Low-frequency signal detection method suitable for blowing identification, control method and electronic device

Abstract

The invention discloses a low-frequency signal detection method suitable for blowing identification, a control method and an electronic device, the low-frequency signal detection method suitable for blowing identification is used for determining whether a low-frequency signal based on a blowing signal exists in an audio signal, and comprises the following steps: collecting the audio signal at a sampling frequency covering a target frequency band; an audio signal to be detected is obtained; determining the signal energy Elow (t) of a target frequency band corresponding to the t moment in the audio signal to be detected; calculating a signal energy difference E (t) in the time difference interval according to the signal energy Elow (t) of the target frequency band; and judging whether the conditions that E (t) is less than-gamma l and the duration of E (t) less than 0 meets a preset requirement or not are met, gamma l is an energy difference threshold value and gamma l is greater than 0, and if not, determining that the audio signal to be detected does not comprise the low-frequency signal. According to the invention, the low-frequency signal generated based on blowing can be detected more accurately and reliably.

Description

Technical Field

[0001] The present invention relates to the technical fields of acoustic detection and electronic device control, and particularly to a low-frequency signal detection method, a control method, and an electronic device suitable for blowing recognition. Background Art

[0002] The current voice-controlled electronic devices on the market mainly adopt the following technical solutions.

[0003] (1) Simple threshold detection method: The technical principle of this method is to trigger by comparing the amplitude of the time-domain signal collected by the microphone with a preset fixed threshold. A typical trigger implementation is to trigger an action when the signal amplitude exceeds the threshold (such as 1.5V). The main defects of this method include:

[0004] Poor sound source recognition ability: It is unable to distinguish sound sources of different natures such as blowing, speech, and collision;

[0005] Weak anti-interference ability: Sudden noises (such as the sound of an item falling) are likely to cause false triggers, and the measured false trigger rate is as high as 18 - 25%;

[0006] Poor environmental adaptability: The fixed threshold cannot adapt to different environmental noise levels, and the missed trigger rate increases by more than 40% in a noisy environment.

[0007] (2) Single frequency domain detection method: The technical principle of this method is to analyze the energy of a specific frequency band (such as 1 kHz) using FFT. Its trigger mode is to trigger when the energy of the target frequency band exceeds the threshold. The main defects of this method include:

[0008] Poor frequency adaptability;

[0009] Insufficient dynamic response: Sensitive to sudden broadband noises (such as the sound of glass breaking), the measured false trigger rate is 15 - 20%, and the response delay generally exceeds 300 ms.

[0010] (3) Hybrid sensor solution: The technical principle of this method is multi-modal detection combining a microphone and a pressure sensor (or an infrared sensor), and it triggers only when both the acoustic signal and the pressure change conditions are met. The main defects of this method include:

[0011] Cost issue: Adding a pressure sensor increases the BOM cost by 35 - 40%;

[0012] Reliability issue: The multi-sensor data fusion algorithm is complex, and the failure rate increases by 2 - 3 times. The pressure sensor is vulnerable to environmental temperature and humidity, and the failure probability increases by 50% in a humid environment;

[0013] User experience issue: The detection distance is limited (usually <15 cm), and the sensitivity drops sharply by 80% beyond the distance.

[0014] In summary, the existing low-frequency signal detection and control methods generally have at least the following deficiencies:

[0015] (1) Unable to distinguish specific frequencies, prone to false triggering due to environmental noise (such as voices, music);

[0016] (2) Lack of anti-interference mechanism, poor stability in complex acoustic environments;

[0017] (3) Poor user experience: high response delay, unable to meet real-time interaction requirements, and lack of natural transition effects in state switching;

[0018] (4) High hardware cost: multi-sensor solutions lead to a BOM cost increase of more than 35%.

[0019] The disclosure of the above background technical content is only used to assist in understanding the inventive concept and technical solution of the present invention. It does not necessarily belong to the prior art of this application, nor will it necessarily provide technical teachings; without clear evidence indicating that the above content was publicly available before the filing date of this application, the above background art should not be used to evaluate the novelty and inventiveness of this application. Summary of the Invention

[0020] The object of the present invention is to provide a low-frequency signal detection method, a control method, and an electronic device suitable for blowing recognition, which can more accurately and reliably detect low-frequency signals generated based on blowing.

[0021] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0022] A low-frequency signal detection method suitable for blowing recognition, used to determine whether there is a low-frequency signal based on a blowing signal in an audio signal, including the following steps:

[0023] Collect the audio signal at a sampling frequency covering the target frequency band to obtain the audio signal to be detected;

[0024] Determine the signal energy E low (t) of the target frequency band corresponding to the t moment in the audio signal to be detected;

[0025] According to the signal energy E low (t) of the target frequency band, calculate the signal energy difference △E(t) within the time difference interval, ΔE(t) = E low (t) - E low (t - Δt), where △t is the preset time difference interval;

[0026] Judge whether the following conditions are met: △E(t) < -Γ l And the duration of △E(t) < 0 meets the preset requirements, where Γ lis the energy difference threshold, Γ l > 0. If not satisfied, it is determined that the audio signal to be detected does not include the low-frequency signal based on the blowing signal.

