A Wind Noise Detection Method and System with Dual Microphones

Abstract

The present invention relates to the technical field of sound detection, and discloses a wind noise detection method and system for dual microphones, including: collecting the time-domain audio signal of the dual microphones deployed, performing signal iterative processing on the time-domain audio signal to obtain a target signal; performing short-time segmentation and short-time frequency domain transformation on the target signal to obtain a time-frequency spectrogram, identifying the background noise spectrum in the time-frequency spectrogram, performing smoothing correction to obtain a corrected spectrum, identifying the spectral components in the time-frequency spectrogram to obtain a target spectrum; performing signal independent component analysis on the target spectrum to identify the wind noise signal, identifying the full amount of wind noise signal of the target spectrum, calculating the signal intensity of the full amount of wind noise signal, and analyzing the frequency characteristics of the full amount of wind noise signal; performing spatio-temporal correlation analysis on the dual microphones to obtain the wind noise propagation law, and using the signal intensity, frequency characteristics and wind noise propagation law to perform real-time wind noise detection on the dual microphones. The present invention can improve the overall effect of wind noise detection of dual microphones.

Description

Technical Field

[0001] The present invention relates to a method and system for detecting wind noise using dual microphones, belonging to the technical field of sound detection. Background Art

[0002] In the field of audio acquisition and processing, wind noise is a key interference factor affecting sound quality. Accurately detecting wind noise is of great significance for improving the clarity, intelligibility of audio signals, and the accuracy of subsequent processing. Due to its unique architecture and working principle, the dual microphone technology shows great potential in wind noise detection. The wind noise detection method using dual microphones can effectively utilize the signal differences collected by the two microphones to accurately identify the wind noise components in the environment. By analyzing and comparing the signals received by the dual microphones, the dynamic changes of wind noise can be tracked in real time, so as to provide accurate wind noise information for the audio processing system to take targeted noise reduction measures, significantly improving the audio quality. This not only helps to improve the effect of voice communication, such as making the voice more clear and natural in scenarios such as teleconferences and voice assistants, but also provides pure audio signals for applications such as audio recording and monitoring, enhancing the performance of related devices and systems.

[0003] Currently, common wind noise detection methods are mostly based on single microphones or simple signal processing algorithms. When detecting wind noise with a single microphone, it is difficult to accurately distinguish wind noise from complex environmental noises due to the lack of a comparison reference, resulting in low detection accuracy. And some simple signal processing algorithms have poor adaptability when facing the variable and complex characteristics of wind noise, and cannot meet the requirements of accuracy and real-time of wind noise detection in practical applications. Especially in a strong wind environment or when there are multiple noises in the audio signal at the same time, these traditional methods often cannot work effectively, making the wind noise detection results unsatisfactory, and thus affecting the overall effect of audio processing. Summary of the Invention

[0004] The present invention provides a method and system for detecting wind noise using dual microphones, and its main purpose is to improve the overall effect of wind noise detection using dual microphones.

[0005] To achieve the above object, a method for detecting wind noise using dual microphones provided by the present invention includes:

[0006] Obtain dual microphones to be detected for wind noise, perform acoustic space deployment on the dual microphones to obtain deployed dual microphones, use pre-configured phase-locked loop synchronization technology to collect the time-domain audio signals of the deployed dual microphones, analyze the desired signals of the time-domain audio signals, and based on the desired signals, perform signal iterative processing on the time-domain audio signals to obtain target signals;

[0007] Segment the target signal in a short time period to obtain a segmented signal, perform a short-time frequency domain transform on the segmented signal to obtain the time-frequency spectrogram of the segmented signal, identify the silent segment spectrum in the time-frequency spectrogram, analyze the spectral characteristics of the silent segment spectrum, based on the spectral characteristics, determine the background noise spectrum of the time-frequency spectrogram, perform smoothing correction on the background noise spectrum to obtain a corrected spectrum, and identify the spectral components in the time-frequency spectrogram based on the corrected spectrum to obtain a target spectrum;

[0008] Calculate the linear transformation matrix of the target spectrum, and based on the linear transformation matrix, perform signal independent component analysis on the target spectrum to identify the wind noise signal in the target spectrum, analyze the wind noise signal characteristics corresponding to the wind noise signal, and based on the wind noise signal characteristics, identify the full wind noise signal in the target spectrum, calculate the signal strength of the full wind noise signal, and analyze the frequency characteristics of the full wind noise signal;

[0009] Query the spatial positions and sampling times of the dual microphones, and based on the spatial positions and the sampling times, perform spatio-temporal correlation analysis on the dual microphones to obtain the wind noise propagation law, and use the signal strength, the frequency characteristics, and the wind noise propagation law to construct the wind noise dynamic threshold of the dual microphones, and perform real-time wind noise detection on the dual microphones based on the wind noise dynamic threshold.

[0010] Optionally, the acoustic spatial deployment of the dual microphones to obtain deployed dual microphones includes:

[0011] Query the sound signal wavelength and the desired spatial resolution of the dual microphones to determine the arrangement spacing of the dual microphones;

[0012] Identify the sound source direction and the sound change range of the dual microphones to determine the arrangement angle of the dual microphones;

[0013] Based on the arrangement spacing and the arrangement angle, perform preliminary spatial deployment on the dual microphones to obtain initially deployed microphones;

[0014] Perform signal synchronization verification on the initially deployed microphones, and when the synchronization result of the signal synchronization verification meets the preset synchronization value, obtain the deployed dual microphones.

[0015] Optionally, the analysis of the desired signal of the time-domain audio signal includes:

[0016] Remove the DC offset of the time-domain audio signal to obtain a preliminarily processed signal;

[0017] Perform frequency screening on the preliminarily processed signal to obtain a screened signal;

[0018] Perform noise estimation on the screening signal to obtain an estimated noise;

[0019] Use the adaptive filtering method to remove the estimated noise to obtain a clean signal;

[0020] Analyze the clean signal characteristics of the clean signal;

[0021] Based on the clean signal characteristics, determine the desired signal of the time-domain audio signal.

[0022] Optionally, the performing short-time frequency-domain transformation on the segmented signal to obtain a spectrogram of the segmented signal includes:

[0023] Perform short-time frequency-domain transformation on the segmented signal using the following formula to obtain a frequency-domain transformed signal:

[0024] ;

[0025] where, represents the frequency-domain transformed signal, represents the segmented signal, w() represents the window function, N represents the length of the window in the window function, m represents the time index, R represents the sliding step of the window, Y represents the frequency index, j represents the imaginary unit, and n represents the time index of the segmented signal in the window function;

[0026] Arrange the frequency-domain transformed signals in chronological order to obtain the spectrogram of the segmented signal.

