Active noise reduction system, control method thereof, and abnormal sound detection method and device

Abstract

The disclosure provides an active noise reduction system and a control method thereof, an abnormal sound detection method and device. The abnormal sound detection method comprises: acquiring a first noise signal in the operation process of the active noise reduction system; the first noise signal comprises a noise signal of a noise source and / or a control signal of the loudspeaker; performing Fourier transform on the first noise signal, and determining a total spectral power amplitude and a noise peak value of the first noise signal according to the Fourier transform result; and performing abnormal sound detection on the active noise reduction system according to the total spectral power amplitude and the noise peak value. The disclosure determines the total spectral power amplitude and the noise peak value related to the abnormal sound in the first noise signal by performing spectral analysis on the first noise signal, and then accurately identifies the abnormal sound according to the total spectral power amplitude and the noise peak value, thereby improving the abnormal sound detection accuracy of the active noise reduction system.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the field of active noise reduction technology, and in particular to an active noise reduction system, a control method thereof, and a method and device for detecting abnormal sound. BACKGROUND

[0002] In an active noise reduction system, the noise reduction system utilizes the superimposed interference effect of the loudspeaker control signal (anti-phase noise) generated according to the input signal (including the noise signal collected and the feedback signal) and the noise signal to achieve the reduction of the noise signal, thereby achieving the purpose of noise reduction. However, in this process, when the feedback signal is greater than the noise signal, a howling phenomenon occurs. Even if the feedback signal is less than the noise signal, when the ratio of the feedback signal to the noise signal is greater than a threshold value, some stable single-frequency abnormal sound will still be emitted. The amplitude of the stable single-frequency abnormal sound is much smaller than the noise when howling occurs, and there is no obvious difference in the amplitude of the loudspeaker control signal generated when the system is normally working and the control signal generated when the stable single-frequency abnormal sound exists. Therefore, through the traditional howling monitoring scheme based on the amplitude of the loudspeaker control signal in the prior art, the stable single-frequency abnormal sound phenomenon cannot be effectively identified. SUMMARY

[0003] The technical problem to be solved by the present disclosure is to overcome the above-mentioned defects in the prior art, and to provide an active noise reduction system, a control method thereof, and a method and device for detecting abnormal sound.

[0004] The present disclosure solves the above-mentioned technical problem by the following technical scheme:

[0005] In a first aspect, a method for detecting abnormal sound is provided, which is applied to an active noise reduction system including a microphone and a loudspeaker. The method includes:

[0006] obtaining a first noise signal in the running process of the active noise reduction system; the first noise signal includes a noise signal of a noise source collected by the microphone and / or a control signal of the loudspeaker;

[0007] performing Fourier transform on the first noise signal, and determining a total spectral power amplitude and a noise peak value of the first noise signal according to the Fourier transform result;

[0008] detecting abnormal sound of the active noise reduction system according to the total spectral power amplitude and the noise peak value.

[0009] Optionally, detecting abnormal sound of the active noise reduction system according to the total spectral power amplitude and the noise peak value includes:

[0010] in response to a difference between the total spectral power amplitude and the noise peak value being less than a difference threshold value, determining that the active noise reduction system has stable single-frequency abnormal sound;

[0011] The difference threshold is determined by the following manner:

[0012] Obtaining abnormal samples and normal samples; the abnormal samples are noise signals containing abnormal sounds, and the normal samples are noise signals not containing abnormal sounds;

[0013] Constructing a target function; the target function is constructed according to at least one of the following parameters: the number of missed abnormal samples, the number of false normal samples, the proportion of missed abnormal samples, and the proportion of false normal samples;

[0014] Optimizing the target function in a preset optimization range of the difference threshold to determine the final difference threshold.

[0015] Optionally, the preset optimization range is [1dB, 10dB];

[0016] And / or, the optimization algorithm includes at least one of the following algorithms: genetic algorithm, simulated annealing algorithm, particle swarm algorithm, Bayesian optimization algorithm, hill climbing algorithm;

[0017] And / or, the target function is the ratio of the number of abnormalities to the total number of samples, and the number of abnormalities is the sum of the number of missed abnormal samples and the number of false normal samples.

[0018] Optionally, if the frequency spectrum of the first noise signal contains at least two peaks, the largest peak is determined as the noise peak;

[0019] And / or, the target sampling frequency is determined according to the abnormal sound frequency and / or the computing power of the analysis device of the first noise signal; the abnormal sound frequency is determined according to the noise sample containing abnormal sound;

[0020] And / or, the first noise signal is subjected to Fourier transform with a target frequency resolution; wherein the target frequency resolution is obtained by optimization, and the optimization target is to accurately identify abnormal sound and minimize the calculation amount of abnormal sound detection.

[0021] Optionally, determining the total amplitude of the frequency spectrum power and the noise peak of the first noise signal according to the Fourier transform result includes: performing A-weighting processing on the Fourier transform result, and determining the total amplitude of the frequency spectrum power and the noise peak of the first noise signal according to the A-weighting processing result;

[0022] And / or, the Fourier transform result is divided into at least two noise segments, and the total amplitude of the frequency spectrum power and the noise peak are determined according to the integration of each noise segment;

[0023] And / or, in response to the difference between the total amplitude of the spectral power and the noise peak being less than a difference threshold, determining that the active noise reduction system has a stable single-frequency abnormal sound, comprising: in response to the number of times that the difference between the total amplitude of the spectral power and the noise peak is less than the difference threshold being greater than a number threshold, determining that the active noise reduction system has a stable single-frequency abnormal sound.

[0024] Optionally, before obtaining the first noise signal during the operation of the active noise reduction system, further comprising:

[0025] obtaining a second noise signal of the noise source when the active noise reduction system is not started;

[0026] segmenting the second noise signal so that the power proportion of each noise segment of the second noise signal is less than a proportion threshold, and determining a segmentation strategy of the second noise signal;

[0027] performing Fourier transform on the first noise signal, comprising:

[0028] segmenting the first noise signal according to the segmentation strategy, and performing Fourier transform on the noise segment obtained by segmentation.

[0029] Optionally, before obtaining the first noise signal during the operation of the active noise reduction system, further comprising:

[0030] obtaining a second noise signal of the noise source when the active noise reduction system is not started;

[0031] performing octave calculation on the second noise signal to determine a segmentation strategy matched with the second noise signal;

[0032] segmenting the second noise signal according to the segmentation strategy, and calculating characteristic values of each noise segment of the second noise signal; the characteristic values include power and / or decibel value;

[0033] determining a target gain of a first target noise segment with a characteristic value greater than a characteristic value threshold; the characteristic value of the first target noise segment after filtering the first target noise segment by a band-pass filter with the target gain is less than or equal to the characteristic value threshold;

[0034] performing Fourier transform on the first noise signal, comprising:

[0035] segmenting the first noise signal to obtain at least two noise segments, filtering a second target noise segment contained in the first noise signal by a band-pass filter with a target gain, and performing Fourier transform on the noise segments obtained by segmentation and the filtered noise segments in the first noise signal respectively; wherein the frequency band of the second target noise segment contains the frequency band of the first target noise segment.

