A noise detection method and a readable storage medium

Abstract

This invention discloses a noise detection method. The method and readable storage medium include: digitizing the audio to be detected and normalizing it to form a time-domain audio vector; performing a Fast Fourier Transform on the time-domain audio vector data, adding each spectral vector to a queue, calculating the similarity index and power dispersion of the first spectral vector in the queue compared to other spectral vectors, and calculating a noise index; determining whether the noise index of the first spectral vector belongs to a noise range, and if so, identifying the audio corresponding to the first spectral vector as noise data. In this invention, based on the highly dispersed spectral distribution and temporal stability of noise, the similarity index and power dispersion are combined into a noise index that can distinguish noise from normal audio, thereby accurately identifying noise signals in broadcast signals.

Description

Technical Field

[0001] This invention relates to the field of broadcast signal detection, and more specifically to a noise detection method and a readable storage medium. Background Technology

[0002] Broadcast signal monitoring mainly refers to the real-time monitoring, analysis, and alarming of signal indicators during broadcast control and transmission. Monitored indicators typically include volume, phase, and channel balance. With increasingly widespread application, higher requirements are being placed on monitoring indicators, one of which is the detection of noise signals.

[0003] Noise in broadcast signals primarily originates from transmitter malfunctions and other transmission equipment failures. When the transmitter experiences a power outage or malfunction, the radio receiver receives a regular "hissing" or "buzzing" noise. The spectral characteristics of this noise are as follows: Figure 2 and Figure 3 As shown. Because the volume of noise is similar to that of normal audio, it is impossible to accurately distinguish between noise and normal sound using indicators such as volume and phase, thus failing to promptly alert relevant personnel. Therefore, there is an urgent need for a method to detect noise that cannot be accurately identified using indicators such as volume and phase, helping on-duty personnel to promptly identify faults and ensure safe broadcasting. Summary of the Invention

[0004] The main objective of this invention is to provide a noise detection method that aims to solve the problem of detecting noise that cannot be accurately determined by indicators such as volume and phase.

[0005] To achieve the above objectives, the present invention proposes a noise detection method, comprising the following steps:

[0006] The audio to be detected is digitally sampled, the digital audio stream is packaged according to audio frames, and each sampling point of each audio frame is normalized to form a time-domain audio vector.

[0007] The time-domain audio vector data is subjected to a fast Fourier transform to convert the time-domain audio vector into a spectrum vector; each spectrum vector is added to a queue, and when the number of spectrum vectors in the queue reaches a preset number, the similarity index between the first spectrum vector in the queue and the other spectrum vectors is calculated.

[0008] Calculate the power dispersion of the first spectrum vector in the queue;

[0009] The noise index of the first spectrum vector is calculated using the similarity index and the power dispersion of the first spectrum vector.

[0010] Determine whether the first spectral vector noise index belongs to a noise interval, wherein the noise interval is the range of the noise index corresponding to the noise audio.

[0011] If so, the audio corresponding to the first spectral vector is identified as noise data.

[0012] Preferably, the function for calculating the similarity index between the first spectral vector and other vectors in the calculation queue includes:

[0013]

[0014] Where: S1 is the similarity index; F i Let be the i-th spectral vector; the dist function is the Euclidean distance between two spectral vectors; is the arithmetic mean of the i-th spectrum vector; n is the preset number of the queue.

[0015] Preferably, the function for calculating the power discreteness of the first spectrum vector in the calculation queue includes:

[0016]

[0017] Where: S2 is the power dispersion; P is the arithmetic mean of the power spectrum. i Let P be the power spectrum of the i-th sampling point. i =f i ×f i ;f i Let be the spectrum of the i-th sampling point; n is the preset number of the queue.

[0018] Preferably, the step of calculating the noise index of the first spectral vector using the similarity index and the power dispersion of the first spectral vector includes:

[0019] The product of the similarity index and the power dispersion is used as the noise index of the current audio frame.