[0027] Further, based on any one of the foregoing technical solutions or a combination of multiple technical solutions, the following steps are further included:

[0028] If it satisfies: △E(t) < -Γ l And the duration of △E(t) < 0 meets the preset requirements, it is determined that the audio signal to be detected includes the low-frequency signal.

[0029] Further, based on any one of the foregoing technical solutions or a combination of multiple technical solutions, the following steps are further included:

[0030] If it satisfies: △E(t) < -Γ l And the duration of △E(t) < 0 meets the preset requirements, calculate the blowing probability. If the blowing probability is greater than the preset probability value, it is determined that the audio signal to be detected includes the low-frequency signal; otherwise, it is determined that the audio signal to be detected does not include the low-frequency signal.

[0031] Further, based on any one of the foregoing technical solutions or a combination of multiple technical solutions, the blowing probability is calculated in the following manner:

[0032]

[0033] where, P blow (t) represents the blowing probability, σ(·) represents the probability conversion function, φ i (ΔE(t)) is the feature extraction function, and w i represents the weight coefficient.

[0034] Further, based on any one of the foregoing technical solutions or a combination of multiple technical solutions, the probability conversion function adopts the Sigmoid function, which is expressed as follows: where, e represents the exponent, and x is the independent variable;

[0035] and / or,

[0036] The feature extraction function adopts the Gabor basis function, which is defined as: where, x is the independent variable, μ i is the center position parameter, f i is the frequency adjustment parameter, ε i is the bandwidth adjustment parameter, cos(·) is the cosine function, π is the circumference ratio, i takes 1, 2, 3, and ψ i is the phase parameter.

[0037] Further, based on any one of the foregoing technical solutions or a combination of multiple technical solutions, it further includes dynamically adjusting the weight coefficient w by the following method i :

[0038] Determine the real-time signal-to-noise ratio SNR(m) by the following method:

[0039]

[0040] where SNR(m) represents the real-time signal-to-noise ratio, ΔE(m) represents the normalized signal energy difference, m is the block index, and σ N represents the noise standard deviation, and ∝ is the proportional symbol;

[0041] Based on the following formula, scale w proportionally according to the real-time signal-to-noise ratio SNR(m) i : w i ←w i ·(1 + α·SNR(m)), where α represents the proportionality coefficient.

[0042] Further, based on any one of the foregoing technical solutions or a combination of multiple technical solutions, determine whether the duration of ΔE(t) < 0 meets the preset requirements by the following method: If N S ≥N th , it is determined that the duration of ΔE(t) < 0 meets the preset requirements, where N S represents the number of consecutive frames satisfying ΔE(t) < 0, and N th represents the preset frame number threshold;

[0043] Or

[0044] Determine whether the duration of ΔE(t) < 0 meets the preset requirements by the following method:

[0045]

[0046] where k is the frame index, K is the preset duration frame number threshold, z is the current starting frame, I(.) is the indicator function, which takes the value of 1 when the condition is satisfied and 0 otherwise, and β is the proportion coefficient, 0.5 ≤ β ≤ 1.

[0047] Further, based on any one of the foregoing technical solutions or a combination of multiple technical solutions, calculate the signal energy E low (t) of the target frequency band in the audio signal to be detected at time t:

[0048]

[0049] where X w(k, t) is the frequency-domain representation of the audio signal to be detected after windowing, and its dimension is the frequency point k × the time frame t; SUM is the total number of FFT points. Corresponding to 1 / 4 bandwidth.

[0050] Furthermore, based on any one of the foregoing technical solutions or a combination of multiple technical solutions, the following steps are further included:

[0051] Use a sliding window to determine the signal energy difference △E(t) within multiple time difference intervals in the audio signal to be detected.

[0052] According to another aspect of the present invention, there is provided a method for controlling an electronic device. Based on the low-frequency signal detection method applicable to blowing identification described in any one of the foregoing technical solutions or a combination of multiple technical solutions, it is determined whether there is a low-frequency signal based on a blowing signal in the audio signal. If so, the electronic device is controlled to switch from a first state to a second state, and the second state is different from the first state.

[0053] Furthermore, based on any one of the foregoing technical solutions or a combination of multiple technical solutions, the first state is an on state, and the second state is an off state.