[0036] In a second aspect, a control method of an active noise reduction system is provided, comprising:

[0037] The abnormal noise detection method according to any one of the first aspect detects abnormal noise of the active noise reduction system;

[0038] In response to the fact that the active noise reduction system has stable single-frequency abnormal noise, the active noise reduction system is restarted.

[0039] In a third aspect, an abnormal noise detection device is provided, which is applied to an active noise reduction system, and the active noise reduction system comprises a microphone and a speaker; the abnormal noise detection device is used to implement the abnormal noise detection method according to any one of the first aspect; and the abnormal noise detection device comprises:

[0040] An acquisition module is configured to acquire a first noise signal in the running process of the active noise reduction system; the first noise signal is a noise signal of a noise source collected by the microphone or a control signal of the speaker;

[0041] A determination module is configured to perform Fourier transform on the first noise signal, and determine a total spectral power amplitude and a noise peak value of the first noise signal according to the Fourier transform result;

[0042] A detection module is configured to determine that the active noise reduction system has stable single-frequency abnormal noise in response to the fact that a difference between the total spectral power amplitude and the noise peak value is less than a difference threshold.

[0043] In a fourth aspect, an active noise reduction system is provided, which comprises a memory, a processor, and a computer program stored in the memory and used to run on the processor, and the processor implements the method according to any one of the first aspect or the second aspect when executing the computer program.

[0044] On the basis of common sense in the art, the above-mentioned preferred conditions can be combined arbitrarily, that is, each preferred example of the present disclosure is obtained.

[0045] The positive progress effect of the present disclosure is that, by performing spectral analysis on the first noise signal, the total spectral power amplitude and the noise peak value related to abnormal noise in the first noise signal are determined, and then the abnormal noise can be accurately identified according to the total spectral power amplitude and the noise peak value, thereby improving the abnormal noise detection accuracy of the active noise reduction system. BRIEF DESCRIPTION OF DRAWINGS

[0046] Figure 1 A flowchart of an abnormal noise detection method provided for an exemplary embodiment of the present disclosure is provided;

[0047] Figure 2A frequency domain signal curve schematic diagram for stabilizing single-frequency abnormal sound, howling, and normal noise is provided for an exemplary embodiment of the present disclosure.

[0048] Figure 3 A time domain signal curve schematic diagram for stabilizing single-frequency abnormal sound, howling, and normal noise is provided for an exemplary embodiment of the present disclosure.

[0049] Figure 4 A flowchart of difference threshold optimization in an abnormal sound detection method is provided for an exemplary embodiment of the present disclosure.

[0050] Figure 5 A curve effect comparison schematic diagram of A weighting processing or not in an abnormal sound detection method is provided for an exemplary embodiment of the present disclosure.

[0051] Figure 6 A flowchart of another abnormal sound detection method is provided for an exemplary embodiment of the present disclosure.

[0052] Figure 7 A flowchart of another abnormal sound detection method is provided for an exemplary embodiment of the present disclosure.

[0053] Figure 8 A module schematic diagram of an abnormal sound detection device is provided for an exemplary embodiment of the present disclosure.

[0054] Figure 9 A module schematic diagram of an active noise reduction system is provided for an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION

[0055] The present disclosure will be further described below by way of examples, but the present disclosure is not limited in the scope of the examples.

[0056] In the embodiments of the present disclosure, the prefix words such as “first”, “second” are merely used to distinguish different description objects, and have no limiting effect on the position, order, priority, quantity or content of the described objects. The use of ordinal words such as ordinal words in the embodiments of the present disclosure does not constitute a limitation on the described objects, and the description of the described objects should be referred to the description of the context in the claims or embodiments, and should not constitute an unnecessary limitation because of the use of such prefix words. In addition, in the description of the embodiments, unless otherwise specified, the meaning of “plurality” is two or more.

[0057] Figure 1 A flowchart of an abnormal sound detection method is provided for an exemplary embodiment of the present disclosure, which is applied to an active noise reduction system including devices such as a loudspeaker, a controller, and a microphone.

[0058] The abnormal noise detected in this embodiment refers to a noise emitted by the speaker of the active noise cancellation system due to certain faults or interference during operation, which affects the user experience. Abnormal noise includes howling and stable single-frequency abnormal noise (or stable single-frequency sound).

[0059] The characteristics of howling during active noise cancellation are: see Figure 2 and Figure 3 The noise spectrum exhibits a prominent peak, with the amplitude at its highest point significantly higher than adjacent frequencies. This peak is generally much larger than the peak of a stable single-frequency abnormal sound, accounting for a large portion of the total noise power. From the moment it is generated, the power amplitude in the time domain gradually increases, with the ratio of maximum power to minimum power generally exceeding 5 times, and the amplitude typically approaching or even exceeding the maximum output power of the loudspeaker itself. Figure 2 The second to fifth columns of the table refer to the amplitude (power) at 94.28Hz, 108.75Hz, 0Hz and 6000Hz respectively. The RMS data in the fifth column refers to the root mean square value of the amplitude from 0 to 6000Hz.

[0060] Unlike howling, stable single-frequency abnormal noise / stable single-frequency sound is characterized by a prominent peak in the noise spectrum. The amplitude of the peak (noise peak value) is higher than the amplitude of the corresponding adjacent frequency, and the power of this peak accounts for a relatively small proportion of the total noise power. That is, the power of the peak accounts for a proportion of the total noise power that is less than or equal to the power threshold, where the power threshold is determined based on experimental data. Figure 2 As shown by the blue curve, the bandwidth of this peak does not exceed 50Hz. See also Figure 3 In stable single-frequency noise, the power amplitude in the time domain is relatively stable from the beginning of its generation, with little power variation. The ratio of the maximum power to the minimum power will not exceed 3 times, and the amplitude is generally much smaller than the maximum sound power of the speaker itself. Figure 3 The left vertical axis represents the power amplitude, and the right vertical axis represents the sound pressure amplitude; the two can be converted to each other.

[0061] This disclosure primarily detects whether a stable single-frequency abnormal sound occurs in the active noise cancellation system. (See also...) Figure 1 The abnormal sound detection method includes the following steps:

[0062] Step 101: Obtain the first noise signal during the operation of the active noise cancellation system.

[0063] The first noise signal can be the noise signal emitted by the noise source collected by the first microphone. This noise signal will be used as the basis for abnormal sound detection. The first microphone is deployed near the noise source. It should be noted that the first noise signal can be the result of a single first microphone or a combination of results from multiple first microphones. In this case, the first noise signal collected by the first microphone is the superposition of the original noise signal emitted by the noise source and the feedback signal.

[0064] Active noise cancellation systems can be used to reduce noise in appliances such as range hoods and washing machines. Taking the noise reduction of a range hood as an example, the noise source can be, for example, the range hood's fan, and a first microphone for collecting noise signals is deployed near the fan.

[0065] The first noise signal can be the speaker's control signal, which will be used as the basis for subsequent abnormal noise detection. It should be noted that the speaker's control signal can be obtained directly from the controller's output, or it can be acquired through a second microphone deployed near the speaker. The first noise signal can be the control signal of a single speaker, or a combination of control signals from multiple speakers.

[0066] The first noise signal can also be a combination of the noise signal from the noise source and the control signal from the loudspeaker. The specific combination method is not particularly limited in the embodiments disclosed herein.

