module 816 extracts the signal features of the voice signal, determines a first wake-up distance between the sound source emitting the voice signal and the main home appliance based on these features, and determines the angle between the propagation direction of the voice signal and the orientation of the home appliance. The main processing module 818 determines the wake-up evaluation value of the main home appliance based on the first wake-up distance and the angle; and determines the home appliance with the highest wake-up evaluation value based on at least one wake-up evaluation value sent from a home appliance, and controls the home appliance with the highest wake-up evaluation value to perform a wake-up operation. The display interaction module 822 processes the data to be displayed on the home appliance and transmits it to the display panel of the home appliance for display.

[0138] Understandably, since the WiFi chip 820, display interaction module 822, voice recognition module 816 and main processing module 818 are integrated into a single system-on-a-chip 802, the device cost and design complexity are reduced.

[0139] Additionally, the home appliance 800 may include one or more of the following components: a power supply component 806, a display panel 808, an audio component 810, an input / output (I / O) interface 812, a sensor component 814, and an antenna 816.

[0140] Memory 804 is configured to store various types of data to support the operation of home appliance 800. Examples of this data include instructions for any application or method used to operate on home appliance 800. Memory 804 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, and flash memory.

[0141] Power supply component 806 provides power to various components of home appliance 800. Power supply component 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to home appliance 800.

[0142] Display panel 808 includes a screen that provides an output interface between the home appliance and the user. In embodiments of this application, the screen includes multiple in-vehicle physical screens. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen can be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors can sense not only the boundaries of touch or swipe actions but also the duration and pressure associated with the touch or swipe operation.

[0143] Audio component 810 is configured to output and / or input audio signals. For example, audio component 810 includes a microphone (MIC) configured to receive external audio signals. The received audio signals may be further stored in memory 804 or transmitted via antenna 816. In some embodiments, audio component 810 also includes a speaker for outputting audio signals.

[0144] I / O interface 812 provides an interface between the system-on-a-chip 802 of the home appliance and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, power buttons, and lock buttons.

[0145] Sensor assembly 814 includes one or more sensors for providing status assessments of various aspects of the home appliance 800. For example, sensor assembly 814 can detect the on / off state of the home appliance 800, the relative positioning of components, changes in the position of the home appliance 800 or a component of the home appliance 800, the presence or absence of user contact with the home appliance 800, the orientation or acceleration / deceleration of the home appliance 800, and temperature changes of the home appliance 800.

[0146] Antenna 816 is configured to facilitate wired or wireless communication between home appliance 800 and other devices. Home appliance 800 can access wireless networks based on communication standards, such as WiFi, 2G, or 3G, or combinations thereof. In one exemplary embodiment, antenna 816 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel.

[0147] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 804 including instructions, which can be executed by the system-on-a-chip 802 of the home appliance 800 to perform the above-described method. For example, the non-transitory computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, or optical data storage device. When the instructions in this non-transitory computer-readable storage medium are executed by the system-on-a-chip 802 of the home appliance, the home appliance 800 is able to perform the home appliance wake-up method provided in the above embodiment.

[0148] In addition, this application also provides a storage medium that, when the instructions in the storage medium are executed by the system-on-a-chip of a home appliance, enables the home appliance to perform the home appliance wake-up method provided in the above embodiments of this application.

[0149] In addition, this application also provides a computer program product, including a computer program that, when executed by a system-on-a-chip, implements the home appliance wake-up method provided in the above embodiments of this application. The computer program product may be pre-written in memory or downloaded and installed in memory as software. 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 flow or function according to the embodiments of this application is generated.

[0150] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can also be implemented in other ways. The apparatus embodiments described above are merely illustrative; for example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0151] In addition, the functional modules in the various embodiments of the present invention can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0152] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks. It should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0153] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the invention should be included within the scope of protection of the invention. It should be noted that similar reference numerals and letters in the following figures denote similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

Claims

1. A method for waking up a home appliance, characterized in that, Applied to a master home appliance, the master home appliance being communicatively connected to at least one slave home appliance, the method includes: When the main home appliance detects a voice signal carrying a wake-up keyword, it extracts the signal features of the voice signal. Based on the signal characteristics of the voice signal, determine the first wake-up distance of the sound source emitting the voice signal relative to the main home appliance, and determine the angle between the propagation direction of the voice signal and the orientation of the home appliance. The wake-up evaluation value of the main home appliance is determined based on the first wake-up distance and the angle between the two surfaces, wherein the wake-up evaluation value is negatively correlated with the first wake-up distance and the angle between the two surfaces, respectively. Receive a wake-up evaluation value from the at least one home appliance; The home appliance with the highest wake-up evaluation value is identified, and the home appliance with the highest wake-up evaluation value is controlled to perform a wake-up operation, wherein the home appliance with the highest wake-up evaluation value is the master home appliance or one of the slave home appliances.

2. The method according to claim 1, characterized in that, The signal feature is a spectral feature. Determining the angle between the propagation direction of the voice signal and the orientation of the home appliance based on the signal feature includes: Based on the spectral characteristics of the speech signal, the direct sound signal and the reverberation signal in the speech signal are separated. The energy of the direct sound signal and the high-frequency attenuation value of the energy of the direct sound signal are determined respectively; Based on the energy of the direct sound signal and the high-frequency attenuation value of the direct sound signal, the angle between the propagation direction of the voice signal and the orientation of the home appliance is determined, wherein the energy of the direct sound signal is negatively correlated with the angle between the direct sound signal and the high-frequency attenuation value of the direct sound signal is positively correlated with the angle between the direct sound signal and the orientation of the home appliance.

