Audio signal denoising method based on improved variational mode decomposition and related product

Abstract

The application provides an audio signal denoising method based on improved variational mode decomposition and a related product. The method comprises the following steps: obtaining a target input audio signal, and using an improved particle swarm optimization algorithm with a local optimal solution jump-out mechanism and a fuzzy entropy search variational mode decomposition to search for a target component number and a target penalty factor; decomposing the target input audio signal according to the target component number and the target penalty factor to obtain a plurality of first target components; calculating the fuzzy entropy of each first target component, and using a wavelet threshold denoising algorithm to denoise the first target components with fuzzy entropy higher than a preset fuzzy entropy threshold to obtain first components; and obtaining a target output audio signal according to the first target components with fuzzy entropy not higher than the preset fuzzy entropy threshold and the first components. The application can solve the problem of poor denoising effect caused by unreasonable parameter setting in the prior art, thereby realizing more efficient and accurate audio signal denoising processing.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of audio denoising, and in particular to an audio signal denoising method based on improved variational mode decomposition, an audio signal denoising system based on improved variational mode decomposition, an electronic device and a computer readable storage medium. BACKGROUND

[0002] With the development of digital communication technology and multimedia applications, the demand for high-quality audio signals is increasing. However, in actual application scenarios, due to the influence of environmental noise, equipment interference and other factors, the original audio signals collected often contain different degrees of noise, which not only affects the auditory experience, but also may interfere with the subsequent information processing process. Although the traditional filter design can improve the signal quality to a certain extent, the effect of noise removal is limited for non-stationary or complex background. In recent years, the variational mode decomposition technology has been widely concerned because it can adaptively decompose the mixed signal into a series of intrinsic mode functions, but there are still challenges in how to select appropriate decomposition parameters to achieve the optimal denoising effect. SUMMARY

[0003] In view of the above problems, an audio signal denoising method based on improved variational mode decomposition, an audio signal denoising system based on improved variational mode decomposition, an electronic device and a computer readable storage medium are provided to overcome the above problems or at least partially solve the above problems, comprising:

[0004] An audio signal denoising method based on improved variational mode decomposition, the method comprising:

[0005] Obtaining a target input audio signal, and using an improved particle swarm optimization algorithm introducing a local optimal solution jump-out mechanism and a fuzzy entropy search variational mode decomposition to search for a target component number and a target penalty factor;

[0006] According to the target component number and the target penalty factor, the target input audio signal is decomposed to obtain a plurality of first target components;

[0007] The fuzzy entropy of each first target component is calculated, and the wavelet threshold denoising algorithm is used to denoise the first target component with fuzzy entropy higher than a preset fuzzy entropy threshold to obtain a first component;

[0008] According to the first target component with fuzzy entropy not higher than the preset fuzzy entropy threshold and the first component, a target output audio signal is obtained.

[0009] Optionally, the improved particle swarm optimization algorithm introducing a local optimal solution jump-out mechanism and the fuzzy entropy search variational mode decomposition to search for a target component number and a target penalty factor, comprising:

[0010] initialize a position and a velocity of each individual particle, and a position of a random walk particle; wherein the position comprises a number of components and a penalty factor;

[0011] calculate a fitness of a current position of each individual particle;

[0012] update the velocity and the position of each individual particle, and record a current global optimal solution fitness of the individual particle;

[0013] if the fitness of the random walk particle is less than the current global optimal solution fitness, replace a position of an individual particle corresponding to the current global optimal solution fitness with the position of the random walk particle;

[0014] when an iteration termination condition is met, output the target number of components and the target penalty factor.

[0015] Optionally, the calculating the fitness of the current position of each individual particle comprises:

[0016] performing decomposition on the target input audio signal according to the current position of each individual particle to obtain a plurality of second target components;

[0017] calculating an average fuzzy entropy of the plurality of second target components, and taking the average fuzzy entropy as the fitness of the individual particle;

[0018] updating the velocity and the position of the individual particle, and updating the position of the random walk particle, and recording the current global optimal solution fitness of the individual particle;

[0019] if the fitness of the random walk particle is less than the current global optimal solution fitness, replace the position of the individual particle corresponding to the current global optimal solution fitness with the position of the random walk particle;

[0020] when the iteration termination condition is met, output the target number of components and the target penalty factor.

[0021] Optionally, the obtaining the target output audio signal according to the first target component with the fuzzy entropy not higher than the preset fuzzy entropy threshold and the first component comprises:

[0022] reconstructing the first target component with the fuzzy entropy not higher than the preset fuzzy entropy threshold and the first component to obtain the target output audio signal after denoising.

[0023] Optionally, the performing the decomposition on the target input audio signal according to the target number of components and the target penalty factor to obtain a plurality of first target components comprises:

[0024] According to the target component number and the target penalty factor, an improved variational mode decomposition algorithm is called to decompose the target input audio signal, and a plurality of first target components are obtained.

[0025] Optionally, the target output audio signal is obtained according to the first target component with the fuzzy entropy not higher than the preset fuzzy entropy threshold and the first component, including:

[0026] A target audio encoding format is determined.

[0027] According to the first target component with the fuzzy entropy not higher than the preset fuzzy entropy threshold and the first component, the target output audio signal in the target audio encoding format is generated.

[0028] Embodiments of the present application also provide an audio signal denoising system based on improved variational mode decomposition, the system comprising:

[0029] An input interface module is configured to receive an input audio signal.

[0030] A preprocessing module is configured to preprocess the input audio signal to obtain a target input audio signal.