[0067] In one embodiment, the first noise signal meets the sampling frequency requirements and total sampling time requirements for Fourier transform processing, such as the sampling frequency f. s ≥2f U , where f U The highest analysis frequency (e.g., 12800Hz) of the active noise reduction system during noise signal acquisition, where the highest analysis frequency is greater than or equal to the highest possible frequency f of the abnormal noise. ANC For example, abnormal sounds may occur at frequencies of 700Hz, 800Hz, and 900Hz. ANC 900Hz, f U ≥900Hz. Total sampling time T ≥ duration of a single Fourier transform T FFT T FFT The frequency resolution Δf required for abnormal sound detection is determined, where Δf ≤ 100Hz. Preferably, Δf ≤ 20Hz.

[0068] The highest analysis frequency for active noise reduction refers to the target control frequency. Active noise reduction only analyzes noise within a specific frequency range and does not control noise outside that range. The maximum analysis frequency for active noise reduction control ranges from [600Hz to 2000Hz]. Preferably, the maximum analysis frequency for active noise reduction control is [600Hz to 1000Hz].

[0069] For example, in a scenario where an active noise cancellation system is used to reduce the noise of a range hood, the noise of the range hood is distributed in the range of 0 to 20000 Hz, while the frequency range of the analysis frequency of active noise cancellation may only be 0 to 1000 Hz. Correspondingly, the abnormal noise caused by the active noise cancellation system can only be in the range of 0 to 1000 Hz. Therefore, setting the highest analysis frequency of active noise cancellation to 1000 Hz is sufficient.

[0070] In one embodiment, a first noise signal during the operation of the active noise cancellation system is acquired at a target sampling frequency. The target sampling frequency is determined based on the abnormal sound frequency and / or the computing power of the analysis device for the first noise signal; the abnormal sound frequency is determined based on noise samples containing abnormal sounds. The noise samples can be obtained experimentally or by superimposing abnormal sounds onto noise samples without abnormal sounds. The analysis device for the first noise signal is the device that analyzes the first noise signal to achieve abnormal sound detection, and may be, for example, an active noise cancellation chip.

[0071] The following describes one method for determining the target sampling frequency:

[0072] The method for determining the minimum target sampling frequency: By analyzing noise samples containing abnormal sounds, it can be seen that the maximum frequency (abnormal sound frequency) of the abnormal sound is generally f. ANC According to the Nyquist sampling theorem, to avoid aliasing, the target sampling frequency should be at least twice the frequency of the different audio frequencies. Therefore, the minimum value of the target sampling frequency f is... s_min =2f ANC .

[0073] The method for determining the maximum value of the sampling frequency: Analyzing the first noise signal includes performing a Fourier transform on it. The computing power of the analysis device can be characterized by the number of data points of the first noise signal in each Fourier transform. The maximum value of the target sampling frequency can be derived from the maximum computing power. Assuming the analysis device allows N data points to be transformed in one Fourier transform, the time of one Fourier transform is 1 / Δf, and the sampling frequency f... s Indicates f per second s Therefore, the number of data points allowed for a single Fourier transform is f. s ·1 / Δf, i.e., N=f s• 1 / Δf, assuming the maximum computing power of the analysis device is N max , then f s_max = N max · Δf.

[0074] Therefore, the target sampling frequency f s_res of the first noise signal is in the range of [f s_min , f s_max ]. In the related art, in order to improve the control accuracy of active noise reduction, the original signal collected by the active noise reduction controller has a high sampling frequency f s , and f s is usually 12000Hz. In this embodiment, the target sampling frequency f s res The first noise signal is obtained.

[0075] In this embodiment, the first noise signal is obtained at the target sampling frequency determined according to the tonal frequency and / or the computing power of the analysis device, which serves as the data basis for tonal detection. On the one hand, it can provide data basis conforming to the tonal characteristics for subsequent steps, eliminate interference, and thus improve the tonal detection accuracy. On the other hand, the first noise signal obtained at the target sampling frequency eliminates part of the interference signal, reduces the data analysis amount, and thus can improve the tonal detection rate and meet the timeliness requirement.

[0076] In step 102, the first noise signal is subjected to Fourier transform, and the frequency spectrum power total amplitude and the noise peak value of the first noise signal are determined according to the Fourier transform result.

[0077] The first noise signal is subjected to Fourier transform to obtain the frequency spectrum data (Fourier transform result) of the first noise signal. Through analysis of the frequency spectrum data, it is found that when the howling and stable single-frequency tonal phenomenon occurs, there is a signal peak value obviously higher than the surrounding frequencies in the frequency spectrum, as shown by the peak value near the red dotted line 95Hz and the peak value near the blue thin solid line 109Hz in FIG. 1. Through the filter playback mode, it is found that the noise corresponding to the peak value frequency is the sound that produces howling and stable single-frequency tonal. The normal signal (the green thick solid line in FIG. 1) has no such obvious noise peak value, so the frequency spectrum analysis of the first noise signal can accurately identify the tonal. Figure 2 Figure 2

[0078] In one embodiment, in order to improve the algorithm rate, the first noise signal is subjected to Fourier transform by using fast Fourier transform (FFT).

[0079] In one embodiment, the number of signal data points for one FFT processing is selected to be T FFT / f s_res ​​the values of D N , N is a positive integer, and ensure that the selected N satisfies 2 N ≥ T FFT / f s_res . The final spectral data D FFT select the results of a FFT, or power average of multiple FFT results. The total power of the spectral power E FFT is the total power of D A . The maximum peak value E P in the spectrum is the maximum value of the entire FFT spectrum analysis data.

[0080] In one embodiment, if the spectrum of the first noise signal contains at least two peaks, the largest peak is determined as the noise peak.

[0081] In one embodiment, the first noise signal is subjected to Fourier transform at a target frequency resolution, and the active noise reduction system is subjected to tone detection according to the Fourier transform result; wherein the target frequency resolution is obtained by optimization, and the optimization target is to accurately identify the tone and minimize the calculation amount of tone detection.

[0082] Said accurate identification of the tone means that the identification accuracy is greater than an accuracy threshold and / or the misidentification rate is less than or equal to a misidentification threshold. The accuracy threshold and the misidentification threshold can be set according to actual conditions.

[0083] The frequency resolution determines the calculation amount of the tone detection process. When the frequency resolution is 1 Hz, it means that a Fourier transform is performed every 1 s. Assuming that the calculation amount of each time is the data amount within 1 s, K1, when the frequency resolution is increased to 4 Hz, the calculation amount of each time is the data amount within 1 / 4 s, K2, K1=4·K2. The data amount when the frequency resolution is 4 Hz is 4 times less than that when the frequency resolution is 1 Hz, greatly accelerating the calculation speed of tone detection.

[0084] With the increase of the frequency resolution, the amplitude-frequency graph amplitude after Fourier transform will increase, that is, when the frequency resolution increases, the number of spectral lines decreases, and the power of each spectral line is the sum of the power of the spectral lines before the increase of the frequency resolution.

[0085] The following example further illustrates the process of determining the target frequency resolution:

[0086] Based on the tone sample, it is known that the frequency band where the tone occurs, such as the stable single-frequency tone at frequencies fa, fb, and fc in the tone sample.