[0054] According to another aspect of the present invention, there is provided an electronic device. The electronic device includes a controller configured to execute the method for controlling an electronic device described in any one of the foregoing technical solutions or a combination of multiple technical solutions. The electronic device is an electronic lamp and / or an electronic audio device, and the electronic lamp includes an electronic candle and an electronic lamp rod.

[0055] Furthermore, based on any one of the foregoing technical solutions or a combination of multiple technical solutions, the electronic device includes an electronic device body, an audio detection module, and a control module;

[0056] Among them, the audio detection module includes an audio acquisition module and a signal processing module. The audio acquisition module is configured to acquire an audio signal at a sampling frequency covering the target frequency band to obtain the audio signal to be detected. The signal processing module is configured to determine whether there is a low-frequency signal based on a blowing signal in the audio signal to be detected and transmit the detection result to the control module;

[0057] The control module is configured to control the state of the electronic device body to change or not change according to the detection result, including:

[0058] If the detection result is that there is a low-frequency signal based on a blowing signal in the audio signal to be detected, then control the state of the electronic device body to change, otherwise do not control the state of the electronic device body to change.

[0059] The beneficial effects brought by the technical solutions provided by the present invention are as follows:

[0060] a. The present invention proposes a dual decision-making mechanism that combines energy mutation detection and verification of the duration of the energy decline signal, which can significantly improve the accuracy of blowing recognition and the robustness of the system. Energy mutation detection can quickly capture the energy decline caused by the blowing operation, while duration verification (the duration of the signal energy decline meets the preset requirements) further confirms the persistence of the blowing signal, ensuring the effectiveness and reliability of the blowing signal. This collaborative detection method not only reduces the false trigger rate but also enhances the stability of the system in complex acoustic environments;

[0061] b. The present invention determines whether the duration of the signal energy decline meets the preset requirements by whether the proportion of the number of frames of the signal with energy decline meets the preset frame threshold, which can avoid short-term interference, especially can eliminate missed triggers and false triggers caused by single-frame noise, ensure the persistence of the blowing signal and the reliability of the blowing signal detection, and further reduce the probability of missed detection and false detection of the blowing signal;

[0062] c. On the basis of the dual judgment mechanism, the present invention further combines the judgment condition of the blowing probability and judges based on dynamically adjusting the signal-to-noise ratio according to the ambient noise level. The energy threshold is dynamically adjusted according to the real-time signal-to-noise ratio (SNR). For example, when it is detected that the ambient noise increases significantly, the threshold is appropriately increased to reduce false triggers; conversely, in a quiet environment, the threshold can be appropriately reduced to improve sensitivity, which can improve the robustness of the detection method and system in complex environments. Description of the Drawings

[0063] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments recorded in the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0064] Figure 1 Flowchart of the first low-frequency signal detection method applicable to blowing recognition provided for an exemplary embodiment of the present invention;

[0065] Figure 2 Flowchart of the first method for determining whether the duration of the signal energy being negative meets the preset requirements provided for an exemplary embodiment of the present invention;

[0066] Figure 3 Flowchart of the second method for determining whether the duration of the signal energy being negative meets the preset requirements provided for an exemplary embodiment of the present invention;

[0067] Figure 4Flow chart of the second low-frequency signal detection method applicable to blowing recognition provided for an exemplary embodiment of the present invention;

[0068] Figure 5 Module diagram of an electronic device controlled based on the low-frequency signal detection result provided for an exemplary embodiment of the present invention. Detailed implementation manners

[0069] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0070] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence. It should be understood that such data used may be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, device, product or equipment including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or equipment.

[0071] In an embodiment of the present invention, a low-frequency signal detection method applicable to blowing recognition is provided. As Figure 1 shown, it is used to determine whether there is a low-frequency signal based on a blowing signal in an audio signal, and includes the following steps:

[0072] Collect an audio signal at a sampling frequency covering the target frequency band to obtain the audio signal to be detected;

[0073] Determine the signal energy E low (t) of the target frequency band corresponding to the t moment in the audio signal to be detected;

[0074] According to the signal energy E low (t) of the target frequency band, calculate the signal energy difference △E(t) within the time difference interval, ΔE(t) = E low (t) - E low (t - Δt), where △t is the preset time difference interval;

[0075] Determine whether the following conditions are met: △E(t) < -Γ l and the duration for which △E(t) < 0 meets a preset requirement, where Γ l is the energy difference threshold, Γ l > 0. If the conditions are not met, it is determined that the audio signal to be detected does not include the low-frequency signal.

