Audio multi-band separation and compression method, device, computer equipment and storage medium

Through the Cascade of the Basworth filter and the delay coefficient optimization, multi-band crossover filter is formed. Combined with the gain curve processing, the distortion problem caused by the audio divider is solved, the audio multi-band separation and gain supplementation are realized, and the sound quality is improved. Loudness and editing effects.

CN114299968BActive Publication Date: 2024-11-08SHENZHEN WONDERSHARE SOFTWARE CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202111647787.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-12-30
Publication Date
2024-11-08
Estimated Expiration
2041-12-30

AI Technical Summary

Technical Problem

Existing audio dividers are prone to audio distortion when processing different frequency segments, and cannot automatically adjust the frequency according to the frequency distribution of the audio, resulting in the impact of the sound quality.

Method used

The Basworth filter cascade method is used to combine the high-pass filter and the low-pass filter into a cross filter, and a multi-band cross filter is formed through the optimization of the delay coefficient, and the time domain waveform data is obtained to convert it into loudness data, and the gain is calculated. The curve performs smooth gain, automatic gain attenuation and linear conversion, and builds a compressor, using a multi-band crossover filter for audio separation and gain supplementation.

Benefits of technology

It effectively avoids the incorrectness of audio frequency division. Through multi-band separation and compression, it solves the distortion problem during single-band operation of audio, and provides better audio editing effects.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114299968B_ABST
    Figure CN114299968B_ABST
Patent Text Reader

Abstract

The present invention discloses an audio multi-band separation and compression method, device, computer equipment and storage medium. The method includes: according to the Basworth filter cascade method, multiple high-pass filters and / or low-pass filters are cascaded into multiple cross filters of different combinations; using a delay coefficient for optimization, and combining the optimized multiple cross filters into a multi-band crossover filter; obtaining the time domain waveform data of the target audio, and converting the time domain waveform data into corresponding loudness data; calculating the gain curve corresponding to the loudness data, and performing smooth gain, automatic gain attenuation control and linear gain conversion on the gain curve in sequence, thereby constructing a compressor; using a multi-band crossover filter to perform multi-band separation on the target audio, and using a compressor to perform gain supplement on the target audio after multi-band separation. The present invention can solve the limitations of the compressor during single-band audio operation and the effect of audio distortion caused by the divider.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of computer software technology, and in particular to an audio multi-band separation and compression method, device, computer equipment and storage medium. Background Art

[0002] Music is a creative way for humans to express personal emotions. Music of different frequencies can bring different auditory experiences. Low-frequency sounds can give people a thick feeling. The rich low-frequency volume can ensure the balance of the sound and make the music more pleasant to listen to. The mid-frequency sound is the most important component in music, which basically includes all instruments and human voices and determines the performance of audio. The high-frequency sound is the most difficult part to grasp and change. Once it is not operated carefully, it will produce a hissing noise.

[0003] Audio frequency division is a commonly used music processing method. Currently, the audio frequency division operation is usually completed by combining low-pass filters and high-pass filters. Compressor is the most commonly used audio range control, which can produce different effects for audio or voice clips and is an important technology for audio post-processing.

[0004] A single frequency divider designed in the prior art can be used to retain most of the audio information in a specific frequency band and adjust the audio information of the corresponding frequency band through a compressor algorithm, but there are still many difficulties, specifically: 1. When a single frequency divider retains the frequency band, it is easy to produce audio distortion after relatively flat filtering, resulting in a relatively large impact on the sound quality of the audio after frequency division and compression; 2. The current frequency divider solution cannot automatically adjust the frequency according to the frequency distribution of the audio. Summary of the invention

[0005] The embodiments of the present invention provide a method, apparatus, computer device and storage medium for audio multi-band separation compression, aiming to solve the limitations of the compressor during audio single-band operation and the audio distortion problem caused by a general frequency divider.

