CN101359902B - Equalizing method and system for audio signal - Google Patents
Equalizing method and system for audio signal Download PDFInfo
- Publication number
- CN101359902B CN101359902B CN2008101493937A CN200810149393A CN101359902B CN 101359902 B CN101359902 B CN 101359902B CN 2008101493937 A CN2008101493937 A CN 2008101493937A CN 200810149393 A CN200810149393 A CN 200810149393A CN 101359902 B CN101359902 B CN 101359902B
- Authority
- CN
- China
- Prior art keywords
- pass filter
- band pass
- audio
- audio signal
- gain
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000005236 sound signal Effects 0.000 title claims abstract description 89
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000001914 filtration Methods 0.000 claims abstract description 58
- 230000003321 amplification Effects 0.000 claims description 10
- 238000003199 nucleic acid amplification method Methods 0.000 claims description 10
- 230000008859 change Effects 0.000 claims description 2
- 238000013461 design Methods 0.000 description 22
- 230000006870 function Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 238000012546 transfer Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000001105 regulatory effect Effects 0.000 description 2
- 230000009711 regulatory function Effects 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000012940 design transfer Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 239000011435 rock Substances 0.000 description 1
- 230000001235 sensitizing effect Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Images
Landscapes
- Tone Control, Compression And Expansion, Limiting Amplitude (AREA)
- Circuit For Audible Band Transducer (AREA)
Abstract
The invention discloses an equilibrium method of audio signals as well as a system thereof, comprising: audio classes are classified according to the audio signal frequency domain in conformity to a psychoacoustic model, a band-pass filter is designed for each audio class, the pass band of each band-pass filter covers the corresponding frequency domain of one audio class; then the pass band of each band-pass filter is adjusted according to the corresponding audio class of each band-pass filter; and all the band-pass filters form a filter bank, the band-pass filters are selected for filtering the audio signals and then stacking the signals after gaining and amplifying the output signals of the band-pass filters. The method and the system solve the problems of obvious energy wave crest and wave trough existing in an EQ curve and reduce the amount of calculation compared with the prior art.
Description
Technical field
The present invention relates to field of audio play, relate in particular to a kind of equalization methods and system of the audio signal in portable audio player.
Background technology
In portable players such as MP3; Audio signal is carried out equilibrium can pass through equalizer (EQ) adjusting realization; Promptly regulate various frequency content signal of telecommunication amplification quantity respectively; Compensate the defective of loud speaker and sound field through the adjusting to the signal of telecommunication of various different frequencies, compensate and modify various sound sources and other special role, the EQ regulatory function can filtering realizes to audio signal through using the multistage filter combination.
The quality factor q of filter is definition like this; The passband of a known band pass filter and center frequency point; Then its bandwidth is the up and down frequency range between two cut-off frequencies of its passband; The ratio of center frequency point and bandwidth is called quality factor q, and the quality factor q of the used filter of prior art is generally between 1-2.5.
At present the equalization methods of existing audio signal is at first simply the audio signal of the 20-20khz of people's ear audible to be divided into equal several frequency bands according to frequency range; To band pass filter of each frequency band design of dividing, this band pass filter center frequency point and bandwidth fixed, Gain Adjustable; Each band pass filter is formed bank of filters; Then decoded audio input signal is carried out filtering,, realize the EQ regulatory function through calculating filter coefficient and the gain of regulating filter.If adopt the fixedly five equilibrium division of this audio signal, need audio signal be divided into 10 equal portions to many five equilibriums such as 31 equal portions at least, just can obtain regulating effect preferably; EQ adjusting as using in the computer software is just used this implementation method, has divided 31 frequency bands altogether, needs 31 band pass filters of design; Amount of calculation is very big; When being applied in the portable player, generally can be divided into 10 equal portions or 5 equal portions, amount of calculation is still bigger when being divided into 10 equal portions; And the quality factor q of each band pass filter that is designed equates that the EQ curve that causes obtaining can lower trough occur at the mid point of two five equilibrium crests, makes adjusting inhomogeneous.Therefore, adopt the EQ curve shake of the equalization methods acquisition that has audio signal now bigger, promptly have tangible energy crest and trough, need bigger amount of calculation simultaneously.
