POP sound suppression method and device, readable storage medium and computer program product

Abstract

The invention discloses a POP sound suppression method and device, a readable storage medium and a computer program product, and relates to the technical field of signal processing, and the method comprises the steps: obtaining a preset number of frames of original signals to be subjected to noise reduction, obtaining associated signals associated with the original signals, and determining the original signals and the associated signals as target signals; calculating a first energy variance of the original signal, and / or calculating a second energy variance of the target signal; if the first energy variance is smaller than a preset threshold value and / or the second energy variance is larger than the preset threshold value, a gain factor is determined based on the first energy variance and / or the second energy variance, the gain factor is in positive correlation with the first energy variance, and / or the gain factor is in negative correlation with the second energy variance; and performing noise reduction processing on the original signal based on the gain factor to obtain a noise reduction processing result. According to the invention, effective suppression of the POP sound in the signal is realized.

Description

Technical Field

[0001] The present application relates to the field of signal processing technology, and in particular to a POP sound suppression method, device, readable storage medium and computer program product. Background Art

[0002] POP sound is also called explosive sound. There are many reasons for POP sound, such as data packet loss, wind noise overflow, short knocking sound, external sound when recording stops, large external sound recorded at the end of the signal, and the sound of pressing the stop button when recording stops is recorded into the audio file as part of the audio, etc. POP sound will reduce the hearing and quality of the signal, and may even cause potential damage to the user's ears. Therefore, it is necessary to suppress POP sound in the signal.

[0003] However, traditional noise reduction algorithms usually use the assumptions of speech intermittency and noise stationarity to estimate the power spectrum of noise, and then estimate the signal-to-noise ratio of the noisy speech signal, and perform noise suppression based on this. However, POP sounds are highly non-stationary, and traditional noise reduction algorithms cannot effectively suppress POP sounds.

[0004] Therefore, how to effectively suppress the POP sound in the signal is a technical problem that needs to be solved urgently in the technical field. Summary of the invention

[0005] The main purpose of the present application is to provide a POP sound suppression method, device, readable storage medium and computer program product, aiming at solving the technical problem of how to effectively suppress the POP sound in the signal.

[0006] To achieve the above object, the present application provides a POP sound suppression method, the POP sound suppression method comprising:

[0007] Acquire a preset number of frames of original signals to be denoised, acquire associated signals associated with the original signals, and determine that the original signals and the associated signals are target signals, wherein the associated signals include at least one frame of signals adjacent to the original signals;

[0008] Calculating a first energy variance of the original signal, and / or calculating a second energy variance of the target signal;

[0009] If the first energy variance is less than a preset threshold and / or the second energy variance is greater than a preset threshold, determining a gain factor based on the first energy variance and / or the second energy variance, wherein the gain factor is positively correlated with the first energy variance and / or the gain factor is negatively correlated with the second energy variance;

[0010] The original signal is subjected to noise reduction processing based on the gain factor to obtain a noise reduction processing result.

[0011] In one implementation, the step of calculating the first energy variance of the original signal includes:

[0012] Determine a signal frequency point in the original signal within a preset frequency range as a first signal frequency point, and calculate a first signal amplitude of each of the first signal frequency points;

[0013] The variance of all the first signal amplitudes is calculated, and the variance of all the first signal amplitudes is determined to be a first energy variance.

[0014] In one embodiment, the step of calculating the second energy variance of the target signal includes:

[0015] Determine a signal frequency point in the target signal within a preset frequency range as a second signal frequency point;

[0016] For each preset frequency point group, calculating the second signal amplitudes of all the second signal frequency points in the preset frequency point group, calculating the variance of all the second signal amplitudes, and determining the variance of all the second signal amplitudes as a third energy variance, wherein each of the preset frequency point groups contains second signal frequency points with the same index in each of the second signal frequency points;

[0017] An average value of all the third energy variances is calculated, and the average value of all the third energy variances is determined as the second energy variance.

[0018] In one embodiment, the step of determining a gain factor based on the first energy variance and the second energy variance comprises:

[0019] For each signal frequency point in the original signal, obtaining a third energy variance corresponding to the signal frequency point based on the second energy variance;

[0020] Determining the third energy variance as a fourth energy variance, or determining the sum of the third energy variance and a preset bias factor as the fourth energy variance;

[0021] Calculating a variance ratio between the first energy variance and the fourth energy variance, and calculating a product between the variance ratio and a preset adjustment coefficient;

[0022] A sum of the product and a preset fixed gain factor is calculated, and a gain factor corresponding to the signal frequency point is determined based on the sum.

[0023] In one embodiment, the step of determining the gain factor corresponding to the signal frequency point based on the sum value includes:

[0024] If the sum is less than or equal to the preset lower limit gain value, determining the gain factor corresponding to the signal frequency point to be the preset lower limit gain value;

[0025] If the sum is greater than or equal to a preset upper limit gain value, determining the gain factor corresponding to the signal frequency point to be the preset upper limit gain value;

[0026] If the sum is less than the preset upper limit gain value and greater than the preset lower limit gain value, the sum is determined to be the gain factor corresponding to the signal frequency point.

[0027] In one embodiment, the step of performing noise reduction processing on the original signal based on the gain factor to obtain a noise reduction processing result includes:

[0028] For each signal frequency point in the original signal, obtaining a gain factor corresponding to the signal frequency point;

[0029] The gain factor corresponding to each of the signal frequency points is used to perform noise reduction processing on the original signal to obtain a noise reduction processing result.

[0030] In one embodiment, the step of calculating the first energy variance of the original signal and / or calculating the second energy variance of the target signal comprises:

[0031] Calculating a first energy variance of the original signal, and if the first energy variance is less than a preset threshold, calculating a second energy variance of the target signal; or,

[0032] A second energy variance of the target signal is calculated, and if the second energy variance is greater than a preset threshold, a first energy variance of the original signal is calculated.