[0076] In an embodiment of the present invention, the audio signal / data is collected at a sampling frequency F s ≥ 8 kHz, and the collected audio signal is windowed to reduce spectral leakage. The signal energy E low (t) of the target frequency band (e.g., 0 - 1 kHz) in the audio signal to be detected at time t is calculated by the following method:

[0077]

[0078] where X w (k,t) is the frequency-domain representation of the windowed audio signal to be detected, obtained through short-time Fourier transform (STFT), and its dimension is the frequency point k × time frame t; SUM is the total number of FFT points, for example corresponding to 1 / 4 bandwidth, i.e., the 0 - 1 / 4 sampling rate range, and the unit of E low (t) is dB or linear amplitude squared.

[0079] The calculation method of the signal energy E low (t) of the above target frequency band can focus on the core frequency band of the blowing signal (i.e., the target low-frequency band) and suppress high-frequency noise interference. In this embodiment, real-time signal energy monitoring is achieved through a sliding time window, which can adapt to the characteristics of non-stationary signals.

[0080] The signal energy difference △E(t) within multiple time differential intervals in the audio signal to be detected is determined using a sliding window. The calculation formula for the signal energy difference △E(t) is as follows:

[0081] ΔE(t) = E low (t) - E low (t - Δt)

[0082] where △t is a preset time differential interval, representing the time interval between two adjacent signal energy calculation times, and is used to calculate the signal energy change rate. △ is the detection time window for determining the energy change, which is usually set according to the sampling frequency and application scenario. For example, △t can take values from 10 ms to 50 ms.

[0083] In this embodiment, E low (t) represents the low-frequency band energy at time t (the current time), that is, at time t, the signal energy of the low-frequency band (such as 0 - 1 kHz) after filtering processing, Elow (t) reflects the intensity of the low-frequency band signal at the current moment and is the basis for calculating the energy change.

[0084] E low (t - Δt) represents the low-frequency band energy at a past moment. Specifically, it represents the energy of the low-frequency band signal after filtering at time t - Δt, E low (t - Δt) is used as a reference value to compare with the signal energy at the current moment to calculate the change amount of the signal energy.

[0085] The signal energy difference △E(t) is a signal energy mutation feature, which represents the change amount of the low-frequency band energy at time t and is used to capture the sudden change of energy. By calculating the change amount of energy, the energy drop caused by the blowing signal can be detected, so as to judge whether it is an effective blowing operation.

[0086] The calculation formula of the signal energy difference △E(t) is used to extract the signal energy mutation feature in the blowing signal. During the blowing operation, the airflow passing through the microphone will cause a sudden drop in the low-frequency band energy. By calculating the energy difference ΔE(t) between the current moment and a past moment, this energy drop can be effectively captured, so as to judge whether it is an effective blowing signal.

[0087] For example, the blowing signal feature can be determined in the following way:

[0088] When blowing, the energy decays rapidly, and ΔE(t) is a significant negative value (such as -30dB);

[0089] The ΔE(t) of ambient noise (such as speech) fluctuates less (such as ±5dB).

[0090] Based on the signal energy mutation feature in the above-mentioned blowing signal, this application proposes △E(t) < -Γ l as one of the conditions for detecting the blowing signal. In addition, in order to improve the effectiveness of the blowing detection signal and reduce the false detection rate of the blowing signal, this application further proposes that the duration of △E(t) < 0 meets the preset requirements as one of the conditions for detecting the blowing signal.

[0091] This application proposes two methods to judge whether the duration of △E(t) < 0 meets the preset requirements. As Figure 2 shown, in the first judgment method:

[0092] If N S ≥ N th (or △t blow ≥ the preset duration), it is judged that the duration of △E(t) < 0 meets the preset requirements, where N S represents the number of consecutive frames that satisfy ΔE(t) < 0, Nth It represents a preset frame number threshold, which is used to determine the number of frames for the duration of energy change.

[0093] Based on this judgment method, the conditions for detecting a valid blowing signal are determined as follows:

[0094] If ΔE(t) < -Γ l and N S ≥ N th , it is determined as a valid blowing signal. Here, -Γl represents a preset negative energy threshold, which is used to determine the threshold for energy decrease.

[0095] As Figure 3 shown, in the second judgment method, different from the first judgment method, it is determined whether the duration of ΔE(t) < 0 meets the preset requirements in the following way:

[0096]

[0097] where k is the frame index, K is a preset continuous frame number threshold, representing the size of the detection window, that is, the number of consecutive frames to be verified, z is the current starting frame, and z + K represents the last frame of the detection window. For example, K = 10 means detecting the energy change within 10 frames. The value of K is usually determined through experiments or experience. For example, the number of frames corresponding to 100 - 300 ms is determined as K, and it is converted to the number of frames through the sampling rate. For example, when the sampling rate is 44.1 kHz and the frame length is 10 ms, K = 15 corresponds to a 150 ms detection window.

[0098] I(.) is an indicator function, which takes the value of 1 when the condition inside () holds, otherwise 0.