Summary of the invention
The embodiment of the invention provides a kind of equalization methods and system of audio signal, and there is the problem of tangible energy crest and trough in the EQ curve in the solution prior art.
The equalization methods of a kind of audio signal of the embodiment of the invention, said method comprises:
Adopt band pass filter to form bank of filters;
According to psychoacoustic model, said audio signal is divided into the audio class that frequency range varies in size, make the passband of each band pass filter in the said bank of filters cover corresponding said audio class;
Select the band pass filter in the said bank of filters that audio signal is carried out filtering;
After the gain amplification of output signal with selected band pass filter, carry out the signal stack again according to selected band pass filter.
Said method also comprises: according to the corresponding audio class of each said band pass filter, adjust the passband of each said band pass filter.
The passband of said each said band pass filter of adjustment further comprises:
The passband of adjusting the band pass filter of the audio class correspondence that comprises 1khz and being adjacent between the corresponding band pass filter of audio class is overlapping at most;
The passband of adjusting between each self-corresponding band pass filter of adjacent audio class is overlapping, makes according to the respective frequencies of said audio class that from the respective frequencies of the said audio class that comprises 1khz to low or high frequency change order, said passband is overlapping fewer and feweri.
Select the band pass filter in the said bank of filters that audio signal is carried out filtering, further comprise:
The gain of each said band pass filter of initialization;
According to the sample rate of audio signal, confirm band pass filter and the coefficient thereof selected; And close remaining band pass filter;
Band pass filter by said selection carries out filtering to audio signal.
The gain of said each said band pass filter of initialization further comprises:
According to the equalizer EQ curve of current requirement, from the corresponding relation of preset EQ curve and band pass filter gain, select the gain of each said band pass filter.
The said preset EQ curve and the corresponding relation of each band pass filter gain further comprise:
According to various EQ curvilinear characteristics, the gain of each said band pass filter is set in conjunction with psychoacoustic model; Perhaps
The gain of each the said band pass filter corresponding with various EQ curves is set by the user.
Said remaining band pass filter of cutting out further comprises:
The gain of remaining band pass filter is made as 0.
Said band pass filter adopts the mode of series connection to form said bank of filters; Perhaps
Said band pass filter adopts the mode of parallel connection to form said bank of filters.
The embodiment of the invention also provides a kind of equalizing system of audio signal, comprising:
Bank of filters, said bank of filters is made up of band pass filter; The passband of each said band pass filter covers the frequency range of corresponding a kind of audio class, and said audio class is to divide kind according to the audio signal frequency scope according to psychoacoustic model, and the frequency range of each audio class of division varies in size; Each band pass filter in the said bank of filters is used for audio signal is carried out filtering;
The filtering control device is used for selecting the band pass filter of said bank of filters that audio signal is carried out filtering;
Signal amplifying apparatus is used for the output signal of selected band pass filter is amplified according to the gain of selected band pass filter;
Signal superposition device is used for the signal after the said signal amplifying apparatus amplification is carried out the signal stack.
Said filtering control device comprises:
The gain initialization unit is used for equalizer EQ curve as requested, the gain of each said band pass filter of initialization;
Selection and coefficient are confirmed the unit, are used for the sample rate according to audio signal, confirm band pass filter and the coefficient thereof selected; And close remaining band pass filter;
The filtering control unit is used to control selected band pass filter audio signal is carried out filtering.
Said bank of filters adopts the mode of series connection to form by said band pass filter; Perhaps
Said bank of filters adopts the mode of parallel connection to form by said band pass filter.
