阅读说明：本技术 音频信号处理方法及设备 (Audio signal processing method and device ) 是由 M·德维科 于 2019-10-22 设计创作，主要内容包括：本发明涉及用于音频链中的音频信号处理的方法和设备/装置,该方法和装置通过以受控方式在音频链中添加非线性,来校正音频链中电声换能器的非线性,也考虑人耳的非线性心理声学特性,以便通过在某些范围内使用二次和五次多项式函数的近似来在再现声音时获得更好的声学图像和更多细节。根据本发明,该方法包括用非线性五次多项式函数近似人耳的心理声学特性,并在音频链中至少一个电声换能器的前面添加至少一个非线性元件(4),从而非线性元件(4)具有在音频链中添加非线性的功能,该非线性校正至少一个电声换能器的非线性和/或针对高达p-Δ的人耳的压力改变的人耳的近似心理声学特性的非线性。本发明的音频信号处理装置(19)包括音频链中的至少一个非线性元件(4),其具有向音频链添加非线性的功能,该非线性校正至少一个电声换能器的非线性和/或针对高达p-Δ的人耳的压力改变的人耳的近似心理声学特性的非线性。本方法和装置(19)通过添加非线性来减少对电声换能器以及人耳的限制,该非线性最终减少整个音频链与人耳的非线性,即在音频链中添加非线性,使得音频链特性减少了人耳多项式近似对压力改变p-Δ＝±1Pa的非线性。(The present invention relates to a method and a device/arrangement for audio signal processing in an audio chain, which method and arrangement correct the non-linearity of an electroacoustic transducer in the audio chain by adding the non-linearity in the audio chain in a controlled manner, taking also into account the non-linear psychoacoustic properties of the human ear, to obtain a correction signal for the non-linearity of the electroacoustic transducer in the audio chainBetter acoustic images and more detail are obtained when reproducing sound by using approximations of quadratic and quintic polynomial functions in certain ranges. According to the invention, the method comprises approximating the psychoacoustic properties of the human ear with a non-linear quintic polynomial function and adding at least one non-linear element (4) in front of at least one electroacoustic transducer in the audio chain, whereby the non-linear element (4) has the function of adding a non-linearity in the audio chain which corrects for the non-linearity of the at least one electroacoustic transducer and/or for up to p Δ The pressure of the human ear changes, the non-linearity of the approximate psychoacoustic properties of the human ear. The audio signal processing device (19) of the invention comprises at least one non-linear element (4) in the audio chain having the function of adding to the audio chain a non-linearity correcting the non-linearity of at least one electroacoustic transducer and/or for up to p Δ The pressure of the human ear changes, the non-linearity of the approximate psychoacoustic properties of the human ear. The present method and apparatus (19) reduces the restriction on the electroacoustic transducer and the human ear by adding non-linearities that ultimately reduce the non-linearities of the entire audio chain with the human ear, i.e., adding non-linearities in the audio chain such that the audio chain characteristics reduce the pressure change p by the human ear polynomial approximation Δ A nonlinearity of ± 1 Pa.)

1. A method of processing an audio signal in an audio chain, said method correcting for non-linearity of an electroacoustic transducer in said audio chain, also taking into account non-linear psychoacoustic properties of a human ear, characterized in that said method comprises:

-approximating the non-linear psychoacoustic characteristics of the human ear by a quintic polynomial function, an

-adding at least one non-linear element (4) before at least one electroacoustic transducer in the audio chain, the non-linear element (4) having the function of adding non-linearities in the audio chain, which non-linearities correct at least one electroacoustic transducer for non-linearities and/or for up to p_ΔOf the human ear changes the non-linearity of the approximate psychoacoustic properties of the human ear.

2. The method according to claim 1, wherein the non-linear element (4) is ax + bx by application²Is used to reduce the non-linearity of the electroacoustic transducer, wherein x is the relative membrane deflection or the relative force on the membrane of the electroacoustic transducer, wherein a and b are positive constants.

3. Method according to claim 1, wherein the non-linear element (4) is reduced by the component x by application²、x³And x⁴Introducing a non-linear function of at least two times to reduce the non-linearity x-ax of the psychoacoustic properties of the human ear²-bx³-cx⁴-dx⁵Wherein the constant a ═Andto within ± 30% tolerance for each constant, where x is the relative pressure of the human ear.

