Method, system, electronic device, medium and program product for identifying broadband aberrant sounds

Abstract

The present disclosure provides a wideband abnormal sound identification method, system, electronic device, medium and program product. The wideband abnormal sound identification method comprises: obtaining to-be-detected frequency domain data corresponding to to-be-detected audio data; obtaining wave crest data and wave trough data corresponding to the to-be-detected frequency domain data; obtaining an actual wideband abnormal sound coefficient based on the wave crest data and the wave trough data; and determining whether the to-be-detected frequency domain data has an abnormal sound in a target frequency range according to the actual wideband abnormal sound coefficient; wherein the target frequency range is constituted by a range between valley frequency points corresponding to two wave trough data adjacent to a peak frequency point corresponding to the wave crest data. The present disclosure does not require subjective determination by a person, and the identification process of the abnormal sound is independent of the test environment. The determination result is more accurate, the accuracy of the abnormal sound identification is improved, and the wideband abnormal sound is automatically, efficiently and accurately identified. Meanwhile, the maintenance personnel can timely detect and upgrade the equipment, and the user experience is improved.

Description

Technical Field

[0001] This disclosure relates to the field of audio recognition technology, and in particular to a method, system, electronic device, medium, and program product for identifying broadband abnormal sounds. Background Technology

[0002] Rotating machinery with many sheet metal parts may produce structural noise due to the resonance of components caused by the excitation source, or aerodynamic noise generated by the structure of the flow channel when the whole machine or components are working. These noises (also known as noise) can be clearly felt by users and affect the sound quality and user experience of the equipment. At the same time, after the product produces abnormal noise, it is necessary to conduct fault detection and upgrade the equipment in a timely manner.

[0003] For single-frequency (i.e. narrowband) abnormal sounds, the existing manual identification methods are to convert the steady-state time domain into a spectrum and manually screen whether the spectrum is abnormal, or to make a subjective judgment by having testers listen to the sound to determine whether there is an abnormal sound, or a combination of both methods can be used for identification.

[0004] For single-frequency abnormal sounds within a frequency range of 20Hz, the existing automated identification method is to filter out the peak values, interpolate, and then compare the sound pressure level difference between the spectrum and the characteristic loudness to identify the abnormal sound.

[0005] Since broadband (i.e., hump) abnormal sounds are abnormal sounds covering a frequency range rather than a single frequency peak, existing automatic abnormal sound identification methods can only identify single-frequency abnormal sounds and cannot identify broadband abnormal sounds. Furthermore, subjective judgment methods are greatly affected by human factors, requiring high standards for the listening environment and personnel, necessitating multiple people and groups to make judgments and compile data, and requiring a significant investment of manpower, resources, and time. Moreover, manual spectrum screening requires solid domain knowledge and extensive experience, making it a highly demanding process.

[0006] Therefore, how to automatically and effectively identify broadband abnormal sounds is a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0007] The technical problem to be solved by this disclosure is to overcome the shortcomings of the prior art in that it cannot automatically and effectively identify broadband abnormal sounds, and to provide a method, system, electronic device, medium and program product for identifying broadband abnormal sounds.

[0008] This disclosure solves the above-mentioned technical problems through the following technical solution:

[0009] Firstly, a method for identifying broadband abnormal sounds is provided, the method comprising:

[0010] Obtain the frequency domain data corresponding to the audio data to be detected;

[0011] Obtain the peak and trough data corresponding to the frequency domain data to be detected;

[0012] The actual broadband acoustic coefficient is obtained based on the peak data and the trough data;

[0013] Based on the actual broadband abnormal sound coefficient, determine whether the frequency domain data to be detected has abnormal sound in the target frequency band range;

[0014] The target frequency band range is defined by the range between the valley frequency points corresponding to two valley data points adjacent to the peak frequency point of the peak data.

[0015] Optionally, the step of determining whether there is abnormal sound in the target frequency band range of the frequency domain data to be detected based on the actual broadband abnormal sound coefficient includes:

[0016] Obtain the preset wideband sound coefficient;

[0017] If the actual broadband noise coefficient is greater than or equal to the preset broadband noise coefficient, then it is determined that there is noise in the target frequency band range.