Beneficial effect of the present invention is following:
The equalization methods of the audio signal that the embodiment of the invention provides is divided audio class according to the audio signal frequency scope according to psychoacoustic model; To band pass filter of audio class design of each division, the passband of each band pass filter covers the frequency range of corresponding a kind of audio class; According to the corresponding audio class of each band pass filter, adjust the passband of each band pass filter again; Each band pass filter is formed bank of filters, and the band pass filter in the selective filter group carries out filtering to audio signal, and after the gain amplification of output signal according to selected band pass filter with selected band pass filter, carries out the signal stack again.Audio signal is divided into 7 frequency bands that do not wait according to psychoacoustic model from the angle of music and people's impression; And have overlapping between the nearby frequency bands; So the Q value of each band pass filter of design is equal entirely as stated above, and the passband between each self-corresponding band pass filter of adjacent audio class have overlapping, therefore; Adopt the equalization methods and the system of audio signal provided by the invention, solved the problem that EQ curve in the prior art exists tangible energy crest and trough; And the present invention is based on psychoacoustic model and analyze the EQ curvilinear characteristic; Confirm the correspondence gain of filter; Realization is done lifting, decay to certain or a few kinds of corresponding frequencies of audio class; Simultaneously in filtering, confirm the band pass filter selected, and close remaining band pass filter, can reduce amount of calculation according to the sample rate of audio signal.
Description of drawings
Fig. 1 is the design flow diagram of non-group such as device such as value filtering such as the Q of grade in the embodiment of the invention;
The equalization methods flow chart of the audio signal that Fig. 2 provides for the embodiment of the invention;
The equalizing system structural representation of the audio signal that Fig. 3 provides for the embodiment of the invention;
The filtering control device structural representation of the equalizing system of the audio signal that Fig. 4 provides for the embodiment of the invention.
Embodiment
The embodiment of the invention provides a kind of equalization methods and system of audio signal; Divide audio class according to the audio signal frequency scope according to psychoacoustic model; Be used to distinguish the frequency range of audio signal; To band pass filter of audio class design of each division, the passband of each band pass filter covers the frequency range of corresponding a kind of audio class; According to the corresponding audio class of each band pass filter, adjust the passband of each band pass filter again; Each band pass filter is formed bank of filters, and the band pass filter in the selective filter group carries out filtering to decoded audio signal, and after the gain amplification of output signal according to selected band pass filter with selected band pass filter, carries out the signal stack again.Adopt the equalization methods of this audio signal can obtain level and smooth EQ curve, avoid occurring tangible energy crest and trough, and amount of calculation is little.
Below in conjunction with accompanying drawing said method of the present invention and system are described in detail.
The present invention implements to realize through following steps:
A kind of equalization methods of audio signal, said method comprises:
Adopt band pass filter to form bank of filters;
According to psychoacoustic model, said audio signal is divided into the audio class of different frequency scope, make the passband of each band pass filter in the said bank of filters cover corresponding said audio class;
Select the band pass filter in the said bank of filters that audio signal is carried out filtering;
After the gain amplification of output signal with selected band pass filter, carry out the signal stack again according to selected band pass filter.
Followingly further specify with reference to accompanying drawing.The embodiment of the invention provides a kind of equalization methods of audio signal, and the design cycle of the non-group of using in the equalization methods of this audio signal such as device such as value filtering such as the Q of grade is as shown in Figure 1, comprising:
Step S101: divide audio class according to psychoacoustic model according to the audio signal frequency scope;
Step S102: to band pass filter of each audio class design;
Step S103: adjust the passband of each band pass filter, confirm center frequency point;
Step S104: each band pass filter is formed bank of filters.
Be described in detail in the face of each step down:
Step S101: divide audio class according to psychoacoustic model according to the audio signal frequency scope.
The embodiment of the invention is according to psychoacoustic model; Promptly divide audio signal from the angle of music and people's impression; For example; By frequency range the audio signal of the 20-20khz of people's ear audible is divided into 7 audio class according to the characteristics of sound from low to high: ultralow frequency audio class, low frequency audio class, medium and low frequency audio class, intermediate frequency audio class, medium-high frequency audio class, high-frequency audio class, hyperfrequency audio class; The corresponding frequency range of concrete each audio class is not fixed; For example; Can be: ultralow frequency audio class [20--100] hz, low frequency audio class [60--500] hz, medium and low frequency audio class [200--1000] hz, intermediate frequency audio class [500--2500] hz, medium-high frequency audio class [2000--8000] hz, high-frequency audio class [4500--14000] hz, hyperfrequency audio class [14000--22000] hz, the corresponding frequency range of adjacent audio class has overlapping part.
Step S102: to band pass filter of each audio class design.