4. Method according to claim 1, wherein the non-linear element (4) is obtained by applying a hyperbolic functionAndto reduce non-linearity of the psychoacoustic properties of the human ear, wherein x is the relative pressure of the human ear.

5. Method according to claim 1, wherein the non-linear element (4) is formed by applying a function x^1.5To reduce non-linearity of psychoacoustic properties of the human ear, wherein x is a relative pressure of the human ear.

6. The method according to any one of claims 1-5, wherein the method comprises the steps of:

(a) routing an input audio signal (1) to a non-isolated portion of the input audio signal (1) and at least one isolated audio signal (1);

(b) modifying at least one isolated audio signal (1) in the non-linear element (4) by adding a non-linearity;

(c) optionally amplifying/attenuating at least one isolated audio signal in an amplifier/attenuator (3) before the non-linear element (4), optionally amplifying/attenuating at least one isolated audio signal in an amplifier/attenuator (5) after the non-linear element (4), optionally filtering at least one isolated audio signal in a filter (2) before the non-linear element (4), optionally filtering at least one isolated audio signal in a filter (6) after the non-linear element (4), and obtaining at least one isolated non-linear audio signal (7); and

(d) combining the non-isolated part of the audio signal (1) and at least one isolated non-linear audio signal (7) in an adder (8) into an output audio signal (9).

7. The method according to any one of claims 1-5, wherein the method comprises the steps of:

(a) -amplifying/attenuating the input signal (15) in an adjustable preamplifier (16);

(b) -performing audio signal processing in a first device (19) by applying hyperbolic non-linearities;

(c) splitting the audio signal into two branches by frequency range in an audio frequency divider (18);

(d) processing the split audio signal in each branch in at least one second device (19) by applying a quadratic non-linearity;

(e) amplifying the power of the split audio signal in each branch in a power amplifier (20), an

(f) The audio signal of each branch is routed to an associated electroacoustic transducer (21).

8. The method according to any one of claims 1-5, wherein the method comprises the steps of:

(a) -amplifying/attenuating an input signal (15) in said adjustable preamplifier (16);

(b) -performing audio signal processing in said first means (19) by applying quadratic and hyperbolic non-linearities;

(c) amplifying the audio signal power in the power amplifier (20);

(d) splitting the audio signal into two branches by frequency range in the audio frequency divider (18); and

(e) the audio signal of each branch is routed to an associated electroacoustic transducer (21).

9. The method according to any one of the preceding claims, wherein the device (19) reduces the non-linearity of the approximated psychoacoustic features of the human ear by a factor of at least 2 and/or reduces the quadratic non-linearity of the electroacoustic transducer by a factor of at least 3.

10. The method according to any of the preceding claims, wherein the pressure of the human ear is changed by up to p_Δ＝±1Pa。

11. The method of any preceding claim 1 to 10, wherein the audio signal is processed in an analogue format.

12. The method of any preceding claim 1 to 10, wherein the audio signal is processed in a digital format.

13. A computer program adapted for execution on a processor and for performing the method steps according to any one of patent applications 1 to 12 when executed on a computer device.

14. An apparatus (19) for implementing the audio signal processing method according to claims 1-12, characterized in that the apparatus comprises a non-linear element (4), the non-linear element (4) having the function of adding a non-linearity in an audio chain, the non-linearity correcting the non-linearity of the audio chain, which corrects the non-linearity of at least one electroacoustic transducer and/or for up to p_ΔOf the human ear varies a non-linearity of an approximate psychoacoustic characteristic of the human ear.

15. The apparatus (19) of claim 14, wherein the apparatus (19) further comprises an adder (8) and an input signal manifold, the input signal manifold routing the input audio signal (1) to the non-isolated part of the audio signal (1) and the at least one isolated audio signal (1), the at least one isolated audio signal (1) being processed by using the non-linear element (4) in the at least one isolated non-linear audio signal (7), and the non-isolated part of the audio signal (1) and the at least one isolated non-linear audio signal (7) being combined/merged into the processed output audio signal (9) in the adder (8).

16. The device (19) according to claim 15, wherein the device (19) comprises a selectable amplifier/attenuator (3) before the non-linear element (4) and a selectable amplifier/attenuator (5) after the non-linear element (4), a selectable filter (2) before the non-linear element (4) and a selectable filter (6) after the non-linear element (4).

17. An audio chain configured to correct non-linearities of an electroacoustic transducer, also taking into account non-linear psychoacoustic characteristics of the human ear, the audio chain comprising at least one means (19) for performing the audio signal processing method according to claims 1-12, characterized in that the audio chain comprises: an adjustable preamplifier (16) of the audio signal (15) connected to a first means (19) for audio signal processing by using hyperbolic non-linearity; an audio frequency divider (18) connected to said first means (19) and dividing the processed audio signal in said first means (19) into several signal branches by frequency range, at least two second means (19) for audio signal processing using quadratic non-linearity being connected to said audio frequency divider (18), each of at least two other said second means (19) being connected to a corresponding power amplifier (20), and at least two electroacoustic transducers (21) connected to said corresponding power amplifier (20).

18. Audio chain according to claim 17, wherein the inverse function of the approximation of the psycho-acoustic properties of the human ear in the device (19) is derived by using non-linearities of electronic components such as diodes, transistors, vacuum tubes.

19. The audio chain according to any one of claims 16 to 18, wherein the device (19) reduces a non-linearity of the approximated psychoacoustic features of the human ear by a factor of at least 2 and/or reduces a quadratic non-linearity of the electroacoustic transducer by a factor of at least 3.

20. The audio chain of any of claims 17 to 19, wherein the pressure change of the human ear is up to p_Δ＝±1Pa。

Technical Field

The present invention relates to an audio signal processing method for enhancing the quality and/or other characteristics of an audio signal. The method corrects for non-linearity of an electroacoustic transducer in an audio chain by taking into account non-linear psychoacoustic properties of the ear by adding non-linearity in the audio chain in a controlled manner. Furthermore, the invention relates to a device/apparatus for implementing the method and audio chain, which is configured to correct the non-linearity of the electroacoustic transducer, also taking into account the non-linear psychoacoustic properties of the human ear. The audio chain comprises at least one means for implementing the audio signal processing method.

Background

Technical problem

Today, the audio chain before the electroacoustic transducer shows no critical features. It is not known why some audio chain components with larger distortion produce better sound than components with lower distortion. Some amplifiers incorporate a vacuum tube to better sound, while others employ a small feedback loop to enhance the non-linearity of the (intensify) component. Distortion of the audio chain in front of the electroacoustic transducer does not mean that it sounds better or worse. Two different electro-acoustic transducers sound good on their audio chain, but sound worse when they swap positions. One reason for this is that the audio chain before the electroacoustic transducer has non-linearity, which reduces its non-linearity, making it sound better than on other audio chains.

The technical problem to be solved by the present invention is a method and a device for audio signal processing in an audio chain, which corrects the non-linearity of an electroacoustic transducer in the audio chain, also taking into account the non-linear psychoacoustic characteristics of the human ear.

Prior Art

The non-linearity of electroacoustic transducers has been known for some time. The nonlinear distortion characterizes the entire electroacoustic reproduction chain, from the recording process on the recording medium up to the sound reproduction of the recording medium, the amplifier and the loudspeaker itself. There are many publications that record these nonlinearities. It has also long been known to apply non-linearities in musical instruments to alter sound. Some nonlinearities in the sound are not perceived by humans, while others are perceived, even though they have the same acoustic energy as described in the article written by Jean Hiraga, "Amplifier music-A Study of Amplifier Harmonic Distortion Spectroscopy" (Amplifier music-A Study). Document US5133015 discloses a process and an apparatus for audio signal processing, more precisely a technique allowing various levels of audio signal distortion, including audio signal distortion to a certain level. Document US2011255701 discloses an electronic circuit and an audio enhancement method, in particular an electronic circuit that can introduce predictable and controllable harmonic distortion that increases with increasing signal amplitude. Document US2015249889 discloses a system and method for digital audio signal processing by extending the loudspeaker frequency response and reducing or eliminating non-linear loudspeaker distortion. Based on the modified loudspeaker frequency response, the audio signal may be expanded by applying a digital linear filter. Based on the inverse parametric model of the loudspeaker, the digital nonlinear filter may eliminate or reduce nonlinear distortion of the loudspeaker.