[0006] In a first aspect, an embodiment of the present invention provides an audio multi-band separation and compression method, comprising:

[0007] According to the Basworth filter cascade method, multiple high-pass filters and / or low-pass filters are cascaded into multiple crossover filters of different combinations;

[0008] The plurality of crossover filters are optimized by using a delay coefficient, and the optimized plurality of crossover filters are combined into a multi-band crossover filter;

[0009] Acquire time-domain waveform data of the target audio, and convert the time-domain waveform data into corresponding loudness data;

[0010] Calculating a gain curve corresponding to the loudness data, and sequentially performing smooth gain, automatic gain attenuation control, and linear gain conversion on the gain curve, thereby constructing a compressor;

[0011] The multi-band crossover filter is used to perform multi-band separation on the target audio, and the compressor is used to perform gain supplement on the target audio after the multi-band separation.

[0012] In a second aspect, an embodiment of the present invention provides an audio multi-band separation and compression device, comprising:

[0013] A cascade unit, used for cascading a plurality of high-pass filters and / or low-pass filters into a plurality of crossover filters of different combinations according to a Basworth filter cascade method;

[0014] An optimization unit, configured to optimize the plurality of crossover filters by using a delay coefficient, and combine the optimized plurality of crossover filters into a multi-band crossover filter;

[0015] A time domain acquisition unit, used to acquire time domain waveform data of the target audio, and convert the time domain waveform data into corresponding loudness data;

[0016] A compressor construction unit, used for calculating a gain curve corresponding to the loudness data, and sequentially performing smooth gain, automatic gain attenuation control and linear gain conversion on the gain curve, thereby constructing a compressor;

[0017] The separation and gain unit is used to perform multi-band separation on the target audio using the multi-band crossover filter, and to perform gain supplement on the target audio after the multi-band separation using the compressor.

[0018] In a third aspect, an embodiment of the present invention provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the audio multi-band separation and compression method as described in the first aspect is implemented.

[0019] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, characterized in that a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the audio multi-band separation and compression method as described in the first aspect is implemented.

[0020] The embodiment of the present invention provides an audio multi-band separation and compression method, device, computer equipment and storage medium. The method comprises: according to the Basworth filter cascade method, cascading multiple high-pass filters and / or low-pass filters into multiple cross filters of different combinations; optimizing the multiple cross filters by using delay coefficients, and combining the optimized multiple cross filters into a multi-band crossover filter; acquiring time domain waveform data of target audio, and converting the time domain waveform data into corresponding loudness data; calculating a gain curve corresponding to the loudness data, and sequentially performing smooth gain, automatic gain attenuation control and linear gain conversion on the gain curve, thereby constructing a compressor; performing multi-band separation on the target audio by using the multi-band crossover filter, and performing gain supplement on the target audio after the multi-band separation by using the compressor. The embodiment of the present invention separates the audio into different frequency bands to avoid incorrect audio frequency division caused by the particularity of the input audio. At the same time, for the audio of different segments, the sound quality and loudness are compressed and adjusted through the constructed compressor to obtain a brand new audio, thereby achieving the effect of solving the limitations of the compressor during the single-band operation of the audio and the audio distortion problem caused by the general divider. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.

[0022] Figure 1 A flowchart of an audio multi-band separation and compression method provided by an embodiment of the present invention;

[0023] Figure 2 A schematic diagram of a sub-process of an audio multi-band separation and compression method provided by an embodiment of the present invention;

[0024] Figure 3 A schematic block diagram of an audio multi-band separation and compression device provided by an embodiment of the present invention;

[0025] Figure 4 A sub-schematic block diagram of an audio multi-band separation and compression device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0027] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0028] It should also be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0029] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0030] See below Figure 1 , Figure 1 A flowchart of an audio multi-band separation and compression method provided by an embodiment of the present invention specifically includes: steps S101 to S105.

[0031] S101, according to the Basworth filter cascade method, cascading multiple high-pass filters and / or low-pass filters into multiple crossover filters of different combinations;

[0032] S102, optimizing the multiple crossover filters by using a delay coefficient, and combining the optimized multiple crossover filters into a multi-band crossover filter;

[0033] S103, obtaining time domain waveform data of the target audio, and converting the time domain waveform data into corresponding loudness data;

[0034] S104, calculating a gain curve corresponding to the loudness data, and sequentially performing smooth gain, automatic gain attenuation control, and linear gain conversion on the gain curve, thereby constructing a compressor;

[0035] S105: Use the multi-band crossover filter to perform multi-band separation on the target audio, and use the compressor to perform gain supplement on the target audio after the multi-band separation.

[0036] In this embodiment, firstly, a plurality of crossover filters are combined according to the Basworth filter cascade method, and further optimized as a multi-band crossover filter, thereby completing the setting of the audio frequency divider part. Then, the loudness data of the target audio is used to calculate the gain curve, and the gain curve is smoothed, automatically subjected to gain attenuation control, and subjected to linear gain conversion, thereby completing the construction of the audio compressor part. Subsequently, the target audio is separated and compressed by the frequency divider and the compressor, so as to obtain the desired music or complete the desired music editing, etc.

[0037] This embodiment separates the audio into different frequency bands to avoid incorrect audio frequency division caused by the particularity of the input audio. At the same time, for the audio of different segments, the sound quality and loudness are compressed and adjusted through the constructed compressor to obtain a brand new audio, thereby achieving the effect of solving the limitations of the compressor when the audio is operated in a single frequency band and the audio distortion problem caused by the general frequency divider. In addition, the method provided by this embodiment can help users perform audio frequency division processing, perform different compressor operations on the data in different frequency bands, so that the edited audio can be used in audio and video editing, and the operation is simple and fast, giving users the convenience of quickly completing audio frequency division compression, and also providing users with self-selected parameters to facilitate adjustment to the sound quality effect they want.

[0038] In one embodiment, the step S101 includes:

[0039] The plurality of high-pass filters and low-pass filters are combined in a low-high-low, low-high-low-high, and low-high combination mode to obtain a low-high-low combination crossover filter, a low-high-low-high combination crossover filter, and a low-high combination crossover filter;

[0040] Acquire audio signals within different cutoff frequencies using the low-high-low combined crossover filter and the low-high-low-high combined crossover filter;

[0041] The high-frequency information and low-frequency information of the target audio are filtered using the low-high combined crossover filter.

[0042] In this embodiment, the Basworth filter cascade method is used to cascade two high-pass filters or two low-pass filters into a group of cross filters, and multiple cross filter groups pre-combined and set such as low-high-low-low-high-low-high are used for different audio filtering situations. Among them, the low-high combination cross filter can be used for audio bands around 20Hz-8kHz such as voice to filter high-frequency information and low-frequency information in the audio, so that the voice information is kept intact to the greatest extent possible and reduce noise interference; the low-high-low combination cross filter and the low-high-low-high combination filter are actually in the form of two parallel cross filters, which filter the audio with the low-high filter and another group of filters as the core, and the purpose is to obtain the audio signal within each cutoff frequency through different cutoff frequencies.

[0043] In one embodiment, the step S102 includes:

[0044] The filter delay is set at the frequency of each crossover filter, and the high-pass and low-pass filters in the crossover filter are used to supplement the low-pass filter.

[0045] In the present embodiment, for each crossover filter obtained by the combination, the delay coefficient is adopted to optimize to produce a dual-band crossover filter, and the specific reasons are as follows: the Basworth filter has a stable amplitude-frequency characteristic inside and outside the passband, but has a long transition band, which is easy to cause distortion on the transition band, so the frequency on the crossover filter has a certain filter delay, that is, when the cycle (frequency and cycle are reciprocal to each other) meets a point, a common low-pass filter and a high-low pass filter will be opened at the same time, that is, the transition band information originally required to be obtained by the common low-pass filter can be supplemented by the high-low pass filter to reduce the phenomenon that the transition band of the Basworth filter itself is easily distorted. A plurality of such filters are combined with each other to obtain a multi-band crossover filter. The multi-band crossover filter can divide the spectrum into a plurality of frequency bands, and make their sum a relatively perfect all-pass filter.

[0046] The purpose of designing a multi-band crossover filter is to achieve audio frequency division, and the sum is an all-pass filter. The all-pass filter is equivalent to not cutting off any input audio signal, but because the compressor algorithm needs to be used to change the mixing operation of the information of each frequency band, a multi-band crossover filter is needed to implement this process, and the results of each audio frequency band obtained by the multi-band crossover filter can be equal to the all-pass effect. This ensures that the audio signal is not lost and completes the task of compressing and mixing each frequency band separately. Therefore, it is equivalent to using a multi-band crossover filter to simulate an all-pass filter with special functions.

[0047] In one embodiment, the step S103 includes:

[0048] The time domain waveform data is converted into corresponding loudness data according to the following formula:

[0049] X db =20*log10(abs(data))

[0050] In the formula, data is the original data of the target audio, X db is the loudness data.

[0051] Furthermore, for the loudness data obtained by the above formula, a related gain curve is calculated, and the gain curve compares the difference between the loudness data and the loudness data compressed under different conditions.

[0052] Specifically, in one embodiment, step S103 includes:

[0053] The loudness data is compressed under different conditions according to the following formula:

[0054]

[0055] Where, X SC is the loudness data after compression, T represents the Threshold, which is used to specify the lowest loudness range processed by the compressor; R represents the Ratio, which is used to compress the loudness data exceeding the threshold with a Ratio:1; W is the Knee Width boundary range, which is used to optimize the value around the threshold;

[0056] Based on the loudness data and the compressed loudness data, the gain curve g is constructed. c :

[0057] g c =X db -X sc .

[0058] In this embodiment, different parameters are used in the gain curve, including: Threshold, which is used to specify the lowest loudness range processed by the compressor, abbreviated as T; Ratio, which is used to compress the loudness of the part exceeding the threshold in a Ratio:1 manner, abbreviated as R; Knee Width, which is used to optimize the value in the range around Threshold so that it is a curve with a relatively slow change within a range, abbreviated as W.

[0059] Further, in one embodiment, if Figure 2 As shown, the step S104 also includes: steps S201 to S203.

[0060] S201. Reduce gain jaggedness by attack time and release time according to the following formula:

[0061]

[0062] In the formula, g c is the gain curve, g S is the curve after smoothing gain, Fs is the sampling rate of the target audio, T A and T R They are attack time and release time;

[0063] S202, dividing the gain curve into a plurality of stable periods according to a preset period, and compressing the high loudness exceeding a preset loudness threshold in each stable period to complete automatic gain attenuation control;

[0064] S203, converting the gain curve after the automatic gain attenuation control into a linear domain, and then applying the linear domain result to the original data of the target audio in a multiple form, and setting the result after the effect as the output result of the compressor.

[0065] In this embodiment, when the gain curve is smoothed, the attack time and release time are used to reduce the jaggedness of the gain, so that the compression of the audio can be more natural, wherein the attack time refers to the gain supplement of the audio before the loudness decreases, so that the audio needs a certain time to reach a certain threshold instead of instantaneous triggering, and the release time is the opposite. When performing automatic gain attenuation control, the automatic gain attenuation control is to prevent the audio from being unnatural in hearing due to the short-term high loudness points that appear during the compressor processing due to the attack time and release time. For high loudness greater than a stable cycle (it is generally believed that a point greater than 6db above the set loudness is a high loudness, which is understood to be that when the sound is transmitted over a distance, the loudness will decrease by 6db for each cycle of distance), a compression is performed again, and the compression threshold is determined by the median loudness value of the two stable cycles before and after, and the compression ratio remains unchanged, and the stable cycle is set to 10ms to 30ms in this embodiment. When performing linear gain conversion, the gain curve after automatic gain attenuation control can be converted to the linear domain according to the following formula:

[0066]

[0067] Finally, the result of the linear domain is applied to the original data of the target audio in the form of multiples (that is, the corresponding original data and g lin Perform element-by-element multiplication) to obtain the audio output data dataOut after being processed by the compressor.

[0068] In one embodiment, the step S105 includes:

[0069] Using the multi-band crossover filter to separate the target audio into low-band data, mid-low-band data, mid-high-band data and high-band data;

[0070] Using a Chebyshev filter to filter noise in the low-frequency band data and the high-frequency band data respectively, and using the compressor to supplement the gain of the filtered low-frequency band data and the high-frequency band data;

[0071] The mid-low frequency band data and the mid-high frequency band data are compressed according to a preset compression threshold.

[0072] In this embodiment, the multi-band crossover filter is used to separate the target audio to obtain four audio data of different frequency bands, namely, low-frequency data, medium-low-frequency data, medium-high-frequency data and high-frequency data. The noise in the low-frequency data is filtered by a Chebyshev filter, and the gain of the compressor is supplemented to the low-frequency data to improve the average loudness of the target audio; the medium-low-frequency data and the medium-high-frequency data are compressed with a given threshold, and a longer knee width is used to make the audio signals of the medium-low-frequency data and the medium-high-frequency data have a soft texture; the Chebyshev filter is also used for the high-frequency data to filter out the noise in the high-frequency data, and the compressor is used to reduce the gain to improve the average loudness of the audio.

[0073] Furthermore, the low-frequency band data, mid-low-frequency band data, mid-high-frequency band data and high-frequency band data after the above processing are numerically superimposed to obtain the audio after multi-band separation and compression processing. In addition, in a specific embodiment, when the target audio is separated by the multi-band crossover filter, the target audio is calculated according to the box criterion to obtain the three intersection points of the multi-band crossover filter, and then the target audio is separated and cut using 4 filters (i.e., low, low-high, low-high, high) based on these three intersection points. That is, after the target audio is subjected to Fourier transformation, the overall proportion of the frequency is calculated, and the upper and lower quarters of the box are calculated through the frequency space where the overall proportion is located, so as to obtain the threshold points for dividing into low-pass and high-pass.

[0074] Figure 3 A schematic block diagram of an audio multi-band separation and compression device 300 provided in an embodiment of the present invention, the device 300 includes:

[0075] The cascade unit 301 is used to cascade multiple high-pass filters and / or low-pass filters into multiple cross filters of different combinations according to the Basworth filter cascade method;

[0076] The optimization unit 302 is used to optimize the multiple crossover filters by using the delay coefficient, and combine the optimized multiple crossover filters into a multi-band crossover filter;

[0077] A time domain acquisition unit 303 is used to acquire time domain waveform data of the target audio and convert the time domain waveform data into corresponding loudness data;

[0078] A compressor construction unit 304, configured to calculate a gain curve corresponding to the loudness data, and sequentially perform smooth gain, automatic gain attenuation control and linear gain conversion on the gain curve, thereby constructing a compressor;

[0079] The separation and gain unit 305 is used to perform multi-band separation on the target audio using the multi-band crossover filter, and to perform gain supplement on the target audio after the multi-band separation using the compressor.

[0080] In one embodiment, the cascade unit 301 includes:

[0081] A combining unit is used to combine a plurality of high-pass filters and low-pass filters in a low-high-low, low-high-low-high, and low-high combination mode to obtain a low-high-low combination crossover filter, a low-high-low-high combination crossover filter, and a low-high combination crossover filter;

[0082] A signal acquisition unit, used for acquiring audio signals within different cutoff frequencies by using the low-high-low combined crossover filter and the low-high-low-high combined crossover filter;

[0083] The information filtering unit is used to filter the high-frequency information and low-frequency information of the target audio using the low-high combined crossover filter.

[0084] In one embodiment, the optimization unit 302 includes:

[0085] The delay setting unit is used to set the filter delay at the frequency of each crossover filter, and to supplement the low-pass filter with the high-pass and low-pass filters in the crossover filter.

[0086] In one embodiment, the time domain acquisition unit 303 includes:

[0087] A data conversion unit is used to convert the time domain waveform data into corresponding loudness data according to the following formula:

[0088] X db =20*log10(abs(data))

[0089] In the formula, data is the original data of the target audio, X db is the loudness data.

[0090] In one embodiment, the compressor construction unit 304 includes:

[0091] The data compression unit is used to compress the loudness data under different conditions according to the following formula:

[0092]

[0093] Where, X SC is the loudness data after compression, T represents the Threshold, which is used to specify the lowest loudness range processed by the compressor; R represents the Ratio, which is used to compress the loudness data exceeding the threshold with a Ratio:1; W is the Knee Width boundary range, which is used to optimize the value around the threshold;

[0094] A curve construction unit, configured to construct the gain curve g based on the loudness data and the compressed loudness data. c :

[0095] g c =X db -X sc .

[0096] In one embodiment, if Figure 4 As shown, the compressor construction unit 304 also includes:

[0097] The aliasing processing unit 401 is used to reduce the gain aliasing by the attack time and the release time according to the following formula:

[0098]

[0099] In the formula, g c is the gain curve, g S is the curve after smoothing gain, Fs is the sampling rate of the target audio, T A and T R They are attack time and release time;

[0100] A cycle division unit 402, configured to divide the gain curve into a plurality of stable cycles according to a preset cycle, and to compress the high loudness exceeding a preset loudness threshold in each stable cycle to complete automatic gain attenuation control;

[0101] The result acting unit 403 is used to convert the gain curve after the automatic gain attenuation control into a linear domain, and then act on the original data of the target audio in a multiple form of the linear domain result, and set the result after the action as the output result of the compressor.

[0102] In one embodiment, the separation and gain unit 305 includes:

[0103] a data separation unit, which uses the multi-band crossover filter to separate the target audio into low-band data, mid-low-band data, mid-high-band data and high-band data;

[0104] A gain supplement unit, used to filter noise in the low-frequency band data and the high-frequency band data respectively by using a Chebyshev filter, and to perform gain supplement on the filtered low-frequency band data and the high-frequency band data by using the compressor;

[0105] The data compression unit is used to compress the medium-low frequency band data and the medium-high frequency band data according to a preset compression threshold.

[0106] Since the embodiments of the apparatus part correspond to the embodiments of the method part, please refer to the description of the embodiments of the method part for the embodiments of the apparatus part, which will not be repeated here.

[0107] The embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed, the steps provided in the above embodiment can be implemented. The storage medium may include: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0108] The embodiment of the present invention also provides a computer device, which may include a memory and a processor, wherein a computer program is stored in the memory, and when the processor calls the computer program in the memory, the steps provided in the above embodiment may be implemented. Of course, the computer device may also include various network interfaces, power supplies and other components.

[0109] The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same and similar parts between the various embodiments can be referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of this application, several improvements and modifications can be made to the present application, and these improvements and modifications also fall within the scope of protection of the claims of this application.

[0110] It should also be noted that, in this specification, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises", "comprising" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.

Claims

1. An audio multi-band separation and compression method, characterized in that: include: According to the Basworth filter cascade method, multiple high-pass filters and / or low-pass filters are cascaded into multiple crossover filters of different combinations; The plurality of crossover filters are optimized by using a delay coefficient, and the optimized plurality of crossover filters are combined into a multi-band crossover filter; Acquire time-domain waveform data of the target audio, and convert the time-domain waveform data into corresponding loudness data; Calculating a gain curve corresponding to the loudness data, and sequentially performing smooth gain, automatic gain attenuation control, and linear gain conversion on the gain curve, thereby constructing a compressor; Using the multi-band crossover filter to perform multi-band separation on the target audio, and using the compressor to perform gain supplement on the target audio after the multi-band separation; The step of calculating a gain curve corresponding to the loudness data and sequentially performing smooth gain, automatic gain attenuation control and linear gain conversion on the gain curve to construct a compressor includes: The loudness data is compressed under different conditions according to the following formula: Where, X SC is the compressed loudness data, X db is the loudness data, T represents the Threshold value, which is used to specify the lowest loudness range processed by the compressor; R represents the Ratio ratio, which is used to compress the loudness data exceeding the threshold in a Ratio:1 manner; W is the Knee Width boundary range, which is used to optimize the value in the range around the threshold; Based on the loudness data and the compressed loudness data, the gain curve g is constructed. c : g c =X db -X sc ; The step of calculating a gain curve corresponding to the loudness data and sequentially performing smooth gain, automatic gain attenuation control and linear gain conversion on the gain curve to construct a compressor further includes: Reduce gain sawing via attack and release times as follows: In the formula, g c is the gain curve, g S is the curve after smoothing gain, Fs is the sampling rate of the target audio, T A and T R They are attack time and release time; Dividing the gain curve into a plurality of stable periods according to a preset period, and compressing the high loudness exceeding a preset loudness threshold in each stable period to complete automatic gain attenuation control; The gain curve after the automatic gain reduction control is converted to the linear domain, and then the linear domain result is applied to the original data of the target audio in a multiple form, and the result after the application is set as the output result of the compressor.

2. The audio multi-band separation and compression method according to claim 1, characterized in that: The method of cascading a plurality of high-pass filters and / or low-pass filters into a plurality of crossover filters of different combinations according to the Basworth filter cascading method comprises: The plurality of high-pass filters and low-pass filters are combined in a low-high-low, low-high-low-high, and low-high combination mode to obtain a low-high-low combination crossover filter, a low-high-low-high combination crossover filter, and a low-high combination crossover filter; Acquire audio signals within different cutoff frequencies using the low-high-low combined crossover filter and the low-high-low-high combined crossover filter; The high-frequency information and low-frequency information of the target audio are filtered using the low-high combined crossover filter.

3. The audio multi-band separation and compression method according to claim 1, characterized in that: The method of optimizing the multiple crossover filters by using the delay coefficient and combining the optimized multiple crossover filters into a multi-band crossover filter comprises: The filter delay is set at the frequency of each crossover filter, and the high-pass and low-pass filters in the crossover filter are used to supplement the low-pass filter.

4. The audio multi-band separation and compression method according to claim 1, characterized in that: The step of acquiring the time domain waveform data of the target audio and converting the time domain waveform data into corresponding loudness data includes: The time domain waveform data is converted into corresponding loudness data according to the following formula: X db =20*log10(abs(data)) In the formula, data is the original data of the target audio, X db is the loudness data.

5. The audio multi-band separation and compression method according to claim 1, characterized in that: The method of using the multi-band crossover filter to perform multi-band separation on the target audio, and using the compressor to perform gain supplement on the target audio after the multi-band separation, comprises: Using the multi-band crossover filter to separate the target audio into low-band data, mid-low-band data, mid-high-band data and high-band data; Using a Chebyshev filter to filter noise in the low-frequency band data and the high-frequency band data respectively, and using the compressor to supplement the gain of the filtered low-frequency band data and the high-frequency band data; The mid-low frequency band data and the mid-high frequency band data are compressed according to a preset compression threshold.

6. An audio multi-band separation and compression device, characterized in that: include: A cascade unit, used for cascading a plurality of high-pass filters and / or low-pass filters into a plurality of crossover filters of different combinations according to a Basworth filter cascade method; An optimization unit, configured to optimize the plurality of crossover filters by using a delay coefficient, and combine the optimized plurality of crossover filters into a multi-band crossover filter; A time domain acquisition unit, used to acquire time domain waveform data of the target audio, and convert the time domain waveform data into corresponding loudness data; A compressor construction unit, used for calculating a gain curve corresponding to the loudness data, and sequentially performing smooth gain, automatic gain attenuation control and linear gain conversion on the gain curve, thereby constructing a compressor; A separation and gain unit, used for performing multi-band separation on the target audio using the multi-band crossover filter, and performing gain supplement on the target audio after the multi-band separation using the compressor; The compressor building unit comprises: The data compression unit is used to compress the loudness data under different conditions according to the following formula: Where, X SC is the compressed loudness data, X db is the loudness data, T represents the Threshold value, which is used to specify the lowest loudness range processed by the compressor; R represents the Ratio ratio, which is used to compress the loudness data exceeding the threshold in a Ratio:1 manner; W is the Knee Width boundary range, which is used to optimize the value in the range around the threshold; A curve construction unit, configured to construct the gain curve g based on the loudness data and the compressed loudness data. c : g c =X db -X sc ; The compressor construction unit also includes: A sawing processing unit for reducing gain sawing via attack time and release time according to the following formula: In the formula, g c is the gain curve, gS is the curve after smoothing the gain, Fs is the sampling rate of the target audio, T A and T R They are attack time and release time; A cycle division unit, used to divide the gain curve into a plurality of stable cycles according to a preset cycle, and compress the high loudness exceeding a preset loudness threshold in each stable cycle to complete automatic gain attenuation control; The result action unit is used to convert the gain curve after the automatic gain attenuation control into a linear domain, and then act on the original data of the target audio in the form of a multiple of the linear domain result, and set the result after the action as the output result of the compressor.

7. A computer device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the audio multi-band separation and compression method as claimed in any one of claims 1 to 5 when executing the computer program.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the audio multi-band separation and compression method according to any one of claims 1 to 5 is implemented.