On the basis that above-mentioned audio signal is divided; Band pass filter of audio class design to each division; The passband of each band pass filter covers the frequency range of its corresponding audio class; For example; According to the results of above-mentioned audio signal, the passband of the filter that each audio class is corresponding can be: ultralow frequency audio class [20--100] hz, low frequency audio class [60--500] hz, medium and low frequency audio class [200--1000] hz, intermediate frequency audio class [500--2500] hz, medium-high frequency audio class [2000--8000] hz, high-frequency audio class [4500--14000] hz, hyperfrequency audio class [14000--22000] hz.
Step S103: adjust the passband of each band pass filter, confirm center frequency point.
The inventor finds in practice process; In order to reach rounding off between each classification of EQ curve sound intermediate frequency signal, avoid occurring unnecessary shake, it is certain overlapping to let each self-corresponding band pass filter passband of adjacent audio class have; Overlapping composition is many more, and the trough of appearance is just not obvious more.Therefore, the embodiment of the invention on the basis of above-mentioned Design of Bandpass, from transition between class and type between the passband of each band pass filter of trough angle adjustment, promptly adjust the overlapping of passband between each self-corresponding band pass filter of adjacent audio class.Specifically adjust according to human auditory's loudness sensitiveness.The a certain frequency sound sensitive property that human auditory's loudness sensitiveness shows as about 1khz is the highest, and more past low frequency sensitiveness is low more, and more past high frequency sensitiveness is also low more; And because in the zone of sensitivity; It is just obvious more that the influence of trough appears in the EQ curve, thus design in the zone of sensitivity, the lap of passband is maximum between each self-corresponding band pass filter of adjacent audio class; Inferior sensitizing range; Lap just lacks, and therefore, the passband of the band pass filter that the band pass filter of adjustment intermediate frequency audio class correspondence is corresponding with being adjacent audio class is overlapping at most; From the intermediate frequency to the low frequency or the order of high frequency, it is overlapping few more to adjust between each self-corresponding band pass filter of adjacent audio class passband according to the respective frequencies of each audio class.
According to the passband of each band pass filter of last design, confirm each band pass filter center frequency point and quality factor q respectively, the embodiment of the invention provides a kind of design of non-group such as device such as value filtering such as the Q of grade according to said method; As shown in table 1; It is thus clear that the quality factor q of each filter of the method design of employing present embodiment is not equate entirely that the passband between the adjacent filter has overlapping, and overlapping at most at the corresponding band pass filter of intermediate frequency class; To low frequency or high frequency direction, overlapping few more.
Table 1: the passband of each band pass filter, center frequency point and quality factor q
| Filter | Passband (unit: Hz) | Center frequency point (unit: Hz) | Quality factor |
| 1 | [40.0，160.0] | 80 | 0.67 |
| 2 | [66.0，606.3] | 200 | 0.37 |
| 3 | [92.5，2575.7] | 488 | 0.196 |
| 4 | [323.3，2970.8] | 980 | 0.37 |
| 5 | [2000.0，8000.0] | 4000 | 0.67 |
| 6 | [4594.8，13928.8] | 8000 | 0.857 |
| 7 | [11313.7，22627.4] | 16000 | 1.41 |
In the concrete design of filter, the present invention does not stress concrete filter form, only needs filter to have bandpass characteristics; Meet the passband requirement of design, for example, use 2 rank infinitely to respond filtering (IIR) filter; This filter has bandpass characteristics, and the design transfer function is:
The alpha of filter, beta, gamma have determined the passband and the center frequency point of this filter.
Step S104: each band pass filter is formed bank of filters.
Each band pass filter of above-mentioned design adopts the mode of series connection to form bank of filters, also can adopt the mode of parallel connection to form bank of filters.
Adopt the mode of series connection to form bank of filters in the present embodiment, decoded audio signal is carried out filtering by above-mentioned bank of filters.
The equalization methods flow chart of the audio signal that provides for the embodiment of the invention shown in Figure 2 comprises:
Step S201:, design non-group such as device such as value filtering such as the Q of grade according to psychoacoustic model;
Step S202: EQ curve as requested, the gain of each band pass filter of initialization;
Step S203:, confirm the band pass filter and the coefficient thereof of selection, and close remaining band pass filter according to the sample rate of audio signal;
Step S204: the band pass filter by selecting carries out filtering to decoded audio signal, after the output signal of selected band pass filter is amplified according to its gain, carries out the signal stack again.