Most known conventional methods related to audio signal processing that aim at enhancing the quality and/or other characteristics of the audio signal do not also take into account the non-linear psychoacoustic characteristics of the ear.

Disclosure of Invention

The present invention relates to an audio signal processing method and apparatus in an audio chain, which corrects nonlinearity of an electroacoustic transducer in the audio chain by adding nonlinearity to the audio chain in a controlled manner, also taking into account nonlinear psychoacoustic characteristics of the human ear, in order to obtain better acoustic images and more details when reproducing sound by using quadratic (quadratic) approximation and a certain range of quintic polynomial functions.

According to the invention, a method comprises approximating non-linear psychoacoustic characteristics of a human ear by a fifth order polynomial and adding at least one non-linear element before at least one electroacoustic transducer in an audio chain, whereby the non-linear element has the function of adding non-linearity in the audio chain that corrects for non-linearity of the at least one electroacoustic transducer and/or for up to p_ΔThe pressure of the human ear changes, the non-linearity of the approximate psychoacoustic properties of the human ear.

According to the invention, the audio chain for implementing the audio signal processing method is configured to correct the non-linearity of the electroacoustic transducer in the audio chain, also taking into account the non-linear psychoacoustic properties of the human ear. The audio chain comprises at least one means for implementing the audio signal processing method. The aforementioned device has the function of adding to the audio chain a non-linearity correcting for non-linearity of at least one electroacoustic transducer and/or for up to p_ΔThe pressure of the human ear changes, the non-linearity of the approximate psychoacoustic properties of the human ear.

The method, apparatus and audio chain of the present invention reduce the restriction on the electroacoustic transducer and the human ear by adding non-linearitiesThe non-linearity ultimately reduces the overall audio chain from non-linearity of the human ear, i.e., adds non-linearity to the audio chain such that the audio chain characteristics reduce the change p of the human ear polynomial approximation to pressure_ΔA nonlinearity of ± 1 Pa.

Drawings

The invention is described in detail below with reference to the attached drawing figures, wherein:

FIG. 1a is a hyperbolic function with an asymptoteA diagram of;

FIG. 1b shows the harmonic spectrum of a distorted sinusoidal signal with amplitude 0.57 of the function shown in FIG. 1 a;

FIG. 2a is a hyperbolic function with an asymptoteA diagram of;

FIG. 2b shows the harmonic spectrum of a distorted sinusoidal signal with amplitude 0.57 of the function shown in FIG. 2 a;

FIG. 3a shows the approximate psychoacoustic properties of the human ear A diagram of;

FIG. 3b shows the harmonic spectrum of a distorted sinusoidal signal with amplitude 2 of the function shown in FIG. 3 a;

FIG. 4a is an inverse approximation psychoacoustic characteristic of a human ear

FIG. 4b shows the harmonic spectrum of a distorted sinusoidal signal with amplitude 2 of the function shown in FIG. 4 a;

FIG. 5a shows a function through a hyperbolaA graph that inversely approximates psychoacoustic characteristics of a human ear;

FIG. 5b shows the harmonic spectrum of a distorted sinusoidal signal with amplitude 2 of the function shown in FIG. 5 a;

FIG. 6a shows the inverse psychoacoustic properties x + ((a-x) of a human ear using a vacuum diode^1.5-a^1.5+1.5·a^0.5x) b, where a-5.31423 and b-0.0366175;

FIG. 6b shows the harmonic spectrum of a distorted sinusoidal signal with amplitude 2 of the function shown in FIG. 6 a;

fig. 7 schematically shows an apparatus for implementing a method of adding non-linearities in an audio signal according to the invention;

FIG. 8 schematically shows the function-ax²A derivation of the non-linear square element of (a);

FIG. 9 schematically shows a functionDerivation of the nonlinear hyperbolic element;

FIG. 10 schematically shows a functionDerivation of the nonlinear hyperbolic element;

FIG. 11 schematically shows an audio chain according to a preferred way of carrying out the invention;

FIG. 12 schematically illustrates an audio chain according to another performance method of the present invention;

FIG. 13 shows one of the embodiments of an apparatus for audio signal processing by using quadratic and hyperbolic nonlinearities according to the invention; and

fig. 14 shows an embodiment of a non-linear element employing a vacuum diode.