[0087] S1. Assuming that the initial frequency resolution is 1 Hz, increase the frequency resolution to 32 Hz, perform FFT on the tone sample, and perform tone detection according to the FFT.

[0088] S2. If the frequency resolution is increased to 32 Hz, it is sufficient to determine the abnormal sound. If the frequency resolution of 32 Hz cannot accurately identify the abnormal sound in the abnormal sound sample at all frequencies fa, fb, fc and / or misjudgment (for example, identifying other frequencies as abnormal sound), the frequency resolution is increased to 32, which reduces the calculation amount, but at this time it is not enough to determine all the active noise reduction abnormal sound points, and the frequency resolution needs to be reduced.

[0089] S3. If the frequency resolution is sufficient to determine the abnormal sound, select 32 Hz as the frequency resolution, and determine the calculation amount at this time. The calculation amount is reduced to about 1 / 32 compared to the initial calculation amount when the frequency resolution is 1 Hz. If the frequency resolution is not sufficient to determine all the active noise reduction abnormal sound points, repeat step 2 until the frequency resolution that accurately identifies the abnormal sound and has the minimum abnormal sound detection calculation amount is found.

[0090] In this embodiment, the target frequency resolution obtained by optimization is used to perform spectral analysis on the first noise signal, which can provide data basis conforming to the characteristics of the abnormal sound, exclude interference, and further improve the accuracy of abnormal sound detection; At the same time, the calculation amount of abnormal sound detection is the smallest, so as to improve the abnormal sound detection rate and meet the timeliness requirement.

[0091] Step 103, according to the total amplitude of the spectral power and the noise peak, the active noise reduction system is detected for abnormal sound.

[0092] In this embodiment, by performing spectral analysis on the first noise signal, the total amplitude of the spectral power and the noise peak related to the abnormal sound in the first noise signal are determined, and then the total amplitude of the spectral power and the noise peak are used to accurately identify the abnormal sound, thereby improving the accuracy of the active noise reduction system.

[0093] In one embodiment, in step 103, in response to the difference between the total amplitude of the spectral power and the noise peak being less than the difference threshold, it is determined that the active noise reduction system has stable single-frequency abnormal sound.

[0094] When the difference between the total amplitude of the spectral power and the noise peak is greater than or equal to the difference threshold, it means that the noise peak accounts for a large part of the entire noise and is an easily heard abnormal sound. This abnormal sound is likely to be caused by the instability of the active noise reduction system.

[0095] The difference threshold can be an empirical value, for example, 4 dB. When the active noise reduction system is working, if the total amplitude of the spectral power of the first noise signal E A is 65 dB(A), and the first noise signal is composed of all frequencies from 0 to 6000 Hz. Through spectral analysis, the maximum peak in the spectrum appears at 500 Hz, and the size of the noise peak E PIt is 62dB(A). E A and E P Convert dB to power: 65dB: Power 1 = 10 (0.1*65) =3162277; 62dB: Power² = 10 (0.1*62) =1584893; Power 2 / Power 1 ≈ 0.5 = 50%. At this point, the sound power at a frequency of 500Hz accounts for 50% of the total amplitude of the spectral power, while the combined sound power of other frequencies only accounts for 50%. Therefore, in the spectrum of this sound, the sound at 500Hz is significantly louder than the sounds at other frequencies. At this point, the active noise cancellation system will emit a very unpleasant single-frequency sound, i.e., an abnormal sound. If the sound is expressed in dB, then E A -E P =65-62=3dB<ΔE PT The presence of a stable single-frequency abnormal sound was confirmed.

[0096] The difference threshold can also be dynamically selected based on actual conditions to improve the accuracy of abnormal sound detection. Below is a method for dynamically selecting the difference threshold; see [link to relevant documentation]. Figure 4 The difference threshold is determined in the following way:

[0097] Steps 100-11: Obtain abnormal signal samples and normal samples.

[0098] Abnormal samples are noise signals containing abnormal sounds, while normal samples are noise signals without abnormal sounds. Abnormal samples can be collected experimentally or are the result of superimposing normal samples with noise signals whose frequency and amplitude match those of the abnormal sounds.

[0099] The number of abnormal and normal samples can be set according to actual needs. Understandably, the more samples there are, the better the difference threshold can be optimized.

[0100] Steps 100-12: Construct the objective function.

[0101] The objective function is constructed based on at least one of the following parameters: the number of missed abnormal samples, the number of false positives among normal samples, the percentage of missed abnormal samples, and the percentage of false positives among normal samples.

[0102] The design principle of the objective function is to find a combination and value of judgment parameters that allows the optimized difference threshold to detect all abnormal sounds while preventing normal signals from being misidentified as abnormal sounds, and also to have good robustness. The objective function for parameter optimization can be set as: (1) (number of missed abnormal samples + number of false normal samples) / total number of samples, where the smaller the objective function, the better; (2) (abnormal sound judgment intensity of normal samples) / (abnormal sound judgment intensity of abnormal signals), where the smaller the objective function, the better.

[0103] The abnormal sound judgment strength = the degree of abnormal sample / normal sample meeting the abnormal sound judgment. If the total abnormal sound threshold is 100, higher than 100 is abnormal sound, if the value of the normal sample is 30, the abnormal sound strength 0.3, which is the abnormal sound judgment strength of the normal sample, the smaller the value, the farther the normal sound from the abnormal sound standard, the greater the probability of the normal sound, and the higher the accuracy of the detection algorithm. If the value of the abnormal sound sample is 200, the abnormal sound strength 2, which is the abnormal sound judgment strength of the abnormal signal, the greater the value, the more likely the sound is abnormal sound, the greater the probability of the abnormal sound, and the higher the accuracy of the detection algorithm. Therefore, the smaller the value of (the abnormal sound judgment strength of the normal sample) / (the abnormal sound judgment strength of the abnormal signal), the higher the accuracy of the detection algorithm, and the better the corresponding parameter combination for detection.

[0104] The abnormal sound judgment strength can also be defined in other ways, for example: the abnormal sound judgment strength of the normal sample = the number of samples of the normal sample misjudged as abnormal sound / the total number of normal samples; the abnormal sound judgment strength of the abnormal signal = the number of samples of the abnormal sample accurately judged as abnormal sound / the total number of abnormal samples.

[0105] In addition to the above objective function, other calculation formulas similar to the design principles of the objective function can also be used, and the embodiments of the present disclosure are not particularly limited.

[0106] Step 100-13, optimizing the objective function in the preset optimization range of the difference threshold value to determine the final difference threshold value.

[0107] In one embodiment, the preset optimization range is [1dB, 10dB], that is, the difference threshold value is optimized in [1dB, 10dB] to find the optimal solution, so that the objective function is minimized.

[0108] In this embodiment, the optimization range of the difference threshold value is provided, so that the optimization meets certain physical characteristics of abnormal sound, and the optimization accuracy and efficiency can be improved.

[0109] In one embodiment, the optimization algorithm includes at least one of the following algorithms: genetic algorithm, simulated annealing algorithm, particle swarm algorithm, Bayesian optimization algorithm, hill climbing algorithm.

[0110] In one embodiment, in step 103, in response to the difference between the total spectral power amplitude E A and the noise peak E A-P being greater than or equal to the difference threshold value, it is determined that the active noise reduction system has a stable single-frequency abnormal sound; wherein the power difference E A-P is the difference between the total spectral power amplitude E A and the noise peak E P .