Be described in detail in the face of each step down:
Step S201:, design non-group such as device such as value filtering such as the Q of grade according to psychoacoustic model.
Adopt the non-Q value filtering device group that waits of said method design.
Step S202: EQ curve as requested, the gain of each band pass filter of initialization.
The corresponding relation of various EQ curves and each band pass filter gain is set in the embodiment of the invention at first in advance; When audio signal is carried out filtering; According to the EQ curve of current requirement, from the corresponding relation of preset EQ curve and band pass filter gain, select the gain of each band pass filter.
For a kind of EQ curve, its characteristic shows as does lifting to some frequency band audio signal, and some frequency band audio signal is done decay, and the amplitude of lifting and decay is respectively much.In conjunction with psychoacoustic model; Can know that it mainly is that the audio signal of the corresponding frequency range of some audio class in 7 audio class is done lifting; The audio signal of the frequency range that some audio class is corresponding is done decay; The amplitude that promotes and decay is respectively much, can also judge that the audio signal of the frequency range which audio class is corresponding need not done Filtering Processing, and band pass filter gain that then can this audio class is corresponding be made as 0.Therefore; The embodiment of the invention is when being provided with the corresponding relation of various EQ curves and each band pass filter gain in advance; The gain of respective filter at first is set according to these judgements; According to the difference of the EQ curve of actual EQ curve that obtains and requirement, trickle adjustment is done in the gain of respective filter then, thereby obtained the gain of each band pass filter corresponding with each EQ curve; Perhaps rule of thumb be worth the gain that preestablishes each band pass filter corresponding with each EQ curve by the user.
Present embodiment can be provided with the gain of many each band pass filters of group in advance, corresponding to multiple specific EQ curve.Table 2 is depicted as in the embodiment of the invention yield value of each band pass filter of design as stated above corresponding to two kinds of specific EQ curves.
Table 2: the gain setting of the corresponding specific EQ curve of each band pass filter
| ? | Filter 1 | Filter 2 | Filter 3 | Filter 4 | Filter 5 | Filter 6 | Filter 7 |
| The supper bass curve | 5.0 | 3.0 | 1.0 | 0 | 0 | 0 | 0 |
| The rock and roll curve | 5.0 | 2.0 | 0 | 0 | 3.0 | 5.0 | 0 |
Step S203:, confirm the band pass filter and the coefficient thereof of selection, and close remaining band pass filter according to the sample rate of audio signal.
In the current audio application; The sample rate of audio signal mainly contains 9 kinds; Be respectively: 48kHz, 44.1kHz, 32kHz, 24kHz, 22.05kHz, 16kHz, 12kHz, 11.025kHz and 8kHz, the present invention can confirm the band pass filter and the coefficient thereof of selection according to these 9 kinds of sample rates.
Because sample rate should be the twice of audio signal highest frequency at least; So no matter what characteristic the EQ curve that requires has; The band pass filter that free transmission range exceeds the signal highest frequency is inoperative for the filtering of this signal, considers the amount of calculation when reducing signal filtering calculates, so select respective band pass filters; And close remaining band pass filter, the band pass filter that concrete corresponding different sample rate is selected is as shown in table 3.
Table 3: the filter that corresponding different sample rates is selected
| Sample rate (kHz) | Filter 1 | Filter 2 | Filter 3 | Filter 4 | Filter 5 | Filter 6 | Filter 7 |
| 48 | Launch | Launch | Launch | Launch | Launch | Launch | Launch |
| 44.1 | Launch | Launch | Launch | Launch | Launch | Launch | Launch |
| 32 | Launch | Launch | Launch | Launch | Launch | Launch | ? |
| 24 | Launch | Launch | Launch | Launch | Launch | ? | ? |
| 22.5 | Launch | Launch | Launch | Launch | Launch | ? | ? |
| 16 | Launch | Launch | Launch | Launch | Launch | ? | ? |
| 12 | Launch | Launch | Launch | Launch | ? | ? | ? |
| 11.025 | Launch | Launch | Launch | Launch | ? | ? | ? |
| 8 | Launch | Launch | Launch | Launch | ? | ? | ? |
Through the gain that remaining band pass filter is set is that remaining band pass filter is closed in 0 realization, reaches the purpose that reduces amount of calculation.