Detailed Description

The method of the present invention takes into account the non-linearity of the electroacoustic transducer and the in-the-ear non-linearity.

According to the present invention, an audio signal processing method in an audio chain which corrects nonlinearity of an electroacoustic transducer in the audio chain, also taking into account nonlinear psychoacoustic characteristics of a human ear, comprises: approximating the psychoacoustic properties of the human ear by a quintic polynomial function, and adding at least one non-linear element 4 in front of at least one electroacoustic transducer in the audio chain, said non-linear element 4 having a function for adding non-linearities in the audio chain which correct for non-linearities of the at least one electroacoustic transducer and/or for up to p_ΔThe pressure of the human ear changes, the non-linearity of the approximate psychoacoustic properties of the human ear. According to the method, the non-linear element 4 reduces the non-linearity of the electroacoustic transducer by applying a quadratic non-linearity, which is ax + bx²Where x is the relative membrane deflection (extension) or the relative force on the membrane of the electroacoustic transducer, and a and b are positive constants.

According to one embodiment of the invention, the nonlinear element 4 is reduced by the component x by application²、x³And x⁴Introducing a function of at least two times the nonlinearity to reduce the nonlinearity of the in-ear psychoacoustic characteristic, and wherein the constant isAndheld within a tolerance of + -30% of each constant, where x is the relative pressure of the human ear.

According to another embodiment of the invention, the non-linear element 4 is formed by applying a hyperbolic functionAndto reduce the non-linearity of the in-ear psychoacoustic properties, where x is the relative pressure of the human ear.

According toIn another embodiment of the invention, the non-linear element 4 is formed by applying a function x^1.5To reduce the non-linearity of the psychoacoustic properties of the human ear, where x is the relative pressure of insertion into the ear.

The method will be described in further detail according to an embodiment of the audio chain of the invention.

The non-linearity within an electroacoustic transducer is defined by an adiabatic process defined as:

p Vⁿ＝const [1]

said non-linearity within the electro-acoustic transducer affects the quality of the sound. In the case of an electroacoustic transducer producing sound by moving a membrane, the air surrounding the membrane changes pressure by an adiabatic process. The volume of air compressed is unknown. However, changes in air pressure may be measured. The larger the volume of air compressed at the same pressure, the larger the membrane deflection required and vice versa. As the gas pressure changes through an adiabatic process, the same membrane deflection in the direction of increasing pressure will produce a greater pressure change than deflection in the opposite direction. Two ideal cases will be considered. In both cases, the mass of the membrane is negligibly small and rigid. In the first case, the membrane deflection is linear and the volume of compressed air varies linearly with the membrane deflection. An adiabatic process of air will be used. The initial pressure is atmospheric pressure. The adiabatic equation for air is:

p V^1.4＝const [2]

as the membrane moves, the volume changes, thereby adiabatically changing the gas pressure:

the air pressure of the membrane was:

wherein, V₀Is the initial volume of compression, and V_ΔIs the change in volume that occurs by moving the membrane. V_ΔHas the advantages ofNegative sign because the volume decreases as the membrane moves forward. The initial conditions would be: const ═ p₀、V₀1 and volume change V_ΔD, wherein p₀Is atmospheric pressure, d_rIs the relative membrane excursion. Thus, it is possible to write:

if the function is expanded as a taylor series according to the relative offset d, the first five components are:

p＝p₀(1+1.4x+1.68x²+1.904x³+2.0944x⁴+...)， [6]

p＝p₀+p_Δ， [7]

wherein p is_ΔIs the pressure change:

p_Δ＝p₀(1.4x+1.68x²+1.904x³+2.0944x⁴+...)。 [8]

for p_ΔThe relative membrane deflection is given by a pressure change of 1 Pa:

if it is placed in a taylor series, the components after the secondary component are negligible:

p₀(1.904x³+2.0944x⁴+...)≈0。 [10]

maximum non-linearity at normal loudness (loudness) is a quadratic function of pressure change

p_Δ≈p₀(1.4x+1.68x²)。 [11]

