An audio dynamic noise reduction method, a computer-readable storage medium, and an electronic device.
By using dynamic audio noise reduction, signal features are acquired and noise reduction is performed, which solves the problem of unsmooth and coherent sound after audio signal processing and achieves low-cost and high-efficiency audio signal processing results.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHENZHEN JAMR TECH CO LTD
- Filing Date
- 2022-01-26
- Publication Date
- 2026-07-03
Smart Images
Figure CN116543782B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of signal processing technology, and in particular to an audio dynamic noise reduction method, a computer-readable storage medium, and an electronic device. Background Technology
[0002] In existing technologies, the collected sound can generally be processed by hardware or software. Hardware processing refers to processing through high-pass, low-pass, and band-pass filter circuits. However, after the audio signal is processed by hardware circuits, some interference signals will still remain. The cost of processing these interference signals by hardware alone will be very high.
[0003] Software processing refers to converting analog audio signals into discrete digital signals, then using algorithms to extract valid audio signals and suppress or eliminate invalid ones. However, when processing audio signals with software, some valid signals, such as detail signals, that are very similar to invalid signals are easily mistaken for invalid signals and eliminated. This can cause the audio signal to appear or disappear suddenly, resulting in a discontinuous sound and a sense of truncation. Summary of the Invention
[0004] The purpose of this invention is to provide an audio dynamic noise reduction method, device, and storage medium to address the shortcomings of existing technologies, thereby solving the technical problem that the sound is not smooth and coherent after audio signal processing.
[0005] In a first aspect, the present invention provides an audio dynamic noise reduction method, comprising:
[0006] Acquire the raw audio signal;
[0007] The original audio signal is subjected to signal feature extraction to obtain a feature audio signal;
[0008] The original audio signal is denoised using an audio dynamic noise reduction algorithm and the characteristic audio signal to obtain a valid audio signal.
[0009] The noise reduction process includes amplifying, preserving, and suppressing the original audio signal;
[0010] Outputs a valid audio signal.
[0011] Secondly, this application also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program and the processor executes the aforementioned audio dynamic noise reduction method.
[0012] Thirdly, this application also provides a computer-readable storage medium storing the computer program, wherein the computer program controls the device where the computer-readable storage medium is located to perform the audio dynamic noise reduction method when it is running.
[0013] In one embodiment, acquiring the original audio signal includes:
[0014] Acquire audio signals;
[0015] Convert the audio signal into a digital signal;
[0016] The digital signal is subjected to DC baseline removal processing to obtain the original audio signal.
[0017] In one embodiment, signal features are extracted from the original audio signal to obtain a feature audio signal.
[0018] Furthermore, a first preset feature value is determined based on the requirement for retaining detailed signals and the signal characteristics of the original audio signal. The first preset feature value is used to distinguish between valid signals and invalid signals.
[0019] Furthermore, a second preset feature value is determined based on the first preset feature value, the required sound loudness, and the signal characteristics of the original audio signal.
[0020] Furthermore, the characteristic audio signal is determined based on the valid signal, invalid signal, and the second preset characteristic value, as distinguished by the first preset characteristic value.
[0021] Furthermore, the characteristic audio signal can be categorized as: a continuous invalid signal, a gradually increasing valid signal, a continuous strong valid signal, a continuous weak valid signal, a gradually decreasing valid signal, and a signal that alternates between valid and invalid signals.
[0022] In one embodiment, the step of performing noise reduction processing on the feature audio signal using an audio dynamic noise reduction algorithm to obtain an effective audio signal includes:
[0023] The noise reduction coefficient corresponding to the current characteristic audio signal is determined by the characteristic audio signal;
[0024] The effective audio signal is determined based on the noise reduction coefficient and the sample superposition formula.
[0025] Furthermore, the noise reduction coefficient is:
[0026] Let the noise reduction coefficient be denoted as K:
[0027] K = k(x) ± C (K >= 0)
[0028] Where k(x) is the noise reduction coefficient of the previous signal, and C is the amplitude of the noise reduction coefficient.
[0029] In the formula K = k(x) ± C (K >= 0), the meaning of the noise reduction coefficient amplitude C is as follows:
[0030] When the original audio signal is converted from an invalid signal to an effective signal, the noise reduction coefficient K gradually increases, that is, K = k(x) + C (K>=0);
[0031] When the original audio signal changes from an effective signal to an invalid signal, the noise reduction coefficient K gradually decreases, i.e., K = k(x) - C (K >= 0).
[0032] Furthermore, the sample superposition formula is as follows:
[0033] D = d(t) * K;
[0034] D represents the denoised audio signal data, d(t) represents the original audio signal data, and K represents the denoising coefficient.
[0035] The noise reduction coefficient K is a variable, and in the formula for superimposing the sample data, the noise reduction coefficient K has the following meaning:
[0036] When K=0, the original audio signal is suppressed and eliminated;
[0037] When K>0 and K<1, the original audio signal is suppressed and reduced;
[0038] When K=1, the original audio signal remains unchanged;
[0039] When K>1, the original audio signal is amplified.
[0040] Furthermore, the effective audio signal is determined based on the noise reduction coefficient and the sample superposition formula:
[0041] When the characteristic audio signal is a continuous invalid signal, K=0, that is, the original audio signal remains in a suppressed and canceled state, and the output effective audio signal is a straight line with a signal amplitude of 0.
[0042] When the characteristic audio signal is a gradually increasing effective signal, that is, when it changes from an invalid signal to an effective signal, K = k(x) + C, that is, the original audio signal is gradually released from suppression, the original audio signal is slowly restored, and the output effective audio signal is a sinusoidal signal with a gradually increasing signal amplitude.
[0043] When the characteristic audio signal is a continuous and strong effective signal, K=1, that is, the original audio signal remains in its original state, and the amplitude of the output effective audio signal is consistent with the amplitude of the original audio signal;
[0044] When the characteristic audio signal is a continuous, weak, and effective signal, the original audio signal needs to be amplified to a certain extent. K = k(x) + C = n, that is, the original audio signal is amplified by n times, and the output effective audio signal is the original audio signal amplified by n times.
[0045] When the characteristic audio signal is a gradually decreasing effective signal, that is, when it changes from an effective signal to an ineffective signal, K = k(x) - C, that is, the original audio signal is gradually suppressed, the original audio signal is slowly suppressed and reduced, and the output effective audio signal is a sine signal with a gradually decreasing amplitude.
[0046] When the characteristic audio signal changes back and forth between a valid signal and an invalid signal, the valid signal is gradually amplified, K = k(x) + C; the invalid signal is gradually suppressed and reduced, K = k(x) - C, where K is a decimal that changes back and forth between 0 and 1. The output valid audio signal is a sinusoidal signal with irregular amplitude changes.
[0047] Specifically, the algorithm determines whether to amplify the original audio signal, which fluctuates between valid and invalid signals, based on the filtering of the acquisition circuit and the signal source of the original audio signal. When the acquisition circuit does not perform overflow amplification or the signal source of the original audio signal is relatively weak, the audio dynamic noise reduction algorithm needs to boost and amplify the valid signal in the original audio signal.
[0048] In one embodiment, the output valid audio signal outputs a valid audio signal to a power amplifier circuit for playing the valid audio signal.
[0049] Compared with the prior art, the audio dynamic noise reduction method disclosed in this embodiment of the invention can process audio signals in a low cost and with flexibility, extract the effective signals, filter out the invalid signals, make the sound sound smooth and mellow, preserve the details of the sound completely, and eliminate noise. It solves the problem that some effective signals are easily treated as invalid signals and eliminated during audio signal processing, and the sound is not smooth and coherent. Attached Figure Description
[0050] To more clearly illustrate the technical solution of this application, the present invention will be further described below in conjunction with the accompanying drawings and embodiments. Obviously, those skilled in the art can obtain other drawings based on these drawings without creative effort.
[0051] Figure 1 This is a flowchart illustrating an audio dynamic noise reduction method in one embodiment of this application;
[0052] Figure 2This is a signal feature map of a fetal heart rate signal in one embodiment of this application;
[0053] Figure 3 This is a flowchart illustrating an audio dynamic noise reduction algorithm in one embodiment of this application;
[0054] Figure 4 This is an internal diagram of an electronic device used to perform the audio dynamic noise reduction method according to an embodiment of this application;
[0055] Figure 5 This is a diagram of a storage unit used to store program code implementing the audio dynamic noise reduction method according to an embodiment of this application. Detailed Implementation
[0056] This section will describe in detail specific embodiments of the present invention. Preferred embodiments of the present invention are shown in the accompanying drawings. The purpose of the drawings is to supplement the textual description with graphics, so that people can intuitively and vividly understand each technical feature and overall technical solution of the present invention, but they should not be construed as limiting the scope of protection of the present invention.
[0057] This invention provides an audio dynamic noise reduction method, device, and storage medium to solve the problem that the sound is not smooth and coherent after audio signal processing in the prior art.
[0058] In one embodiment, the audio dynamic noise reduction method is applied to a Doppler fetal heart rate monitor medical device to reduce noise in the fetal heart rate audio, with a sampling frequency of 1kHz.
[0059] Reference Appendix Figure 1 , attached Figure 1 This is a flowchart illustrating an audio dynamic noise reduction method in one embodiment of this application. Taking the application of this method to the aforementioned device as an example, the method includes the following steps:
[0060] Step S101, acquire the raw audio signal, including:
[0061] Acquire audio signals;
[0062] Convert the audio signal into a digital signal;
[0063] The digital signal is subjected to DC baseline removal processing to obtain the original audio signal.
[0064] In one embodiment, the acquisition of the audio signal is achieved by acquiring the fetal heart rate audio signal through the acquisition circuit of an ADC (Audio Signal Analog-to-Digital Converter), and the acquired fetal heart rate audio analog signal is converted into a digital signal through the ADC.
[0065] In this process, the fetal heart rate audio signal is acquired by the ADC and the analog signal is converted into a digital signal. Since the fetal heart rate audio signal acquired by the ADC is a DC signal, while the signal processed by the audio dynamic noise reduction method is an AC signal, the DC baseline of the fetal heart rate audio signal acquired by the ADC needs to be removed to obtain the original fetal heart rate signal.
[0066] Step S102: Extract signal features from the original audio signal to obtain feature audio signals.
[0067] The signal characteristics include one or more of the following: frequency, amplitude, average value, and time.
[0068] Based on the signal characteristics and the original audio signal, the characteristic audio signal can be divided into the following six categories: continuous invalid signal, gradually increasing valid signal, continuous strong valid signal, continuous weak valid signal, gradually decreasing valid signal, and signal that changes back and forth between valid and invalid signals.
[0069] In one embodiment, signal features are extracted from the original fetal heart rate signal, including frequency and amplitude signal features.
[0070] Optionally, a first preset feature value is determined based on the need to preserve as much detail of the fetal heart rate signal as possible and the signal characteristics of the original audio signal:
[0071] When the amplitude of the original audio signal begins to gradually increase or the amplitude increases significantly, the signal characteristic value with the maximum amplitude before the amplitude of the original audio signal begins to gradually increase or the amplitude increases significantly is taken as the first preset characteristic value.
[0072] The first preset feature value is used to distinguish between valid and invalid signals:
[0073] The original audio signal is valid when its amplitude value is greater than the first preset characteristic value; it is valid when its amplitude value is less than the first preset characteristic value.
[0074] It should be noted that the first preset feature value can be adjusted according to the strength of the noise signal:
[0075] When the noise signal is strong, the first preset feature value takes a larger amplitude value;
[0076] When the noise signal is weak, the first preset feature value takes a smaller amplitude value.
[0077] Reference Appendix Figure 2 , attached Figure 2This is a signal feature diagram of a fetal heart rate signal in one embodiment of this application. It can be seen that the invalid signal has the characteristics of high frequency and small vibration amplitude, while the valid signal has the characteristics of low frequency and large vibration amplitude.
[0078] Among them, the appendix Figure 2 The detail signal is strictly speaking a valid signal. It is a transition signal from no sound to sound and from sound to no sound. The characteristic of the detail signal is that it gradually increases or decreases. Or, with the addition of some external interference signals, the detail signal becomes a signal that fluctuates around the boundary between valid and invalid signals and gradually increases or decreases.
[0079] Optionally, a second preset feature value is determined based on the first preset feature value, the fetal heart rate requirement for high volume, and the signal characteristics of the original audio signal:
[0080] When the amplitude of the effective signal remains greater than the amplitude of the first characteristic value and the amplitude difference is not significant, the second preset characteristic value is the characteristic value of the signal with the largest amplitude within the aforementioned signal characteristic range.
[0081] Furthermore, a characteristic audio signal is determined based on the valid signal, invalid signal, and the second preset characteristic value, distinguished by the first preset characteristic value.
[0082] The characteristic audio signal can be divided into:
[0083] When the original fetal heart rate signal is characterized by a continuous high frequency and small amplitude, the characteristic audio signal is a continuous invalid signal.
[0084] When the original fetal heart rate signal exhibits a state of gradually decreasing frequency and gradually increasing amplitude, the original audio signal becomes an effective signal that gradually increases in size, and the original fetal heart rate signal changes from an invalid signal to an effective signal.
[0085] When the original fetal heart rate signal is characterized by a sustained frequency lower than the second preset frequency and a amplitude greater than the second preset amplitude, the characteristic audio signal is a sustained and relatively strong effective signal.
[0086] When the original fetal heart rate signal is characterized by a sustained frequency higher than the second preset frequency and a amplitude lower than the second preset amplitude, the characteristic audio signal is a sustained, weak, and effective signal.
[0087] When the original fetal heart rate signal exhibits a state of gradually increasing frequency and gradually decreasing amplitude, the characteristic audio signal becomes a valid signal, and the original fetal heart rate signal changes from a valid signal to an invalid signal.
[0088] When the original fetal heart rate signal is characterized by large changes in frequency and amplitude over a short period of time, the characteristic audio signal alternates between valid and invalid signals.
[0089] It should be noted that the second preset frequency and the second preset amplitude are set based on experimental data.
[0090] Through the above steps, compared with the prior art, the technical solution proposed in this application can distinguish between valid signals and invalid signals to the greatest extent, and can obtain characteristic audio signals based on the characteristics of the distinguished valid signals and detailed signals.
[0091] Step S103: The original audio signal is denoised using an audio dynamic noise reduction algorithm and the characteristic audio signal to obtain an effective audio signal.
[0092] The noise reduction coefficient corresponding to the current characteristic audio signal is determined by the characteristic audio signal;
[0093] The effective audio signal is determined based on the noise reduction coefficient and the sample superposition formula.
[0094] Optionally, in one embodiment, the audio dynamic noise reduction method is applied to a Doppler fetal heart rate monitoring medical device. There is no excessive noise when sampling the fetal heart rate, the sampling frequency is low, and the boundary between the effective signal and the invalid signal is obvious. The coefficient amplitude C can be defined as a larger constant, that is, the noise reduction coefficient amplitude C = 0.2.
[0095] According to the above embodiments, refer to the appendix Figure 2 The attached figure is a flowchart illustrating an audio dynamic noise reduction algorithm in one embodiment of this application.
[0096] Step S201: Determine the noise reduction coefficient corresponding to the current characteristic audio signal through the characteristic audio signal.
[0097] The noise reduction coefficient K is:
[0098] K = k(x) ± 0.2 (K >= 0)
[0099] Where k(x) is the noise reduction coefficient of the previous signal.
[0100] In the formula K = k(x) ± 0.2 (K >= 0), the meaning of the amplitude of the noise reduction coefficient is as follows:
[0101] As the original audio signal changes from an invalid signal to an effective signal, the noise reduction coefficient K gradually increases, i.e., K = k(x) + 0.2 (K >= 0);
[0102] When the original audio signal changes from an effective signal to an invalid signal, the noise reduction coefficient K gradually decreases, i.e., K = k(x) - 0.2 (K >= 0).
[0103] Step 202: Determine the effective audio signal based on the noise reduction coefficient and the sample superposition formula.
[0104] The formula for sample stacking is:
[0105] D = d(t) * K;
[0106] D represents the denoised audio signal data, d(t) represents the original audio signal data, and K represents the denoising coefficient.
[0107] The noise reduction coefficient K is a variable, and in the formula for superimposing the sample data, the noise reduction coefficient K has the following meaning:
[0108] When K=0, the original audio signal is suppressed and eliminated;
[0109] When K>0 and K<1, the original audio signal is suppressed and reduced;
[0110] When K=1, the original audio signal remains unchanged;
[0111] When K>1, the original audio signal is amplified.
[0112] In one embodiment, the original audio signal data is denoised based on the characteristic audio signal and the audio dynamic noise reduction algorithm.
[0113] The noise reduction process includes amplifying, preserving, and suppressing the original audio signal:
[0114] When the characteristic audio signal is a continuous invalid signal, K = 0, that is, the d(t) data remains in a suppressed and canceled state, and the valid audio signal is represented by a straight line with a signal amplitude of 0;
[0115] When the characteristic audio signal is a gradually increasing effective signal, that is, when it changes from an invalid signal to an effective signal, K = k(x) + 0.2, that is, the d(t) data is gradually released from suppression, the original audio signal is slowly restored, and the amplitude of the output effective audio signal is a sine signal with a gradually increasing signal amplitude.
[0116] When the characteristic audio signal is a continuous and strong effective signal, K=1, the d(t) data is kept, that is, the original state of the original audio signal is preserved, and the amplitude of the output effective audio signal is consistent with the original audio signal;
[0117] When the characteristic audio signal is a continuous, weak, and effective signal, the original audio signal needs to be amplified to a certain extent, K = k(x) + 0.2 = n, that is, the d(t) data is amplified by n times;
[0118] When the characteristic audio signal is a gradually decreasing effective signal, that is, when it changes from an effective signal to an ineffective signal, K = k(x) - 0.2, that is, the d(t) data is gradually increased and suppressed, and the output effective audio signal is the original audio signal amplified by n times;
[0119] When the characteristic audio signal switches back and forth between a valid signal and an invalid signal, the d(t) data should be unsuppressed when the original audio signal is a valid signal, and the d(t) data should be suppressed and reduced when the original audio signal is an invalid signal, i.e., K = k(x) ± 0.2, where K is a decimal that varies between 0 and 1.
[0120] Optionally, in one embodiment, the signal is amplified by a factor of 1.2 (n = 1.2) according to the signal characteristics of the weaker effective signal and the sustained stronger effective signal, so that the output effective audio signal is clear and complete.
[0121] Optionally, in one embodiment, the decision on whether to amplify the valid signal in the back-and-forth conversion between valid and invalid signals of the characteristic audio signal is determined based on the acquisition circuit of the ADC and the signal source of the original audio signal.
[0122] Optionally, when the ADC acquisition circuit does not perform gain amplification or the signal source of the original audio signal is relatively weak, the audio dynamic noise reduction algorithm needs to boost and amplify the effective signal.
[0123] In one embodiment, the audio dynamic noise reduction method is actually applied to the fetal heart rate monitoring scenario. The fetal heart rate signal is a very weak signal. In order to better process the audio signal, it is necessary to boost and amplify the audio signal before processing.
[0124] Through the above steps, compared with the prior art, the technical solution proposed in this application performs corresponding audio signal processing based on the distinguished characteristic audio signals, which can preserve the detail signals to the greatest extent and make the sound smooth and coherent.
[0125] Step S104: Output a valid audio signal.
[0126] In one embodiment, the output valid audio signal outputs the valid audio signal to the power amplifier circuit, which is used to receive the processed valid audio signal of the fetal heart rate and output the fetal heart rate audio.
[0127] Optionally, the processed fetal heart rate audio can be played by connecting a playback device such as a speaker, headphones, or audio system to the power amplifier circuit.
[0128] Reference Appendix Figure 4 The diagram shows a structural block diagram of an electronic device 10 provided in this application embodiment. The electronic device 10 in this application may include one or more of the following components: a processor 11, a memory 12, and one or more computer programs stored in the memory 12 and executable on the processor 11. When the processor 11 executes the computer program, it implements the steps in the above-described audio dynamic noise reduction method embodiment, for example, as shown in the attached diagram. Figure 1 Steps S101 to S104 are shown in the figure.
[0129] The processor 11 may include one or more processing cores. The processor 11 connects to various parts of the electronic device 10 using various interfaces and lines, and performs various functions and processes data of the electronic device 10 by running or executing instructions, programs, code sets or instruction sets stored in the memory 12, and calling data stored in the memory 12.
[0130] Optionally, the electronic device 10 may be a processing device such as an MCU, a computer, or a cloud server. The electronic device 10 may include, but is not limited to, a processor 11 and a memory 12. Those skilled in the art will understand that the schematic diagram is merely an example of the electronic device 10 and does not constitute a limitation on the electronic device 10. It may include more or fewer components than illustrated, or combine certain components, or use different components.
[0131] Reference Appendix Figure 5 As shown in the figure, an embodiment of this application provides a computer-readable storage medium 20. This computer-readable storage medium 20 stores a computer program 21, which can be invoked by the processor 11 to execute the methods described in the above method embodiments.
[0132] The computer-readable storage medium 20 may be an electronic memory such as flash memory, EEPROM (erasable programmable read-only memory), EPROM, hard disk, or ROM. Optionally, the computer-readable storage medium 20 includes a non-transitory computer-readable storage medium. The computer-readable storage medium 20 has storage space for a computer program that performs any of the method steps of the audio dynamic noise reduction method. This program code can be read from or written to one or more computer program products.
[0133] Compared with the prior art, the present invention discloses an audio dynamic noise reduction method, a computer-readable storage medium and an electronic device that can process audio signals in a low cost and with flexibility, extract effective signals, filter out invalid signals and make the sound sound smooth and mellow, preserve the details of the sound completely and eliminate noise, and solve the problem that some effective signals are easily treated as invalid signals and eliminated during audio signal processing, resulting in an unsmooth and coherent sound.
[0134] The above embodiments are only used to illustrate the technical solutions of this application and are not intended to limit this application. Although the embodiments have described this application in detail, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. These modifications or substitutions can be made without departing from the principle of this invention, and these modifications and substitutions are also within the protection scope of this invention.
Claims
1. An audio dynamic noise reduction method, characterized in that, The audio dynamic noise reduction method includes: Acquire the raw audio signal; The original audio signal is subjected to signal feature extraction to obtain feature audio signals. The obtained feature audio signals are: continuous invalid signals, gradually increasing valid signals, continuous strong valid signals, continuous weak valid signals, gradually decreasing valid signals, and signals that change back and forth between valid and invalid signals. The original audio signal is denoised using an audio dynamic noise reduction algorithm and the characteristic audio signal to obtain a valid audio signal. The step of denoising the characteristic audio signal using the audio dynamic noise reduction algorithm to obtain a valid audio signal includes: The noise reduction coefficient corresponding to the current characteristic audio signal is determined by the characteristic audio signal; The effective audio signal is determined based on the noise reduction coefficient and the sample superposition formula. The noise reduction coefficient is: The noise reduction coefficient is denoted as K: K = k(x) ± C (K>=0), where k(x) is the noise reduction coefficient of the previous signal, and C is the amplitude of the noise reduction coefficient. The sample superposition formula is: D = d(t). K; D represents the denoised audio signal data, d(t) represents the original audio signal data, and K represents the denoising coefficient; The noise reduction process includes amplifying, preserving, and suppressing the original audio signal; Outputs a valid audio signal.
2. The audio dynamic noise reduction method according to claim 1, characterized in that, The steps for acquiring the original audio signal include: Acquire audio signals; Convert the audio signal into a digital signal; The digital signal is subjected to DC baseline removal processing to obtain the original audio signal.
3. The audio dynamic noise reduction method according to claim 1, characterized in that, The signal characteristics include one or more of frequency, amplitude, average value, and time.
4. The audio dynamic noise reduction method according to claim 1, characterized in that, The original audio signal is denoised using an audio dynamic noise reduction algorithm and the characteristic audio signal to obtain a valid audio signal. When the original audio signal is a continuous invalid signal, K=0, and the original audio signal is suppressed and eliminated; When the original audio signal is an effective signal that gradually increases in size, K = k(x) + C, and the suppression of the original audio signal is gradually released. When the characteristic audio signal is a continuous and strong effective signal, K=1, and the original audio signal remains in its original state; When the characteristic audio signal is a continuous, weak, effective signal, K = k(x) + C = n, amplifying the original audio signal by n times; When the characteristic audio signal is a gradually decreasing effective signal, K = k(x) - C, the original audio signal is suppressed and reduced; When the characteristic audio signal alternates between valid and invalid signals, K = k(x) ± C, amplifying the valid signal of the original audio signal and suppressing the invalid signal of the original audio signal.
5. An electronic device, characterized in that: It includes a memory and a processor, the memory storing a computer program, and the processor executing an audio dynamic noise reduction method as described in any one of claims 1 to 4.
6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein, when the computer program is executed, it controls the device on which the computer-readable storage medium is located to perform the steps of the audio dynamic noise reduction method as described in any one of claims 1 to 4.