[0020] Preferably, the noise range is 25-50.

[0021] Preferably, the step of digitally sampling the audio to be detected, packaging the digital audio stream according to audio frames, and normalizing each sampling point of each audio frame to form a time-domain audio vector includes:

[0022] Stereo and multi-channel audio are uniformly converted into mono audio.

[0023] The sampling frequency is 48kHz, and the bit depth is 24-bit or 16-bit; the digital audio stream is packaged into audio frames, with each audio frame containing m sample points.

[0024] Normalize each sample point of each audio frame.

[0025] Preferably, after the step of identifying the first audio frame corresponding to the spectrum vector as noise data, the method further includes:

[0026] Recalculate the noise index of the first said spectrum vector;

[0027] Determine whether the new noise index falls within the noise range.

[0028] Preferably, after the step of identifying the first audio frame corresponding to the spectrum vector as noise data, the method further includes:

[0029] The first spectrum vector is dequeued from the queue, and the spectrum vectors corresponding to subsequent audio frames are enqueued to form a new queue and the first frequency domain spectrum.

[0030] Calculate the noise index for the new first frequency domain spectrum until the audio to be detected ends.

[0031] The present invention also includes a readable storage medium on which a computer program is stored, the computer program being executed by a processor to implement a noise detection method as described in any of the preceding claims.

[0032] In the technical solution of this invention, based on the highly dispersed spectral distribution and temporal stability of noise, the dispersion index is represented by the statistical term dispersion, and the spectral similarity of audio is analyzed. The similarity index and the power dispersion are combined to form a noise index that can distinguish noise from normal audio. The noise range corresponding to the noise is obtained through experiments, thereby accurately judging the noise signal in the broadcast signal through the noise index. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0034] Figure 1 This is a schematic flowchart of a noise detection method according to the present invention;

[0035] Figure 2 This is a short-time spectrum diagram of the noise.

[0036] Figure 3 This is a spectral distribution diagram of the noise.

[0037] Figure 4 This is a schematic diagram illustrating the specific process of a noise detection method according to the present invention.

[0038] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0039] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0040] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indication will also change accordingly.

[0041] Furthermore, in this invention, descriptions involving "first," "second," etc., are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0042] In this invention, unless otherwise explicitly specified and limited, the terms "connection," "fixed," etc., should be interpreted broadly. For example, "fixed" can mean a fixed connection, a detachable connection, or an integral part; it can mean a mechanical connection or an electrical connection; it can mean a direct connection or an indirect connection through an intermediate medium; it can mean the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0043] Furthermore, the technical solutions of the various embodiments of the present invention can be combined with each other, but only if they are feasible for those skilled in the art. If the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection claimed by the present invention.

[0044] Please refer to Figures 1-4To achieve the above objectives, the present invention proposes a noise detection method, which includes the following steps:

[0045] S100 performs digital sampling on the audio to be detected, packages the digital audio stream according to audio frames, and normalizes each sampling point of each audio frame to form a time-domain audio vector.

[0046] S200, perform a fast Fourier transform on the time-domain audio vector data to convert the time-domain audio vector into a spectrum vector; add each spectrum vector to a queue, and when the number of spectrum vectors in the queue reaches a preset number, calculate the similarity index between the first spectrum vector in the queue and other spectrum vectors.

[0047] S300, calculate the power dispersion of the first said spectrum vector in the queue;

[0048] S400, calculate the noise index of the first spectrum vector using the similarity index and the power dispersion of the first spectrum vector;

[0049] S500, determine whether the first spectral vector noise index belongs to a noise interval, wherein the noise interval is the interval range of the noise index corresponding to the noise audio.

[0050] If so, then execute S600, and identify the audio corresponding to the first said spectrum vector as noise data.

[0051] In the technical solution of this invention, based on the highly dispersed spectral distribution and temporal stability of noise, the dispersion index is represented by the statistical term dispersion, and the spectral similarity of audio is analyzed. The similarity index and the power dispersion are combined to form a noise index that can distinguish noise from normal audio. The noise range corresponding to the noise is obtained through experiments, thereby accurately judging the noise signal in the broadcast signal through the noise index.