[0033] In addition, to achieve the above-mentioned purpose, the present application also provides a POP sound suppression device, which includes: a memory, a processor, and a computer program stored on the memory and executable on the processor, and the computer program is configured to implement the steps of the POP sound suppression method as described above.

[0034] In addition, to achieve the above-mentioned purpose, the present application also provides a readable storage medium, which is a computer-readable storage medium, and the computer-readable storage medium stores a program for implementing the POP sound suppression method, and the program for implementing the POP sound suppression method is executed by a processor to implement the steps of the POP sound suppression method as described above.

[0035] The present application also provides a computer program product, including a computer program, which implements the steps of the POP sound suppression method as described above when executed by a processor.

[0036] One or more technical solutions proposed in this application have at least the following technical effects:

[0037] Acquire a preset number of frames of original signals to be denoised, acquire associated signals associated with the original signals, determine that the original signals and the associated signals are target signals, wherein the associated signals include at least one frame of signals adjacent to the original signals; calculate a first energy variance of the original signal, and / or calculate a second energy variance of the target signal; if the first energy variance is less than a preset threshold and / or the second energy variance is greater than a preset threshold, determine a gain factor based on the first energy variance and / or the second energy variance, wherein the gain factor is positively correlated with the first energy variance, and / or the gain factor is negatively correlated with the second energy variance; perform denoising on the original signal based on the gain factor to obtain a denoising result. Considering that POP sound usually has the characteristics of short duration and uniform energy distribution, while speech has obvious harmonic structure, so the energy distribution is not uniform, and at the same time it shows stability in the long time domain (that is, long-term stability), thus, in the embodiment of the present application, the first energy variance of the original signal is calculated, and / or the second energy variance of the target signal is calculated. It is easy to understand that the first energy variance is the energy variance of the original signal of a preset number of frames, so that the first energy variance can characterize the energy distribution difference of the original signal itself in the frequency band, and the second energy variance is the energy variance of the original signal and the associated signal adjacent to the original signal, so that the second energy variance can characterize the energy distribution difference between the original signal and the associated signal, that is, the energy distribution difference of the original signal in the time period, and the POP sound is short-term and has uniform energy distribution, so its first energy variance is small, while the second energy variance is large, and the second energy variance is large, while the speech energy distribution is uneven and has long-term stability, so its first energy variance is large, while the second energy variance is small. Based on this, when the first energy variance is less than a preset threshold and / or the second energy variance is greater than a threshold, it indicates that there may be POP sounds in the original signal. At this time, a gain factor is determined based on the first energy variance and / or the second energy variance, and the original signal is subjected to noise reduction processing based on the gain factor, thereby effectively suppressing the POP sounds in the signal. In addition, the suppression of POP sounds is realized at the software level, without adding additional hardware circuits, and only the energy variance needs to be calculated, which has a small amount of calculations and low computational complexity, thus reducing the implementation cost of POP sounds and facilitating embedded deployment. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0039] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0040] Figure 1 It is a simulation diagram of POP sound spectrum characteristics;

[0041] FIG2( a ) is a diagram showing an exemplary signal spectrum simulation of a signal having a POP sound due to wind noise overflow;

[0042] FIG2( b ) is a signal spectrum simulation diagram corresponding to FIG2( a );

[0043] FIG3( a ) is an exemplary signal spectrum simulation diagram showing a signal having a POP tone due to data packet loss;

[0044] FIG3( b ) is a signal spectrum simulation diagram corresponding to FIG3( a );

[0045] Figure 4 This is a flow chart of the first embodiment of the POP sound suppression method of the present application;

[0046] Figure 5 This is an experimental simulation diagram of a denoising result after a signal is denoised using a known signal denoising algorithm;

[0047] Figure 6 To adopt the POP sound suppression method of the present application Figure 5 Experimental simulation diagram of the noise reduction result after processing the signal in;

[0048] Figure 7 This is a schematic diagram of the equipment structure of the hardware operating environment involved in the POP sound suppression device in the embodiment of the present application.

[0049] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0050] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.

[0051] Audio noise reduction is one of the most classic and important research directions in the field of signal processing. It requires analyzing, transforming and processing noisy speech signals to estimate and suppress noise components. There are many types of noise, such as wind noise, pink noise, whistle noise and other types of noise and noise. They have obvious differences in spectrum distribution, some of which are stable and some are non-stationary. The presentation of POP sound on the spectrum is as follows: Figure 1 As shown, Figure 1 The figure shows an exemplary POP sound spectrogram, where the horizontal axis represents time and the vertical axis represents frequency.

[0052] POP sound is a common type of noise in call noise reduction. It may be caused by wind noise overflow, as shown in Figure 2(a) and Figure 2(b), or by packet loss, as shown in Figure 3(a) and Figure 3(b), or by short tapping. Figure 2(a) and Figure 3(a) are both exemplary signal spectrograms with POP sound, where the horizontal axis represents frequency and the vertical axis represents signal amplitude. Figure 2(b) is the spectrogram of the signal shown in Figure 2(a), and Figure 3(b) is the spectrogram of the signal shown in Figure 3(a). The horizontal axis of Figure 2(b) and Figure 3(b) represents time, and the vertical axis represents signal frequency. Due to the characteristics of POP sound itself, many traditional noise reduction algorithms are basically ineffective and it is difficult to suppress POP sound well.

[0053] Based on this, the main solution of the present application is: obtain a preset number of frames of original signals to be denoised, obtain associated signals associated with the original signals, determine that the original signals and the associated signals are target signals, wherein the associated signals include at least one frame of signals adjacent to the original signals; calculate the first energy variance of the original signal, and / or calculate the second energy variance of the target signal; if the first energy variance is less than a preset threshold and / or the second energy variance is greater than a preset threshold, determine a gain factor based on the first energy variance and / or the second energy variance, wherein the gain factor is positively correlated with the first energy variance, and / or the gain factor is negatively correlated with the second energy variance; perform denoising on the original signal based on the gain factor to obtain a denoising result.