The passband of known band pass filters, center frequency point and transfer function form confirm that according to the sample rate of audio signal the coefficient of this band pass filter is a prior art, does not do detailed description at this.
Step S204: the band pass filter by selecting carries out filtering to decoded audio signal, after the output signal of selected band pass filter is amplified according to its gain, carries out the signal stack again.
The embodiment of the invention is before carrying out filtering by the band pass filter of selecting to decoded audio signal; At first decoded audio signal is carried out preliminary treatment; The digital audio and video signals that is about to input 4 inputs as bank of filters that move to right, purpose are to prevent that data produce in filtering computational process to overflow.
Form bank of filters owing to adopt series system in the embodiment of the invention; And the transfer function form of each band pass filter is identical; Therefore, obtaining by each band pass filter audio input signal to be carried out filtering one by one from low to high from center frequency point through behind the pretreated audio input signal.
When corresponding each band pass filter carries out filtering calculating, comprising:
Whether the gain of at first judging this band pass filter is 0, if gain is 0, then this band pass filter has been closed in expression, does not do filtering and calculates, and carries out the filtering of next band pass filter and calculates, if gain is not 0, then this band pass filter has been launched in expression;
Carry out filtering according to the coefficient of above-mentioned definite this band pass filter then and calculate, for example, adopt the transfer function form of above-mentioned filter to calculate; If be input as x [n]; Be output as y [n], then can calculate y [n]=α (x [n]-x [n-2])+γ y [n-1]-β y [n-2], again y [n] multiply by its corresponding gain according to its transfer function; Obtain product, as the output of this band pass filter;
Next judge the number of times passed through Filtering Processing (gain of the band pass filter that judgement is closed be 0 also count once),, then carry out the filtering of next band pass filter and calculate that if equal 7 times, then whole Filtering Processing have been passed through in expression if less than 7 times;
The output stack of each band pass filter that at last aforementioned calculation is obtained is as the output of this bank of filters.
Owing to before audio input signal has been carried out preliminary treatment, therefore after the output that obtains bank of filters, should numeral output move to left 4, obtain audio signal through the EQ processing.
Among another embodiment, each filter can be the parallel connection relation, and promptly each filter can carry out Filtering Processing simultaneously.
Based on same inventive concept, the equalization methods of the audio signal that provides according to the above embodiment of the present invention, correspondingly, the embodiment of the invention also provides a kind of equalizing system of audio signal, and the system configuration sketch map is as shown in Figure 3, and is specific as follows,
Bank of filters S301, bank of filters S301 is made up of band pass filter; The passband of each band pass filter covers the frequency range of corresponding a kind of audio class, and this audio class is to divide kind according to the audio signal frequency scope according to psychoacoustic model; Each band pass filter among the bank of filters S301 is used for decoded audio signal is carried out filtering.
Filtering control device S302, the band pass filter that is used for selective filter group S301 carries out filtering to decoded audio signal;
Signal amplifying apparatus S303 is used for the output signal of selected band pass filter is amplified according to the gain of selected band pass filter;
Signal superposition device S304 is used for the signal after the said signal amplifying apparatus S303 amplification is carried out the signal stack.
Above-mentioned bank of filters S301 can adopt the mode of series connection to form by band pass filter; Perhaps adopt the mode of parallel connection to form by band pass filter.
Above-mentioned filtering control device S302 structure is as shown in Figure 4, comprising:
Gain initialization unit S401 is used for EQ curve as requested, the gain of each band pass filter of initialization;
Selection and coefficient are confirmed cell S 402, are used for the sample rate according to audio signal, confirm band pass filter and the coefficient thereof selected; And close remaining band pass filter;
Filtering control unit S403 is used to control selected band pass filter decoded audio signal is carried out filtering.