[0052] Specifically, each spectrum vector F is added to the queue Q, i.e., Q = {F1, F2, ..., Fn}, where Fi =<fi1,fi2,...,fim> When the number of spectral vectors in queue Q reaches the preset number n (n = 48000 × 2 / m).

[0053] According to a first embodiment of the noise detection method of the present invention, and a second embodiment of the noise detection method of the present invention, the function for calculating the similarity index between the first spectral vector and other vectors in the queue in step S200 includes:

[0054]

[0055] Where: S1 is the similarity index; Fi Let be the i-th spectral vector; the dist function is the Euclidean distance between two spectral vectors; is the arithmetic mean of the i-th spectrum vector; n is the preset number of the queue.

[0056] Specifically, the dist function is the Euclidean distance between two vectors, that is:

[0057]

[0058]

[0059] According to a first embodiment of the noise detection method of the present invention, and a third embodiment of the noise detection method of the present invention, the function for calculating the power dispersion of the first spectrum vector in the queue in step S200 includes:

[0060]

[0061] Where: S2 is the power dispersion; P is the arithmetic mean of the power spectrum. i Let P be the power spectrum of the i-th sampling point. i =f i ×f i ;f i Let be the spectrum of the i-th sampling point; n is the preset number of the queue.

[0062] According to a first embodiment of the noise detection method of the present invention, and a fourth embodiment of the noise detection method of the present invention, step S400 includes:

[0063] S410 uses the product of the similarity index and the power dispersion as the noise index of the current audio frame.

[0064] According to a fourth embodiment of the noise detection method of the present invention, and a fifth embodiment of the noise detection method of the present invention, the noise range is 25-50.

[0065] Specifically, based on the spectral characteristics of the noise, such as Figure 2 As shown, its spectral distribution is highly dispersed. In statistics, the degree of dispersion can be represented by the dispersion. The lower the dispersion, the more uniform the dispersion, while the higher the dispersion, the more concentrated the distribution. Experiments have shown that taking the square of the spectrum, i.e., the power spectrum, to calculate the dispersion makes it easier to distinguish noise. The power dispersion of noise (i.e., S2 calculated in the previous step) is generally between 1.7 and 2.5.

[0066] Another characteristic of the noise spectrum is its temporal stability. The short-term spectra of the same type of noise are relatively similar at any time. Through experiments, the spectral similarity S1 of the noise vector is generally between 15 and 20.

[0067] Here, we take S = S1 * S2 as the noise index for the current audio frame. When 25 < S < 50 (1.7 * 15 = 25.5; 2.5 * 20 = 50), it is considered noise.

[0068] Based on the first embodiment of a noise detection method of the present invention, in the sixth embodiment of a noise detection method of the present invention, the step of S100 includes:

[0069] S110, uniformly convert stereo audio and multi-channel audio into mono audio;

[0070] S120, the sampling frequency is 48KHz, and the bit depth is 24bit or 16bit; pack the digital audio stream according to audio frames, and each audio frame contains m sampling points.

[0071] S130, normalize each sampling point of each audio frame.

[0072] Specifically, each audio frame contains m (usually m = 1024) sampling points; normalize each audio frame (i.e., normalize each sampling point to between (0, 1)); perform a fast Fourier transform (FFT) to convert the time-domain audio vector into a spectral vector, denoted as F = <f1, f2,..., fm>.

[0073] Based on the first embodiment of a noise detection method of the present invention, in the seventh embodiment of a noise detection method of the present invention, after the step of S600, it includes:

[0074] S610, recalculate the noise index of the first spectral vector;

[0075] S611, determine whether the new noise index belongs to the noise interval.

[0076] Specifically, to prevent misjudgment, recalculate the noise index of the first spectral vector, or recalculate the noise index of the first spectral vector multiple times and make a judgment.