[0054] The present application calculates the first energy variance of the original signal, and / or calculates the second energy variance of the target signal, and it is easy to understand that the first energy variance is the energy variance of the original signal of a preset number of frames, so that the first energy variance can characterize the energy distribution difference of the original signal itself in the frequency band, and the second energy variance is the energy variance of the original signal and the associated signal adjacent to the original signal, so that the second energy variance can characterize the energy distribution difference between the original signal and the associated signal, that is, the energy distribution difference of the original signal in the time period, and the POP sound is short and the energy distribution is uniform, so its first energy variance is small, while the second energy variance is large, and the speech energy distribution is uneven and has long-term stability, so its first energy variance is large, while the second energy variance is small. Based on this, when the first energy variance is less than the preset threshold and / or the second energy variance is greater than the threshold, it means that there may be a POP sound in the original signal. At this time, the gain factor is determined based on the first energy variance and / or the second energy variance, and the original signal is subjected to noise reduction processing based on the gain factor, so that the POP sound in the signal can be effectively suppressed. In addition, POP sound suppression is achieved at the software level without adding additional hardware circuits. Only energy variance needs to be calculated, which reduces the amount of calculation and the complexity of calculation, thus reducing the implementation cost of POP sound and facilitating embedded deployment.

[0055] It should be noted that the execution subject of each embodiment of the POP sound suppression method of the present application can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or a POP sound suppression device capable of realizing the above functions, such as headphones, AR (Augmented Reality) glasses, VR (Virtual Reality) glasses, etc. The embodiments of the POP sound suppression method of the present application do not make specific restrictions on this.

[0056] Based on this, the present application proposes a POP sound suppression method of the first embodiment, please refer to Figure 4 The POP sound suppression method comprises steps S10 to S40:

[0057] Step S10, obtaining a preset number of frames of original signals to be denoised, obtaining associated signals associated with the original signals, and determining that the original signals and the associated signals are target signals, wherein the associated signals include at least one frame of signals adjacent to the original signals;

[0058] The original signal may be a signal initially picked up without any processing, or may be a preprocessed signal, and this embodiment does not impose any specific restrictions on this. The preprocessing includes but is not limited to noise reduction processing based on a known noise reduction algorithm, and the known noise reduction algorithm may specifically be spectral subtraction, OMLSA algorithm (optimally-modified log-spectral amplitude estimation algorithm), MMSE algorithm (Minimum Mean Squared Error), etc., and this embodiment does not impose any specific restrictions on this.

[0059] In a preferred embodiment, the original signal is a signal that has been subjected to noise reduction processing based on a known noise reduction algorithm to reduce the influence of other noises in the signal on the POP sound suppression effect.

[0060] It should be noted that in this embodiment, POP sound suppression processing is performed on the original signal in the frequency domain. If the original signal is a time domain signal, the original signal can be converted to the frequency domain before subsequent processing. Specifically, the time domain signal can be converted to the frequency domain through Fourier transform.

[0061] A preset number of frame original signals are obtained, so that the POP sound suppression processing is performed with the preset number of frame signals as a period. The preset number can be a number set according to actual conditions. Since the duration of the POP sound is usually relatively stable, the preset number can be set based on the duration of the POP sound, such as the duration of the preset number of frame signals is less than or equal to the duration of the POP sound. Preferably, the preset number is one to improve the suppression accuracy of the POP sound. Exemplarily, the various embodiments of the present application are described and illustrated with the preset number being one.

[0062] The associated signal is at least one frame signal adjacent to the original signal. Specifically, the associated signal may be at least one frame signal before the original signal and / or at least one frame signal after the original signal. For example, if the original signal currently acquired is the 10th frame signal, the 5th, 6th, 7th, 8th, and 9th frame signals may be selected as the associated signal.

[0063] Furthermore, the specific number of frames of signals selected as associated signals can be set in advance by relevant personnel. It can usually be set based on the duration of the POP sound, such as the duration of the target signal is greater than the duration of the POP sound. For example, the duration of the POP sound is usually 50 milliseconds, and the duration of a frame signal is 5 milliseconds, then at least 10 frames of signals can be selected as associated signals, so that the target signal has at least 11 frames, that is, 55 milliseconds, which is greater than 50 milliseconds, so that when the original signal is a POP sound, the probability of speech in the target signal is increased, so that the second energy variance when the original signal is speech (at this time, since the associated signal is generally speech, the target signal is generally speech, the energy is relatively stable, and the second energy variance is small) and the second energy variance when the original signal is a POP sound (at this time, there are both speech and POP sounds in the target signal, the energy distribution difference is large, and the second energy variance is large) There is a large difference, thereby improving the second energy variance for the identification effect of POP sound.

[0064] Furthermore, a buffer area can be set up, and the signal after the POP sound suppression is completed can be stored in the buffer area, that is, the noise reduction processing result is cached in the buffer area, so that the associated signal can be obtained from the buffer area. The buffer area can specifically be a stack structure, and at least one frame of signal is obtained from the buffer area to obtain the associated signal. In this way, it can be ensured that the associated signal obtained is the signal after the POP sound is suppressed, and the associated signal is basically voice without POP sound, thereby ensuring that when the original signal is POP sound, the energy distribution difference between it and the associated signal is large (the second energy variance is large), and when the original signal is voice, the energy distribution difference between it and the associated signal is small (the second energy variance is small), which further improves the second energy variance for POP sound identification effect.