[0027] Optionally, the identifying the silent segment spectrum in the spectrogram includes:

[0028] Calculate the spectral energy of the time window in the spectrogram using the following formula:

[0029] ;

[0030] where, represents the spectral energy, represents the i-th window in the time window, represents the k-th frequency in the spectrogram, represents the spectrogram at and the spectral amplitude at, K represents the number of frequencies in the spectrogram;

[0031] Construct the silent energy threshold of the spectrogram;

[0032] Based on the silent energy threshold and the spectral energy, determine the silent segment spectrum in the spectrogram.

[0033] Optionally, analyzing the spectral characteristics of the silent segment spectrum includes:

[0034] Constructing a spectral curve of the silent segment spectrum;

[0035] Using the spectral curve to analyze the spectral shape characteristics of the silent segment spectrum;

[0036] Identifying the energy distribution state of the silent segment spectrum on the frequency axis;

[0037] Based on the energy distribution state, identifying the frequency distribution characteristics of the silent segment spectrum;

[0038] Calculating the statistical value of the silent segment spectrum;

[0039] Based on the statistical value, analyzing the spectral dispersion of the silent segment spectrum;

[0040] Based on the spectral dispersion, identifying the spectral statistical characteristics of the silent segment spectrum;

[0041] Calculating the average value of the absolute values of the amplitude differences between adjacent frequency points in the silent segment spectrum to analyze the spectral smoothness characteristics of the silent segment spectrum,

[0042] Based on the spectral shape characteristics, the frequency distribution characteristics, the spectral statistical characteristics, and the spectral smoothness characteristics, determining the spectral characteristics of the silent segment spectrum.

[0043] Optionally, based on the wind noise signal characteristics, identifying the full - volume wind noise signal of the target spectrum includes:

[0044] Querying the time - domain characteristics, frequency - domain characteristics, and high - order statistical characteristics of the wind noise signal to obtain comprehensive characteristics;

[0045] Based on the comprehensive characteristics, performing wind noise feature matching on the target spectrum to obtain matching characteristics;

[0046] Using the weighting method to construct a wind noise judgment rule for the matching characteristics;

[0047] Based on the wind noise judgment rule, making a preliminary judgment on the full - volume wind noise signal of the target spectrum to obtain a preliminary full - volume wind noise signal;

[0048] Performing continuous region integration analysis on the preliminary full - volume wind noise signal to determine the full - volume wind noise signal of the target spectrum.

[0049] Optionally, calculating the signal intensity of the full - volume wind noise signal includes:

[0050] Calculating the mean square value of the full - volume wind noise signal using the following formula:

[0051] ;

[0052] Among them, RMS represents the root mean square value, M represents the number of sequences of the total wind noise signal, represents the b-th signal in the total wind noise signal,

[0053] Analyze the time-domain signal intensity of the total wind noise signal by using the root mean square value;

[0054] Calculate the sum of the squares of the spectral amplitudes of the total wind noise signal by using the following formula:

[0055] ;

[0056] Among them, V represents the sum of the squares of the spectral amplitudes, H represents the number of frequency points of the total wind noise signal, represents the spectral amplitude at the frequency point h,

[0057] Analyze the frequency-domain signal intensity of the total wind noise signal by using the sum of the squares of the spectral amplitudes;

[0058] Determine the signal intensity of the total wind noise signal according to the time-domain signal intensity and the frequency-domain signal intensity.

[0059] Optionally, the spatio-temporal correlation analysis of the dual microphones based on the spatial position and the sampling time to obtain the wind noise propagation law includes:

[0060] Query the signal sequences of the dual microphones based on the sampling time;

[0061] Align the signal sequences on the time axis to obtain aligned signals;

[0062] Analyze the signal delay of the aligned signals;

[0063] Calculate the propagation distance of the dual microphones based on the spatial position;

[0064] Analyze the propagation speed and propagation direction of the dual microphones based on the signal delay and the propagation distance;

[0065] Query the signal intensity of the dual microphones;

[0066] Analyze the signal attenuation trend of the dual microphones based on the signal intensity and the propagation speed;

[0067] Determine the wind noise propagation law of the dual microphones based on the propagation speed, the propagation direction and the signal attenuation trend.

[0068] To solve the above problems, the present invention also provides a wind noise detection system for dual microphones, and the system includes:

[0069] A signal processing module, which is used to obtain a dual microphone to be wind noise detected, acoustically deploy the dual microphone to obtain a deployed dual microphone, collect the time-domain audio signals of the deployed dual microphone by using a pre-configured phase-locked loop synchronization technology, analyze the desired signals of the time-domain audio signals, and perform signal iterative processing on the time-domain audio signals based on the desired signals to obtain target signals;

[0070] A spectrum component identification module, which is used to perform short-time segmentation on the target signals to obtain segmented signals, perform short-time frequency-domain transformation on the segmented signals to obtain the time-frequency spectrogram of the segmented signals, identify the silent-segment spectra in the time-frequency spectrogram, analyze the spectral characteristics of the silent-segment spectra, determine the background noise spectrum of the time-frequency spectrogram based on the spectral characteristics, perform smoothing correction on the background noise spectrum to obtain a corrected spectrum, and identify the spectrum components in the time-frequency spectrogram based on the corrected spectrum to obtain target spectra;

[0071] A signal analysis module, which is used to calculate the linear transformation matrix of the target spectra, perform signal independent component analysis on the target spectra based on the linear transformation matrix to identify the wind noise signals in the target spectra, analyze the wind noise signal characteristics corresponding to the wind noise signals, identify the full amount of wind noise signals in the target spectra based on the wind noise signal characteristics, calculate the signal intensity of the full amount of wind noise signals, and analyze the frequency characteristics of the full amount of wind noise signals;

[0072] A wind noise detection module, which is used to query the spatial positions and sampling times of the dual microphones, perform spatio-temporal correlation analysis on the dual microphones based on the spatial positions and the sampling times to obtain the wind noise propagation law, construct the wind noise dynamic threshold of the dual microphones by using the signal intensity, the frequency characteristics and the wind noise propagation law, and perform real-time wind noise detection on the dual microphones based on the wind noise dynamic threshold.

[0073] Compared with the problems described in the background art, in the embodiments of the present invention, the dual microphones are acoustically deployed in space. The arrangement spacing is determined by querying the wavelength of the sound signal and the desired spatial resolution, and the arrangement angle is clarified based on the sound source direction and the variation range. After the preliminary deployment, signal synchronization verification is performed on the initially deployed microphones to ensure that the dual microphones can effectively capture information such as the time difference and intensity difference of the sound. Further, in the embodiments of the present invention, short-time frequency domain transformation is performed on the segmented signal, and the segmented signal is converted into a frequency domain transformation signal using a specific formula, and then arranged in chronological order to obtain a spectrogram, which helps to discover the variation law of the signal frequency components over time and analyze the spectral characteristics of the silent segment, and is quantified in terms of spectral shape, frequency distribution, spectral dispersion, spectral smoothness, etc., providing a basis for estimating the background noise spectrum. Further, the present invention calculates the linear transformation matrix of the target spectrum (which can be obtained through the ICA algorithm), decomposes the target spectrum signal into different independent components, creates conditions for identifying the wind noise signal, calculates the signal intensity of the total wind noise signal, calculates the mean square value and the time-frequency domain signal intensity respectively through formulas, and provides quantitative indicators for formulating subsequent noise reduction strategies and evaluating system performance. At the same time, Fourier transform is performed on the total wind noise signal, the spectrogram is observed and the energy proportion of each frequency interval is calculated, and its frequency characteristics are analyzed to provide a basis for designing noise reduction technologies. Furthermore, the present invention queries the spatial positions and sampling moments of the dual microphones to provide geometric and time dimension information for wind noise propagation analysis, and performs spatio-temporal correlation analysis on the dual microphones to obtain the propagation speed and direction of the wind noise, combines the signal intensity to analyze the signal attenuation trend, and determines the wind noise propagation law to help users understand the propagation characteristics of the wind noise in the actual environment, providing a physical basis for subsequent wind noise processing and detection, and performing real-time wind noise detection on the dual microphones based on the dynamic threshold of the wind noise, adjusting the detection standard according to the real-time change of the wind noise, improving the accuracy and real-time performance of the detection, and thus improving the overall effect of the wind noise detection of the dual microphones. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 It is a schematic flowchart of a method for detecting wind noise of a dual microphone provided by an embodiment of the present invention;

[0075] Figure 2 It is a schematic diagram of the functional modules for implementing the system for detecting wind noise of a dual microphone provided by an embodiment of the present invention.

[0076] The implementation, functional features and advantages of the present invention will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0077] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0078] The embodiments of the present application provide a method for detecting wind noise using dual microphones. The execution subject of the method for detecting wind noise using dual microphones includes, but is not limited to, at least one of electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided by the embodiments of the present application. In other words, the method for detecting wind noise using dual microphones can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to: a single server, a server cluster, a cloud server, or a cloud server cluster, etc.

[0079] Embodiment 1: Refer to Figure 1 As shown, it is a schematic flowchart of a method for detecting wind noise using dual microphones provided by an embodiment of the present invention. In this embodiment, the method for detecting wind noise using dual microphones includes:

[0080] S1. Obtain the dual microphones to be detected for wind noise, perform acoustic spatial deployment on the dual microphones to obtain deployed dual microphones, collect the time-domain audio signals of the deployed dual microphones using a pre-configured phase-locked loop synchronization technology, analyze the desired signals of the time-domain audio signals, and based on the desired signals, perform signal iterative processing on the time-domain audio signals to obtain target signals.

[0081] In the embodiments of the present invention, by obtaining the dual microphones to be detected for wind noise, the hardware devices for wind noise detection can be determined, providing a basis for subsequent detection work. Among them, the dual microphones refer to a combination of two microphones used in an audio acquisition system. These two microphones are independent of each other in space but work together functionally to collect sound signals in the surrounding environment. They can be microphones of the same model or a combination of different types of microphones selected according to specific application scenarios and user requirements.

[0082] Furthermore, in the embodiments of the present invention, by performing acoustic spatial deployment on the dual microphones to obtain deployed dual microphones, it is possible to make the dual microphones better capture information such as the time difference and intensity difference of sound arrival, which helps subsequent analysis of the sound source direction and characteristics.

[0083] As an embodiment of the present invention, performing acoustic spatial deployment on the dual microphones to obtain deployed dual microphones includes: querying the sound signal wavelength and desired spatial resolution of the dual microphones to determine the arrangement spacing of the dual microphones, identifying the sound source direction and sound change range of the dual microphones to determine the arrangement angle of the dual microphones, based on the arrangement spacing and the arrangement angle, performing preliminary spatial deployment on the dual microphones to obtain initially deployed microphones, performing signal synchronization verification on the initially deployed microphones, and when the synchronization result of the signal synchronization verification meets a preset synchronization value, obtaining deployed dual microphones.

[0084] Among them, the sound signal wavelength refers to a mechanical wave that has a certain wavelength when propagating in a medium, and the expected spatial resolution refers to the minimum spatial interval or angular range that we expect to be able to resolve the wind noise source in wind noise detection.

[0085] Optionally, the process of querying the sound signal wavelength and the expected spatial resolution of the dual microphones to determine the arrangement spacing of the dual microphones is as follows: According to the frequency range of the wind noise signals expected to be collected by the dual microphones, use the acoustic wave calculation formula to calculate the sound signal wavelength. Then, in combination with the expected spatial resolution, set the arrangement spacing to a certain proportion (such as 1 / 4 or 1 / 3 wavelength) of the corresponding wavelength according to the required detection accuracy. The process of identifying the sound source direction and the sound change range of the dual microphones to determine the arrangement angle of the dual microphones is as follows: According to the direction from which the wind may blow, such as whether it is omnidirectional or within a specific direction range, and the possible change range of the sound intensity and frequency, then adjust the arrangement angle of the dual microphones to an angle that can effectively capture wind noise signals from different directions. For example, for an omnidirectional wind source, it can be set to an included angle of 90° or 120°. The signal synchronization verification of the initial deployed microphones can be carried out by collecting the signals collected by the dual microphones through signal acquisition technology, and determining the signal synchronization value by comparing the time difference or phase difference of the signals collected by the two microphones. When the synchronization result in the signal synchronization verification meets the preset synchronization value, it means that the time difference or the phase difference is within the preset nanosecond-level synchronization value, and then it is considered that the synchronization result meets the requirements. At this time, the dual microphones are the deployed dual microphones.

[0086] By using the pre-configured phase-locked loop synchronization technology to collect the time-domain audio signals of the deployed dual microphones in the embodiment of the present invention, it can ensure the temporal consistency of the collected signals, avoid errors introduced by asynchronous sampling, and make the subsequent analysis and processing of the signals more accurate.

[0087] Among them, the phase-locked loop synchronization technology refers to a technology widely used in electronic systems for frequency synthesis and signal synchronization. In the context of dual microphone wind noise detection, it is mainly used to ensure the high temporal synchronization of the audio signals collected by the two microphones.

[0088] By analyzing the expected signal of the time-domain audio signal in the embodiment of the present invention, the direction and target of subsequent signal processing can be determined, that is, it is clear in which direction the original time-domain audio signal is to be processed.