[0099] means traversing each frame of the signal within the time window (such as from the z-th frame to the z + K-th frame), and is used to check frame by frame whether the energy change ΔE(k) is negative. A blowing signal usually causes a sudden drop in energy, and ΔE(k) < 0 is a key feature of the mutation.

[0100] β is a proportion coefficient, 0.5 ≤ β ≤ 1. In this embodiment, taking β = 0.5 as an example, when it is determined that the duration of △E(t) < 0 meets the preset requirements.

[0101] Based on this judgment method, the conditions for detecting a valid blowing signal are determined as follows:

[0102] If ΔE(t) < -Γ l and it is determined as a valid blowing signal.

[0103] Based on the condition of detecting a valid blowing signal, taking β = 0.5 as an example, it is required that the number of frames with energy decrease ≥ K / 2 (that is, at least half of the frames need to meet the condition). Assume K = 10. While satisfying △E(t) < -Γ l if only 3 frames out of 10 frames have energy decrease then the energy decrease signal is determined as noise; if 7 frames out of 10 frames have energy decrease then the energy decrease signal is determined as noise.

[0104] By this method, short-term interference can be avoided, especially single-frame noise can be excluded to prevent missed triggers and false triggers, ensuring the continuity of the blowing signal and the reliability of blowing signal detection, and further reducing the probability of missed detection and false detection of the blowing signal.

[0105] In an embodiment of the present invention, a low-frequency signal detection method applicable to blowing recognition is proposed. As Figure 4 shown, the low-frequency signal detection method applicable to blowing recognition provided in this embodiment further includes the following steps on the basis of the above embodiment:

[0106] If it satisfies: △E(t) < -Γ l and the duration of △E(t) < 0 meets the preset requirement, then calculate the blowing probability. If the blowing probability is greater than the preset probability value, it is determined that the audio signal to be detected includes the low-frequency signal; otherwise, it is determined that the audio signal to be detected does not include the low-frequency signal.

[0107] In this embodiment, the blowing probability is calculated in the following way:

[0108]

[0109] where P blow (t) represents the blowing probability, σ(·) represents the probability conversion function, φ i (ΔE(t)) is the feature extraction function, and w i represents the weight coefficient.

[0110] The preset signal energy threshold in the above embodiment is usually determined based on a large amount of experimental data statistics, or determined according to experience, or determined by using an AI algorithm based on standard learning samples, ensuring that the blowing signal and noise can be effectively distinguished in typical application scenarios. For example, by analyzing the blowing data under different environmental noise conditions, an initial threshold is determined so that both the false trigger rate and the missed trigger rate are at a relatively low level, that is, reaching an ideal level.

[0111] Based on the above embodiments, this embodiment proposes a dynamic adjustment mechanism. During the detection process, the energy threshold is dynamically adjusted according to the real-time signal-to-noise ratio or other environmental parameters. For example, when it is detected that the environmental noise increases significantly, the threshold is appropriately increased to reduce false triggers; conversely, in a quiet environment, the threshold can be appropriately reduced to improve sensitivity.

[0112] Preferably, the probability conversion function adopts the Sigmoid function, which is used to map the eigenvalue to the interval [0,1] and output the blowing probability. The probability conversion function σ(x) is expressed as follows:

[0113]

[0114] where e represents the exponential, and x is the independent variable.

[0115] Preferably, the feature extraction function adopts the Gabor basis function, which is defined as:

[0116]

[0117] where x is the independent variable, μ i is the center position parameter, f i is the frequency adjustment parameter, ε i is the bandwidth adjustment parameter, cos(·) is the cosine function, π is the pi, i takes 1, 2, 3, and ψ i represents the phase parameter, and its physical meaning is to control the initial phase of the cosine function. By adjusting ψ i , the phase characteristics of the signal can be changed, thereby affecting the time-frequency representation of the signal. It should be noted that the feature extraction function φ i (x) can be replaced by other time-frequency analysis basis functions such as Morlet wavelet and Hermite function in addition to the Gabor basis function.

[0118] In this embodiment, the physical meanings, typical values, and determination methods of the relevant parameters involved in the dynamic adjustment mechanism are shown in Table 1.

[0119] Table 1 Relevant parameters of the dynamic adjustment mechanism

[0120] Parameter Physical meaning Typical value Determination method <![CDATA[μ i > Center position parameter 0.5，1.0，1.5 Match the peak value of the blowing signal feature <![CDATA[ε i > Bandwidth adjustment parameter 0.2～0.5 Optimize through cross-validation <![CDATA[f i > Frequency adjustment parameter 10Hz, 20Hz Set as the harmonic of the fundamental frequency of the blowing signal

[0121] Preferably, the weight coefficient w is dynamically adjusted in the following manner i to adapt to different detection environments.