[0065] Step S20, calculating a first energy variance of the original signal, and / or calculating a second energy variance of the target signal;

[0066] The energy of the frequency domain signal is the integral (for continuous frequency domain signals) or sum (for discrete frequency domain signals) of the amplitude of the signal in the frequency domain, and the energy variance of the frequency domain signal is the expected value of the square of the difference between the energy of the frequency domain signal and its average energy. Based on this, the energy variance of the original signal, i.e., the first energy variance, and the energy variance of the target signal, i.e., the second energy variance, can be calculated.

[0067] Step S30, if the first energy variance is less than a preset threshold and / or the second energy variance is greater than a preset threshold, determining a gain factor based on the first energy variance and / or the second energy variance, wherein the gain factor is positively correlated with the first energy variance and / or the gain factor is negatively correlated with the second energy variance;

[0068] If only the first energy variance is calculated, when the first energy variance is less than a preset threshold, it is determined that there is a POP sound in the original signal. At this time, a gain factor is determined based on the first energy variance, and the gain factor is positively correlated with the first energy variance.

[0069] If only the second energy variance is calculated, when the second energy variance is greater than a preset threshold, it is determined that there is a POP sound in the original signal. At this time, a gain factor is determined based on the second energy variance, and the gain factor is negatively correlated with the second energy variance.

[0070] If the first energy variance and the second energy variance are calculated, when the first energy variance is less than a preset threshold and the second energy variance is greater than the preset threshold, it is determined that there is a POP sound in the original signal. At this time, the gain factor is determined based on the first energy variance and the second energy variance, and the gain factor is positively correlated with the first energy variance and negatively correlated with the second energy variance.

[0071] It should be noted that preset thresholds corresponding to the first energy variance and the second energy variance can be set respectively, that is, the first energy variance and the second energy variance are compared with the corresponding preset thresholds respectively to obtain the size relationship between them and the preset thresholds. The preset threshold corresponding to the first energy variance is recorded as the first preset threshold, and the preset threshold corresponding to the second energy variance is recorded as the second preset threshold. If the first energy variance is less than the preset threshold and / or the second energy variance is greater than the preset threshold, it can be expressed as if the first energy variance is less than the first preset threshold and / or the second energy variance is greater than the second preset threshold. The first preset threshold and the second preset threshold may be the same or different, and this embodiment does not impose specific restrictions on this.

[0072] It can be understood that the smaller the first energy variance is, the greater the probability that there is a POP sound in the original signal. At this time, the smaller the gain factor is used to perform noise reduction processing on the original signal, the suppression effect of the original signal can be improved, so that if there is a POP sound in the original signal, the suppression effect of the POP sound can be improved. Similarly, the larger the second energy variance is, the greater the probability that there is a POP sound in the original signal. At this time, the smaller the gain factor is used to perform noise reduction processing on the original signal, the suppression effect of the original signal can be improved, so that if there is a POP sound in the original signal, the suppression effect of the POP sound can be improved.

[0073] In a preferred embodiment, the first energy variance of the original signal and the second energy variance of the target signal are calculated. If the first energy variance is less than a preset threshold and the second energy variance is greater than the preset threshold, a gain factor is determined based on the first energy variance and the second energy variance to determine whether the original signal has POP sound in combination with the first energy variance and the second energy variance. When both the first energy variance and the second energy variance meet the conditions, that is, when the original signal exhibits the characteristics of POP sound in both time period and frequency band, it is determined that the original signal has POP sound, thereby achieving high-accuracy identification of POP sound.

[0074] Further, if the gain factor is calculated based on the first energy variance and the second energy variance, as a first implementation, the first energy variance may be divided by the second energy variance to obtain a ratio, and if the ratio is greater than or equal to a preset upper limit gain value, the preset upper limit gain value is determined to be the gain factor, if the ratio is less than or equal to a preset lower limit gain value, the preset lower limit gain value is determined to be the gain factor, and if the ratio is less than the preset upper limit gain value and greater than the preset lower limit gain value, the ratio is determined to be the gain factor.

[0075] As a second implementation, the sum of the second energy variance and the preset bias factor can be calculated, and the first energy variance is divided by the sum to obtain a ratio. If the ratio is greater than or equal to the preset upper limit gain value, the preset upper limit gain value is determined to be the gain factor. If the ratio is less than or equal to the preset lower limit gain value, the preset lower limit gain value is determined to be the gain factor. If the ratio is less than the preset upper limit gain value and greater than the preset lower limit gain value, the ratio is determined to be the gain factor.

[0076] As a third implementation, the sum of the second energy variance and the preset bias factor can be calculated, the first energy variance is divided by the sum to obtain a ratio, the product between the ratio and the preset adjustment coefficient is calculated, and the sum between the product and the preset fixed factor is calculated. If the sum is greater than or equal to the preset upper limit gain value, the preset upper limit gain value is determined to be the gain factor; if the sum is less than or equal to the preset lower limit gain value, the preset lower limit gain value is determined to be the gain factor; if the sum is less than the preset upper limit gain value and greater than the preset lower limit gain value, the sum is determined to be the gain factor.

[0077] It should be noted that the above are only three feasible implementation methods for determining the gain factor provided in this embodiment. Relevant personnel can set the specific method for determining the gain factor according to actual conditions. This embodiment does not specifically limit the specific implementation method for determining the gain factor.

[0078] Step S40: performing noise reduction processing on the original signal based on the gain factor to obtain a noise reduction processing result.

[0079] The original signal is subjected to noise reduction processing based on the gain factor. Specifically, the original signal can be subjected to noise reduction processing based on a known noise reduction processing method based on the gain factor, such as adjusting the signal amplitude of the original signal based on the gain factor, such as updating the signal amplitude of the original signal by multiplying the signal amplitude of the original signal by the gain factor, thereby obtaining a noise reduction processing result. This embodiment does not impose any specific limitation on this.