[0111] The difference threshold value in the embodiment can be an empirical value or dynamically selected by target optimization.

[0112] In one embodiment, in step 103, it is determined that the active noise reduction system has the stable single-frequency abnormal sound in response to the ratio of the first decibel value to the second decibel value being less than the ratio threshold value. The first decibel value is the dB conversion result of the total amplitude of the spectral power, and the second decibel value is the dB conversion result of the noise peak value.

[0113] Similar to the difference threshold value, the ratio threshold value ΔE PT may be an empirical value. If the first decibel value of the total amplitude of the spectral power of the first noise signal is 3.16x10 6 , and the second decibel value of the noise peak value is 1.58x10 6 , then ΔE A / E P = 3.16x10 6 / 1.58x10 6 = 2 < ΔE PT , it is determined that the stable single-frequency abnormal sound exists. The embodiment provides another judgment condition for abnormal sound detection, in which the ratio threshold value can be dynamically selected by target optimization to improve the accuracy of abnormal sound detection. The target optimization process of the ratio threshold value is similar to that of the difference threshold value, which will not be described here.

[0114] In one embodiment, the step of determining the total amplitude of the spectral power and the noise peak value of the first noise signal according to the Fourier transform result includes: performing A-weighting processing on the Fourier transform result, and determining the total amplitude of the spectral power and the noise peak value of the first noise signal according to the A-weighting processing result.

[0115] Currently, the definition of abnormal sound is mainly based on the subjective feeling of a person, so some sounds of some frequencies can be ignored even if they are loud, because the human ear is not sensitive to the sound of this frequency band (such as low-frequency sound). In the abnormal sound detection process, the Fourier transform result of the first noise signal is first A-weighted based on the sensitivity of the human ear, and then the A-weighted result is detected for abnormal sound. Since the A-weighting can better reflect the hearing perception of the human ear, this detection method is more likely to detect the real abnormal sound, has better detection effect, and is more in line with the feeling and judgment of the user on the abnormal sound.

[0116] Referring to Figure 5 , the green solid line represents the spectrum of the first noise signal without A-weighting processing, in which there are many low-frequency components that are not sensitive to the human ear below 100 Hz. The sound signal of the first noise signal after A-weighting is shown by the red dotted line in Figure 5 , and the total power ratio of the low-frequency components that are not sensitive to the human ear is reduced. The total amplitude of the spectral power of the first noise signal E A -E PThe value of the difference between the total amplitude of the spectrum power and the noise peak value is changed from 3.2 dB to 2.5 dB, which is more consistent with the tonal judgment E A -E P When the difference is less than 5 dB, the detection is more accurate, and more consistent with the feeling of the human ear.

[0117] It should be noted that, since it is easier to determine the tonal after A-weighting processing, the difference threshold value after A-weighting processing can be smaller than the difference threshold value without A-weighting processing, for example, the difference threshold value without A-weighting processing is 5 dB, and the difference threshold value after A-weighting processing is 4 dB.

[0118] In one embodiment, the tonal is detected by multiple judgments, and step 103 comprises: in response to the number of times that the difference between the total amplitude of the spectrum power and the noise peak value is less than the difference threshold value being greater than the number threshold value, determining that the active noise reduction system has a stable single-frequency tonal.

[0119] When the spectrum data of the first noise signal is used for tonal detection, when the collected first noise signal contains noise interference signal, or the first noise signal used for detection is not a complete steady-state signal, there is a certain fluctuation characteristic with time. These reasons cause the difference between the spectrum data obtained at different times, and further cause the misjudgment or wrong judgment phenomenon of tonal detection.

[0120] In order to reduce or avoid the misjudgment and wrong judgment phenomenon, in the embodiment, each time the difference between the total amplitude of the spectrum power and the noise peak value is less than the difference threshold value, it is determined that the active noise reduction system has an abnormality, and the number of abnormality occurrences is counted by 1, and when the number of occurrences is greater than the number threshold value, it is determined that the active noise reduction system has a stable single-frequency tonal. Through multiple judgments, the misjudgment and wrong judgment phenomenon can be effectively reduced or avoided.

[0121] The number threshold value can be set according to actual needs, which can be an empirical value, for example, set to 3 times; or dynamically selected by target optimization to improve the accuracy of tonal detection. The target optimization process of the number threshold value is similar to that of the difference threshold value, which will not be described here.

[0122] Many tonal frequencies have a certain fluctuation, and the fluctuation has a great influence on the amplitude of the corresponding frequency. When the spectrum is used for tonal recognition, there is misjudgment and omission. Moreover, the feeling and judgment of a person to a tonal is the overall feeling of the power in a frequency band, rather than the power size at a single frequency. Therefore, the noise spectrum is integrated according to a certain rule, and the power size of the corresponding frequency band after integration is used for tonal judgment, which is more consistent with the actual feeling of a person.

[0123] Based on this, in one embodiment, the Fourier transform result (spectrum) of the first noise signal is segmented and processed to obtain at least two noise segments, and the total amplitude of the spectrum power and the noise peak value are determined according to the integration of each noise segment. The total amplitude of the spectrum power and the noise peak value here are the total amplitude of the spectrum power and the noise peak value of the first noise signal globally.

[0124] The segmentation algorithm can be, but is not limited to, one of the following: 1 / n octave method, Bark bandwidth method, equal bandwidth method, etc. The first two are preferred, where n in 1 / n octave is a non-zero natural number such as 1, 2, 3, etc., taking into account detection accuracy and computing power, and n=3 is preferred.

[0125] In one embodiment, each noise segment is subjected to abnormal sound detection, specifically: the Fourier transform result (spectrum) of the first noise signal is segmented and processed, the total amplitude of the spectrum power and the noise peak value of each noise segment are calculated, and each noise segment is subjected to abnormal sound detection. The total amplitude of the spectrum power and the noise peak value here are the local total amplitude of the spectrum power and the noise peak value of each noise segment.

[0126] In one embodiment, before the Fourier transform result of the first noise signal is segmented and processed, the Fourier transform result is subjected to A-weighting processing, and then the A-weighting processing result is segmented.

[0127] The device equipped with the active noise reduction system, such as an extractor hood, may also produce sound with spectral characteristics similar to the stable single-frequency sound caused by the active noise reduction system, which we call "normal abnormal sound". This "normal abnormal sound" may also make users feel uncomfortable, but it is different from the stable single-frequency sound caused by the active noise reduction system and is not a "byproduct" of the active noise reduction system. It cannot be reduced by turning off or adjusting the active noise reduction system.

[0128] Based on this, in one embodiment, if the detection result of step 103 is that there is a stable single-frequency abnormal sound, in order to exclude that the abnormal sound is caused by the device equipped with the active noise reduction system rather than the active noise reduction system, the active noise reduction system is first turned off, a second noise signal during the process of turning off the active noise reduction system is obtained, the Fourier transform of the second noise signal is performed, the total amplitude of the spectrum power and the noise peak value are calculated, and it is determined whether there is still a stable single-frequency abnormal sound; if the determination result is that the second noise signal has a stable single-frequency abnormal sound and the abnormal sound frequency of the second noise signal is the same as or similar to that of the first noise signal, it is determined that the abnormal sound is caused by the device rather than the active noise reduction system; otherwise, it is determined that the active noise reduction system has a stable single-frequency abnormal sound.