[0122] The real-time signal-to-noise ratio SNR(m) is determined according to the following formula: where SNR(m) represents the real-time signal-to-noise ratio, ΔE(m) represents the normalized signal energy difference, m is the block index, and σ NIt represents the noise standard deviation, which is obtained through the environmental noise estimation module. ∝ is the proportionality symbol. The larger the absolute value of ΔE(m), the more significant the signal energy change and the higher the signal-to-noise ratio SNR.

[0123] The calculation formula of ΔE(m) is:

[0124]

[0125] Among them, σ E represents the moving average, and μ E represents the moving standard deviation. Differential feature extraction can standardize the energy change, improve the robustness of the features, make them not affected by the absolute energy level, and thus capture the energy change more effectively. By calculating the mean and standard deviation through a sliding window, the energy features are normalized in real time, enhancing the anti-interference ability of the system.

[0126] E l (m) is calculated as follows:

[0127]

[0128] Among them, L is the block length (such as 256 samples), which is normalized by the moving average and standard deviation to enhance the robustness of the detection system. ΔE(m) is the normalized signal energy difference based on discrete signal blocks, used for normalized feature extraction, and enhances the anti-interference ability through statistical normalization. The core role of ΔE(m) is to normalize the energy change through the moving average and standard deviation, applicable to the input of probability models (such as the calculation of P blow (t)) and the robust decision-making in complex environments (such as when dynamically adjusting the threshold), which can eliminate the influence of the absolute energy level and adapt to the dynamic environment (such as the change of the signal-to-noise ratio). It is not suitable for real-time mutation detection because its calculation depends on block statistics (such as 256 samples / block), which will introduce delay.

[0129] Scale w proportionally according to the real-time signal-to-noise ratio SNR(m) i : w i ←w i ·(1 + α·SNR(m)), where 1 + α·SNR(m) is the adjustment factor, used to dynamically scale the weight coefficient according to the signal-to-noise ratio; 1 ensures that the weight coefficient at least maintains its original value and will not decrease due to a negative signal-to-noise ratio (although the signal-to-noise ratio is usually non-negative), α represents the proportionality coefficient, which determines the influence degree of the signal-to-noise ratio on the weight coefficient, and α can be adjusted according to the requirements of system design. SNR represents the signal-to-noise ratio, that is, the ratio of the signal power to the noise power. In practical applications, multiple groups of blowing data can be collected in a quiet environment, and w i can be fitted by the least squares method. In addition, the weight coefficient w i can also be dynamically optimized through an online learning algorithm (such as LMS adaptive filtering).

[0130] In this embodiment, the weight coefficient w is dynamically adjusted. i The parameter meanings of the related formulas involved are shown in Table 2.

[0131] Table 2 Parameter Meanings in the Related Formulas for Dynamically Adjusting the Weight Coefficient

[0132] Parameter Define Describe <![CDATA[E l (m)]]> The energy of the m-th block Represents the sum of the signal energy from the time point mL to (m + 1)L - 1 m Block index Used to identify the currently processed signal block L Block length The number of samples contained in each signal block n Sample index The sample position in the signal sequence x(n) Signal sample value The amplitude of the input signal at the time point n

[0133] Since the SNR signal-to-noise ratio needs to rely on the block statistical results, the formula based on ΔE(m) can assist in explaining the data calculation source of the signal-to-noise ratio in the dynamic adjustment of the weight coefficient. The direct input of the probability model is the real-time feature ΔE(t), while ΔE(m) is used for system-level parameter optimization (such as weight adjustment) and does not participate in the probability calculation. That is, the signal-to-noise ratio SNR may rely on the block statistical results of ΔE(m), thereby indirectly affecting the output of P blow (t).

[0134] As described above, the low-frequency signal detection method for blowing identification proposed by the present invention proposes a dual decision-making mechanism combining energy mutation detection and duration verification, which can significantly improve the accuracy of blowing identification and the robustness of the system. Energy mutation detection can quickly capture the energy drop caused by the blowing operation, while duration verification further confirms the persistence of the blowing signal, ensuring the effectiveness and reliability of the blowing signal. This collaborative working method not only reduces the false trigger rate but also enhances the stability of the system in a complex acoustic environment.

[0135] For example, in a specific embodiment, the joint determination condition can be set as follows:

[0136]

[0137] Where:

[0138] Trigger off (t) represents that the blowing signal is detected by triggering;

[0139] is an indicator function used to determine whether the condition is satisfied;

[0140] Δt blow is the duration of the blowing signal, used to ensure the persistence of the blowing signal;

[0141] [T min ,T max is the duration range of the blowing signal, determined by experimental statistics, and the typical value is 100 - 300 ms.