[0080] Further, after obtaining the noise reduction processing result, if a buffer area is set, the noise reduction processing result can be stored in the buffer area. It should be noted that when the first energy variance is greater than or equal to the threshold or the second energy variance is less than or equal to the preset threshold, that is, when it is determined that there is no POP sound in the original signal, the original signal of this frame can also be directly stored in the buffer area, so that the continuous frame associated signal before the original signal can be obtained later.

[0081] This embodiment obtains a preset number of frames of original signals to be denoised, obtains associated signals associated with the original signals, determines that the original signals and the associated signals are target signals, wherein the associated signals include at least one frame of signals adjacent to the original signals; calculates a first energy variance of the original signal, and / or calculates a second energy variance of the target signal; if the first energy variance is less than a preset threshold and / or the second energy variance is greater than a preset threshold, determines a gain factor based on the first energy variance and / or the second energy variance, wherein the gain factor is positively correlated with the first energy variance, and / or the gain factor is negatively correlated with the second energy variance; performs denoising on the original signal based on the gain factor to obtain a denoising result. Considering that POP sound usually has the characteristics of short duration and uniform energy distribution, while speech has obvious harmonic structure, so the energy distribution is not uniform, and it shows stability in the long time domain (that is, long-term stability). Therefore, in this embodiment, the first energy variance of the original signal is calculated, and / or the second energy variance of the target signal is calculated. It is easy to understand that the first energy variance is the energy variance of the original signal of a preset number of frames, so that the first energy variance can characterize the energy distribution difference of the original signal itself in the frequency band, and the second energy variance is the energy variance of the original signal and the associated signal adjacent to the original signal. Therefore, the second energy variance can characterize the energy distribution difference between the original signal and the associated signal, that is, the energy distribution difference of the original signal in the time period. The POP sound is short-term and has uniform energy distribution, so its first energy variance is small, while the second energy variance is large. The speech energy distribution is not uniform and has long-term stability, so its first energy variance is large, while the second energy variance is small. Based on this, when the first energy variance is less than a preset threshold and / or the second energy variance is greater than a threshold, it indicates that there may be POP sounds in the original signal. At this time, a gain factor is determined based on the first energy variance and / or the second energy variance, and the original signal is subjected to noise reduction processing based on the gain factor, thereby effectively suppressing the POP sounds in the signal. In addition, the suppression of POP sounds is realized at the software level, without adding additional hardware circuits, and only the energy variance needs to be calculated, which has a small amount of calculations and low computational complexity, thus reducing the implementation cost of POP sounds and facilitating embedded deployment.

[0082] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above-mentioned embodiment 1 can be referred to the above introduction, and will not be repeated in the following. On this basis, the step of calculating the first energy variance of the original signal includes:

[0083] Step A10, determining the signal frequency points in the original signal within a preset frequency range as first signal frequency points, and calculating the first signal amplitude of each of the first signal frequency points;

[0084] The preset frequency range may be a frequency range set in advance, and the preset frequency range may include all frequencies of the original signal, or may only include local frequencies of the original signal, so as to reduce the amount of calculation. Preferably, the preset frequency range may be a frequency range where the energy distribution of speech and POP sound in the frequency band is greatly different. For example, if relevant personnel have found through experiments that the energy distribution of speech and POP sound is greatly different in 4K Hz to 8K Hz, the preset frequency range may be set to 4K Hz to 8K Hz. In this way, the amount of calculation can be reduced, and the accuracy of the first energy variance in identifying POP sound can be improved.

[0085] In order to facilitate the distinction of the preset frequency range for the subsequent calculation of the second energy variance, the preset frequency range when calculating the first energy variance is referred to as the first energy frequency range.

[0086] The first signal amplitude is the signal amplitude of the first signal frequency point. It can be understood that after the N sampling points are Fourier transformed, the Fourier transform results of the N signal points can be obtained. Each sampling point corresponds to a signal frequency point after the Fourier transform. The value after the Fourier transform corresponding to each signal frequency point (recorded as FT value) is the modulus of the FT value corresponding to the signal frequency point. For example, assuming that the FT value of a certain signal frequency point is x=a+bi, where a is the real part of the signal, b is the imaginary part of the signal, and i is the imaginary unit, the signal amplitude of the signal frequency point is

[0087] Step A20: Calculate the variance of all the first signal amplitudes, and determine the variance of all the first signal amplitudes as a first energy variance.

[0088] Each signal frequency point in the frequency domain signal represents a certain frequency and has a unique index. The frequency represented by the signal frequency point refers to the actual frequency value corresponding to the position of the signal frequency point in the spectrum. For example, a signal with a sampling frequency of Fs=1000Hz and a 1024-point FFT (Fast Fourier Transform) is performed on it, then the spectrum resolution is Δf=1000Hz / 1024≈0.9765625Hz. If the signal frequency point indexed as k in the FFT output, then the frequency represented by the signal frequency point is f=k*Δf. Based on this, when the spectrum resolution is known, the corresponding index can be calculated based on the frequency, and the frequency represented by the signal frequency point can also be calculated based on the index.

[0089] The original signal is denoted as Y, the frame number of the original signal is 1, and the first energy variance is δ F (l), then the first energy variance can be expressed as in, k represents the index of the signal frequency point, k1 is equal to kF1 and represents the index corresponding to the upper limit value of the first preset frequency range, k2 is equal to kF2 and represents the index corresponding to the lower limit value of the first preset frequency range, K represents the total number of signal frequency points within the index range of [kF1, kF2], (Y(k,l) represents the first signal amplitude of the first signal frequency point with index k in the original signal.