In sum; The scheme that the embodiment of the invention provides; Divide audio class according to psychoacoustic model, to the audio class designing filter group of dividing, the band pass filter in the selective filter group carries out filtering to decoded audio signal; And after the gain amplification of output signal with selected band pass filter, carry out the signal stack again according to selected band pass filter.Solved that there is the problem of tangible energy crest and trough in the EQ curve in the prior art, and can reduce amount of calculation.
Obviously, those skilled in the art can carry out various changes and modification to the present invention and not break away from the spirit and scope of the present invention.Like this, belong within the scope of claim of the present invention and equivalent technologies thereof if of the present invention these are revised with modification, then the present invention also is intended to comprise these changes and modification interior.
Claims (11)
1. the equalization methods of an audio signal is characterized in that, said method comprises:
Adopt band pass filter to form bank of filters;
According to psychoacoustic model, said audio signal is divided into the audio class that frequency range varies in size, make the passband of each band pass filter in the said bank of filters cover corresponding said audio class;
Select the band pass filter in the said bank of filters that audio signal is carried out filtering;
After the gain amplification of output signal with selected band pass filter, carry out the signal stack again according to selected band pass filter.
2. the method for claim 1 is characterized in that, said method also comprises: according to the corresponding audio class of each said band pass filter, adjust the passband of each said band pass filter.
3. method as claimed in claim 2 is characterized in that, the passband of said each said band pass filter of adjustment further comprises:
The passband of adjusting the band pass filter of the audio class correspondence that comprises 1khz and being adjacent between the corresponding band pass filter of audio class is overlapping at most;
The passband of adjusting between each self-corresponding band pass filter of adjacent audio class is overlapping, makes according to the respective frequencies of said audio class that from the respective frequencies of the said audio class that comprises 1khz to low or high frequency change order, said passband is overlapping fewer and feweri.
4. the method for claim 1 is characterized in that, selects the band pass filter in the said bank of filters that audio signal is carried out filtering, further comprises:
The gain of each said band pass filter of initialization;
According to the sample rate of audio signal, confirm band pass filter and the coefficient thereof selected; And close remaining band pass filter;
Band pass filter by said selection carries out filtering to audio signal.
5. method as claimed in claim 4 is characterized in that, the gain of said each said band pass filter of initialization specifically comprises:
According to the equalizer EQ curve of current requirement, from the corresponding relation of preset EQ curve and band pass filter gain, select the gain of each said band pass filter.
6. method as claimed in claim 5 is characterized in that, the said preset EQ curve and the corresponding relation of each band pass filter gain further comprise:
According to various EQ curvilinear characteristics, the gain of each said band pass filter is set in conjunction with psychoacoustic model; Perhaps
The gain of each the said band pass filter corresponding with various EQ curves is set by the user.
7. method as claimed in claim 4 is characterized in that, said remaining band pass filter of cutting out further comprises:
The gain of remaining band pass filter is made as 0.
8. like the arbitrary described method of claim 1-7, it is characterized in that said band pass filter adopts the mode of series connection to form said bank of filters; Perhaps
Said band pass filter adopts the mode of parallel connection to form said bank of filters.
9. the equalizing system of an audio signal is characterized in that, comprising:
Bank of filters, said bank of filters is made up of band pass filter; The passband of each said band pass filter covers the frequency range of corresponding a kind of audio class, and said audio class is to divide kind according to the audio signal frequency scope according to psychoacoustic model, and the frequency range of each audio class of division varies in size; Each band pass filter in the said bank of filters is used for audio signal is carried out filtering;
The filtering control device is used for selecting the band pass filter of said bank of filters that audio signal is carried out filtering;
Signal amplifying apparatus is used for the output signal of selected band pass filter is amplified according to the gain of selected band pass filter;
Signal superposition device is used for the signal after the said signal amplifying apparatus amplification is carried out the signal stack.
10. system as claimed in claim 9 is characterized in that, said filtering control device comprises:
The gain initialization unit is used for equalizer EQ curve as requested, the gain of each said band pass filter of initialization;
Selection and coefficient are confirmed the unit, are used for the sample rate according to audio signal, confirm band pass filter and the coefficient thereof selected; And close remaining band pass filter;
The filtering control unit is used to control selected band pass filter audio signal is carried out filtering.