[0089] Among them, the expected signal refers to the audio signal that meets the user's expectations and has been ideally processed.

[0090] As an embodiment of the present invention, the analysis of the desired signal of the time-domain audio signal includes: removing the DC offset of the time-domain audio signal to obtain a preliminary processed signal, performing frequency screening on the preliminary processed signal to obtain a screened signal, performing noise estimation on the screened signal to obtain an estimated noise, using an adaptive filtering method to remove the estimated noise to obtain a clean signal, analyzing the clean signal characteristics of the clean signal, and based on the clean signal characteristics, determining the desired signal of the time-domain audio signal.

[0091] Optionally, the removal of the DC offset of the time-domain audio signal to obtain a preliminary processed signal can be achieved by calculating the average value of the time-domain audio signal and subtracting the average value from each sample value of the original signal, so that the signal fluctuates around zero mean; the frequency screening of the preliminary processed signal to obtain a screened signal can be achieved by using a filter (such as a low-pass, high-pass or band-pass filter); the noise estimation of the screened signal to obtain an estimated noise can be achieved by statistically analyzing the silent segments in the screened signal (which can be judged according to the signal energy), such as calculating statistical quantities such as the mean and variance of these silent segments, or using an adaptive method such as the least mean square (LMS) algorithm to estimate the noise component in the signal; the analysis of the clean signal characteristics of the clean signal extracts the characteristics of the clean signal from the time domain and the frequency domain, such as the signal envelope and pulse duration in the time domain, and the spectral peak and bandwidth in the frequency domain, etc.; the desired signal can determine the clean signal with the above characteristics such as the signal envelope, pulse duration, spectral peak and bandwidth in the frequency domain as the desired signal.

[0092] Furthermore, in the embodiment of the present invention, by performing signal iterative processing on the time-domain audio signal based on the desired signal to obtain a target signal, the signal quality can be gradually improved, so that the finally obtained target signal better meets the requirements of wind noise detection.

[0093] Optionally, the process of performing signal iterative processing on the time-domain audio signal based on the desired signal to obtain a target signal is as follows: First, according to the characteristics of the desired signal, set the initial parameters for the signal processing algorithm (such as the adaptive filtering algorithm). For example, when using the least mean square (LMS) algorithm, initialize the filter coefficients as small random values and set an appropriate step size factor. Then, use the time-domain audio signal as the input and perform iterative processing using the selected signal processing algorithm (such as the LMS algorithm). In each iteration, update the processing parameters according to the difference between the desired signal and the current processing result, continuously adjust the processing method, and gradually optimize the signal until the preset convergence condition is met or the predetermined number of iterations is reached, and finally obtain the target signal.

[0094] S2. Segment the target signal in a short time period to obtain segmented signals, perform short-time frequency domain transformation on the segmented signals to obtain the time-frequency spectrogram of the segmented signals, identify the silent segment spectra in the time-frequency spectrogram, analyze the spectral characteristics of the silent segment spectra, determine the background noise spectrum of the time-frequency spectrogram based on the spectral characteristics, perform smoothing correction on the background noise spectrum to obtain a corrected spectrum, and identify the spectral components in the time-frequency spectrogram based on the corrected spectrum to obtain the target spectrum.

[0095] In the embodiment of the present invention, by segmenting the target signal in a short time period to obtain segmented signals, the target signal can be divided into a series of sub-signals in short time periods, so as to facilitate subsequent detailed analysis of the signal characteristics at different moments.

[0096] Optionally, the segmented signals can be obtained by selecting an appropriate window length (such as 256 or 512 sample points) and a window function (such as a Hamming window), starting from the starting position of the target signal, successively intercepting the signal according to the window length, and setting a certain overlapping part (such as 50% overlap) for adjacent windows.

[0097] In the embodiment of the present invention, by performing short-time frequency domain transformation on the segmented signals to obtain the time-frequency spectrogram of the segmented signals, it can help users discover the variation law of the signal frequency components at different moments.

[0098] As an embodiment of the present invention, performing short-time frequency domain transformation on the segmented signals to obtain the time-frequency spectrogram of the segmented signals includes: performing short-time frequency domain transformation on the segmented signals using the following formula to obtain a frequency domain transformation signal:

[0099] ;

[0100] where represents the frequency domain transformation signal, represents the segmented signal, w() represents the window function, N represents the length of the window in the window function, m represents the time index, R represents the sliding step of the window, Y represents the frequency index, j represents the imaginary unit, and n represents the time index of the segmented signal in the window function;

[0101] Arrange the frequency domain transformation signals in chronological order to obtain the time-frequency spectrogram of the segmented signals.

[0102] Furthermore, in the embodiment of the present invention, identifying the silent segment spectra in the time-frequency spectrogram provides a key data source for extracting background noise information, and provides a basis for subsequent noise processing and signal purification, thereby achieving effective noise reduction and signal separation.

[0103] As an embodiment of the present invention, the identification of the silent segment spectrum in the time-frequency spectrogram includes: calculating the spectral energy of the time window in the time-frequency spectrogram using the following formula:

[0104] ;

[0105] where, represents the spectral energy, represents the i-th window in the time window, represents the k-th frequency in the time-frequency spectrogram, represents the spectral amplitude of the time-frequency spectrogram at and , and K represents the number of frequencies in the time-frequency spectrogram;

[0106] Construct the silent energy threshold of the time-frequency spectrogram, and determine the silent segment spectrum in the time-frequency spectrogram based on the silent energy threshold and the spectral energy.

[0107] Optionally, the silent energy threshold is determined by analyzing the overall energy distribution of the time-frequency spectrogram or based on past experience to determine a suitable silent energy threshold (specifically, it needs to be set in combination with actual applications). The determination of the silent segment spectrum in the time-frequency spectrogram based on the silent energy threshold and the spectral energy can compare the spectral energy calculated for each time window with the silent energy threshold, and the spectrum corresponding to the time window with energy lower than the threshold is determined as the silent segment spectrum.

[0108] By analyzing the spectral characteristics of the silent segment spectrum in the embodiment of the present invention, the basic characteristics of the background noise can be grasped, such as the main frequency range of the background noise and the energy level at different frequencies, providing specific quantitative indicators for subsequent estimation of the background noise spectrum and helping to accurately describe the nature of the background noise.