[0031] A core processing module is configured to use an improved particle swarm optimization algorithm with a local optimal solution jump-out mechanism and a target component number and a target penalty factor searched by fuzzy entropy to decompose the target input audio signal according to the target component number and the target penalty factor, to obtain a plurality of first target components, to calculate the fuzzy entropy of each first target component, to use a wavelet threshold denoising algorithm to denoise the first target component with the fuzzy entropy higher than a preset fuzzy entropy threshold to obtain a first component, and to obtain an output audio signal according to the first target component with the fuzzy entropy not higher than the preset fuzzy entropy threshold and the first component.

[0032] An output interface module is configured to generate a target output audio signal in a target audio encoding format according to the output audio signal.

[0033] Optionally, the core processing module comprises:

[0034] A parameter optimization module is configured to use an improved particle swarm optimization algorithm with a local optimal solution jump-out mechanism and a target component number and a target penalty factor searched by fuzzy entropy for variational mode decomposition.

[0035] A signal decomposition module is configured to decompose the target input audio signal according to the target component number and the target penalty factor to obtain a plurality of first target components.

[0036] The noise reduction module is used to calculate the fuzzy entropy of each first target component and use the wavelet threshold denoising algorithm to denoise the first target components whose fuzzy entropy is higher than the preset fuzzy entropy threshold to obtain the first component.

[0037] The signal reconstruction module is used to obtain the output audio signal based on a first target component whose fuzzy entropy is not higher than a preset fuzzy entropy threshold and the first component.

[0038] This invention also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and capable of running on the processor. When the computer program is executed by the processor, it implements the above-described audio signal denoising method based on improved variational mode decomposition.

[0039] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described audio signal denoising method based on improved variational mode decomposition.

[0040] The embodiments of the present invention have the following advantages:

[0041] In this embodiment of the invention, a target input audio signal is acquired, and an improved particle swarm optimization algorithm with a local optimum escape mechanism and a fuzzy entropy search variational mode decomposition are used to determine the number of target components and the target penalty factor. Based on the number of target components and the target penalty factor, the target input audio signal is decomposed to obtain multiple first target components. The fuzzy entropy of each first target component is calculated, and a wavelet threshold denoising algorithm is used to denoise the first target components whose fuzzy entropy is higher than a preset fuzzy entropy threshold, resulting in a first component. Based on the first target components whose fuzzy entropy is not higher than the preset fuzzy entropy threshold and the first components, the target output audio signal is obtained. This embodiment of the invention solves the problem of poor denoising effect caused by unreasonable parameter settings in the prior art, thereby achieving more efficient and accurate audio signal denoising processing. Attached Figure Description

[0042] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the description of the present invention will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a flowchart illustrating the steps of an audio signal denoising method based on improved variational mode decomposition according to an embodiment of the present invention.

[0044] Figure 2is a step flow chart of another audio signal denoising method based on improved variational mode decomposition according to an embodiment of the present application;

[0045] Figure 3 is a step flow chart of still another audio signal denoising method based on improved variational mode decomposition according to an embodiment of the present application;

[0046] Figure 4 is a structural schematic diagram of an audio signal denoising system based on improved variational mode decomposition according to an embodiment of the present application;

[0047] Figure 5 is a partial structural schematic diagram of an audio signal denoising system based on improved variational mode decomposition according to an embodiment of the present application. DETAILED DESCRIPTION

[0048] In order to make the above objectives, characteristics and advantages of the present application more apparent, comprehensible and easier to understand, the present application will be further described in detail below with reference to the drawings and specific embodiments. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0049] The embodiments of the present application provide an audio signal denoising method based on improved variational mode decomposition, aiming to solve the problem of poor denoising effect caused by unreasonable parameter setting in the prior art, and to realize more efficient and accurate audio signal denoising processing. Specifically, the audio signal denoising method based on improved variational mode decomposition can include the following steps: Figure 1 , a step flow chart of an audio signal denoising method based on improved variational mode decomposition according to an embodiment of the present application is shown.

[0050] As shown in Figure 1 , the audio signal denoising method based on improved variational mode decomposition can include the following steps:

[0051] Step 101, obtaining a target input audio signal, and using an improved particle swarm optimization algorithm with a local optimal solution jump-out mechanism and a fuzzy entropy search variational mode decomposition to search for a target component number and a target penalty factor.

[0052] Firstly, an audio signal sensor can collect an audio signal; an audio signal denoising system can obtain a target input audio signal from the audio signal sensor and perform denoising processing thereon.

[0053] Specifically, to solve the problems that the existing audio signal denoising method often relies on manual parameter setting, which easily leads to poor denoising effect or over-denoising, and to solve the problems of mode aliasing, overfitting, information loss and other problems caused by improper determination of the number of decompositions and penalty factors by the variational mode decomposition algorithm, and to solve the problem that the traditional optimization algorithm is easy to fall into a local optimal solution, the embodiments of the present application can first use the improved particle swarm optimization algorithm with a local optimal solution jump-out mechanism and fuzzy entropy search to search for the number of target components and the target penalty factor of variational mode decomposition, thereby improving the global search ability of the PSO (Particle Swarm Optimization) algorithm and avoiding the problem that the traditional PSO algorithm is easy to fall into a local optimal solution. And by using the optimal super parameter combination searched by the improved particle swarm optimization algorithm for VMD (Variational Mode Decomposition), different frequency components in the signal can be more effectively separated, and the decomposition precision is improved. This not only helps to remove noise, but also preserves useful information in the signal, ensuring that the audio quality after denoising is higher.