[0090] For example, if the first preset frequency range is [f1, f2], then

[0091] In a possible implementation, the step of calculating the second energy variance of the target signal includes:

[0092] Step B10, determining a signal frequency point in the target signal within a preset frequency range as a second signal frequency point;

[0093] The preset frequency range may be a frequency range set in advance, and the preset frequency range may include all frequencies of the original signal, or may only include local frequencies of the original signal, so as to reduce the amount of calculation. Preferably, the preset frequency range may be a frequency range where the energy distribution of speech and POP sound in a time period is relatively different. For example, if relevant personnel have found through experiments that the energy distribution of speech and POP sound is relatively different in the range of 6K Hz to 8K Hz, the preset frequency range may be set to 6K Hz to 8K Hz. In this way, the amount of calculation can be reduced, and the accuracy of the second energy variance in identifying POP sound can be improved.

[0094] The preset frequency range when calculating the second energy variance is referred to as the second preset frequency range. Relevant personnel may respectively set the first preset frequency range and the second preset frequency range in advance. The two may be the same or different. This embodiment does not impose any specific limitation on this.

[0095] Step B20: for each preset frequency point group, calculating the second signal amplitudes of all the second signal frequency points in the preset frequency point group, calculating the variance of all the second signal amplitudes, and determining the variance of all the second signal amplitudes as a third energy variance, wherein each of the preset frequency point groups contains second signal frequency points with the same index in each of the second signal frequency points;

[0096] The indexes of the second signal frequency points in the same preset frequency point group are the same, and the indexes of the second signal frequency points in different preset frequency point groups are different.

[0097] The index of the second signal frequency point in the preset frequency point group is used as the index of the preset frequency point group, and the third energy variance of the preset frequency point group with index k is δ T (k,l), then the third energy variance can be expressed as in, N represents the total number of frames of the target signal, and |buff(k,n) represents the second signal amplitude of the second signal frequency point with index k in the nth frame of the target signal.

[0098] Step B30, calculating an average value of all the third energy variances, and determining the average value of all the third energy variances as the second energy variance.

[0099] The second energy variance is δ Tavg (l), then the second energy variance can be expressed as Among them, kT1 represents the index corresponding to the lower limit value of the second preset frequency range, kT2 represents the index corresponding to the upper limit value of the second preset frequency range, kT1 and kF1 may be the same or different, and kT2 and kF2 may be the same or different.

[0100] For example, if the second preset frequency range is [f3, f4], then

[0101] In a possible implementation, the step of calculating the first energy variance of the original signal and / or calculating the second energy variance of the target signal includes:

[0102] Step C10, calculating a first energy variance of the original signal, and if the first energy variance is less than a preset threshold, calculating a second energy variance of the target signal; or,

[0103] Step C20, calculating the second energy variance of the target signal, and if the second energy variance is greater than a preset threshold, calculating the first energy variance of the original signal.

[0104] It should be noted that when calculating the first energy variance of the original signal and the second energy variance of the target signal, the first energy variance of the original signal can be calculated first, and when the first energy variance is less than the first preset threshold, the second energy variance of the target signal can be calculated; or the second energy variance of the target signal can be calculated first, and when the second energy variance is greater than the second preset threshold, the first energy variance of the original signal can be calculated; or, the first energy variance of the original signal and the second energy variance of the target signal can be calculated.

[0105] In this embodiment, an energy variance is calculated first. When this energy variance meets a given condition, that is, when it meets the condition for the existence of POP sound, another energy variance is calculated. When the first calculated energy variance does not meet the condition for the existence of POP sound, it can be considered that there is no POP sound in the original signal, and there is no need to further calculate another energy variance. The current POP sound suppression process is directly ended, and the process returns to the initial step to enter the next POP sound suppression process. In this way, unnecessary calculation of energy variance can be reduced, thereby reducing the amount of calculation.

[0106] Based on the first embodiment and / or the second embodiment of the present application, in the third embodiment of the present application, the same or similar contents as those in the first and second embodiments can be referred to the above description and will not be described in detail later. On this basis, the step of determining the gain factor based on the first energy variance and the second energy variance includes:

[0107] Step D10: for each signal frequency point in the original signal, obtaining a third energy variance corresponding to the signal frequency point based on the second energy variance;

[0108] Obtain the third energy variance corresponding to the signal frequency point. Specifically, the third energy variance can be obtained based on the index of the signal frequency point. The index of the signal frequency point is recorded as the target index. The third energy variance is the energy variance of all second signal frequency points in the target signal whose index is the target index.

[0109] It can be understood that the second energy variance is the average value of the third energy variances of multiple preset frequency groups. Therefore, each third energy variance of the second energy variance and the index corresponding to the third energy variance (actually the index of the signal frequency in the preset frequency band grouping corresponding to the third energy variance) can be recorded and calculated, so that the third energy variance corresponding to each frequency in the original signal can be obtained based on the index. Specifically, the index of each signal frequency in the original signal can be obtained, and the third energy variance corresponding to the index is obtained, that is, the third energy variance corresponding to the signal frequency.

[0110] It should be noted that if the third energy variance acquisition fails, the gain factor corresponding to the signal frequency point may be determined to be a preset fixed gain factor. The preset fixed gain factor may be a gain factor set by relevant personnel according to actual needs, and is not specifically limited here.

[0111] Step D20, determining the third energy variance as a fourth energy variance, or determining the sum of the third energy variance and a preset bias factor as the fourth energy variance;

[0112] The preset gain factor may be any positive value greater than 0. Further, in order to reduce the influence of the preset bias factor on the result, a smaller positive value is set as the preset bias factor, such as 0.01, 0.02, etc.

[0113] In a preferred embodiment, the sum of the third energy variance and a preset bias factor is determined as the fourth energy variance to avoid division by zero error.

[0114] Step D30, calculating a variance ratio between the first energy variance and the fourth energy variance, and calculating a product between the variance ratio and a preset adjustment coefficient;

[0115] A variance ratio between the first energy variance and the fourth energy variance is calculated. Specifically, the variance ratio may be the first energy variance divided by the fourth energy variance.

[0116] The preset adjustment coefficient may be a coefficient set in advance by relevant personnel, and this embodiment does not impose any specific restrictions on the value.