[0142] Through this collaborative working mechanism, the present invention can, while capturing an energy mutation, verify whether its duration conforms to the characteristics of a blowing signal, thereby effectively distinguishing the blowing signal from environmental noise or other interference signals. This dual-verification mechanism significantly improves the accuracy of blowing signal recognition and the robustness of the system.

[0143] In an embodiment of the present invention, a method for controlling an electronic device is provided. Based on the low-frequency signal detection method applicable to blowing signal recognition described in any one of the above, it is determined whether there is a low-frequency signal based on a blowing signal in an audio signal. If so, the electronic device is controlled to switch from a first state to a second state, where the second state is different from the first state.

[0144] In practical applications, the electronic device can be electronic lights such as electronic candles and electronic audio devices. For the first state and the second state, one of the states can be off and the other can be on, or they can be two states with different brightness, display patterns, or light colors.

[0145] In an embodiment of the present invention, an electronic device is provided. The electronic device includes a controller configured to execute the method for controlling an electronic device described in the above embodiment. The electronic device is an electronic light and / or an electronic audio device, and the electronic light includes an electronic candle and an electronic light stick.

[0146] In this embodiment, as Figure 5 shown, the electronic device includes an electronic device body, an audio detection module, a control module, and a visual feedback module;

[0147] Among them, the audio detection module includes an audio acquisition module and a signal processing module. The audio acquisition module is configured to acquire an audio signal at a sampling frequency covering a target frequency band to obtain the audio signal to be detected. The signal processing module is configured to determine whether there is a low-frequency signal based on a blowing signal in the audio signal to be detected and transmit the detection result to the control module;

[0148] The control module is configured to control the state of the electronic device body to change or not change according to the detection result, including:

[0149] If the detection result is that there is a low-frequency signal based on a blowing signal in the audio signal to be detected, the state of the electronic device body is controlled to change; otherwise, the state of the electronic device body is not controlled to change;

[0150] The visual feedback module includes a display screen. If the detection result is that there is a low-frequency signal based on a blowing signal in the audio signal to be detected, the display screen is controlled to display a pattern according to a preset rule.

[0151] In a specific embodiment of the present invention, the hardware configuration of the electronic device is shown in Table 3.

[0152] Table 3 Hardware Configuration of the Electronic Candle

[0153] Module Parameter Example configuration Microphone Omnidirectional microphone Frequency response: 20Hz - 20kHz ADC ≥12-bit resolution Sampling rate: 44.1kHz Main control chip Microcontroller supporting FFT calculation Operating frequency ≥100MHz

[0154] The performance test data of the electronic device are shown in Table 4.

[0155] Table 4 Performance Test Data of the Electronic Candle

[0156] Index Test condition Result Detection accuracy Target frequency sine wave, 60dB ambient noise >99％ False trigger rate Voice and music interference (<10kHz) <0.5％ Response delay From the acoustic wave input to the execution trigger ≤200ms

[0157] The parameter optimization range of the electronic device is shown in Table 5.

[0158] Table 5 Parameter Optimization Range of the Electronic Device

[0159] Parameter Meaning Preferred range <![CDATA[F s > Sampling frequency 8k - 48kHz N Number of FFT points 256-1024 Δf Frequency tolerance 50 - 500Hz M Continuous detection times 5 - 20 times <![CDATA[T c > Cooling time 500 - 2000ms

[0160] The present invention relates to the control technology of low-frequency signal detection and intelligent electronic devices. Through dual decision logic, high-robustness non-contact interactive control is realized, which is especially suitable for the high-precision triggering requirements in scenarios such as electronic candles and intelligent lighting.

[0161] In a specific embodiment of the present invention, the electronic device is an electronic candle with an LED screen. When blowing air on the electronic candle, the microphone detects the low-frequency signal generated by the blowing air, and then controls the electronic candle to go out. The specific extinguishing method can be to gradually fade out from top to bottom in a simple way of combining row scanning masks with non-linear brightness on the LED screen, or to gradually dim to extinction in a specific pattern, or to play music while extinguishing the electronic candle.

[0162] It should be noted that the above-mentioned electronic device control method, electronic device embodiment and low-frequency signal detection method embodiment applicable to blowing air recognition are based on the same inventive concept. By reference, all the contents of the low-frequency signal detection method embodiment applicable to blowing air recognition are incorporated into the electronic device control method and electronic device embodiment.

[0163] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprising", "including" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the said element.

[0164] The above are only specific embodiments of the present application. It should be pointed out that for those of ordinary skill in the art, without departing from the principle of the present application, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present application.

Claims

1. A low-frequency signal detection method applicable to blowing recognition, characterized in that, To determine whether there is a low-frequency signal based on a blowing signal in an audio signal, the following steps are included: Collect the audio signal at a sampling frequency covering the target frequency band to obtain the audio signal to be detected; Determine the signal energy E of the target frequency band corresponding to the t-th moment in the audio signal to be detected low (t); According to the signal energy E low (t) of the target frequency band, calculate the signal energy difference ΔE(t) within the time difference interval, ΔE(t) = E low (t) - E low (t - Δt), where Δt is the preset time difference interval; Determine whether the following conditions are met: △E(t) < -Γ l and the duration for which △E(t) < 0 meets a preset requirement, where Γ l is an energy difference threshold, Γ l > 0. If the conditions are not met, it is determined that the audio signal to be detected does not include the low-frequency signal.

2. The low-frequency signal detection method applicable to blowing identification according to claim 1, wherein The following steps are further included: If the following conditions are met: △E(t) < -Γ l and the duration for which △E(t) < 0 meets the preset requirements, it is determined that the audio signal to be detected includes the low-frequency signal.

3. The low-frequency signal detection method applicable to blowing identification according to claim 1, characterized in that, The following steps are further included: If the following condition is met: ΔE(t) < -Γ l and the duration for which ΔE(t) < 0 meets a preset requirement, then calculate the blowing probability. If the blowing probability is greater than a preset probability value, it is determined that the audio signal to be detected includes the low-frequency signal; otherwise, it is determined that the audio signal to be detected does not include the low-frequency signal.

4. The low-frequency signal detection method applicable to blowing identification according to claim 3, characterized in that Calculate the blowing probability in the following manner: Among them, P blow (t) represents the blowing probability, σ(·) represents the probability conversion function, and φ i (ΔE(t)) is the feature extraction function, and w i represents the weight coefficient.

5. The low-frequency signal detection method applicable to blowing recognition according to claim 4, characterized in that, The probability conversion function adopts the Sigmoid function, which is expressed as follows: where e represents the exponent and x is the independent variable; and / or The feature extraction function uses the Gabor basis function, which is defined as: where x is the independent variable, μ i is the central position parameter, f i is the frequency adjustment parameter, ε i is the bandwidth adjustment parameter, cos(·) is the cosine function, π is the pi, i takes 1, 2, 3, and ψ i is the phase parameter.

6. The low-frequency signal detection method applicable to blowing recognition according to claim 4, characterized in that It also includes dynamically adjusting the weight coefficient w in the following manner i :[[]]END]] Determine the real-time signal-to-noise ratio SNR(m) in the following manner: Among them, SNR(m) represents the real-time signal-to-noise ratio, ΔE(m) represents the normalized signal energy difference, m is the block index, and σ N represents the noise standard deviation, and ∝ is the proportional symbol; Scale \(w\) proportionally according to the real-time signal-to-noise ratio \(SNR(m)\) based on the following formula i : \(w\) i ← \(w\) i ·(1 + α·SNR(m)), where α represents the proportionality coefficient.

7. The low-frequency signal detection method applicable to blowing recognition according to claim 1, wherein Judge whether the duration of △E(t) < 0 meets the preset requirements in the following way: If N S ≥N th , it is judged that the duration of △E(t) < 0 meets the preset requirements, where N S represents the number of consecutive frames satisfying ΔE(t) < 0, and N th represents the preset number-of-frames threshold; Or Judge whether the duration of △E(t) < 0 meets a preset requirement in the following manner: Where k is the frame index, K is the preset threshold of the continuous frame number, z is the current starting frame, I(.) is the indicator function, which takes the value of 1 when the condition is satisfied and 0 otherwise, and β is the proportion coefficient, 0.5 ≤ β ≤ 1.

8. The low-frequency signal detection method applicable to blowing recognition according to claim 1, wherein Calculate the signal energy E of the target frequency band in the audio signal to be detected at time t in the following way low (t): where X w (k,t) is the frequency-domain representation after windowing the audio signal to be detected, with dimensions of frequency point k × time frame t; SUM is the total number of FFT points, corresponding to 1 / 4 bandwidth.

9. The low-frequency signal detection method applicable to blowing recognition according to claim 1, wherein The following steps are further included: Use a sliding window to determine the signal energy difference △E(t) within multiple time difference intervals in the audio signal to be detected.

10. A method for controlling an electronic device, characterized in that, Based on the low-frequency signal detection method for blowing recognition according to any one of claims 1 to 9, determine whether there is a low-frequency signal based on a blowing signal in the audio signal. If so, control the electronic device to switch from the first state to the second state, and the second state is different from the first state.

11. An electronic device, characterized in that, The electronic device includes a controller, the controller is configured to execute the electronic device control method according to claim 10, the electronic device is an electronic lamp and / or an electronic sound, and the electronic lamp includes an electronic candle and an electronic light stick.

Images