[0077] Please refer to FIG. Figure 4 , based on the first embodiment of a noise detection method of the present invention, in the eighth embodiment of a noise detection method of the present invention, after the step of S600, it includes:

[0078] S620, the first spectrum vector is dequeued from the queue, and the spectrum vectors corresponding to subsequent audio frames are enqueued to form a new queue and the first frequency domain spectrum;

[0079] S621, calculate the noise index of the new first frequency domain spectrum until the audio to be detected ends.

[0080] Specifically, when the audio is long, the judgment is made sequentially by dequeuing and enqueuing.

[0081] To achieve the above objectives, the present invention also provides a readable storage medium on which a computer program is stored, wherein the computer program, when executed by a processor, implements a noise detection method as described in any of the preceding embodiments.

[0082] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a computer-readable storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device to enter the methods described in the various embodiments of the present invention.

[0083] In the description of this specification, references to terms such as "one embodiment," "another embodiment," "other embodiments," or "first embodiment to Xth embodiment," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, method steps, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0084] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0085] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0086] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A noise detection method, characterized by, The method comprises the following steps: digitally sampling the audio to be detected, packing the digital audio stream according to audio frames, and normalizing each sampling point of each audio frame to form a time-domain audio vector; performing fast Fourier transform on the time-domain audio vector data to convert the time-domain audio vector to a frequency spectrum vector; adding each frequency spectrum vector to a queue; and calculating the similarity index of the first frequency spectrum vector and other frequency spectrum vectors in the queue when the number of frequency spectrum vectors in the queue reaches a preset number; the function of the similarity index comprises: ； wherein: is a similarity index; is the i-th spectral vector; the dist function is the Euclidean distance between two spectral vectors; is the arithmetic mean of the i-th spectral vector; n is a predetermined number of the queue; calculating the power dispersion of the first frequency spectrum vector in the queue; the function of the power dispersion comprises: ； wherein: is a power dispersion; is a power spectrum arithmetic mean; is a power spectrum of the i-th sample point, ; is a frequency spectrum of the i-th sample point; n is a preset number of the queue; calculating the noise index of the first frequency spectrum vector through the similarity index and the power dispersion of the first frequency spectrum vector; and taking the product of the similarity index and the power dispersion as the noise index of the current audio frame; judging whether the noise index of the first frequency spectrum vector belongs to a noise interval, wherein the noise interval is the interval range of the noise index corresponding to noise audio; if yes, identifying the audio corresponding to the first frequency spectrum vector as noise data.

2. The noise detection method of claim 1, wherein, The range of the noise interval is 25-50.

3. The noise detection method of claim 1, wherein, The step of digitally sampling the audio to be detected, packing the digital audio stream according to audio frames, and normalizing each sampling point of each audio frame to form a time-domain audio vector comprises: unifying stereo audio and multi-channel audio into single-channel audio; the sampling frequency is 48KHz, and the bit depth is 24bit or 16bit; the digital audio stream is packed according to audio frames, and each audio frame contains m sampling points, each sampling point of each audio frame is normalized.

4. The noise detection method of claim 1, wherein, After the step of identifying the audio frame corresponding to the first frequency spectrum vector as noise data, the following steps are included: recalculating the noise index of the first frequency spectrum vector; judging whether the new noise index belongs to the noise interval.

5. The method of claim 1, wherein, After the step of identifying the audio frame corresponding to the first frequency spectrum vector as noise data, the following steps are included: dequeueing the first frequency spectrum vector from the queue, enqueuing the frequency spectrum vectors corresponding to subsequent audio frames to form a new queue and the first frequency spectrum vector; calculating the noise index of the new first frequency spectrum vector until the audio to be detected ends.

6. A readable storage medium characterized by A computer program is stored on a computer-readable storage medium, and the computer program is executed by a processor to implement the noise detection method according to any one of claims 1 to 5.

Images