[0117] Step D40, calculating the sum of the product and a preset fixed gain factor, and determining the gain factor corresponding to the signal frequency based on the sum.

[0118] A gain factor corresponding to the signal frequency point is determined based on the sum value. Specifically, the gain factor may be positively correlated with the sum value.

[0119] In this embodiment, for each signal frequency point in the original signal, the third energy variance corresponding to the signal frequency point is obtained, and the gain factor corresponding to the signal frequency point is calculated based on the third energy variance. In this way, the gain factor corresponding to each signal frequency point is determined with the signal frequency point as the atom, so that the original signal can be subsequently denoised with the signal frequency point as the atom, thereby suppressing the POP sound in the original signal in a finer granularity and improving the suppression accuracy of the POP sound.

[0120] In a possible implementation manner, the step of determining the gain factor corresponding to the signal frequency point based on the sum value includes:

[0121] Step E10: if the sum is less than or equal to a preset lower limit gain value, determining the gain factor corresponding to the signal frequency point to be the preset lower limit gain value;

[0122] It should be noted that the preset lower limit gain value may be a minimum gain value set in advance. In a preferred embodiment, the preset lower limit gain value is 0.

[0123] Step E20: if the sum is greater than or equal to a preset upper limit gain value, determining the gain factor corresponding to the signal frequency point to be the preset upper limit gain value;

[0124] It should be noted that the preset upper limit gain value may be a maximum gain value set in advance. In a preferred embodiment, the preset upper limit gain value is 1.

[0125] Step E30: If the sum value is less than the preset upper limit gain value and greater than the preset lower limit gain value, determine the sum value as the gain factor corresponding to the signal frequency point.

[0126] Exemplarily, when the preset lower limit gain value is 0 and the preset upper limit gain value is 1, assuming that the index corresponding to the signal frequency point is k, and the gain factor corresponding to the signal frequency point is g(k,l), then g(k,l) can be expressed by the formula: Among them, c represents the preset adjustment coefficient, g0 represents the preset fixed gain factor, and ε represents the preset bias factor.

[0127] In a possible implementation manner, the step of performing noise reduction processing on the original signal based on the gain factor to obtain a noise reduction processing result includes:

[0128] Step F10, for each signal frequency point in the original signal, obtaining a gain factor corresponding to the signal frequency point;

[0129] Step F20, performing noise reduction processing on the original signal using the gain factor corresponding to each of the signal frequency points to obtain a noise reduction processing result.

[0130] The original signal is subjected to noise reduction processing using the gain factor corresponding to each signal frequency point, and the original signal can be adjusted based on the known gain factor to adjust the signal. In a specific embodiment, the noise reduction process can be expressed by the formula X(k,l)=Y(k,l)*g(k,l), where X(k,l) represents the noise reduction processing result of the signal frequency point indexed as k, and Y(k,l) is the FT value corresponding to the signal frequency point indexed as k in the original signal. As the FT value can usually be expressed in complex form, Y(k,l)*g(k,l) can be specifically the real part and the imaginary part of the FT value multiplied by the gain factor.

[0131] In this embodiment, the original signal is denoised using the gain factor corresponding to each signal frequency point in the original signal, that is, the signal frequency point is used as an atom for denoising, thereby suppressing the POP sound in the original signal in a finer granularity and improving the suppression accuracy of the POP sound.

[0132] For example, refer to Figures 5 and 6 As shown, Figure 5 is an exemplary signal after noise reduction processing based on a known noise reduction processing algorithm, Figure 6 The POP suppression method using this embodiment combined with the first embodiment and the second embodiment is Figure 5 The signal in the signal is processed by POP sound suppression. Figure 5 and Figure 6 It can be seen from the comparison that the POP sound suppression method proposed in the embodiment of the present application can further suppress the POP sound in the signal and obtain a cleaner high-quality signal output.

[0133] In addition, an embodiment of the present application also proposes a POP sound suppression device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the POP sound suppression method as described above.

[0134] In addition, reference Figure 7 , which shows a schematic diagram of the structure of a POP sound suppression device suitable for implementing the embodiment of the present application. The POP sound suppression device in the embodiment of the present application may also include but is not limited to mobile terminals such as headphones, AR glasses, VR glasses, mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 7 The POP sound suppression device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0135] like Figure 7 As shown, the POP sound suppression device may include a processing device 1001 (e.g., a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. Various programs and data required for the operation of the POP sound suppression device are also stored in the RAM 1004. The processing device 1001, the ROM 1002, and the RAM 1004 are connected to each other via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems may be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 1003 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 1009. The communication device 1009 may allow the POP sound suppression device to communicate wirelessly or wired with other devices to exchange data. Although the POP sound suppression device with various systems is shown in the figure, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems may be implemented or have alternatively.

[0136] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0137] The POP sound suppression device provided in the embodiment of the present application adopts the POP sound suppression method in the above embodiment, which can solve the technical problem of how to effectively suppress the POP sound in the signal. Compared with the prior art, the beneficial effects of the POP sound suppression device provided in the present application are the same as the beneficial effects of the POP sound suppression method provided in the above embodiment, and other technical features in the POP sound suppression device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0138] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0139] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0140] In addition, to achieve the above-mentioned purpose, an embodiment of the present application also provides a readable storage medium having computer-readable program instructions (ie, computer program) stored thereon, and the computer-readable program instructions are used to execute the POP sound suppression method in the above-mentioned embodiment.

[0141] The computer-readable storage medium provided in the embodiment of the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM: Random Access Memory), a read-only memory (ROM: Read Only Memory), an erasable programmable read-only memory (EPROM: Erasable Programmable Read Only Memory or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM: CD-Read Only Memory), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency: Radio Frequency), etc., or any suitable combination of the above.