[0109] As an embodiment of the present invention, the analysis of the spectral characteristics of the silent segment spectrum includes: constructing the spectral curve of the silent segment spectrum, using the spectral curve to analyze the spectral shape characteristics of the silent segment spectrum, identifying the energy distribution state of the silent segment spectrum on the frequency axis, based on the energy distribution state, identifying the frequency distribution characteristics of the silent segment spectrum, calculating the statistical value of the silent segment spectrum, analyzing the spectral dispersion of the silent segment spectrum based on the statistical value, identifying the spectral statistical characteristics of the silent segment spectrum based on the spectral dispersion, calculating the average value of the absolute values of the amplitude differences between adjacent frequency points in the silent segment spectrum to analyze the spectral smoothness characteristics of the silent segment spectrum, and determining the spectral characteristics of the silent segment spectrum based on the spectral shape characteristics, the frequency distribution characteristics, the spectral statistical characteristics, and the spectral smoothness characteristics.

[0110] Among them, the spectrum curve refers to a visual representation curve for showing the amplitude change of the silent segment spectrum in the frequency dimension. The energy distribution state refers to the distribution of the energy of the silent segment spectrum at different frequencies. The spectrum dispersion degree is an index for measuring the dispersion degree of the amplitude change of the silent segment spectrum at different frequency points.

[0111] Optionally, the spectrum curve can use frequency as the horizontal axis and the amplitude of the silent segment spectrum at each frequency point (which can be the average amplitude, peak amplitude, etc.) as the vertical axis, and connect the data points to form it. The spectrum shape feature can be judged by observing the shape of the spectrum curve whether it is flat (possibly white noise), has peaks (possibly narrowband noise), or other irregular shapes, and analyze the undulation of the curve and the position and number of peaks, etc., to determine the spectrum shape feature. The energy distribution state can calculate the energy of each frequency point or frequency interval (such as the sum of squared amplitudes), observe the high and low distribution of the energy on the frequency axis, and determine which frequency ranges the energy is mainly concentrated in and which frequency ranges have lower energy. The frequency distribution feature can find the frequency region with higher energy as the main frequency distribution range according to the energy distribution. If the energy is concentrated near a certain specific frequency, it can be judged as narrowband frequency distribution. If the energy is relatively evenly distributed in a wider frequency range, it is broadband frequency distribution. The spectrum dispersion degree can be calculated by calculating the statistical values of the spectrum amplitude, such as variance and standard deviation. Among them, the larger the variance or standard deviation, the higher the spectrum dispersion degree, indicating that the fluctuation of the spectrum amplitude at different frequency points is larger, and vice versa, the smaller the fluctuation. The spectrum statistical feature can identify the statistical feature according to the value of the spectrum dispersion degree and in combination with other factors such as the spectrum shape. For example, high dispersion degree and irregular spectrum shape may indicate complex and variable noise, and low dispersion degree and relatively regular spectrum shape may be relatively stable noise. The spectrum smoothness feature can be obtained by calculating the absolute value of the amplitude difference between adjacent frequency points in turn and then finding the average value. Among them, a smaller average value indicates high spectrum smoothness and small amplitude change between adjacent frequency points, and a larger average value indicates low spectrum smoothness and large amplitude change.

[0112] By determining the background noise spectrum of the time-frequency spectrogram based on the spectrum feature in the embodiment of the present invention, it can help lay a foundation for removing noise from the noisy signal and extracting useful signals, and further can more accurately separate the target signal (such as wind noise, speech, etc.) from the mixed signal.

[0113] Optionally, based on the spectrum feature, to determine the background noise spectrum of the time-frequency spectrogram, the amplitude of the silent segment spectrum can be statistically averaged in frequency according to the analyzed spectrum feature of the silent segment spectrum. For each frequency point, add up the spectrum amplitudes of all silent segments at this frequency point and then divide by the number of silent segments to obtain the background noise spectrum amplitude at this frequency point, and so on to determine the background noise spectrum of the entire time-frequency spectrogram.

[0114] In the embodiment of the present invention, by smoothing and correcting the background noise spectrum to obtain a corrected spectrum, the background noise spectrum can be made more stable and continuous, the reliability of noise estimation can be improved, the spectrum estimation deviation caused by local noise or measurement error can be reduced, and the accuracy and stability of subsequent processing can be further improved.

[0115] Optionally, the smoothing and correction of the background noise spectrum to obtain a corrected spectrum can be achieved by using the median filtering method.

[0116] In the embodiment of the present invention, by identifying the spectral components in the time-frequency spectrogram based on the corrected spectrum to obtain a target spectrum, the spectrum of the target signal (such as wind noise signal, speech signal, etc.) can be presented more clearly, and the spectral characteristics of the target signal can be reflected more clearly.

[0117] Optionally, the identification of the spectral components in the time-frequency spectrogram based on the corrected spectrum to obtain a target spectrum can be achieved by the spectral subtraction method.

[0118] S3. Calculate the linear transformation matrix of the target spectrum. Based on the linear transformation matrix, perform independent component analysis on the target spectrum to identify the wind noise signal in the target spectrum, analyze the wind noise signal characteristics corresponding to the wind noise signal, based on the wind noise signal characteristics, identify the full amount of wind noise signal in the target spectrum, calculate the signal intensity of the full amount of wind noise signal, and analyze the frequency characteristics of the full amount of wind noise signal.

[0119] In the embodiment of the present invention, by calculating the linear transformation matrix of the target spectrum, a new set of basis vectors can be found, and the signals in the target spectrum can be decomposed into different independent components, and then the complex signal can be decomposed into relatively independent parts, creating conditions for finding the wind noise signal. Among them, the linear transformation matrix refers to a matrix used to transform the representation of the original signal from one space to another space.

[0120] Optionally, the linear transformation matrix can be calculated by the ICA algorithm.

[0121] In the embodiment of the present invention, by performing independent component analysis on the target spectrum based on the linear transformation matrix to identify the wind noise signal in the target spectrum, it can help the user extract the wind noise signal from the target spectrum, provide a more pure wind noise signal source for subsequent wind noise analysis and processing, and improve the identifiability and analyzability of the wind noise signal.

[0122] Optionally, the process of performing independent component analysis on the target spectrum based on the linear transformation matrix to identify the wind noise signal in the target spectrum is as follows: First, multiply the target spectrum data by the linear transformation matrix for linear transformation to obtain a transformed signal matrix; then, calculate the statistical characteristics (such as kurtosis) of each independent component and compare them with the known wind noise characteristics to screen out the wind noise signal; finally, verify the results through methods such as cross-validation. If it is inaccurate, adjust the index threshold or improve the identification process to optimize the wind noise signal identification.

[0123] In the embodiment of the present invention, analyzing the wind noise signal characteristics corresponding to the wind noise signal helps the user to more accurately grasp the essence of the wind noise signal.

[0124] Optionally, the wind noise signal characteristics can be obtained by analyzing through a deep learning model.