[0054] The number of target components refers to the number of components of the objective function in the optimization problem. In a multi-objective optimization problem, the objective function is usually composed of multiple components, each of which represents a different optimization objective.

[0055] The target penalty factor is used to introduce a penalty term in the objective function to constrain or penalize certain undesirable behaviors or states. The target penalty factor can be a constant or a function, which is used to adjust the value of the objective function so that it is more reasonable or meets certain requirements in certain cases.

[0056] Step 102, decompose the target input audio signal according to the number of target components and the target penalty factor to obtain a plurality of first target components.

[0057] After determining the optimal number of target components and the optimal target penalty factor, the target input audio signal can be decomposed using the optimal number of target components and the optimal target penalty factor, thereby obtaining a plurality of first target components. The first target component refers to a feature extracted from the target input audio signal.

[0058] Step 103, calculate the fuzzy entropy of each first target component, and use a wavelet threshold denoising algorithm to denoise the first target component whose fuzzy entropy is higher than a preset fuzzy entropy threshold, to obtain a first component.

[0059] Next, the fuzzy entropy of each first target component can be calculated, and the size relationship between the fuzzy entropy corresponding to each first target component and the preset fuzzy entropy threshold can be compared respectively. The fuzzy entropy is a method for measuring the complexity and irregularity of a time series.

[0060] After determining the first target component with the fuzzy entropy higher than the preset fuzzy entropy threshold, the wavelet threshold denoising algorithm can be used to denoise the first target component to obtain the first component.

[0061] The wavelet threshold denoising algorithm is a widely used denoising method in the fields of signal processing and image processing. It utilizes the characteristics of wavelet transform to decompose the signal into different frequency subbands, then removes the noise components by setting a threshold, and finally reconstructs the signal through wavelet inverse transform.

[0062] Step 104: obtaining a target output audio signal according to the first target component with the fuzzy entropy not higher than the preset fuzzy entropy threshold and the first component.

[0063] After obtaining the first component, the target output audio signal can be generated and obtained according to the first target component with the fuzzy entropy not higher than the preset fuzzy entropy threshold and the first component, and the target output audio signal can be output.

[0064] In the embodiment of the present application, the target input audio signal is obtained, and the improved particle swarm optimization algorithm with the introduction of local optimal solution jump-out mechanism and the target component number and target penalty factor of fuzzy entropy search variational mode decomposition are used; the target input audio signal is decomposed according to the target component number and the target penalty factor to obtain a plurality of first target components; the fuzzy entropy of each first target component is calculated, and the wavelet threshold denoising algorithm is used to denoise the first target component with the fuzzy entropy higher than the preset fuzzy entropy threshold to obtain the first component; and the target output audio signal is obtained according to the first target component with the fuzzy entropy not higher than the preset fuzzy entropy threshold and the first component. Through the embodiment of the present application, the problem of poor denoising effect caused by unreasonable parameter setting in the prior art can be solved, so that more efficient and accurate audio signal denoising processing is realized.

[0065] Referring to Figure 2 , another step flowchart of the audio signal denoising method based on improved variational mode decomposition is shown, which can include the following steps:

[0066] Step 201: obtaining a target input audio signal.

[0067] First, the audio signal sensor can collect the audio signal; the audio signal denoising system can obtain the target input audio signal from the audio signal sensor and perform denoising processing thereon.

[0068] Exemplarily, the audio denoising system can obtain an original input audio signal from an audio signal sensor line, and pre-process the original input audio signal to obtain a target input audio signal.

[0069] Step 202, initializing the position and speed of each individual particle, and the position of a random walk particle; wherein the position includes a component number and a penalty factor.

[0070] To solve the limitation that the PSO is easy to fall into a local optimal solution, a local optimal solution jump-out mechanism is introduced, that is, a random walk particle not participating in position iteration is added to increase the possibility of jumping out of a local optimal solution.

[0071] Specifically, the position and speed of each individual particle, and the position of a random walk particle can be initialized first; wherein the position of the individual particle includes a component number and a penalty factor, and the position of the random walk particle also includes a corresponding component number and a penalty factor.

[0072] Exemplarily, the initialization can be set according to actual conditions, and embodiments of the present application do not limit this,

[0073] Step 203, calculating the fitness of the current position of each individual particle.

[0074] In some feasible embodiments, the fitness of the current position of each individual particle can be calculated; the fitness is used to evaluate the position of the particle in the search space, thereby guiding the movement and optimization process of the particle.

[0075] In an embodiment of the present application, step 203 can be implemented through the following sub-steps:

[0076] Sub-step 11, decomposing the target input audio signal according to the current position of each individual particle to obtain a plurality of second target components.

[0077] In some feasible embodiments, the target input audio signal can be decomposed according to the current position of each individual particle first, and the decomposition can obtain a plurality of second target components.

[0078] Sub-step 12, calculating the average fuzzy entropy of the plurality of second target components, and taking the average fuzzy entropy as the fitness of the individual particle.

[0079] Then, the average fuzzy entropy of each second target component can be calculated, and the average fuzzy entropy is taken as the fitness of the individual particle.

[0080] Sub-step 13, updating the speed and position of the individual particle, and updating the position of the random walk particle, and recording the current global optimal solution fitness of the individual particle.

[0081] Next, the velocity and position of the individual particle can be updated, and the position of the random walk particle can be updated, and the current global optimal solution fitness of the individual particle is recorded.

[0082] Sub-step 14, if the fitness of the random walk particle is less than the current global optimal solution fitness, the position of the individual particle corresponding to the current global optimal solution fitness is replaced by the position of the random walk particle.

[0083] In some possible embodiments, if the fitness of the random walk particle is less than the current global optimal solution fitness, the position of the individual particle corresponding to the current global optimal solution fitness can be replaced by the position of the random walk particle.

[0084] Sub-step 15, when the iteration termination condition is met, the target number of components and the target penalty factor are output.

[0085] In some possible embodiments, the iteration termination condition can be preset; when the iteration process meets the iteration termination condition, the current number of components and the penalty factor (i.e. the position corresponding to the current global optimal solution fitness) can be output, and the output number of components is taken as the target number of components, and the output penalty factor is taken as the target penalty factor.

[0086] Step 204, the velocity and position of each individual particle are updated, and the current global optimal solution fitness of the individual particle is recorded.

[0087] In the particle swarm optimization algorithm, the velocity and position of each individual particle can be updated to constantly explore and optimize in the search space, and finally find the current global optimal solution fitness of the individual particle, i.e. the fitness of the current optimal solution.

[0088] Step 205, if the fitness of the random walk particle is less than the current global optimal solution fitness, the position of the individual particle corresponding to the current global optimal solution fitness is replaced by the position of the random walk particle.

[0089] If the fitness of the random walk particle is less than the current global optimal solution fitness, the position of the individual particle corresponding to the current global optimal solution fitness can be replaced by the position of the random walk particle.

[0090] Step 206, when the iteration termination condition is met, the target number of components and the target penalty factor are output.

[0091] In some possible embodiments, the iteration termination condition can be preset; when the iteration process meets the iteration termination condition, the current number of components and the penalty factor can be output, and the output number of components is taken as the target number of components, and the output penalty factor is taken as the target penalty factor.

[0092] Exemplarily, the iteration termination condition can be set based on the number of iterations, or based on iteration time, etc., and embodiments of the present application do not limit this.

[0093] Step 207: According to the target component number and the target penalty factor, an improved variational mode decomposition algorithm is called to decompose the target input audio signal to obtain a plurality of first target components.

[0094] After obtaining the target component number and the target penalty factor, the improved variational mode decomposition algorithm can be called to decompose the target input audio signal, so as to obtain a plurality of first target components based on the target input audio signal.

[0095] Step 208: The fuzzy entropy of each first target component is calculated, and a wavelet threshold denoising algorithm is used to denoise the first target component whose fuzzy entropy is higher than a preset fuzzy entropy threshold to obtain a first component.

[0096] After obtaining a plurality of first target components, the fuzzy entropy of each first target component can be calculated respectively, and the size relationship between the fuzzy entropy corresponding to each first target component and the preset fuzzy entropy threshold can be compared respectively.

[0097] After determining the first target component whose fuzzy entropy is higher than the preset fuzzy entropy threshold, a wavelet threshold denoising algorithm can be used to denoise the first target component to obtain a first component.

[0098] Step 209: According to the first target component whose fuzzy entropy is not higher than the preset fuzzy entropy threshold and the first component, a target output audio signal is obtained.

[0099] After obtaining the first component, the target output audio signal can be generated and obtained according to the first target component whose fuzzy entropy is not higher than the preset fuzzy entropy threshold and the first component, and the target output audio signal is output.

[0100] In an embodiment of the present application, step 209 can be implemented through the following sub-steps:

[0101] Sub-step 21: The first target component whose fuzzy entropy is not higher than the preset fuzzy entropy threshold and the first component are reconstructed to obtain a denoised target output audio signal.

[0102] In some feasible embodiments, the first target component whose fuzzy entropy is not higher than the preset fuzzy entropy threshold and the first component can be reconstructed to generate and obtain a denoised target output audio signal. By reconstructing the denoised IMF (Intrinsic Mode Function, intrinsic mode function) component, a high-quality denoised audio signal can be obtained. This method not only improves the denoising effect, but also ensures the integrity of the signal, and is suitable for various complex audio signal processing scenarios.

[0103] In an embodiment of the present application, step 209 can also be implemented by the following sub-steps:

[0104] Sub-step 31, determining the target audio encoding format.

[0105] In some feasible embodiments, multiple mainstream audio encoding format export can be supported. Specifically, the target audio encoding format of the audio signal to be output can be determined first.

[0106] Sub-step 32, generating the target output audio signal in the target audio encoding format according to the first target component and the first component whose fuzzy entropy is not higher than the preset fuzzy entropy threshold.

[0107] Then, the target output audio signal in the target audio encoding format can be generated according to the first target component and the first component whose fuzzy entropy is not higher than the preset fuzzy entropy threshold, and the target output audio signal in the target audio encoding format can be output.

[0108] In some feasible embodiments, the audio denoising system can support download function, i.e. allowing to download the target output audio signal in the target audio encoding format; and can also support online playing, i.e. the target output audio signal in the target audio encoding format can be directly listened to in the audio denoising system, and the embodiments of the present application do not limit this.

[0109] For example, as shown in FIG. 6, the target input audio signal is first obtained from the sensor. The number of particles of PSO and the iteration termination condition are set, and the position K, β) of each individual particle and the random walk particle is initialized. Figure 3

[0110] Next, the fitness of the current position of each particle can be calculated, and the speed and position of each particle are updated, and the historical optimal solution of the individual particle and the fitness of the current global optimal solution are recorded.

[0111] For each individual particle, the target input audio signal can be first decomposed by VMD according to the current position of the individual particle, to obtain multiple second target components. Then, the fuzzy entropy corresponding to each second target component is calculated, and the average fuzzy entropy of all second target components is taken as the fitness of the individual particle.

[0112] If the fitness of the random walk particle is less than the fitness of the current global optimal solution, the position of the individual particle corresponding to the fitness of the current global optimal solution is replaced by the position of the random walk particle.

[0113] When the iteration termination condition is met, the number of target components and the target penalty factor are output.

[0114] ​According to the target component number and the target penalty factor, a VMD algorithm is called to decompose the target input audio signal to obtain a plurality of first target components;

[0115] The fuzzy entropy of each of the plurality of first target components is calculated, a fuzzy entropy threshold is set, if the fuzzy entropy is less than the fuzzy entropy threshold, it indicates that the component carries more useless information (such as white noise), then a wavelet threshold denoising algorithm is applied to process it, otherwise it remains unchanged;

[0116] All effective components processed by screening are recombined into a complete target output audio signal and output.

[0117] Specifically, step one: the improved particle swarm optimization algorithm with local optimal solution jump-out mechanism and fuzzy entropy search can be used to search the optimal component number K and the penalty factor β;

[0118] Step two: the optimal decomposition number K and the penalty factor β obtained can be used to decompose the extracted audio signal;

[0119] Step three: the fuzzy entropy of each first target component obtained by decomposition can be calculated, if it is higher than the fuzzy entropy threshold, the wavelet threshold denoising algorithm is used to denoise the first target component, if it is lower than the fuzzy entropy threshold, the first target component is not processed;

[0120] Step four: the denoised audio signal can be obtained according to component reconstruction.

[0121] In the above step one, the mathematical model of the improved particle swarm optimization algorithm with local optimal solution jump-out mechanism for searching the best super parameter combination of VMD is as follows:

[0122] To solve the limitation of PSO that it is easy to fall into local optimal solution, a local optimal solution jump-out mechanism is introduced, that is, a random walk particle that does not participate in position iteration is added to increase the possibility of jumping out of local optimal solution.

[0123] The super parameter combination (K, β) composed of the component number K and the penalty factor β is regarded as a particle, each particle only has two attributes: speed and position. Each particle searches for the optimal solution in the search space, and the current coordinates of the i-th particle are recorded as X i =(K i ,β i ), the flight speed is V i =(v i1 ,v i2 ), and the average fuzzy entropy of the second target component obtained by each particle is taken as the fitness value.

[0124] The introduced random walk particles only have position attributes, but do not have speed attributes. For the i th particle, the position with the minimum average fuzzy entropy obtained in the iteration process is regarded as the individual historical optimal position, that is, the current local optimal solution The initial value of which is the particle itself, and Pbest i is shared with the entire particle swarm, and the best position of the entire particle swarm at present is found as the current global optimal solution. According to Pbest i , Gbest and the current speed V i , the speed and position of the particle are constantly updated, and the iteration formula is as follows:

[0125]

[0126] Wherein, i = 1, 2, …, m, m represents the total number of particles in the group, r1 and r2 are random numbers between 0 and 1, l1 and l2 are learning factors, and ω is an inertia factor. The position of the walk particle is randomly initialized each time, and if the fitness value of the walk particle is less than the fitness value of Gbest, the position of Gbest is replaced by the position of the walk particle.

[0127] In the above step two, the algorithm principle of variational modal decomposition is as follows:

[0128] VMD converts the original signal f into a constraint optimization problem by constructing a variational problem, and then converts it into a solution of an unconstrained optimization problem through the augmented Lagrange function, as follows:

[0129]

[0130] Wherein, K is the number of modes to be decomposed, {u k}, {v k} correspond to the k th mode component and the center frequency after decomposition respectively, wherein k = 1, 2, …, K. δ(t) is Dirac distribution, * is convolution operator, λ is Lagrange multiplier; β is a penalty factor, which is used to reduce the interference of Gaussian noise.

[0131] The iteration formula of {u k}, v k} and λ is solved by using the alternating direction multiplier iteration algorithm combined with Parseval theorem and Fourier transform, as follows:

[0132]

[0133] In the above step three, the calculation steps of the fuzzy entropy of each first target component are as follows:

[0134] For a group of time series X = (x1, x2, …, x n), m-dimensional space reconstruction is carried out, and m is generally set to 2. The reconstructed time sequence Y is as follows:

[0135] Y(i)=(x i ,x i+1 ,…,x i+m-1 )-x i0

[0136] wherein i=1, 2, …, n-m+1,

[0137] The distance between two time sequences Y(i) and Y(j) is defined as:

[0138]

[0139] A fuzzy membership function is introduced, and the similarity between the time sequences Y(i) and Y(j) is calculated by using the fuzzy function:

[0140]

[0141] wherein r is a similarity tolerance, ω is a fuzzy index, i, j=1, 2, …, n-m+1, and i≠j.

[0142] The function is defined as:

[0143]

[0144] Therefore, the fuzzy entropy of the original time sequence is:

[0145] Fuzzy(m,r,n)=lnΦ m (r)-lnΦ m+1 (r)。

[0146] In the embodiment of the application, by introducing a local optimal solution jumping out mechanism, the global search ability of the PSO algorithm is improved, and the problem that the traditional PSO algorithm is prone to falling into a local optimum is avoided. As an evaluation index, the fuzzy entropy can more accurately reflect the complexity of the signal and the noise level, so that the parameter search is more accurate, and the efficiency and accuracy of parameter optimization are significantly improved, thereby improving the final denoising effect.

[0147] By using the optimal hyperparameter combination searched by the improved particle swarm optimization algorithm for VMD decomposition, different frequency components in the signal can be more effectively separated, and the decomposition accuracy is improved. This not only helps to remove noise, but also preserves useful information in the signal, ensuring that the audio quality after denoising is higher.

[0148] The use of fuzzy entropy makes noise identification more accurate, avoiding the problems of excessive denoising or insufficient denoising. For the first target component with high fuzzy entropy, the wavelet threshold denoising algorithm can effectively remove the noise therein, while the first target component with low fuzzy entropy remains unchanged, thereby maximizing the retention of useful signals while ensuring the denoising effect. This method can better balance the denoising effect and signal fidelity.

[0149] By reconstructing the denoised IMF components, a high-quality denoised audio signal can be obtained. This method not only improves the denoising effect, but also ensures the integrity of the signal, and is suitable for various complex audio signal processing scenarios.

[0150] In the embodiment of the application, a target input audio signal is obtained; the position and speed of each individual particle and the position of a random walk particle are initialized; the position includes the number of components and a penalty factor; the fitness of the current position of each individual particle is calculated; the speed and position of each individual particle are updated, and the current global optimal solution fitness of the individual particle is recorded; if the fitness of the random walk particle is less than the current global optimal solution fitness, the position of the individual particle corresponding to the current global optimal solution fitness is replaced by the position of the random walk particle; when the iteration termination condition is met, the target component number and the target penalty factor are output; the target input audio signal is decomposed by calling the improved variational mode decomposition algorithm according to the target component number and the target penalty factor, to obtain a plurality of first target components; the fuzzy entropy of each first target component is calculated, and the wavelet threshold denoising algorithm is used to denoise the first target component with fuzzy entropy higher than a preset fuzzy entropy threshold, to obtain a first component; the target output audio signal is obtained according to the first target component with fuzzy entropy not higher than the preset fuzzy entropy threshold and the first component. Through the embodiment of the application, the problem of poor denoising effect caused by unreasonable parameter setting in the prior art can be solved, thereby realizing more efficient and accurate audio signal denoising processing.

[0151] It should be noted that, for the method embodiments, they are all described as a series of action combinations for the sake of simple description, but those skilled in the art should know that the embodiments of the present application are not limited to the action order described, because according to the embodiments of the present application, certain steps can be performed in other order or simultaneously. Secondly, those skilled in the art should know that the embodiments described in the specification all belong to preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present application.

[0152] Referring to Figure 4 , a structure schematic diagram of an audio signal denoising system based on improved variational mode decomposition is shown, which can include:

[0153] The input interface module 410 is configured to receive an input audio signal.

[0154] The preprocessing module 420 is configured to preprocess the input audio signal to obtain a target input audio signal.

[0155] The core processing module 430 is configured to utilize the improved particle swarm optimization algorithm with a local optimal solution jump-out mechanism and a target component number and a target penalty factor of fuzzy entropy search variational mode decomposition, decompose the target input audio signal according to the target component number and the target penalty factor to obtain a plurality of first target components, calculate the fuzzy entropy of each first target component, and utilize a wavelet threshold denoising algorithm to denoise the first target component with the fuzzy entropy higher than a preset fuzzy entropy threshold to obtain a first component, and obtain an output audio signal according to the first target component with the fuzzy entropy not higher than the preset fuzzy entropy threshold and the first component.

[0156] The output interface module 440 is configured to generate a target output audio signal in a target audio coding format according to the output audio signal.

[0157] As shown in Figure 5 The core processing module 430 includes:

[0158] The parameter optimization module 431 is configured to utilize the improved particle swarm optimization algorithm with a local optimal solution jump-out mechanism and a target component number and a target penalty factor of fuzzy entropy search variational mode decomposition.

[0159] The signal decomposition module 432 is configured to decompose the target input audio signal according to the target component number and the target penalty factor to obtain a plurality of first target components.

[0160] The denoising module 433 is configured to calculate the fuzzy entropy of each first target component, and utilize a wavelet threshold denoising algorithm to denoise the first target component with the fuzzy entropy higher than a preset fuzzy entropy threshold to obtain a first component.

[0161] The signal reconstruction module 434 is configured to obtain an output audio signal according to the first target component with the fuzzy entropy not higher than the preset fuzzy entropy threshold and the first component.

[0162] As shown in Figure 4 and Figure 5 The audio denoising system shown in the figures, the input interface module 410 is responsible for receiving the audio file to be processed transmitted by the external sensor or data source.

[0163] It supports multiple audio formats (such as WAV, MP3, AAC, etc.); It also supports providing a file upload function, supporting local files and network links; It also supports preliminary checking of the input audio file to ensure file integrity and correct format.

[0164] The pre-processing module 420 can perform necessary format conversion, sampling rate adjustment, and the like. It can convert audio files of different formats into a standard format (such as WAV) supported by the system; it can also adjust the sampling rate of the audio signal to meet the requirements of subsequent processing (for example, 44.1 kHz or 48 kHz); and it can also normalize the audio signal to ensure that the signal amplitude is within a suitable range.

[0165] The core processing module 430 integrates all key algorithm logic and is the operation center of the entire system.

[0166] In the core processing module 430, the parameter optimization module 431 can use an improved particle swarm optimization algorithm to combine the optimal component number K and the penalty factor β of fuzzy entropy search variational mode decomposition. It can introduce a local optimal solution escape mechanism to avoid falling into a local optimum; it can also include the average fuzzy entropy of all components as an evaluation index into the objective function; it can also initialize the particle swarm, set the iteration number termination condition, and record the K and β values that minimize the objective function.

[0167] The signal decomposition module 432 uses the determined optimal K and β values to call the improved variational mode decomposition algorithm to decompose the audio signal. It applies the improved variational mode decomposition algorithm to decompose the audio signal into multiple components; the signal decomposition module 432 can also output each component.

[0168] The noise reduction module 433 is used to calculate the fuzzy entropy of each component and determine whether to perform wavelet threshold denoising according to the fuzzy entropy threshold. It can calculate the fuzzy entropy of each component; it can also set a reasonable fuzzy entropy threshold T; if the fuzzy entropy of a component is higher than the threshold T, it applies a wavelet threshold denoising algorithm to the component for noise reduction; otherwise, it remains unchanged.

[0169] The signal reconstruction module 434 is used to recombine all valid components that have undergone screening processing into a complete audio signal. It can recombine all processed components; it can also generate the final denoised audio signal.

[0170] The output interface module 440 is used to output the final result and supports multiple mainstream audio encoding formats for export. It can convert the processed audio signal into a specified format (such as WAV, MP3, etc.); it can also provide a download function to allow downloading of the denoised audio file; it also supports online playback, which can directly listen to the processed audio effect in the system.

[0171] In the embodiment of the present application, the audio signal denoising system comprises: an input interface module, configured to receive an input audio signal; a preprocessing module, configured to preprocess the input audio signal to obtain a target input audio signal; a core processing module, configured to utilize an improved particle swarm optimization algorithm with a local optimal solution jump-out mechanism and a target component number and a target penalty factor of fuzzy entropy search variational mode decomposition; decompose the target input audio signal according to the target component number and the target penalty factor to obtain a plurality of first target components; calculate the fuzzy entropy of each first target component, and utilize a wavelet threshold denoising algorithm to denoise the first target component with the fuzzy entropy higher than a preset fuzzy entropy threshold to obtain a first component; obtain an output audio signal according to the first target component with the fuzzy entropy not higher than the preset fuzzy entropy threshold and the first component; and an output interface module, configured to generate a target output audio signal in a target audio coding format according to the output audio signal. Through the embodiment of the present application, the problem of poor denoising effect caused by unreasonable parameter setting in the prior art can be solved, so that more efficient and accurate audio signal denoising processing is realized.

[0172] The embodiment of the present application also provides an electronic device, comprising a processor, a memory, and a computer program stored on the memory and capable of running on the processor, and the computer program is executed by the processor to realize the audio signal denoising method based on improved variational mode decomposition as above.

[0173] The embodiment of the present application also provides a computer readable storage medium, and the computer readable storage medium stores a computer program, and the computer program is executed by the processor to realize the audio signal denoising method based on improved variational mode decomposition as above.

[0174] For the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the related parts are described in the part of the method embodiment.

[0175] Each embodiment in the specification is described in a progressive manner, and each embodiment focuses on the difference from other embodiments, and the same and similar parts of each embodiment can be referred to each other.

[0176] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a device, or a computer program product. Therefore, the embodiments of the present application can adopt a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the embodiments of the present application can adopt a computer program product in the form of one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program codes.

[0177] The embodiments of the present application are described with reference to the flowchart illustrations and / or block diagrams of the methods, terminal devices (systems) and computer program products according to the embodiments of the present application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing terminal devices to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal devices, create means for implementing the functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams.

[0178] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal devices to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams.

[0179] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal devices, such that a series of operational steps are carried out on the computer or other programmable terminal devices to produce a computer implemented process so that the instructions executed on the computer or other programmable terminal devices provide steps for implementing the functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams.

[0180] Although preferred embodiments of the present application have been described, those skilled in the art will be able to make additional modifications and variations to the embodiments without departing from the scope of the present application. Accordingly, the appended claims are intended to encompass all such modifications and variations as falling within the scope of the embodiments of the present application.

[0181] Finally, it is to be understood that the phraseology or terminology such as "first" and "second" etc. used herein is merely intended to differentiate one entity or operation from another entity or operation, without necessarily requiring or implying any actual such relationship or order between such entities or operations. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises... a" does not, without more constraints, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0182] The above provides a kind of audio signal denoising method based on improved variational modal decomposition, an audio signal denoising system based on improved variational modal decomposition, an electronic device and a computer readable storage medium, are introduced in detail, the principle and implementation mode of the present application are described in this paper with specific examples, the above example is only for helping to understand the method of the present application and its core idea;For those skilled in the art, according to the idea of the present application, there will be changes in specific implementation mode and application range, as described above, the content of the specification should not be understood as the limitation of the present application.

Claims

1. An audio signal denoising method based on improved variational mode decomposition, characterized in that, The method comprises: acquiring a target input audio signal, and utilizing an improved particle swarm optimization algorithm with a local optimal solution jump-out mechanism and a target component number and a target penalty factor of fuzzy entropy search variational mode decomposition; decomposing the target input audio signal according to the target component number and the target penalty factor to obtain a plurality of first target components; calculating the fuzzy entropy of each first target component, and utilizing a wavelet threshold denoising algorithm to denoise the first target component with fuzzy entropy higher than a preset fuzzy entropy threshold to obtain a first component; obtaining a target output audio signal according to the first target component with fuzzy entropy not higher than the preset fuzzy entropy threshold and the first component; wherein the improved particle swarm optimization algorithm with the local optimal solution jump-out mechanism and the target component number and the target penalty factor of fuzzy entropy search variational mode decomposition comprise: initializing the position and speed of each individual particle and the position of a random walk particle; wherein the position comprises a component number and a penalty factor; calculating the fitness of the current position of each individual particle; updating the speed and position of each individual particle and recording the current global optimal solution fitness of the individual particle; if the fitness of the random walk particle is less than the current global optimal solution fitness, replacing the position of the individual particle corresponding to the current global optimal solution fitness with the position of the random walk particle; when the iteration termination condition is met, outputting the target component number and the target penalty factor; wherein the calculation of the fitness of the current position of each individual particle comprises: decomposing the target input audio signal according to the current position of each individual particle to obtain a plurality of second target components; calculating the average fuzzy entropy of the plurality of second target components and taking the average fuzzy entropy as the fitness of the individual particle; updating the speed and position of the individual particle and the position of the random walk particle, and recording the current global optimal solution fitness of the individual particle; if the fitness of the random walk particle is less than the current global optimal solution fitness, replacing the position of the individual particle corresponding to the current global optimal solution fitness with the position of the random walk particle; when the iteration termination condition is met, outputting the target component number and the target penalty factor.

2. The method of claim 1, wherein, The obtaining of the target output audio signal according to the first target component with fuzzy entropy not higher than the preset fuzzy entropy threshold and the first component comprises: reconstructing the first target component with fuzzy entropy not higher than the preset fuzzy entropy threshold and the first component to obtain the target output audio signal after denoising.

3. The method of claim 1, wherein, The decomposition of the target input audio signal according to the target component number and the target penalty factor to obtain a plurality of first target components comprises: calling an improved variational mode decomposition algorithm to decompose the target input audio signal according to the target component number and the target penalty factor to obtain a plurality of first target components.

4. The method of claim 1, wherein, The obtaining of the target output audio signal according to the first target component with fuzzy entropy not higher than the preset fuzzy entropy threshold and the first component comprises: determining a target audio encoding format; According to the first target component with the fuzzy entropy not higher than a preset fuzzy entropy threshold and the first component, the target output audio signal in the target audio coding format is generated.

5. An audio signal denoising system based on improved variational mode decomposition, characterized in that, The system comprises: an input interface module configured to receive an input audio signal; a preprocessing module configured to preprocess the input audio signal to obtain a target input audio signal; a core processing module configured to utilize an improved particle swarm optimization algorithm with a local optimal solution jump-out mechanism and a fuzzy entropy search variational mode decomposition to search for a target component number and a target penalty factor; decompose the target input audio signal according to the target component number and the target penalty factor to obtain a plurality of first target components; calculate the fuzzy entropy of each first target component, and utilize a wavelet threshold denoising algorithm to denoise the first target component with the fuzzy entropy higher than a preset fuzzy entropy threshold to obtain a first component; obtain an output audio signal according to the first target component with the fuzzy entropy not higher than the preset fuzzy entropy threshold and the first component; initialize the position and speed of each individual particle and the position of a random walk particle; wherein the position comprises a component number and a penalty factor; decompose the target input audio signal according to the current position of each individual particle to obtain a plurality of second target components; calculate the average fuzzy entropy of the plurality of second target components, and take the average fuzzy entropy as the fitness of the individual particle; update the speed and position of the individual particle, and update the position of the random walk particle, and record the current global optimal solution fitness of the individual particle; if the fitness of the random walk particle is less than the current global optimal solution fitness, replace the position of the individual particle corresponding to the current global optimal solution fitness with the position of the random walk particle; when the iteration termination condition is met, output the target component number and the target penalty factor; update the speed and position of each individual particle, and record the current global optimal solution fitness of the individual particle; if the fitness of the random walk particle is less than the current global optimal solution fitness, replace the position of the individual particle corresponding to the current global optimal solution fitness with the position of the random walk particle; when the iteration termination condition is met, output the target component number and the target penalty factor; an output interface module configured to generate a target output audio signal in a target audio coding format according to the output audio signal.

6. The system of claim 5, wherein, The core processing module comprises: a parameter optimization module configured to utilize an improved particle swarm optimization algorithm with a local optimal solution jump-out mechanism and a fuzzy entropy search variational mode decomposition to search for a target component number and a target penalty factor; a signal decomposition module configured to decompose the target input audio signal according to the target component number and the target penalty factor to obtain a plurality of first target components; a denoising module configured to calculate the fuzzy entropy of each first target component, and utilize a wavelet threshold denoising algorithm to denoise the first target component with the fuzzy entropy higher than a preset fuzzy entropy threshold to obtain a first component; a signal reconstruction module configured to obtain the output audio signal according to the first target component with the fuzzy entropy not higher than the preset fuzzy entropy threshold and the first component.

7. An electronic device, comprising: An apparatus comprising a processor, a memory, and a computer program stored on the memory and executable on the processor, the computer program, when executed by the processor, implementing the method for audio signal denoising based on improved variational modal decomposition according to any one of claims 1 to 4.

8. A computer-readable storage medium, characterized in that, A computer readable storage medium storing a computer program, the computer program, when executed by a processor, implementing the method for audio signal denoising based on improved variational modal decomposition according to any one of claims 1 to 4.

Images