[0142] The computer-readable storage medium may be included in the POP sound suppression device; or may exist independently without being assembled into the POP sound suppression device.

[0143] The above-mentioned computer-readable storage medium carries one or more programs. When the above-mentioned one or more programs are executed by the POP sound suppression device, the POP sound suppression device: obtains a preset number of frames of original signals to be denoised, obtains associated signals associated with the original signals, and determines that the original signals and the associated signals are target signals, wherein the associated signals include at least one frame signal adjacent to the original signal; calculates the first energy variance of the original signal, and / or calculates the second energy variance of the target signal; if the first energy variance is less than a preset threshold and / or the second energy variance is greater than a preset threshold, determines a gain factor based on the first energy variance and / or the second energy variance, wherein the gain factor is positively correlated with the first energy variance, and / or the gain factor is negatively correlated with the second energy variance; performs noise reduction processing on the original signal based on the gain factor to obtain a noise reduction processing result.

[0144] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0145] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0146] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0147] The readable storage medium provided in the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned POP sound suppression method, and can solve the technical problem of how to effectively suppress the POP sound in the signal. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in the present application are the same as the beneficial effects of the POP sound suppression method provided in the above-mentioned embodiment, and will not be repeated here.

[0148] In addition, an embodiment of the present application also proposes a computer program product, including a POP sound suppression program, which implements the steps of the POP sound suppression method described above when executed by a processor.

[0149] The specific implementation of the computer program product of the present application is basically the same as the above-mentioned embodiments of the POP sound suppression method, and will not be repeated here.

[0150] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0151] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0152] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software sensor, which is stored in a storage medium (such as ROM / RAM, disk, CD) as described above, including a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0153] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A POP sound suppression method, characterized in that: The POP sound suppression method comprises the following steps: Acquire a preset number of frames of original signals to be denoised, acquire associated signals associated with the original signals, and determine that the original signals and the associated signals are target signals, wherein the associated signals include at least one frame of signals adjacent to the original signals; Calculating a first energy variance of the original signal, and / or calculating a second energy variance of the target signal; If the first energy variance is less than a preset threshold and / or the second energy variance is greater than a preset threshold, determining a gain factor based on the first energy variance and / or the second energy variance, wherein the gain factor is positively correlated with the first energy variance and / or the gain factor is negatively correlated with the second energy variance; The original signal is subjected to noise reduction processing based on the gain factor to obtain a noise reduction processing result.

2. The POP sound suppression method according to claim 1, characterized in that: The step of calculating the first energy variance of the original signal comprises: Determine a signal frequency point in the original signal within a preset frequency range as a first signal frequency point, and calculate a first signal amplitude of each of the first signal frequency points; The variance of all the first signal amplitudes is calculated, and the variance of all the first signal amplitudes is determined to be a first energy variance.

3. The POP sound suppression method according to claim 1, characterized in that: The step of calculating the second energy variance of the target signal comprises: Determine a signal frequency point in the target signal within a preset frequency range as a second signal frequency point; For each preset frequency point group, calculating the second signal amplitudes of all the second signal frequency points in the preset frequency point group, calculating the variance of all the second signal amplitudes, and determining the variance of all the second signal amplitudes as a third energy variance, wherein each of the preset frequency point groups contains second signal frequency points with the same index in each of the second signal frequency points; An average value of all the third energy variances is calculated, and the average value of all the third energy variances is determined as the second energy variance.

4. The POP sound suppression method according to claim 1, characterized in that: The step of determining a gain factor based on the first energy variance and the second energy variance comprises: For each signal frequency point in the original signal, obtaining a third energy variance corresponding to the signal frequency point based on the second energy variance; Determining the third energy variance as a fourth energy variance, or determining the sum of the third energy variance and a preset bias factor as the fourth energy variance; Calculating a variance ratio between the first energy variance and the fourth energy variance, and calculating a product between the variance ratio and a preset adjustment coefficient; A sum of the product and a preset fixed gain factor is calculated, and a gain factor corresponding to the signal frequency point is determined based on the sum.

5. The POP sound suppression method according to claim 4, characterized in that: The step of determining the gain factor corresponding to the signal frequency point based on the sum value comprises: If the sum is less than or equal to the preset lower limit gain value, determining the gain factor corresponding to the signal frequency point to be the preset lower limit gain value; If the sum is greater than or equal to a preset upper limit gain value, determining the gain factor corresponding to the signal frequency point to be the preset upper limit gain value; If the sum is less than the preset upper limit gain value and greater than the preset lower limit gain value, the sum is determined to be the gain factor corresponding to the signal frequency point.

6. The POP sound suppression method according to any one of claims 1 to 5, characterized in that: The step of performing noise reduction processing on the original signal based on the gain factor to obtain a noise reduction processing result comprises: For each signal frequency point in the original signal, obtaining a gain factor corresponding to the signal frequency point; The gain factor corresponding to each of the signal frequency points is used to perform noise reduction processing on the original signal to obtain a noise reduction processing result.

7. The POP sound suppression method according to any one of claims 1 to 5, characterized in that: The step of calculating the first energy variance of the original signal and / or calculating the second energy variance of the target signal comprises: Calculating a first energy variance of the original signal, and if the first energy variance is less than a preset threshold, calculating a second energy variance of the target signal; or, A second energy variance of the target signal is calculated, and if the second energy variance is greater than a preset threshold, a first energy variance of the original signal is calculated.

8. A POP sound suppression device, characterized in that: The POP sound suppression device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the POP sound suppression method according to any one of claims 1 to 7.

9. A readable storage medium, characterized in that: The readable storage medium is a computer-readable storage medium, and a computer program is stored on the readable storage medium. When the computer program is executed by a processor, the steps of the POP sound suppression method according to any one of claims 1 to 7 are implemented.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the POP sound suppression method according to any one of claims 1 to 7 are implemented.

Images