[0125] In the embodiment of the present invention, by identifying all the wind noise signals in the target spectrum based on the wind noise signal characteristics, the locally identified wind noise signals can be extended to the entire target spectrum to obtain a complete wind noise signal distribution, so as to comprehensively grasp the wind noise situation in the target spectrum.

[0126] As an embodiment of the present invention, identifying all the wind noise signals in the target spectrum based on the wind noise signal characteristics includes: querying the time-domain characteristics, frequency-domain characteristics, and high-order statistical characteristics of the wind noise signal to obtain comprehensive characteristics, performing wind noise feature matching on the target spectrum based on the comprehensive characteristics to obtain matching characteristics, constructing a wind noise judgment rule for the matching characteristics by using the weighting method, making a preliminary judgment on all the wind noise signals in the target spectrum based on the wind noise judgment rule to obtain a preliminary all-wind noise signal, and performing continuous region integration analysis on the preliminary all-wind noise signal to determine all the wind noise signals in the target spectrum.

[0127] Optionally, the matching characteristics can be obtained by comparing each part of the characteristics of the target spectrum with the comprehensive characteristics one by one, judging the similarity, and determining the degree of conformity of each part with the wind noise characteristics; the wind noise judgment rule can be obtained by assigning weights to the matching characteristics corresponding to the time domain, frequency domain, and high-order statistics, such as a frequency domain weight of 0.5, a time domain weight of 0.3, and a high-order statistics weight of 0.2, calculating a comprehensive score according to the weights, and setting the score threshold as the wind noise judgment rule; the preliminary all-wind noise signal can be obtained by calculating the comprehensive score for each part of the target spectrum according to the wind noise judgment rule, and determining that the score higher than the threshold is a wind noise signal.

[0128] In the embodiment of the present invention, calculating the signal intensity of all the wind noise signals can provide a quantitative index for subsequent noise reduction strategy formulation and system performance evaluation.

[0129] As an embodiment of the present invention, calculating the signal strength of the full wind noise signal includes: calculating the mean square value of the full wind noise signal using the following formula:

[0130] ;

[0131] where RMS represents the mean square value, M represents the number of sequences of the full wind noise signal, represents the b-th signal in the full wind noise signal,

[0132] using the mean square value to analyze the time-domain signal strength of the full wind noise signal;

[0133] calculating the sum of the squares of the spectral amplitudes of the full wind noise signal using the following formula:

[0134] ;

[0135] where V represents the sum of the squares of the spectral amplitudes, H represents the number of frequency points of the full wind noise signal, represents the spectral amplitude at the frequency point h,

[0136] using the sum of the squares of the spectral amplitudes to analyze the frequency-domain signal strength of the full wind noise signal;

[0137] determining the signal strength of the full wind noise signal according to the time-domain signal strength and the frequency-domain signal strength.

[0138] Furthermore, by analyzing the frequency characteristics of the full wind noise signal in the embodiment of the present invention, the generation mechanism and propagation characteristics of wind noise can be further understood, providing a basis for designing more effective wind noise suppression filters and other noise reduction technologies, and can also provide more detailed information for the classification and evaluation of wind noise.

[0139] Optionally, analyzing the frequency characteristics of the full wind noise signal can be obtained by performing a Fourier transform on the full wind noise signal to obtain a spectrogram, observing its shape, frequency range, and spectral peak position; secondly, dividing the frequency axis into multiple intervals and calculating the energy ratio of each interval to determine.

[0140] S4. Query the spatial positions and sampling times of the dual microphones, based on the spatial positions and the sampling times, perform spatio-temporal correlation analysis on the dual microphones to obtain the wind noise propagation law, use the signal strength, the frequency characteristics, and the wind noise propagation law to construct the wind noise dynamic threshold of the dual microphones, and perform real-time wind noise detection on the dual microphones based on the wind noise dynamic threshold.

[0141] In the embodiments of the present invention, querying the spatial positions and sampling times of the dual microphones can provide geometric information for subsequent wind noise propagation analysis, understand their relative positions in space, and ensure accurate correspondence of signal data at each moment in subsequent analysis, providing a basis for spatio-temporal correlation analysis in the time dimension.

[0142] Furthermore, in the embodiments of the present invention, spatio-temporal correlation analysis of the dual microphones based on the spatial positions and sampling times to obtain the wind noise propagation law can help users understand the propagation characteristics of wind noise in the actual environment, providing a physical basis for subsequent wind noise processing and detection.

[0143] As an embodiment of the present invention, the spatio-temporal correlation analysis of the dual microphones based on the spatial positions and sampling times to obtain the wind noise propagation law includes: querying the signal sequences of the dual microphones based on the sampling times, aligning the signal sequences on the time axis to obtain aligned signals, analyzing the signal delay of the aligned signals, calculating the propagation distance of the dual microphones based on the spatial positions, analyzing the propagation speed and propagation direction of the dual microphones based on the signal delay and the propagation distance, querying the signal intensities of the dual microphones, analyzing the signal attenuation trend of the dual microphones based on the signal intensities and the propagation speed, and determining the wind noise propagation law of the dual microphones based on the propagation speed, the propagation direction, and the signal attenuation trend.

[0144] Optionally, aligning the signal sequences on the time axis to obtain aligned signals can be achieved by finding signal feature points (such as peaks, zero-crossing points) and calibrating the two signal sequences on the time axis; analyzing the signal delay of the aligned signals can be calculated and analyzed using the cross-correlation function (the time offset corresponding to the peak of the cross-correlation function is the time difference between the wind noise signal reaching the dual microphones, that is, the signal delay); the propagation distance can be obtained using the spatial distance formula based on the spatial position coordinates of the dual microphones; the propagation speed can be obtained by dividing the propagation distance by the signal delay, and the propagation direction can be determined based on the order of the positions of the dual microphones and the signal delay. It should be noted that the side of the microphone that receives the signal first is the incoming direction; the signal attenuation trend can be determined by obtaining the signal intensities at the dual microphones, combining the propagation speed, and fitting a function (such as inverse square or exponential function) by comparing the intensity changes at different distances.

[0145] In the embodiments of the present invention, constructing the dynamic wind noise threshold of the dual microphones using the signal intensities, the frequency characteristics, and the wind noise propagation law can more accurately reflect the true level of wind noise using the dynamic wind noise threshold, avoid misjudgment that may be caused by a fixed threshold, and provide a flexible standard for more accurate wind noise detection.

[0146] Among them, the wind noise dynamic threshold refers to a judgment criterion that dynamically changes according to various characteristics of wind noise (such as signal strength, frequency characteristics, propagation law, etc.). It is not a fixed value, but will be adaptively adjusted as the wind noise changes under different time and space conditions.

[0147] Optionally, the process of constructing the wind noise dynamic threshold of the dual microphone by using the signal strength, the frequency characteristics and the wind noise propagation law is as follows: First, according to the wind noise propagation law, if the wind noise propagates rapidly from a specific direction and the signal strength is large, it indicates that the influence of the wind noise is serious, and the dynamic threshold can be appropriately increased; Second, considering the frequency characteristics, if the wind noise is concentrated in the high frequency band, a lower threshold is set for the high frequency signal and a relatively higher threshold is set for the low frequency signal. For example, when the signal strength of the wind noise significantly increases in a certain period and the high frequency components increase, it is judged according to the propagation law that the wind noise is approaching. At this time, the dynamic threshold is greatly reduced in the high frequency band and slightly reduced in the low frequency band to accurately detect the wind noise.

[0148] Furthermore, the real-time wind noise detection of the dual microphone by the embodiment of the present invention based on the wind noise dynamic threshold can adjust the detection standard according to the real-time characteristics of the wind noise, improve the accuracy and real-time performance of the wind noise detection, ensure the stability and reliability of the system in different environments and different times, improve the system performance, and reduce the adverse effects of the wind noise on the subsequent signal processing and applications.

[0149] Embodiment 2: As Figure 2 shown, it is a schematic diagram of the functional modules of a wind noise detection system of a dual microphone of the present invention.

[0150] The wind noise detection system 200 of the dual microphone of the present invention can be installed in an electronic device. According to the functions achieved, the wind noise detection system of the dual microphone can include a signal processing module 201, a spectrum component identification module 202, a signal analysis module 203 and a wind noise detection module 204. The modules of the present invention can also be referred to as units, which refer to a series of computer program segments that can be executed by a processor of an electronic device and can complete fixed functions, and are stored in the memory of the electronic device.

[0151] In the embodiment of the present invention, the functions of each module / unit are as follows:

[0152] The signal processing module 201 is used to obtain the dual microphone to be detected for wind noise, perform acoustic space deployment on the dual microphone to obtain a deployed dual microphone, collect the time-domain audio signal of the deployed dual microphone by using a pre-configured phase-locked loop synchronization technology, analyze the desired signal of the time-domain audio signal, and perform signal iterative processing on the time-domain audio signal based on the desired signal to obtain a target signal;

[0153] The spectrum component identification module 202 is configured to perform short-time segmentation on the target signal to obtain segmented signals, perform short-time frequency-domain transformation on the segmented signals to obtain the time-frequency spectrogram of the segmented signals, identify the silent-segment spectra in the time-frequency spectrogram, analyze the spectral characteristics of the silent-segment spectra, determine the background noise spectrum of the time-frequency spectrogram based on the spectral characteristics, perform smoothing correction on the background noise spectrum to obtain a corrected spectrum, and identify the spectrum components in the time-frequency spectrogram based on the corrected spectrum to obtain a target spectrum;

[0154] The signal analysis module 203 is configured to calculate the linear transformation matrix of the target spectrum, and perform signal independent component analysis on the target spectrum based on the linear transformation matrix to identify the wind noise signal in the target spectrum, analyze the wind noise signal characteristics corresponding to the wind noise signal, identify the full wind noise signal in the target spectrum based on the wind noise signal characteristics, calculate the signal intensity of the full wind noise signal, and analyze the frequency characteristics of the full wind noise signal;

[0155] The wind noise detection module 204 is configured to query the spatial positions and sampling times of the dual microphones, perform spatio-temporal correlation analysis on the dual microphones based on the spatial positions and the sampling times to obtain the wind noise propagation law, construct the dynamic wind noise threshold of the dual microphones by using the signal intensity, the frequency characteristics, and the wind noise propagation law, and perform real-time wind noise detection on the dual microphones based on the dynamic wind noise threshold.

[0156] Specifically, each module in the wind noise detection system 200 of a dual microphone in the embodiment of the present invention adopts the same technical means as those in the Figure 1 a wind noise detection method of a dual microphone described above, and can produce the same technical effects, which will not be elaborated here.

[0157] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms.

[0158] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A dual-microphone wind noise detection method, characterized in that: The method comprises: Acquire dual microphones for wind noise detection, perform acoustic spatial deployment on the dual microphones to obtain deployed dual microphones, collect time domain audio signals of the deployed dual microphones using a preconfigured phase-locked loop synchronization technology, analyze expected signals of the time domain audio signals, and perform signal iterative processing on the time domain audio signals based on the expected signals to obtain target signals; The target signal is segmented into short time periods to obtain segmented signals, the segmented signals are transformed into short-time frequency domain to obtain a time-frequency spectrum of the segmented signal, a silent segment spectrum in the time-frequency spectrum is identified, a spectrum feature of the silent segment spectrum is analyzed, a background noise spectrum of the time-frequency spectrum is determined based on the spectrum feature, the background noise spectrum is smoothed and corrected to obtain a corrected spectrum, and a spectrum component in the time-frequency spectrum is identified based on the corrected spectrum to obtain a target spectrum; Calculating a linear transformation matrix of the target spectrum, performing signal independent component analysis on the target spectrum based on the linear transformation matrix to identify a wind noise signal in the target spectrum, analyzing a wind noise signal feature corresponding to the wind noise signal, identifying a full wind noise signal of the target spectrum based on the wind noise signal feature, calculating a signal strength of the full wind noise signal, and analyzing a frequency characteristic of the full wind noise signal; The spatial positions and sampling times of the dual microphones are queried, and based on the spatial positions and the sampling times, spatiotemporal correlation analysis is performed on the dual microphones to obtain a wind noise propagation law, and a wind noise dynamic threshold of the dual microphones is constructed by using the signal strength, the frequency characteristics, and the wind noise propagation law, and real-time wind noise detection is performed on the dual microphones based on the wind noise dynamic threshold.

2. A dual-microphone wind noise detection method as claimed in claim 1, characterized in that: The step of performing acoustic spatial deployment on the dual microphones to obtain deployed dual microphones includes: Querying the sound signal wavelength and expected spatial resolution of the dual microphones to determine the arrangement spacing of the dual microphones; Identifying the sound source direction and sound change range of the dual microphones to determine the arrangement angle of the dual microphones; Based on the arrangement spacing and the arrangement angle, the dual microphones are preliminarily spatially arranged to obtain initial arranged microphones; Signal synchronization verification is performed on the initially deployed microphones, and when a synchronization result of the signal synchronization verification satisfies a preset synchronization value, dual microphones are deployed.

3. The dual-microphone wind noise detection method according to claim 1, characterized in that: The analyzing the expected signal of the time-domain audio signal comprises: Removing a DC offset of the time-domain audio signal to obtain a preliminary processed signal; Performing frequency screening on the preliminary processed signal to obtain a screened signal; Performing noise estimation on the screening signal to obtain estimated noise; The estimated noise is removed by using an adaptive filtering method to obtain a clean signal; analyzing a clean signal feature of the clean signal; Based on the clean signal characteristics, a desired signal of the time-domain audio signal is determined.

4. The dual-microphone wind noise detection method according to claim 1, characterized in that: The performing short-time frequency domain transformation on the segmented signal to obtain a time-frequency spectrum diagram of the segmented signal includes: The segmented signal is transformed into a frequency domain in a short time period using the following formula to obtain a frequency domain transformed signal: ； in, represents the frequency domain transformed signal, represents the segmentation signal, w() represents the window function, N represents the length of the window in the window function, m represents the time index, R represents the sliding step of the window, Y represents the frequency index, j represents the imaginary unit, and n represents the time index of the segmentation signal in the window function; The frequency domain transformed signals are arranged in time sequence to obtain a time-frequency spectrum diagram of the segmented signals.

5. The dual-microphone wind noise detection method according to claim 1, characterized in that: The step of identifying the silent segment spectrum in the time-spectrogram comprises: The spectrum energy of the time window in the time-spectrogram is calculated using the following formula: ； in, represents the spectrum energy, represents the i-th window in the time window, represents the kth frequency in the time-spectrum diagram, The spectrum diagram is and The spectrum amplitude at , K represents the number of frequencies in the time spectrum graph; constructing a silent energy threshold of the time-spectrogram; Based on the silence energy threshold and the spectrum energy, a silence segment spectrum in the time-spectrogram is determined.

6. The dual-microphone wind noise detection method according to claim 1, characterized in that: The analyzing the spectrum characteristics of the silent segment spectrum includes: Constructing a spectrum curve of the silent segment spectrum; Analyzing the spectrum shape characteristics of the silent segment spectrum using the spectrum curve; Identifying the energy distribution state of the silent segment spectrum on the frequency axis; Based on the energy distribution state, identifying the frequency distribution characteristics of the silent segment spectrum; Calculating the statistical value of the silent segment spectrum; Analyze the spectrum dispersion of the silent segment spectrum based on the statistical value; Identifying the spectrum statistical features of the silent segment spectrum based on the spectrum dispersion; Calculating the average value of the absolute value of the amplitude difference between adjacent frequency points in the silent segment spectrum to analyze the spectrum smoothness characteristics of the silent segment spectrum; The spectrum characteristics of the silent segment spectrum are determined based on the spectrum shape characteristics, the frequency distribution characteristics, the spectrum statistical characteristics and the spectrum smoothness characteristics.

7. The dual-microphone wind noise detection method according to claim 1, characterized in that: The identifying the full wind noise signal of the target spectrum based on the wind noise signal feature includes: Querying the time domain features, frequency domain features and high-order statistical features of the wind noise signal to obtain comprehensive features; Perform wind noise feature matching on the target spectrum based on the comprehensive feature to obtain a matching feature; Constructing a wind noise judgment rule of the matching features by using a weighting method; Based on the wind noise judgment rule, a preliminary judgment is made on the full wind noise signal of the target spectrum to obtain a preliminary full wind noise signal; The preliminary full wind noise signal is subjected to continuous regional integration analysis to determine the full wind noise signal of the target spectrum.

8. The dual-microphone wind noise detection method according to claim 1, characterized in that: The calculating the signal strength of the full wind noise signal includes: The mean square value of the total wind noise signal is calculated using the following formula: ； Among them, RMS represents the mean square value, M represents the sequence number of the full wind noise signal, represents the bth signal in the full wind noise signal, Analyzing the time domain signal strength of the full wind noise signal by using the mean square value; The sum of the squares of the spectrum amplitudes of the full wind noise signal is calculated using the following formula: ； Among them, V represents the sum of squares of spectrum amplitude, H represents the number of frequency points of the full wind noise signal, represents the spectrum amplitude at frequency point h, Analyzing the frequency domain signal strength of the full wind noise signal by using the sum of squares of the spectrum amplitude; The signal strength of the full wind noise signal is determined according to the time domain signal strength and the frequency domain signal strength.

9. The dual-microphone wind noise detection method according to claim 1, characterized in that: The performing of spatiotemporal correlation analysis on the dual microphones based on the spatial position and the sampling time to obtain the wind noise propagation law includes: Query the signal sequence of the dual microphones based on the sampling time; Aligning the signal sequences on the time axis to obtain an alignment signal; analyzing a signal delay of the alignment signal; Calculating the propagation distance of the dual microphones based on the spatial position; Analyzing the propagation speed and propagation direction of the dual microphones based on the signal delay and the propagation distance; Query the signal strength of the dual microphones; Analyzing the signal attenuation trend of the dual microphones based on the signal strength and the propagation speed; The wind noise propagation law of the dual microphones is determined based on the propagation speed, the propagation direction and the signal attenuation trend.

10. A dual-microphone wind noise detection system, characterized in that: The system comprises: A signal processing module is used to obtain dual microphones to be detected for wind noise, perform acoustic spatial deployment on the dual microphones to obtain deployed dual microphones, collect time domain audio signals of the deployed dual microphones using a preconfigured phase-locked loop synchronization technology, analyze an expected signal of the time domain audio signal, and perform signal iterative processing on the time domain audio signal based on the expected signal to obtain a target signal; a spectrum component identification module, for performing short-time segmentation on the target signal to obtain segmented signals, performing short-time frequency domain transformation on the segmented signals to obtain a time-frequency spectrum of the segmented signals, identifying a silent segment spectrum in the time-frequency spectrum, analyzing a spectrum feature of the silent segment spectrum, determining a background noise spectrum of the time-frequency spectrum based on the spectrum feature, performing smoothing correction on the background noise spectrum to obtain a corrected spectrum, identifying spectrum components in the time-frequency spectrum based on the corrected spectrum to obtain a target spectrum; a signal analysis module, configured to calculate a linear transformation matrix of the target spectrum, perform signal independent component analysis on the target spectrum based on the linear transformation matrix to identify a wind noise signal in the target spectrum, analyze a wind noise signal feature corresponding to the wind noise signal, identify a full wind noise signal of the target spectrum based on the wind noise signal feature, calculate a signal strength of the full wind noise signal, and analyze a frequency characteristic of the full wind noise signal; The wind noise detection module is used to query the spatial position and sampling time of the dual microphones, perform spatiotemporal correlation analysis on the dual microphones based on the spatial position and the sampling time, obtain the wind noise propagation law, use the signal strength, the frequency characteristics and the wind noise propagation law to construct the wind noise dynamic threshold of the dual microphones, and perform real-time wind noise detection on the dual microphones based on the wind noise dynamic threshold.

Images