11., it is characterized in that said bank of filters adopts the mode of series connection to form by said band pass filter like claim 9 or 10 described systems; Perhaps
Said bank of filters adopts the mode of parallel connection to form by said band pass filter.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008101493937A CN101359902B (en) | 2008-09-25 | 2008-09-25 | Equalizing method and system for audio signal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008101493937A CN101359902B (en) | 2008-09-25 | 2008-09-25 | Equalizing method and system for audio signal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101359902A CN101359902A (en) | 2009-02-04 |
| CN101359902B true CN101359902B (en) | 2012-03-21 |
Family
ID=40332259
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2008101493937A Active CN101359902B (en) | 2008-09-25 | 2008-09-25 | Equalizing method and system for audio signal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101359902B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106024008A (en) * | 2013-04-05 | 2016-10-12 | 杜比实验室特许公司 | Companding apparatus and method to reduce quantization noise using advanced spectral extension |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103457559B (en) * | 2013-08-26 | 2017-07-28 | Tcl通力电子（惠州）有限公司 | Audio adjusting method and system |
| CN105550391B (en) * | 2015-11-16 | 2019-01-04 | 深圳市冠旭电子股份有限公司 | A kind of EQ curve generation method and device |
| CN108204857B (en) * | 2016-12-17 | 2020-11-03 | 中国石油天然气股份有限公司 | Vibration signal detection system and assembly method thereof |
| CN106601268B (en) * | 2016-12-26 | 2020-11-27 | 腾讯音乐娱乐（深圳）有限公司 | Multimedia data processing method and device |
| CN109819376B (en) * | 2019-01-21 | 2020-11-06 | 北京小唱科技有限公司 | Method and device for dynamically and balancedly adjusting audio frequency |
| CN115567844B (en) * | 2022-09-28 | 2024-05-03 | 浙江越扬电子有限公司 | Self-adaptive equalizer compensation method and system for vehicle-mounted sound equipment |
-
2008
- 2008-09-25 CN CN2008101493937A patent/CN101359902B/en active Active
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106024008A (en) * | 2013-04-05 | 2016-10-12 | 杜比实验室特许公司 | Companding apparatus and method to reduce quantization noise using advanced spectral extension |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101359902A (en) | 2009-02-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101359902B (en) | Equalizing method and system for audio signal | |
| CN102017402B (en) | System for adjusting perceived loudness of audio signals | |
| CN102414746B (en) | Adjusting the loudness of an audio signal with perceived spectral balance preservation | |
| CN100583856C (en) | Equalizer bank with interference suppression | |
| US9431982B1 (en) | Loudness learning and balancing system | |
| CN102378085B (en) | Equal-loudness contour-based loudness compensation method and device, and audio processing system | |
| US8428276B2 (en) | Frequency-warped audio equalizer | |
| CN106020765B (en) | A kind of effect adjusting method and device | |
| US20170207762A1 (en) | Correction of unknown audio content | |
| AU2014340178A1 (en) | System and method for digital signal processing | |
| EP3061091A1 (en) | System and method for digital signal processing | |
| CN104410379A (en) | A volume adjusting method | |
| EP2172930A1 (en) | Audio signal processing device and audio signal processing method | |
| CN204948329U (en) | Unit four move iron earphone frequency dividing circuit | |
| CN101183859A (en) | Audio compressor | |
| CN102915741A (en) | Equal loudness contour based method for automatically recovering tone of voice signal according to volume adjustment | |
| CN101577848A (en) | Supper bass boosting method and system | |
| CN104869503A (en) | Equal loudness curve-based dynamic low-frequency enhancement method and system | |
| CN104900238A (en) | Audio real-time comparison method based on sensing filtering | |
| EP3258704B1 (en) | Determination method and apparatus for pre-set set parameters of audio equalizer (aeq) | |
| WO2011163642A2 (en) | Method and device for optimizing audio quality | |
| CN106713794A (en) | Method for adjusting audio equalizing and audio system for providing equalizing adjustment | |
| CN104363510A (en) | Terminal | |
| CN115691543A (en) | Adaptive equalization method and system for acoustic system | |
| CN111181516A (en) | A sound equalization method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |