Audio noise suppression method, device and equipment

Abstract

The present invention relates to an audio noise suppression method, apparatus and device. The method includes: receiving an audio input signal, performing envelope detection to obtain the amplitude of the audio signal, recording the change in the amplitude of the audio signal and the amplitude of the silent segment, tracking the amplitude of the noise for the audio input signal, and dynamically obtaining the amplitude of the noise signal; determining the closing threshold of the noise gate according to the amplitude of the noise signal; judging whether the change in the amplitude of the audio signal within a preset detection time parameter is greater than a preset voice segment signal amplitude increment parameter, and if it is greater, the noise gate is opened; judging whether the change in the amplitude of the audio signal within the preset detection time parameter continues to be lower than the closing threshold, and if it continues to be lower, the noise gate is closed; outputting the current gain according to the opening and closing state of the noise gate and a preset gain attenuation parameter when closing; filtering the current gain, and applying the filtered output gain to the audio input signal to obtain and output a target audio signal, effectively improving the noise suppression effect of the noise gate.

Description

Technical Field

[0001] The present invention belongs to the technical field of noise suppression, and particularly relates to an audio noise suppression method, device and equipment. Background Art

[0002] The main use of a noise gate is to attenuate signals below a threshold, and this attenuated signal is usually noise. In a recording studio or a home recording studio, a noise gate can solve problems such as the background noise of recording equipment, the sound of a computer fan, the sound of an air conditioner working, instrument crosstalk, and leakage sound of monitoring headphones. A noise gate does not really remove the noise in the signal itself, but only attenuates the signal below the threshold (a large attenuation amount will form a muting effect), and the signal above the threshold will be completely retained, including the background noise.

[0003] The working mode of a traditional noise gate is that when the signal exceeds the threshold, the noise gate is in the open state (Gate Open), and the signal is output normally. When the signal is below the threshold, the noise gate is in the closed state (Gate Close), and the signal is attenuated to the target level. Since most noise gates are based on peak processing mode and are quite sensitive, once the peak of the signal frequently swings back and forth across the threshold, it will cause the noise gate to open and close frequently, resulting in chattering distortion. To solve the distortion problem, the concept of a hysteresis interval is introduced into the noise gate. The hysteresis interval is used to reduce the opening and closing frequency, so that the state of the signal tends to be more "stable". The hysteresis interval can be understood as two thresholds, one is the opening threshold and the other is the closing threshold. When the signal exceeds the opening threshold, the noise gate is in the open state, and when the signal is below the closing threshold, the noise gate is in the closed state.

[0004] The disadvantage of the hysteresis interval technology is that it is difficult to set the hysteresis threshold. If the hysteresis interval is too small, it cannot solve the chattering distortion. If the hysteresis threshold is set larger, a higher opening threshold or a lower closing threshold is required. However, the higher the opening threshold, the greater the signal loss, and the lower the closing threshold, the worse the noise suppression effect. Therefore, it is easy to choose the two thresholds improperly, resulting in a poor noise gate effect. Summary of the Invention

[0005] In order to at least solve the above problems existing in the prior art, the present invention provides an audio noise suppression method, device and equipment to improve the noise suppression effect of the noise gate.

[0006] The technical solution provided by the present invention is as follows:

[0007] On the one hand, an audio noise suppression method includes:

[0008] Receiving an audio input signal;

[0009] Perform envelope detection on the audio input signal to obtain the amplitude of the audio signal, and record the change in the amplitude of the audio signal and the amplitude of the silent segment;

[0010] Based on the amplitude of the silent segment, perform noise amplitude tracking on the audio input signal to dynamically obtain the amplitude of the noise signal; determine the closing threshold of the noise gate according to the amplitude of the noise signal;

[0011] Judge whether the change in the amplitude of the audio signal within the preset detection time parameter is greater than the preset speech segment signal amplitude increment parameter. If it is greater, the noise gate is opened;

[0012] Judge whether the change in the amplitude of the audio signal within the preset detection time parameter is continuously lower than the closing threshold. If it is continuously lower, the noise gate is closed;

[0013] Output the current gain according to the opening and closing states of the noise gate and the preset gain attenuation parameter when closing the door;

[0014] Filter the current gain and apply the filtered output gain to the audio input signal to obtain and output the target audio signal.