[0129] Another implementation mode for excluding the interference of the abnormal sound of the device on the abnormal sound detection of the active noise reduction system is provided below.

[0130] In one embodiment, referring to Figure 6, before acquiring the first noise signal in the running process of the active noise reduction system, further comprising:

[0131] Step 100, acquiring a second noise signal of the noise source when the active noise reduction system is not started, and segmenting the second noise signal to make the power proportion of each noise segment of the second noise signal less than a proportion threshold, and determining a segmentation strategy of the second noise signal.

[0132] The second noise signal is collected by the first microphone.

[0133] The number of noise segments obtained by segmenting the first noise signal is determined according to the power proportion of each noise segment. If the power proportion of each noise segment is not less than the proportion threshold, re-segmentation is required. The frequency bands of each noise segment can be the same or different. Segmenting the second noise signal makes the power proportion of each noise segment of the second noise signal less than the proportion threshold, that is, separating the frequency band with high power of the range hood itself, so as to exclude the influence of the frequency band characteristics of the range hood itself on the abnormal sound judgment.

[0134] The proportion threshold can be set according to the actual situation. The value range of the proportion threshold can be [40%, 50%] for example. Preferably, the proportion threshold is 40%. The value 40% is taken to leave a margin to avoid misjudgment caused by being too close to 50%. This is because: since the criterion for judging abnormal sound in the subsequent process is that more than 50% of the proportion threshold is judged as stable single-frequency abnormal sound, since it cannot be guaranteed that the power proportion of a certain noise segment is an accurate value, there may be a fluctuation of 5% up and down, so if it is too close to 50%, it may cause misjudgment of abnormal sound due to fluctuation error. When the proportion threshold is set to 40%, if the power proportion is less than 40%, it means that the power of the noise segment at this time has been reduced to a low enough level to avoid misjudgment caused by power fluctuation of the noise segment.

[0135] The segmentation strategy includes the following parameters: the number of segments, the upper and lower limit frequencies of each noise segment.

[0136] The calculation method of the power proportion is introduced as follows:

[0137] H = 1 - e i / E;

[0138]

[0139] Wherein, H represents the power proportion; e i represents the power corresponding to the i-th noise segment in the noise signal; E represents the sum of the powers of each noise segment contained in the noise signal; X j (t) represents the amplitude of the j-th sampling point of the noise signal in the time domain; N represents the number of sampling points of the noise segment contained in the noise signal. e iThe calculation manner is similar to E.

[0140] In step 102, the first noise signal is segmented according to the segmentation strategy, and the noise segments obtained by segmentation are subjected to Fourier transform, or the Fourier transform results are segmented according to the segmentation strategy. Subsequently, the global spectral power total amplitude and the noise peak value of the first noise signal are determined according to the integral of each noise segment, or the spectral power total amplitude and the noise peak value of each noise segment, that is, the local spectral power total amplitude and the noise peak value, are determined.

[0141] In this embodiment, the abnormal sound of the active noise reduction system can be quickly monitored, the influence of the frequency band characteristics of the extractor itself on the abnormal sound judgment is excluded, and the active noise reduction abnormal sound of the extractor can be accurately monitored.

[0142] Another implementation manner of excluding the interference of the equipment abnormal sound on the active noise reduction system abnormal sound detection is provided below.

[0143] In one embodiment, referring to Figure 7 Before acquiring the first noise signal in the running process of the active noise reduction system, the method further includes:

[0144] Step 100-21, acquiring a second noise signal of the noise source when the active noise reduction system is not started.

[0145] Step 100-22, performing octave calculation on the second noise signal to determine a segmentation strategy matched with the second noise signal.

[0146] n in 1 / n octave is a non-zero natural number such as 1, 2, 3, etc., which takes into account the detection accuracy and computing power, and n=3 is preferred.

[0147] Step 100-23, segmenting the second noise signal according to the segmentation strategy, and calculating characteristic values of each noise segment of the second noise signal.

[0148] The characteristic values include power and / or decibel values.

[0149] Step 100-24, determining a target gain of a first target noise segment with a characteristic value greater than a characteristic value threshold.

[0150] The characteristic value of the first target noise segment after being filtered by a bandpass filter with the target gain is less than or equal to the characteristic value threshold.

[0151] The characteristic value of the first target noise segment is greater than the characteristic value threshold, indicating that the first target noise segment contains abnormal sound, and the active noise reduction system was not started when the first target noise segment was collected, meaning that the abnormal sound contained in the first target noise segment is caused by the device carrying the active noise reduction system, rather than by the active noise reduction system. In order to exclude device abnormal sound interference during abnormal sound detection, the second target noise segment contained in the first noise signal needs to be filtered to filter the abnormal sound part caused by the device. The frequency band of the second target noise segment contains the frequency band of the first target noise segment.

[0152] Taking the characteristic value as a decibel value as an example, it is assumed that the decibel value of the first target noise segment before adjustment is P, and the decibel value after adjustment is P' (less than or equal to the characteristic value threshold). Here, P' is set to be greater than the decibel value of the adjacent frequency band by 1 dB. According to the formula: P' = 20lg(E' / E0), the power E' of the first target noise segment after adjustment can be calculated. By knowing the power E of the first target noise segment before adjustment and the power E' after adjustment, the target gain / through rate of the first target noise segment is calculated as α = E' / E.

[0153] It should be noted that P' is set to be greater than the decibel value of the adjacent frequency band by 1 dB, rather than being equal to the reason that makes the abnormal sound monitoring more sensitive. The former is more sensitive to abnormal sound judgment than the latter. When the judgment condition is 3 dB higher than the adjacent frequency band, it needs less power to judge abnormal sound than when it is equal to the adjacent frequency band. At this time, the abnormal sound judgment sensitivity is higher while the accuracy is guaranteed.

[0154] In step 102, the first noise signal is segmented to obtain at least two noise segments, and a target gain band-pass filter is used to filter the second target noise segment contained in the first noise signal. The noise segments obtained by segmenting the first noise signal and the filtered noise segments are respectively subjected to Fourier transform.

[0155] Among them, the noise segments obtained by segmenting the first noise signal are other noise segments except the second target noise segment.

[0156] The target gain band-pass filter is used to filter the second target noise segment, that is, the amplitude of the second target noise segment is multiplied by α to filter the abnormal sound part caused by the device in the second target noise segment. The amplitudes of other noise segments in the second noise signal except the second target noise segment are not attenuated, that is, the power through rate of other noise segments is 100%.

[0157] In this embodiment, the target gain band-pass filter is used to filter the noise segments of the target noise segments in the first noise signal, which can exclude the interference of device abnormal sound on the abnormal sound detection of the active noise reduction system, and further improve the accuracy of abnormal sound detection.

[0158] The embodiments of the present disclosure further provide a control method of an active noise reduction system, which comprises: performing abnormal sound detection on the active noise reduction system according to the abnormal sound detection method provided in any of the above embodiments; and restarting the active noise reduction system in response to the existence of stable single-frequency abnormal sound in the active noise reduction system.