3. The method according to claim 1, characterized in that, The signal feature is a spectral feature. Determining the first wake-up distance of the sound source emitting the voice signal relative to the main home appliance based on the signal feature of the voice signal includes: Based on the spectral characteristics of the speech signal, the direct sound signal and the reverberation signal in the speech signal are separated. Determine the energy of the direct sound signal and the energy of the reverberant sound signal respectively; Based on the ratio of the energy of the direct sound signal to the energy of the reverberation sound signal and a preset mapping relationship, a first wake-up distance is determined relative to the main home appliance from which the sound source emitting the voice signal is emitted, wherein the ratio of the energy of the direct sound signal to the energy of the reverberation sound signal is positively correlated with the first wake-up distance.

4. The method according to claim 3, characterized in that, The method further includes: The main home appliance collects Wi-Fi signals when it detects a voice signal carrying a wake-up keyword. The CSI data is input into a pre-trained human position awareness model to determine the second wake-up distance of the sound source emitting the voice signal relative to the main home appliance. The human position awareness model is trained based on multiple training samples input into a neural network, and each training sample includes historical CSI data and the corresponding historical wake-up distance. According to the formula D3=K1D1+K2D2, the first wake-up distance of the sound source emitting the voice signal relative to the main home appliance is corrected, where D1 is the first wake-up distance before correction, D2 is the second wake-up distance, D3 is the first wake-up distance after correction, K1 is the first weighting coefficient, and K2 is the second weighting coefficient.

5. The method according to claim 1, characterized in that, When the main home appliance detects a voice signal carrying a wake-up keyword, it collects CSI data of the Wi-Fi signal. The CSI data is input into a pre-trained human position awareness model to determine the direction and distance of the sound source emitting the voice signal relative to the main home appliance. The human position awareness model is trained based on multiple training samples input into a neural network. Each training sample includes historical CSI data and the corresponding historical wake-up distance and direction. Based on the direction and distance of the sound source emitting the voice signal relative to the main home appliance, the preset three-dimensional space model of the house, and the position of the main home appliance in the three-dimensional space model of the house, it is determined whether the sound source emitting the voice signal and the main home appliance are located in the same room, and the identification result is obtained. The step of determining the wake-up evaluation value of the main home appliance based on the first wake-up distance and the angle between the two surfaces includes: Based on the first wake-up distance, the included angle of the face, and the recognition result, the wake-up evaluation value of the main home appliance is determined. Wherein, when the recognition result indicates that the sound source emitting the voice signal is located in the same room as the main home appliance, the recognition result is positively correlated with the wake-up evaluation value. When the recognition result indicates that the sound source emitting the voice signal is located in the same room as the main home appliance, the recognition result is negatively correlated with the wake-up evaluation value.

6. The method according to claim 1, characterized in that, Before the main home appliance detects a voice signal carrying a wake-up keyword and extracts the signal features of the voice signal, the method further includes: The target home appliance receives IP addresses from other home appliances; The target home appliance sorts its own IP address and the IP addresses of the other home appliances according to their size relationship. If the IP address of the target home appliance is at its maximum or minimum, the target home appliance determines itself as the master home appliance and the other home appliances as slave home appliances.

7. A method for waking up a home appliance, characterized in that, The method, applied to a slave home appliance that is communicatively connected to a master home appliance, includes: When a home appliance detects a voice signal carrying a wake-up keyword, the signal features of the voice signal are extracted. Based on the signal characteristics of the voice signal, determine the first wake-up distance of the sound source emitting the voice signal relative to the home appliance, and determine the angle between the propagation direction of the voice signal and the orientation of the home appliance. The wake-up evaluation value of the home appliance is determined based on the first wake-up distance and the angle between the two surfaces, wherein the wake-up evaluation value is negatively correlated with the first wake-up distance and the angle between the two surfaces, respectively. Send a wake-up assessment value to the main home appliance; The system receives a wake-up notification sent by the master home appliance when the wake-up evaluation value of the slave home appliance is at its maximum, and performs a wake-up operation in response to the wake-up notification.

8. A household appliance, characterized in that, include: System-on-a-chip (SoC); Memory used to store executable instructions of the system-on-a-chip; The system-on-a-chip is configured to execute the instructions to implement the home appliance wake-up method as described in any one of claims 1 to 7.

9. The household appliance according to claim 8, characterized in that, The system-on-a-chip integrates a WiFi chip, a display and interaction module, a voice recognition module, and a main processing module. The WiFi chip is used to collect CSI data of WiFi signals and input the CSI data into a pre-trained human position perception model to determine the second wake-up distance of the sound source emitting the voice signal relative to the main home appliance. The voice recognition module is used to extract the signal features of the voice signal, and based on the signal features of the voice signal, determine the first wake-up distance of the sound source emitting the voice signal relative to the main home appliance, and determine the angle between the propagation direction of the voice signal and the orientation of the home appliance. The main processing module determines the wake-up evaluation value of the main home appliance based on the first wake-up distance and the angle between the two surfaces; and determines the home appliance with the largest wake-up evaluation value based on the wake-up evaluation values ​​sent from at least one home appliance, and controls the home appliance with the largest wake-up evaluation value to perform a wake-up operation. The display interaction module is used to process the data to be displayed by the home appliance and transmit it to the display panel of the home appliance for display.

10. A storage medium, wherein when instructions in the storage medium are executed by a system-on-a-chip of a home appliance, the home appliance is enabled to perform the home appliance wake-up method as described in any one of claims 1 to 7.

Images