[0015] Optionally, the above-mentioned audio noise suppression method further includes:

[0016] Perform a delay process on the audio input signal that matches the preset detection time parameter.

[0017] Optionally, the above-mentioned performing noise amplitude tracking on the audio input signal to dynamically obtain the amplitude of the noise signal includes:

[0018] Set the audio signal processing block length;

[0019] Based on the audio signal processing block length, divide the audio input signal into multiple audio blocks in chronological order;

[0020] Determine that the signal envelope mean value of the first audio block is the current noise floor;

[0021] Record the minimum value of the envelope mean values of all audio blocks within the preset time;

[0022] Based on a preset formula and the minimum value, update and filter the current noise floor, and output the amplitude of the noise signal.

[0023] Optionally, the above-mentioned determining the closing threshold of the noise gate according to the amplitude of the noise signal includes:

[0024] Amplify the amplitude of the noise signal by a preset multiple and use the amplified value as the closing threshold.

[0025] Optionally, the above-mentioned preset detection time parameter is 30 ms - 100 ms, and the preset voice segment signal amplitude increment parameter is 15 dB.

[0026] Optionally, the envelope detection of the audio input signal includes:

[0027] Performing envelope detection on the audio input signal based on the envelope peak power; and / or;

[0028] Performing envelope detection on the audio input signal based on the spectral envelope.

[0029] On the other hand, an audio noise suppression device includes:

[0030] A receiving module, configured to receive an audio input signal;

[0031] An envelope detection module, configured to perform envelope detection on the audio input signal, obtain the amplitude of the audio signal, and record the change in the amplitude of the audio signal and the amplitude of the silent segment;

[0032] A noise tracking module, configured to perform noise amplitude tracking on the audio input signal based on the amplitude of the silent segment, and dynamically obtain the amplitude of the noise signal;

[0033] A switch door control module, configured to determine the closing threshold of the noise gate according to the amplitude of the noise signal; determine whether the change in the amplitude of the audio signal within the preset detection time parameter is greater than the preset voice segment signal amplitude increment parameter, and if it is greater, the noise gate is opened; determine whether the change in the amplitude of the audio signal within the preset detection time parameter is continuously lower than the closing threshold, and if it is continuously lower, the noise gate is closed;

[0034] A gain calculation module, configured to output the current gain according to the open / close state of the noise gate and the preset gain attenuation parameter when closing the door;

[0035] An output module, configured to filter the current gain and apply the filtered output gain to the audio input signal to obtain a target audio signal and output it.

[0036] Optionally, the above-mentioned audio noise suppression device further includes a matching delay module, configured to:

[0037] Perform a delay process on the audio input signal that matches the preset detection time parameter.

[0038] Optionally, the above-mentioned noise tracking module is specifically configured to:

[0039] Set the block length of audio signal processing;

[0040] Divide the audio input signal into a plurality of audio blocks in chronological order based on the audio signal processing block length;

[0041] Determine the signal envelope mean of the first audio block as the current noise floor;

[0042] Record the minimum value of the envelope means of all audio blocks within a preset time;

[0043] Update and filter the current noise floor based on a preset formula and the minimum value, and output the amplitude of the noise signal.

[0044] In another aspect, an audio noise suppression device includes: a processor, and a memory connected to the processor;

[0045] The memory is used to store a computer program, and the computer program is at least used to execute the audio noise suppression method described in any one of the above;

[0046] The processor is used to call and execute the computer program in the memory.

[0047] The beneficial effects of the present invention are:

[0048] An audio noise suppression method, device and equipment provided by the present invention, the method includes: receiving an audio input signal; performing envelope detection on the audio input signal to obtain the amplitude of the audio signal, and recording the amplitude change of the audio signal and the amplitude of the silent segment; based on the amplitude of the silent segment, performing noise amplitude tracking on the audio input signal to dynamically obtain the amplitude of the noise signal; determining the closing threshold of the noise gate according to the amplitude of the noise signal; judging whether the amplitude change of the audio signal within a preset detection time parameter is greater than a preset voice segment signal amplitude increment parameter, if it is greater, the noise gate is opened; judging whether the amplitude change of the audio signal within a preset detection time parameter is continuously lower than the closing threshold, if it is continuously lower, the noise gate is closed; outputting the current gain according to the opening and closing state of the noise gate and a preset gain attenuation parameter when closing the door; filtering the current gain, and applying the filtered gain to the audio input signal to obtain and output a target audio signal, suppressing noise through the noise gate, avoiding tremor distortion caused by breathing sounds, etc., and can automatically and dynamically determine the opening and closing threshold according to the specific signal amplitude, effectively alleviating the distortion problem of fixed opening and closing thresholds, and effectively improving the noise suppression effect of the noise gate. Description of the Drawings

[0049] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0050] Figure 1 is a flowchart of an audio noise suppression method provided by an embodiment of the present invention;

[0051] Figure 2 is a schematic structural diagram of an audio noise suppression device provided by an embodiment of the present invention;

[0052] Figure 3 is a schematic structural diagram of an audio noise suppression device provided by an embodiment of the present invention. Detailed Embodiments

[0053] To make the objectives, technical solutions, and advantages of the present invention clearer, the following will describe the technical solutions of the present invention in detail. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope protected by the present invention.

[0054] Figure 1 is a flowchart of an audio noise suppression method provided by an embodiment of the present invention.

[0055] As Figure 1 shown, an audio noise suppression method provided in this embodiment includes the following steps:

[0056] S11. Receive an audio input signal.

[0057] Taking a voice signal as an example for illustration, the received audio input signal is a voice signal, and the voice signal includes a target voice signal and noises such as breathing sounds.

[0058] S12. Perform envelope detection on the audio input signal, obtain the amplitude of the audio signal, and record the change in the amplitude of the audio signal and the amplitude of the silent segment.

[0059] For the received voice signal, envelope detection can be performed on the audio input signal based on the peak power of the envelope (Peak); and / or; envelope detection can be performed on the audio input signal based on the spectral envelope (RMS), where RMS calculates the RMS value of the first N samples as the level of the current sample. Then record the change in the amplitude of the audio signal and the amplitude of the silent segment, and the silent segment is the smallest voice block within a period of time.

[0060] S13. Based on the amplitude of the silent segment, perform noise amplitude tracking on the audio input signal, dynamically obtain the amplitude of the noise signal, and determine the closing threshold of the noise gate according to the amplitude of the noise signal.

[0061] Specifically, for noise tracking, that is, to track the amplitude of the audio input signal in real time. Since there are silent segments in the speech signal, the silent segment is considered as the smallest speech block within a period of time, and then continuously extract the envelope amplitude of the silent segment signal as the update target of the noise amplitude, that is, the dynamic determination of the noise amplitude is realized. The specific process is to set the audio signal processing block length, which can usually be (1ms - 10ms). Based on the audio signal processing block length, divide the audio input signal into multiple audio blocks in chronological order, for example, n audio blocks; determine the signal envelope mean n1 of the first audio block as the current noise floor n0; record the minimum value n2 of the envelope means of all audio blocks within the preset time. For example, the preset time can be 3 - 10s. Based on the preset formula and the minimum value, update and filter the current noise floor, update n1 = n0, and output the amplitude of the noise signal. For example, the preset formula is n0 = n1 * (1 - r) + n2 * r, where 0 < r < 1. The finally output noise amplitude is n0, and the tracking speed is determined by r. The larger r is, the faster the tracking speed; the smaller r is, the slower the tracking speed. Through the cyclic update of n0, the dynamic acquisition of the noise amplitude is realized, improving the accuracy of amplitude acquisition.

[0062] Determine the closing threshold of the noise gate according to the amplitude of the noise signal. In the specific implementation process, the speech signal can have a large amplitude increment (20dB or even larger) within a short time (100ms), while the amplitude of the breathing sound increases slowly. Therefore, a preset detection time parameter can be set, for example, it can be 30ms - 100ms. Within the preset detection time parameter, the closing threshold can be determined according to the amplitude of the target audio signal. Specifically, to determine the closing threshold of the noise gate according to the amplitude of the noise signal, it includes: amplifying the amplitude of the noise signal by a preset multiple and using the amplified value as the closing threshold, for example, 2 * n0.

[0063] S14. Judge whether the amplitude change of the audio signal within the preset detection time parameter is greater than the preset speech segment signal amplitude increment parameter. If it is greater, the noise gate is opened.

[0064] Specifically, the preset speech segment signal amplitude increment parameter can be 15dB, that is, within the preset time parameter of 30ms - 100ms, if the amplitude change of the audio signal has an amplitude increment of more than 15dB, the noise gate is opened.

[0065] S15. Judge whether the amplitude change of the audio signal within the preset detection time parameter is continuously lower than the closing threshold. If it is continuously lower, the noise gate is closed.

[0066] Specifically, in the above steps, a closing threshold is determined, and then it is judged whether the change in the audio amplitude continuously remains below the closing threshold within the preset detection time parameter. If it continuously remains below the closing threshold, the noise gate is closed. Continuously remaining below can mean continuously being below the closing threshold within a certain period of time, and the determination of the specific duration can be considered for revision. Thus, by judging the change in the audio amplitude signal within the preset detection time parameter, the opening or closing of the noise gate is dynamically determined.

[0067] S16. Output the current gain according to the opening / closing state of the noise gate and the preset gain attenuation parameter when closing the door.

[0068] After dynamically controlling the opening and closing of the noise gate, the current gain is output according to the opening / closing state of the noise gate and the preset gain attenuation parameter when closing the door. When the noise gate is open, the output gain is 0 dB, and when the noise gate is closed, the output gain is the gain value given by the gain attenuation parameter.

[0069] Theoretically speaking, the noise gate is a special form of a low-level expander. The expander is a device for dynamically processing sound signals, and it can expand the dynamic range of sound signals. The low-level expander is such that when the input signal is less than the threshold, the input signal is expanded, and when the input signal is higher than the threshold, the relationship between the output signal and the input signal still remains at 1:1, that is, normal output. The noise gate is made using the principle of the low-level expander and is a special form of the low-level expander. It requires the expansion ratio of the low-level expander to be ∞:1 or a larger ratio (above 3:1). If the expansion ratio is too small, the low-level expander cannot well act as a noise gate. Therefore, a low-level expander can be used instead of the noise gate. Opening the door is equivalent to reducing the gain, and closing the door is equivalent to increasing the gain. Outputting the current gain can be calculating the gain according to the expansion slope control parameter. The smaller the signal gain, the smaller the gain, and the larger the signal gain, the larger the gain.

[0070] S17. Filter the current gain and apply the filtered output gain to the audio input signal to obtain and output the target audio signal.

[0071] Filter the current gain to prevent the gain from jumping between 0 dB and the configured gain attenuation parameter, and the gain value after passing through the filter output changes smoothly. The establishment time and release time can also be set. The establishment time refers to the opening speed of the noise gate. For example, for a -60 dB gain attenuation, when the noise gate is opened, the time it takes for the attenuation amount to change from -60 dB to 0 dB is the establishment time of the noise gate. Correspondingly, the release time is the closing speed of the noise gate. When the noise gate is closed, the time it takes for the attenuation amount to change from 0 dB to -60 dB is the release time of the noise gate. Apply the gain output after filtering to the speech signal, i.e., the audio input signal, to obtain the target audio signal, and then output it.

[0072] An audio noise suppression method, device, and equipment provided in this embodiment. The method includes: receiving an audio input signal; performing envelope detection on the audio input signal to obtain the amplitude of the audio signal, and recording the amplitude change of the audio signal and the amplitude of the silent segment; based on the amplitude of the silent segment, performing noise amplitude tracking on the audio input signal to dynamically obtain the amplitude of the noise signal; determining the closing threshold of the noise gate according to the amplitude of the noise signal; judging whether the amplitude change of the audio signal within the preset detection time parameter is greater than the preset speech segment signal amplitude increment parameter. If it is greater, the noise gate is opened; judging whether the amplitude change of the audio signal within the preset detection time parameter continues to be lower than the closing threshold. If it continues to be lower, the noise gate is closed; outputting the current gain according to the opening and closing states of the noise gate and the preset gain attenuation parameter when closing; filtering the current gain and applying the filtered output gain to the audio input signal to obtain the target audio signal and output it. Noise suppression is performed through the noise gate to avoid tremor distortion caused by breathing sounds, etc., and the opening and closing thresholds can be automatically and dynamically determined according to the specific signal amplitude, effectively alleviating the distortion problem of fixed opening and closing thresholds and effectively improving the noise suppression effect of the noise gate.

[0073] Further, after determining the preset detection time in this embodiment, it further includes: performing a delay process on the audio input signal that matches the pre-detection time. By performing a delay process on the audio input signal that matches the pre-detection time to delay the input audio signal, it is possible to prevent damage to the signal starting point caused by the noise gate not opening in time during the detection process and trailing noise caused by not closing in time.

[0074] Based on the same general inventive concept, this application also protects an audio noise suppression device.

[0075] Figure 2 It is a schematic structural diagram of an audio noise suppression device provided by an embodiment of the present invention.

[0076] As Figure 2 shown, an audio noise suppression device provided in this embodiment includes:

[0077] A receiving module 10 for receiving an audio input signal;

[0078] An envelope detection module 20 for performing envelope detection on the audio input signal, obtaining the amplitude of the audio signal, and recording the change in the amplitude of the audio signal and the amplitude of the silent segment;

[0079] A noise tracking module 30 for performing noise amplitude tracking on the audio input signal based on the amplitude of the silent segment and dynamically obtaining the amplitude of the noise signal;

[0080] A switch door control module 40 for determining the closing threshold of the noise gate according to the amplitude of the noise signal; judging whether the change in the amplitude of the audio signal within a preset detection time parameter is greater than a preset voice segment signal amplitude increment parameter, if it is greater, the noise gate is opened; judging whether the change in the amplitude of the audio signal within the preset detection time parameter is continuously lower than the closing threshold, if it is continuously lower, the noise gate is closed;

[0081] A gain calculation module 50 for outputting the current gain according to the opening and closing state of the noise gate and a preset gain attenuation parameter when closing;

[0082] An output module 60 for filtering the current gain and applying the filtered output gain to the audio input signal to obtain and output a target audio signal.

[0083] An audio noise suppression device provided in this embodiment includes: receiving an audio input signal; performing envelope detection on the audio input signal, obtaining the amplitude of the audio signal, and recording the change in the amplitude of the audio signal and the amplitude of the silent segment; performing noise amplitude tracking on the audio input signal based on the amplitude of the silent segment and dynamically obtaining the amplitude of the noise signal; determining the closing threshold of the noise gate according to the amplitude of the noise signal; judging whether the change in the amplitude of the audio signal within a preset detection time parameter is greater than a preset voice segment signal amplitude increment parameter, if it is greater, the noise gate is opened; judging whether the change in the amplitude of the audio signal within the preset detection time parameter is continuously lower than the closing threshold, if it is continuously lower, the noise gate is closed; outputting the current gain according to the opening and closing state of the noise gate and a preset gain attenuation parameter when closing; filtering the current gain and applying the filtered output gain to the audio input signal to obtain and output a target audio signal. By suppressing noise through the noise gate, it avoids tremor distortion caused by breathing sounds, etc., and can automatically and dynamically determine the opening and closing thresholds according to the specific signal amplitude, effectively alleviating the distortion problem of fixed opening and closing thresholds and effectively improving the noise suppression effect of the noise gate.

[0084] Further, the audio noise suppression device in this embodiment further includes a matching delay module for:

[0085] Perform a delay process that matches the pre-detection time for the audio input signal.

[0086] Further, the noise tracking module 30 is specifically configured to:

[0087] Set the block length of audio signal processing;

[0088] Based on the block length of audio signal processing, divide the audio input signal into multiple audio blocks in chronological order;

[0089] Determine that the signal envelope mean value of the first audio block is the current noise floor;

[0090] Record the minimum value of the envelope mean values of all audio blocks within a preset time;

[0091] Based on a preset formula and the minimum value, update and filter the current noise floor, and output the amplitude of the noise signal.

[0092] Regarding the embodiments of the device part, detailed introductions and explanations have been made in the corresponding method embodiments. Therefore, specific elaborations will not be made in the corresponding device part, and they can be understood by referring to each other.

[0093] Based on the same general inventive concept, the present application also protects an audio noise suppression device.

[0094] Figure 3 It is a schematic structural diagram of an audio noise suppression device provided by an embodiment of the present invention.

[0095] As Figure 3 shown, an audio noise suppression device provided in this embodiment includes: a processor 100, and a memory 200 connected to the processor;

[0096] The memory is used to store a computer program, and the computer program is at least used to execute the audio noise suppression method of any of the above embodiments;

[0097] The processor is used to call and execute the computer program in the memory.

[0098] An embodiment of the present application also protects a storage medium that stores a computer program. When the computer program is executed by a processor, it implements each step of the audio noise suppression method described in any of the above embodiments.

[0099] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, and all should be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claimed rights.

[0100] It is understandable that the same or similar parts in the above embodiments can be referred to each other, and for the content not detailed in some embodiments, reference can be made to the same or similar content in other embodiments.

[0101] It should be noted that in the description of the present invention, terms such as "first", "second", etc. are only used for descriptive purposes and cannot be construed as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, the meaning of "a plurality of" refers to at least two.

[0102] Any process or method description shown in the flowchart or described in other ways herein can be understood as representing a module, segment, or part of code including one or more executable instructions for implementing a specific logical function or process. The scope of the preferred embodiments of the present invention includes additional implementations, where the functions can be executed in a substantially simultaneous manner or in a reverse order according to the involved functions, rather than in the order shown or discussed, which should be understood by those skilled in the technical field of the embodiments of the present invention.

[0103] It should be understood that each part of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one of the following well-known technologies in the art or a combination thereof can be used: discrete logic circuits with logic gate circuits for implementing logical functions on data signals, application-specific integrated circuits with appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0104] Those of ordinary skill in the technical field of the present invention can understand that all or part of the steps carried by the methods of the above embodiments can be completed by instructing relevant hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiments.

[0105] In addition, each functional unit in various embodiments of the present invention can be integrated into a processing module, or each unit can exist physically alone, or two or more units can be integrated into one module. The above-mentioned integrated module can be implemented in the form of hardware or in the form of a software functional module. When the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0106] The above-mentioned storage medium can be a read-only memory, a magnetic disk, an optical disk, etc.

[0107] In the description of this specification, the descriptions referring to terms such as "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in a suitable manner in any one or more embodiments or examples.

[0108] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. An audio noise suppression method, characterized in that, Comprising: Receiving an audio input signal; Performing envelope detection on the audio input signal to obtain the amplitude of the audio signal, and recording the change in the amplitude of the audio signal and the amplitude of the silent segment; Based on the amplitude of the silent segment, performing noise amplitude tracking on the audio input signal to dynamically obtain the amplitude of the noise signal; Determining the closing threshold of the noise gate according to the amplitude of the noise signal; Judging whether the change in the amplitude of the audio signal within a preset detection time parameter is greater than a preset speech segment signal amplitude increment parameter. If it is greater, the noise gate is opened; Judging whether the change in the amplitude of the audio signal within a preset detection time parameter is continuously lower than the closing threshold. If it is continuously lower, the noise gate is closed; Outputting the current gain according to the opening and closing state of the noise gate and a preset gain attenuation parameter when closing the gate; Filtering the current gain and applying the filtered output gain to the audio input signal to obtain and output a target audio signal; The performing noise amplitude tracking on the audio input signal to dynamically obtain the amplitude of the noise signal includes: Setting the block length of audio signal processing; Based on the block length of the audio signal, dividing the audio input signal into multiple audio blocks in chronological order; Determining the signal envelope mean value of the first audio block as the current noise floor; Recording the minimum value of the envelope mean values of all audio blocks within a preset time; Updating and filtering the current noise floor based on a preset formula and the minimum value, and outputting the amplitude of the noise signal.

2. The audio noise suppression method according to claim 1, wherein Further comprising: Performing a delay process on the audio input signal that matches the preset detection time parameter.

3. The audio noise suppression method according to claim 1, characterized in that, The determining the closing threshold of the noise gate according to the amplitude of the noise signal includes: Amplifying the amplitude of the noise signal by a preset multiple and using the amplified value as the closing threshold.

4. The audio noise suppression method according to claim 1, wherein, The preset detection time parameter is 30 ms - 100 ms, and the preset speech segment signal amplitude increment parameter is 15 dB.

5. The audio noise suppression method according to claim 1, wherein The performing envelope detection on the audio input signal includes: Performing envelope detection on the audio input signal based on envelope peak power; and / or; Performing envelope detection on the audio input signal based on spectral envelope.

6. An audio noise suppression device, characterized in that, Applied to the audio noise suppression method as described in any one of claims 1 to 5, the audio noise suppression device includes: A receiving module, configured to receive an audio input signal; An envelope detection module, configured to perform envelope detection on the audio input signal to obtain the amplitude of the audio signal, and record the change in the amplitude of the audio signal and the amplitude of the silent segment; A noise tracking module, configured to perform noise amplitude tracking on the audio input signal based on the amplitude of the silent segment to dynamically obtain the amplitude of the noise signal; An opening and closing control module, configured to determine the closing threshold of the noise gate according to the amplitude of the noise signal; judge whether the change in the amplitude of the audio signal within a preset detection time parameter is greater than a preset speech segment signal amplitude increment parameter. If it is greater, the noise gate is opened; judge whether the change in the amplitude of the audio signal within a preset detection time parameter is continuously lower than the closing threshold. If it is continuously lower, the noise gate is closed; A gain calculation module, configured to output the current gain according to the opening and closing state of the noise gate and a preset gain attenuation parameter when closing the gate; An output module, configured to filter the current gain and apply the filtered output gain to the audio input signal to obtain and output a target audio signal.

7. The audio noise suppression device according to claim 6, wherein It further includes a matching delay module, configured to: Perform a delay process on the audio input signal that matches the preset detection time parameter.

8. The audio noise suppression device according to claim 6, wherein The noise tracking module is specifically configured to: Set the audio signal processing block length; Based on the audio signal processing block length, divide the audio input signal into multiple audio blocks in chronological order; Determine that the signal envelope mean value of the first audio block is the current noise floor; Record the minimum value of the envelope mean values of all audio blocks within a preset time; Based on a preset formula and the minimum value, update and filter the current noise floor, and output the amplitude of the noise signal.

9. An audio noise suppression device, characterized in that, It includes: A processor and a memory connected to the processor; The memory is used to store a computer program, and the computer program is at least used to execute the audio noise suppression method according to any one of claims 1-5; The processor is used to call and execute the computer program in the memory.

Images