[0018] Optionally, the step of obtaining the frequency domain data to be detected corresponding to the audio data to be detected includes:

[0019] The audio data to be detected is subjected to a preset time-frequency conversion operation to obtain initial frequency domain data;

[0020] The initial frequency domain data is subjected to glitch filtering to obtain the frequency domain data to be detected.

[0021] Optionally, the step of obtaining the peak data and trough data corresponding to the frequency domain data to be detected includes:

[0022] The peak data and the trough data are obtained by performing a derivative operation on the frequency domain data to be detected.

[0023] Optionally, the step of performing a derivative operation on the frequency domain data to be detected to obtain the peak data and the trough data includes:

[0024] Perform a differentiation operation on the frequency domain data to be detected to obtain all frequency points where the first derivative is zero and the second derivative is positive or negative;

[0025] The amplitude corresponding to the frequency point where the first derivative is zero and the second derivative is negative is taken as the peak data;

[0026] The amplitude corresponding to two adjacent frequency points where the first derivative is zero and the second derivative is positive is taken as the trough data.

[0027] Optionally, the actual broadband acoustic coefficient is obtained based on the peak data and the trough data;

[0028] Integrating the frequency domain data to be detected within the target frequency band range yields frequency domain integral data.

[0029] The ratio of the peak data to the frequency domain integral data is calculated to obtain the actual broadband acoustic coefficient.

[0030] Secondly, a broadband abnormal sound identification system is provided, the identification system comprising:

[0031] The data acquisition module is used to acquire the frequency domain data to be detected corresponding to the audio data to be detected.

[0032] The peak data acquisition module is used to acquire the peak data and trough data corresponding to the frequency domain data to be detected;

[0033] The sound coefficient acquisition module is used to acquire the actual broadband sound coefficient based on the peak data and the trough data;

[0034] Anomaly detection module is used to determine whether there is anomaly in the target frequency band range of the frequency domain data to be detected based on the actual broadband anomaly coefficient;

[0035] The target frequency band range is defined by the range between the valley frequency points corresponding to two valley data points adjacent to the peak frequency point of the peak data.

[0036] Thirdly, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and used to run on the processor, wherein the processor executes the computer program to implement the broadband abnormal sound identification method described above.

[0037] Fourthly, a computer-readable storage medium is provided, on which a computer program is stored, wherein the computer program, when executed by a processor, implements the broadband abnormal sound identification method described above.

[0038] Fifthly, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the aforementioned method for identifying broadband abnormal sounds.

[0039] Based on common knowledge in the field, the above-mentioned preferred conditions can be combined arbitrarily to obtain various preferred embodiments of this disclosure.

[0040] The positive and progressive effects of this disclosure are as follows:

[0041] The disclosed broadband abnormal noise identification method, system, electronic device, medium, and program product acquires the target frequency domain data corresponding to the audio data to be detected, acquires the peak and trough data corresponding to the target frequency domain data, obtains the actual broadband abnormal noise coefficient based on the peak and trough data, and then determines whether there is abnormal noise in the target frequency domain data within the target frequency band based on the actual broadband abnormal noise coefficient. The target frequency band is defined as the range between the trough frequency points corresponding to two adjacent peak frequency points of the peak data. This method eliminates the need for subjective human judgment, and the abnormal noise identification process is independent of the testing environment, resulting in more accurate judgments and improved accuracy. It achieves automatic, efficient, and accurate broadband abnormal noise identification, while also facilitating timely fault detection and upgrades by maintenance personnel, thus enhancing the user experience. Attached Figure Description

[0042] Figure 1 This is a first flowchart illustrating a broadband abnormal sound identification method provided in Embodiment 1 of this disclosure;

[0043] Figure 2 This is a second flowchart illustrating a broadband abnormal sound identification method provided in Embodiment 1 of this disclosure;

[0044] Figure 3 This is a schematic diagram of the third process of a broadband abnormal sound identification method provided in Embodiment 1 of this disclosure;

[0045] Figure 4 This is a schematic diagram of the initial frequency domain data in a broadband abnormal sound identification method provided in Embodiment 1 of this disclosure;

[0046] Figure 5 This is a schematic diagram of the frequency domain data to be detected obtained after the initial frequency domain data is filtered out by a glitch removal operation in the broadband abnormal sound identification method provided in Embodiment 1 of this disclosure.

[0047] Figure 6 This is a schematic diagram of the fourth process of a broadband abnormal sound identification method provided in Embodiment 1 of this disclosure;

[0048] Figure 7 This is a schematic diagram of the fifth step in a broadband noise identification method provided in Embodiment 1 of this disclosure;

[0049] Figure 8 This is a sixth flowchart illustrating a broadband abnormal sound identification method provided in Embodiment 1 of this disclosure;

[0050] Figure 9 This is a schematic diagram of the structure of a broadband abnormal sound recognition system provided in Embodiment 2 of this disclosure;

[0051] Figure 10This is a schematic diagram of the structure of the electronic device provided in Embodiment 3 of this disclosure. Detailed Implementation

[0052] The present disclosure is further illustrated below by way of embodiments, but the present disclosure is not limited to the scope of the embodiments described herein.

[0053] The prefixes such as "first" and "second" used in this disclosure are merely for distinguishing different descriptive objects and do not limit the position, order, priority, quantity, or content of the described objects. The use of ordinal numbers and other prefixes used to distinguish descriptive objects in this disclosure does not constitute a limitation on the described objects. The description of the described objects is given in the claims or the context of the embodiments, and should not be construed as an unnecessary limitation. Furthermore, in the description of this embodiment, unless otherwise stated, "multiple" means two or more.

[0054] Example 1

[0055] This embodiment provides a method for identifying broadband abnormal sounds, such as... Figure 1 As shown, the identification method includes:

[0056] S1. Obtain the frequency domain data to be detected corresponding to the audio data to be detected.

[0057] S2. Obtain the peak and trough data corresponding to the frequency domain data to be detected.

[0058] S3. Obtain the actual broadband acoustic coefficient based on peak and trough data.

[0059] S4. Determine whether there is abnormal sound in the target frequency band range of the frequency domain data to be detected based on the actual broadband abnormal sound coefficient.

[0060] The target frequency band range is defined by the range between the valley frequencies of two adjacent valley data points corresponding to the peak frequency point of the peak data. Peak data can also be called peak value data, and valley data can also be called peak-valley data.

[0061] The broadband abnormal noise identification method in this embodiment obtains the frequency domain data corresponding to the audio data to be detected, obtains the peak and trough data corresponding to the frequency domain data to be detected, obtains the actual broadband abnormal noise coefficient based on the peak and trough data, and then determines whether there is abnormal noise in the target frequency band based on the actual broadband abnormal noise coefficient. It does not require subjective human judgment, and the abnormal noise identification process is independent of the test environment, resulting in more accurate judgment and improved accuracy of abnormal noise identification. It realizes automatic, efficient, and accurate broadband abnormal noise identification, while facilitating timely fault detection and upgrades of equipment by maintenance personnel, thus improving the user experience.

[0062] In an alternative implementation, such as Figure 2 As shown, step S4 above includes:

[0063] S41. Obtain the preset wideband sound coefficient.

[0064] S42. If the actual broadband noise coefficient is greater than or equal to the preset broadband noise coefficient, then it is determined that there is noise in the target frequency band.

[0065] Specifically, the preset broadband noise coefficient can be obtained based on experimental verification, and the preset broadband noise coefficient is related to frequency and amplitude.

[0066] If the actual broadband noise coefficient is greater than or equal to the preset broadband noise coefficient, then it is determined that there is noise in the target frequency band.

[0067] For example, the target frequency range is represented as (x0-1, x0+1), where x0 is the peak frequency point corresponding to the peak data, and x0-1 and x0+1 are the valley frequencies corresponding to the two valley data points adjacent to the peak frequency point x0. If the preset broadband noise coefficient is 2, the actual broadband noise coefficient is represented as β. When β≥2, that is, when the actual broadband noise coefficient is greater than or equal to the preset broadband noise coefficient, it is considered that there is noise from frequency x0-1 to frequency x0+1.

[0068] Broadband noise refers to noise within a frequency band (e.g., the target frequency band), not noise corresponding to a single frequency peak.

[0069] The broadband noise identification method in this embodiment identifies broadband noise based on the relationship between the actual broadband noise coefficient and the preset broadband noise coefficient. It eliminates the need for subjective human judgment, and the noise identification process is independent of the testing environment, resulting in more accurate judgment and improved noise identification accuracy. This achieves automatic, efficient, and accurate broadband noise identification, while also facilitating timely fault detection and upgrades by maintenance personnel, thus enhancing the user experience.

[0070] In an alternative implementation, such as Figure 3 As shown, step S1 above includes:

[0071] S11. Perform a preset time-frequency conversion operation on the audio data to be detected to obtain the initial frequency domain data.

[0072] S12. Perform glitch filtering on the initial frequency domain data to obtain the frequency domain data to be detected.

[0073] The audio data to be detected is time-domain data. Through a preset time-frequency conversion operation, the time-domain data is converted into frequency-domain data to obtain the initial frequency-domain data corresponding to the audio data to be detected. The initial frequency-domain data can also be called the linear self-power spectrum or the original spectrum.

[0074] Specifically, the preset time-frequency conversion operation can be a Fourier transform operation or other forms of time-frequency conversion operation. This disclosure does not limit the specific form of the preset time-frequency conversion operation.

[0075] For example, the initial frequency domain data is represented by a fitting function f(x), such as Figure 4 As shown, the horizontal axis represents frequency, the left vertical axis represents the sound decibel value, and the right vertical axis represents the amplitude.

[0076] Specifically, the fitting function ;

[0077] Where x represents the frequency, and j and k represent segments in the function. Let j be the coefficient of the j-th segment. Let j be the spline function of the j-th segment. It can be a second-order, third-order, or higher-order spline function; this disclosure does not cover... The specific form is restricted.

[0078] Because the original spectrum has many single-frequency peak spikes, a peak spike filtering operation is performed to eliminate their interference. The filtering function is Q(x), and the filtered spectrum (i.e., the detected frequency domain data) is F(x). Therefore, F(x) = f(x) * Q(x). The filtered spectrum is smooth and spike-free. Figure 5 As shown, Figure 5 To Figure 4 The frequency domain data to be detected is obtained by performing peak glitch filtering on the corresponding initial frequency domain data; where the horizontal axis is frequency, the left vertical axis is sound decibel value, and the right vertical axis is amplitude.

[0079] Specifically, the formula for the filtering function Q(x) is:

[0080] ;

[0081] In the filtering operation, the number of data points is 1-n. Let f(x) be the frequency of the i-th data point in the initial frequency domain data. for The corresponding amplitude, Let be the weight coefficient for the i-th data point. For smoothing parameters, To remove curvature, It is the fitted amplitude of the i-th data point.

[0082] In an alternative implementation, such as Figure 6 As shown, step S2 above includes:

[0083] S21. Perform a derivative operation on the frequency domain data to be detected to obtain peak data and trough data.

[0084] In an alternative implementation, such as Figure 7 As shown, step S21 above includes:

[0085] S211. Perform a derivative operation on the frequency domain data to be detected to obtain all frequency points where the first derivative is zero and the second derivative is positive or negative.

[0086] S212. The amplitude corresponding to the frequency point where the first derivative is zero and the second derivative is negative is taken as the peak data.

[0087] S213. Take the amplitude corresponding to two adjacent frequency points where the first derivative is zero and the second derivative is positive as the trough data.

[0088] For example, assuming the first derivative is 0 and the second derivative is negative at x0, there is a maximum value near x0. Similarly, find two adjacent points where the first derivative is 0 and the second derivative is positive. These two points have local minima: x0-1 and x0+1. Then, a peak packet is found, with the peak amplitude (i.e., peak data) being F(x0) and the trough amplitudes (i.e., trough data) being F(x0-1) and F(x0+1) respectively. Then, x0 is the peak frequency point corresponding to the peak data, and x0-1 and x0+1 are the two trough frequency points corresponding to the peak frequency point x0. The target frequency band range is (x0-1, x0+1).

[0089] The above data is for illustrative purposes only. Those skilled in the art can also use other methods, such as peak and valley detection algorithms, to obtain peak and valley data.

[0090] The broadband abnormal sound identification method of this embodiment obtains peak and trough data by performing a derivative operation on the frequency domain data to be detected, thus laying a data foundation for subsequent broadband abnormal sound identification.

[0091] In an alternative implementation, such as Figure 8 As shown, step S3 above includes:

[0092] S31. Perform an integration operation on the frequency domain data to be detected within the target frequency band to obtain frequency domain integral data.

[0093] S32. Calculate the ratio of peak data to frequency domain integral data to obtain the actual broadband acoustic coefficient.

[0094] Taking the aforementioned example, the target frequency band range is (x0-1, x0+1), F(x0) is the peak data within the target frequency band range, and F(x) is the detected frequency domain data. Integrating F(x) over (x0-1, x0+1) yields the frequency domain integral data A, i.e.: The actual broadband exotic sound coefficient is... If the actual broadband noise coefficient β is greater than or equal to the preset broadband noise coefficient, then it is determined that there is noise in the target frequency band range (x0-1, x0+1).

[0095] In practical applications, there may be multiple target frequency bands. Therefore, the actual broadband noise coefficient is calculated for each target frequency band to determine whether there is noise in each target frequency band.

[0096] Example 2

[0097] Corresponding to the aforementioned embodiments of broadband abnormal sound identification methods, this embodiment provides a broadband abnormal sound identification system, such as... Figure 9 As shown, the broadband abnormal sound identification system includes:

[0098] The detection data acquisition module 1 is used to acquire the detection frequency domain data corresponding to the detection audio data;

[0099] Peak data acquisition module 2 is used to acquire peak data and trough data corresponding to the frequency domain data to be detected;

[0100] The exotic sound coefficient acquisition module 3 is used to acquire the actual broadband exotic sound coefficient based on peak and trough data;

[0101] The abnormal sound recognition module 4 is used to determine whether there is abnormal sound in the target frequency band range of the frequency domain data to be detected based on the actual broadband abnormal sound coefficient;

[0102] The target frequency band range is defined by the range between the valley frequency points corresponding to two adjacent valley data points that correspond to the peak frequency point of the peak data.

[0103] In an optional implementation, the abnormal sound recognition module 4 includes:

[0104] The preset coefficient acquisition unit 41 is used to acquire the preset wideband heterophony coefficient;

[0105] The abnormal sound determination unit 42 is used to determine that there is abnormal sound in the target frequency band if the actual broadband abnormal sound coefficient is greater than or equal to the preset broadband abnormal sound coefficient.

[0106] In an optional implementation, the data acquisition module 1 includes:

[0107] Data conversion unit 11 is used to perform a preset time-frequency conversion operation on the audio data to be detected to obtain initial frequency domain data;

[0108] The glitch filtering unit 12 is used to perform glitch filtering on the initial frequency domain data to obtain the frequency domain data to be detected.

[0109] In an optional implementation, the peak data acquisition module 2 is used to perform a derivative operation on the frequency domain data to be detected to obtain peak data and trough data.

[0110] In an optional implementation, the peak data acquisition module 2 includes:

[0111] The frequency point acquisition unit 21 is used to perform differentiation operations on the frequency domain data to be detected, and to acquire all frequency points where the first derivative is zero and the second derivative is positive or negative.

[0112] The peak data acquisition unit 22 is used to take the amplitude corresponding to the frequency point where the first derivative is zero and the second derivative is negative as the peak data;

[0113] The trough data acquisition unit 23 is used to take the amplitude corresponding to two adjacent frequency points where the first derivative is zero and the second derivative is positive as trough data.

[0114] In an optional implementation, the abnormal sound coefficient acquisition module 3 includes:

[0115] The integral data acquisition unit 31 is used to perform an integral operation on the frequency domain data to be detected within the target frequency band range to obtain frequency domain integral data.

[0116] The actual coefficient acquisition unit 32 is used to calculate the ratio of peak data to frequency domain integral data to obtain the actual broadband acoustic coefficient.

[0117] The broadband abnormal noise identification system in this embodiment acquires the target frequency domain data corresponding to the audio data to be detected, obtains the peak and trough data corresponding to the target frequency domain data, and obtains the actual broadband abnormal noise coefficient based on the peak and trough data. Then, it determines whether there is abnormal noise in the target frequency domain data within the target frequency band based on the actual broadband abnormal noise coefficient. It eliminates the need for subjective human judgment, and the abnormal noise identification process is independent of the testing environment, resulting in more accurate judgment and improved accuracy of abnormal noise identification. It achieves automatic, efficient, and accurate broadband abnormal noise identification, while facilitating timely fault detection and upgrades by maintenance personnel, thus improving the user experience.

[0118] As the system implementation is basically the same as the method implementation, the relevant parts can be referred to in the description of the method implementation, and will not be repeated here.

[0119] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs.

[0120] Example 3

[0121] Figure 10 This is a schematic diagram of the structure of an electronic device according to an example embodiment of the present disclosure. The electronic device includes a memory, a processor, and a computer program stored in the memory and used to run on the processor. When the processor executes the computer program, it implements the broadband abnormal sound identification method of the above embodiment. Figure 10 The electronic device 90 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments disclosed herein.

[0122] like Figure 10 As shown, the electronic device 90 can be manifested as a general-purpose computing device, such as a server device. The components of the electronic device 90 may include, but are not limited to: at least one processor 91, at least one memory 92, and a bus 93 connecting different system components (including memory 92 and processor 91).

[0123] Bus 93 includes a data bus, an address bus, and a control bus.

[0124] The memory 92 may include volatile memory, such as random access memory (RAM) 921 and / or cache memory 922, and may further include read-only memory (ROM) 923.

[0125] The memory 92 may also include a program tool 925 (or utility) having a set (at least one) program module 924, such program module 924 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment.

[0126] The processor 91 executes various functional applications and data processing by running computer programs stored in the memory 92, such as the broadband noise identification method provided in the above embodiments.

[0127] Electronic device 90 can also communicate with one or more external devices 94 (e.g., keyboard, pointing device, etc.). This communication can be performed through input / output (I / O) interface 95. Furthermore, electronic device 90 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public network, such as the Internet) via network adapter 96. As shown, network adapter 96 communicates with other modules of electronic device 90 via bus 93. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with electronic device 90, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, and data backup storage systems.

[0128] It should be noted that although several units / modules or sub-units / modules of the electronic device have been mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to embodiments of this disclosure, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided and embodied by multiple units / modules.

[0129] Example 4

[0130] This disclosure also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the broadband abnormal sound identification method provided in the above embodiments.

[0131] The readable storage medium may be more specifically adopted, including but not limited to: portable disk, hard disk, random access memory, read-only memory, erasable programmable read-only memory, optical storage device, magnetic storage device, or any suitable combination thereof.

[0132] Example 5

[0133] This disclosure also provides a computer program product, including a computer program that, when executed by a processor, implements the broadband abnormal sound identification method provided in the above embodiments.

[0134] The program code for executing the computer program product disclosed herein can be written in any combination of one or more programming languages. The program code can be executed entirely on a user device, partially on a user device, as a stand-alone software package, partially on a user device and partially on a remote device, or entirely on a remote device.

[0135] While specific embodiments of this disclosure have been described above, those skilled in the art should understand that these are merely illustrative examples, and the scope of protection of this disclosure is defined by the appended claims. Those skilled in the art can make various changes or modifications to these embodiments without departing from the principles and essence of this disclosure, but all such changes and modifications fall within the scope of protection of this disclosure.

Claims

1. A method for identifying broadband abnormal sounds, characterized in that, The identification method includes: Obtain the frequency domain data corresponding to the audio data to be detected; Obtain the peak and trough data corresponding to the frequency domain data to be detected; The actual broadband acoustic coefficient is obtained based on the peak data and the trough data; Based on the actual broadband abnormal sound coefficient, determine whether the frequency domain data to be detected has abnormal sound in the target frequency band range; The target frequency band range is defined by the range between the valley frequency points corresponding to two valley data points adjacent to the peak frequency point of the peak data. The step of determining whether there is abnormal sound in the target frequency band range of the frequency domain data to be detected based on the actual broadband abnormal sound coefficient includes: Obtain the preset wideband sound coefficient; If the actual broadband noise coefficient is greater than or equal to the preset broadband noise coefficient, then it is determined that there is noise in the target frequency band range; The step of obtaining the actual broadband exotic sound coefficient based on the peak data and the trough data includes: Integrating the frequency domain data to be detected within the target frequency band range yields frequency domain integral data. The ratio of the peak data to the frequency domain integral data is calculated to obtain the actual broadband acoustic coefficient.

2. The identification method according to claim 1, characterized in that, The step of obtaining the frequency domain data corresponding to the audio data to be detected includes: The audio data to be detected is subjected to a preset time-frequency conversion operation to obtain initial frequency domain data; The initial frequency domain data is subjected to glitch filtering to obtain the frequency domain data to be detected.

3. The identification method according to claim 1, characterized in that, The steps of obtaining the peak data and trough data corresponding to the frequency domain data to be detected include: The peak data and the trough data are obtained by performing a derivative operation on the frequency domain data to be detected.

4. The identification method according to claim 3, characterized in that, The step of performing a derivative operation on the frequency domain data to be detected to obtain the peak data and the trough data includes: Perform a differentiation operation on the frequency domain data to be detected to obtain all frequency points where the first derivative is zero and the second derivative is positive or negative; The amplitude corresponding to the frequency point where the first derivative is zero and the second derivative is negative is taken as the peak data; The amplitude corresponding to two adjacent frequency points where the first derivative is zero and the second derivative is positive is taken as the trough data.

5. A broadband abnormal sound identification system, characterized in that, The identification system includes: The data acquisition module is used to acquire the frequency domain data to be detected corresponding to the audio data to be detected. The peak data acquisition module is used to acquire the peak data and trough data corresponding to the frequency domain data to be detected; The sound coefficient acquisition module is used to acquire the actual broadband sound coefficient based on the peak data and the trough data; Anomaly detection module is used to determine whether there is anomaly in the target frequency band range of the frequency domain data to be detected based on the actual broadband anomaly coefficient; The target frequency band range is defined by the range between the valley frequency points corresponding to two valley data points adjacent to the peak frequency point of the peak data. The abnormal sound recognition module includes: The preset coefficient acquisition unit is used to acquire the preset wideband abnormal sound coefficient; An abnormal sound determination unit is used to determine that there is abnormal sound in the target frequency band range if the actual broadband abnormal sound coefficient is greater than or equal to the preset broadband abnormal sound coefficient. The heterophony coefficient acquisition module includes: An integral data acquisition unit is used to perform an integral operation on the frequency domain data to be detected within the target frequency band range to obtain frequency domain integral data. The actual coefficient acquisition unit is used to calculate the ratio of the peak data to the frequency domain integral data to obtain the actual broadband acoustic coefficient.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and for running on the processor, characterized in that, When the processor executes the computer program, it implements the method for identifying broadband abnormal sounds as described in any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method for identifying broadband abnormal sounds as described in any one of claims 1 to 4.

8. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the method for identifying broadband abnormal sounds as described in any one of claims 1 to 4.

Images