[0159] According to experience, the abnormal sound of the active noise reduction system is caused by some faults or interference, and restarting the active noise reduction system can weaken or eliminate the faults or interference, improve the effectiveness of the active noise reduction system, and further improve the user experience.

[0160] Corresponding to the above-mentioned abnormal sound detection method embodiments, the present disclosure further provides embodiments of an abnormal sound detection device.

[0161] Figure 8 A module schematic diagram of an abnormal sound detection device provided for an exemplary embodiment of the present disclosure is provided, and the device comprises:

[0162] The acquisition module 81 is configured to acquire a first noise signal in the running process of the active noise reduction system; the first noise signal comprises a noise signal of a noise source and / or a control signal of the loudspeaker;

[0163] The determination module 82 is configured to perform Fourier transform on the first noise signal, and determine a total spectral power amplitude and a noise peak value of the first noise signal according to the Fourier transform result;

[0164] The detection module 83 is configured to determine that the active noise reduction system has stable single-frequency abnormal sound in response to the difference between the total spectral power amplitude and the noise peak value being less than a difference threshold value.

[0165] Optionally, the detection module 83 is specifically configured to:

[0166] determine that the active noise reduction system has stable single-frequency abnormal sound in response to the difference between the total spectral power amplitude and the noise peak value being less than a difference threshold value.

[0167] The device further comprises the difference threshold value determination module, which is configured to:

[0168] acquire abnormal samples and normal samples; the abnormal samples are noise signals containing abnormal sound, and the normal samples are noise signals not containing abnormal sound;

[0169] construct a target function; the target function is constructed according to at least one of the following parameters: the number of missed abnormal samples, the number of false detection of normal samples, the proportion of missed abnormal samples, and the proportion of false detection of normal samples;

[0170] perform optimization on the target function in a preset optimization range of the difference threshold value to determine the final difference threshold value.

[0171] Optionally, the preset optimization range is [1dB, 10dB].

[0172] And / or, the optimization algorithm includes at least one of the following algorithms: genetic algorithm, simulated annealing algorithm, particle swarm algorithm, Bayesian optimization algorithm, hill climbing algorithm.

[0173] And / or, the target function is the ratio of the number of anomalies to the total number of samples, and the number of anomalies is the sum of the number of missed detection of abnormal samples and the number of misdetected normal samples.

[0174] Optionally, if the frequency spectrum of the first noise signal contains at least two peaks, the largest peak is determined as the noise peak value.

[0175] And / or, the target sampling frequency is determined according to the abnormal sound frequency and / or the computing power of the analysis device of the first noise signal; the abnormal sound frequency is determined according to the noise sample containing abnormal sound.

[0176] And / or, the first noise signal is subjected to Fourier transform with a target frequency resolution; wherein the target frequency resolution is obtained by optimization, and the optimization target is to accurately identify abnormal sound and minimize the calculation amount of abnormal sound detection.

[0177] Optionally, the determination module is specifically configured to: perform A-weighting processing on the Fourier transform result, and determine the total amplitude of the frequency spectrum power and the noise peak value of the first noise signal according to the A-weighting processing result.

[0178] And / or, the determination module is specifically configured to: divide the Fourier transform result into at least two noise segments, and determine the total amplitude of the frequency spectrum power and the noise peak value according to the integral of each noise segment.

[0179] And / or, the detection module is specifically configured to: in response to the occurrence frequency of the difference between the total amplitude of the frequency spectrum power and the noise peak value being less than a difference threshold value being greater than a frequency threshold value, determine that the active noise reduction system has stable single-frequency abnormal sound.

[0180] Optionally, the acquisition module is further configured to acquire a second noise signal of a noise source when the active noise reduction system is not started.

[0181] The device further comprises a segmentation module configured to segment the second noise signal to make the power proportion of each noise segment of the second noise signal less than a proportion threshold value, and determine a segmentation strategy of the second noise signal.

[0182] The determination module is specifically configured to: segment the first noise signal according to the segmentation strategy, and perform Fourier transform on the segmented noise segment.

[0183] Optionally, the acquisition module is further configured to acquire a second noise signal of the noise source when the active noise reduction system is not started.

[0184] The device further includes a calculation module configured to perform an octave calculation on the second noise signal to determine a segmentation strategy matching the second noise signal.

[0185] The segmentation module is configured to perform segmentation processing on the second noise signal according to the segmentation strategy and calculate feature values of each noise segment of the second noise signal; the feature values include power and / or decibel values.

[0186] The gain module is configured to determine a target gain of a first target noise segment with a feature value greater than a feature value threshold; and filter the first target noise segment with a band-pass filter of the target gain, so that the feature value of the first target noise segment after the filtering is less than or equal to the feature value threshold.

[0187] The determination module is specifically configured to:

[0188] segment the first noise signal to obtain at least two noise segments, filter a second target noise segment contained in the first noise signal with a band-pass filter of the target gain, and perform Fourier transform on the noise segments obtained by the segmentation and the filtered noise segment in the first noise signal, respectively; the frequency band of the second target noise segment contains the frequency band of the first target noise segment.

[0189] For the device embodiment, since it basically corresponds to the method embodiment, the related parts are described in the part of the method embodiment. The device embodiments described above are only illustrative, and the units described as separate components can or can not be physically separated, and the components of the unit can or can not be physical units, i.e., they can be located in one place or distributed on multiple network units. According to actual needs, part or all of the modules can be selected to achieve the purpose of the present disclosure.

[0190] Figure 9 A structural schematic diagram of an active noise reduction system is shown for an example embodiment of the present disclosure, which includes a memory, a processor, and a computer program stored on the memory and used to run on the processor, and the processor implements the heterophonic detection method or control method described in any of the above embodiments when executing the computer program. Figure 9 The displayed active noise reduction system 90 is only an example, and should not impose any limitation on the functions and use range of the embodiments of the present disclosure.

[0191] As Figure 9As shown, the components of the active noise reduction system 90 can include, but are not limited to, the at least one processor 91, the at least one memory 92, and a bus 93 that connects the various system components including the memory 92 and the processor 91.

[0192] The bus 93 includes a data bus, an address bus, and a control bus.

[0193] The memory 92 can include volatile memory, such as random access memory (RAM) 921 and / or cache memory 922, and can further include non-volatile memory, such as read-only memory (ROM) 923.

[0194] The memory 92 can also include a program tool 925 (or utility tool) having a set of (at least one) program modules 924, including but not limited to, an operating system, one or more application programs, other program modules, and program data, and each of such examples, or some combination thereof, can include implementation of a network environment.

[0195] The processor 91, through the operating system, executes interactive software applications and data processing, such as the sound anomaly detection method or control method provided by any of the embodiments described above, by running the computer programs stored in the memory 92.

[0196] The active noise reduction system 90 can also communicate with one or more external devices 94 (such as a range hood processor, etc.) via an input / output (I / O) interface 95. Further, the active noise reduction system 90 can communicate with one or more networks, such as a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet, via a network adapter 96. As shown, the network adapter 96 communicates with the other modules of the active noise reduction system 90 via the bus 93. It should be appreciated that although not shown, other hardware and / or software modules could be used in conjunction with the active noise reduction system 90, including but not limited to, microcode, device drivers, redundant processing units, external disk drive arrays, RAID (Redundant Array of Independent Disks) systems, tape drives, and data archival storage systems, etc.

[0197] It should be noted that although several units / modules or sub-units / modules of the active noise reduction system are mentioned in the foregoing detailed description, such division is merely exemplary and not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided into multiple units / modules.

[0198] The embodiment of the present disclosure further provides a computer readable storage medium, which stores a computer program, and the program is executed by a processor to implement the abnormal sound detection method or the control method provided by any of the above embodiments.

[0199] More specifically, the readable storage medium can include, but is not limited to, a portable disc, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0200] The embodiment of the present disclosure further provides a computer program product, which comprises a computer program, and the computer program is executed by a processor to implement the abnormal sound detection method or the control method.

[0201] The program code for executing the computer program product of the present disclosure can be written in any combination of one or more programming languages, and can be executed entirely on a user device, partially on a user device, as a standalone software package, partially on a user device and partially on a remote device, or entirely on a remote device.

[0202] Although the specific embodiments of the present disclosure are described above, those skilled in the art should understand that this is only an example, and the protection scope of the present disclosure is defined by the appended claims. Those skilled in the art can make various changes or modifications to these embodiments without departing from the principles and essence of the present disclosure, and these changes and modifications all fall within the protection scope of the present disclosure.

Claims

1. A method for detecting abnormal sounds, characterized in that, The method is applied to an active noise cancellation system, which includes a microphone and a speaker; the abnormal sound detection method includes: Acquire a first noise signal during the operation of the active noise cancellation system; the first noise signal includes noise signals from noise sources collected by the microphone and / or control signals from the speaker; Perform a Fourier transform on the first noise signal, and determine the total amplitude of the spectral power and the peak noise value of the first noise signal based on the Fourier transform result; Based on the total amplitude of the spectral power and the peak noise value, the active noise cancellation system performs abnormal noise detection; Before acquiring the first noise signal during the operation of the active noise cancellation system, the method further includes: When the active noise reduction system is not activated, a second noise signal from the noise source is acquired; The second noise signal is segmented to ensure that the power percentage of each noise segment is less than the percentage threshold, and the segmentation strategy of the second noise signal is determined. Performing a Fourier transform on the first noise signal includes: The first noise signal is segmented according to the segmentation strategy, and the resulting noise segments are subjected to Fourier transform.

2. The abnormal sound detection method according to claim 1, characterized in that, Based on the total amplitude of the spectral power and the peak noise value, the active noise cancellation system performs abnormal noise detection, including: If the difference between the total amplitude of the spectral power and the peak noise is less than the difference threshold, it is determined that there is a stable single-frequency abnormal sound in the active noise cancellation system; The difference threshold is determined in the following way: Obtain abnormal samples and normal samples; the abnormal samples are noise signals containing abnormal sounds, and the normal samples are noise signals without abnormal sounds; Construct an objective function; the objective function is constructed based on at least one of the following parameters: the number of missed abnormal samples, the number of false positives of normal samples, the proportion of missed abnormal samples, and the proportion of false positives of normal samples; The objective function is optimized within a preset optimization range of the difference threshold to determine the final difference threshold.

3. The abnormal sound detection method according to claim 2, characterized in that, The preset optimization range is [1dB, 10dB]; And / or, the optimization algorithm includes at least one of the following algorithms: genetic algorithm, simulated annealing algorithm, particle swarm optimization algorithm, Bayesian optimization algorithm, hill climbing algorithm; And / or, the objective function is the ratio of the number of anomalies to the total number of samples, wherein the number of anomalies is the sum of the number of missed anomaly samples and the number of false positives among normal samples.

4. The abnormal sound detection method according to claim 1, characterized in that, If the spectrum of the first noise signal contains at least two peaks, the largest peak is determined as the noise peak. And / or, acquire a first noise signal during the operation of the active noise cancellation system at a target sampling frequency; the target sampling frequency is determined based on the heterophonic frequency and / or the computing power of the analysis device for the first noise signal; the heterophonic frequency is determined based on noise samples containing heterophonic sounds; And / or, perform a Fourier transform on the first noise signal with a target frequency resolution; wherein the target frequency resolution is obtained through optimization, and the optimization objective is to accurately identify abnormal sounds while minimizing the computational load of abnormal sound detection.

5. The abnormal sound detection method according to claim 1, characterized in that, Determining the total amplitude of the spectral power and the peak noise of the first noise signal based on the Fourier transform result includes: performing A-weighting processing on the Fourier transform result, and determining the total amplitude of the spectral power and the peak noise of the first noise signal based on the A-weighting processing result; And / or, divide the Fourier transform result into at least two noise segments, and determine the total amplitude of the spectral power and the peak noise value based on the integral of each noise segment; And / or, in response to the difference between the total amplitude of the spectral power and the peak noise being less than a difference threshold, determining that the active noise cancellation system has a stable single-frequency abnormal sound includes: in response to the number of times the difference between the total amplitude of the spectral power and the peak noise being less than a difference threshold being greater than a number of occurrences threshold, determining that the active noise cancellation system has a stable single-frequency abnormal sound.

6. The abnormal sound detection method according to any one of claims 1-5, characterized in that, Before acquiring the first noise signal during the operation of the active noise cancellation system, the method further includes: When the active noise reduction system is not activated, a second noise signal from the noise source is acquired; The second noise signal is subjected to octave band calculation to determine a segmentation strategy that matches the second noise signal; The second noise signal is segmented according to the segmentation strategy, and the characteristic values ​​of each noise segment of the second noise signal are calculated; the characteristic values ​​include power and / or decibel values; Determine the target gain of a first target noise segment whose eigenvalue is greater than a eigenvalue threshold; after filtering the first target noise segment with a bandpass filter of the target gain, the eigenvalue of the first target noise segment is less than or equal to the eigenvalue threshold; Performing a Fourier transform on the first noise signal includes: The first noise signal is segmented to obtain at least two noise segments, and a bandpass filter with a target gain is used to filter the second target noise segment contained in the first noise signal. Fourier transforms are performed on the segmented noise segments and the filtered noise segments in the first noise signal, respectively; wherein the frequency band of the second target noise segment includes the frequency band of the first target noise segment.

7. A control method for an active noise reduction system, characterized in that, include: The abnormal noise detection method according to any one of claims 1-6 is used to detect abnormal noise in the active noise cancellation system; In response to the presence of a stable single-frequency abnormal sound in the active noise cancellation system, the active noise cancellation system is restarted.

8. An abnormal sound detection device, characterized in that, The device is applied to an active noise cancellation system, which includes a microphone and a speaker; the abnormal sound detection device is used to implement the abnormal sound detection method according to any one of claims 1-6. The abnormal sound detection device includes: The acquisition module is used to acquire a first noise signal during the operation of the active noise cancellation system; the first noise signal includes the noise signal of the noise source collected by the microphone and / or the control signal of the speaker; The determination module is used to perform a Fourier transform on the first noise signal and determine the total amplitude of the spectral power and the peak noise value of the first noise signal based on the Fourier transform result. The detection module is used to determine that there is a stable single-frequency abnormal sound in the active noise cancellation system when the difference between the total amplitude of the spectral power and the peak noise is less than the difference threshold.

9. An active noise cancellation system, comprising a memory, a processor, and a computer program stored in the memory and for running on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 7.

Images