In the second case, there is a force acting on the electroacoustic transducer membrane and a volume of air that varies linearly with membrane deflection. For easier calculation, an isothermal process defined as the ideal gas as follows will be used:

p V＝const。 [12]

the force on the membrane is the sum of the forces on both sides of the membrane. Since the sound is heard from only one side of the membrane, the pressure will be monitored on that side. The force at the membrane surface was:

F＝A₀(p₁-p₂)， [13]

wherein p is₁Is the pressure on the side facing our membrane, p₂Is the pressure on the other side of the membrane, A₀Is the surface of the film and is constant. The pressures p1, p2 are:

wherein, V₀Is the initial volume of compression, and V_ΔIs the change in volume that occurs by moving the membrane. The initial condition would be const ═ p₀，V₀1 and V_ΔD, wherein p₀Is atmospheric pressure, and d_rIs the relative film shift in the listening direction. The equations of p1 and p2 are obtained:

the forces on the membrane were:

if it is assumed that the relative force is F_r＝F/(A₀p₀) Then it is:

and the relative film offset is:

the pressure on the listening side is then p₁＝p₀/(1-d), the result is:

according to relative force F_rA taylor series was developed, resulting in a pressure on the side facing our membrane:

wherein the pressure p on the side facing our membrane₁Is disclosed as:

p₁＝p₀+P_Δ [21]

while the pressure on the listening side changes p_ΔComprises the following steps:

for p_ΔThe relative membrane deflection is given by a pressure change of 1 Pa:

for such small relative forces, the effect of the larger member in the taylor series can be neglected:

maximum non-linearity at normal loudness is a quadratic function of pressure change

In both cases, the function can be determined by a quadratic function ax + bx²To approximate the change in pressure on the membrane, whereinAnd x is the relative membrane deflection in the first case or the relative pressure on the membrane in the second case. If one considers a normal loudness with a pressure change of the human ear of ± 1Pa, the pressure on the membrane is greater, since the pressure decreases with distance. The smaller the surface of the electroacoustic transducer membrane, the same other parameters, the greater the pressure on it at the same distance and the same loudness. Suppose that at 2 meters from the electroacoustic transducer, the pressure difference is 1Pa, and the electroacoustic transducer has a surface of 1.27 Pa²πcm²And there is no ideal dispersion of sound reflections in all directions, the acoustic power on the membrane is equal to the power at the sphere at some distance from the membrane. Passing through the sphere at a distance of 2 m, this is 4.2²πm²This gives 160000 π cm². The sound power is:

P＝I·A＝const [26]

where P is power, I is intensity, and A is surface area. If the intensity I is proportional to the square of the change in pressureThen can write outMeaning that the pressure on the membrane in the listening direction is

As the pressure on the membrane increases, the electroacoustic transducer operates in a nonlinear region, affecting the quality of the sound heard. For calculated loudness p_Δ＝p₀(ax+bx²)

Secondary component bx²The ratio of the linear component ax is

The first case is a 1.4, b 1.68 and p_ΔThe quadratic component is 0.27% of the linear component, which is not negligible at 314.96 Pa. The second case is a-1/2, b-1/4, p_Δ314.96Pa and the quadratic component is 0.31% of the linear component, which is also not negligible. In order to reduce the quadratic non-linearity of the electroacoustic transducers in the previous chain, a non-linear element correcting the non-linearity of the audio chain following it is incorporated:

y＝a(x+bx²) [30]

where a and b are positive constants. The simplest way to correct for the non-linearity of electro-acoustic transducers is to use an approximate inverse function x + bx²The nonlinear element of (a), such that:

development into Taylor series to obtain x-bx²+2b²x³-5b³x⁴+...

The first two components of the taylor series will be taken:

y^-1≈x-bx²， [32]

the remaining components will be ignored because their effect is negligible when x is very small. To obtain the characteristics of the nonlinear element and the audio chain following it, at a (x + bx)²) In, replace x with x-bx²And obtaining a (x-2 b)²x³+b³x⁴) Wherein when x is very small | -2b²x³+b³x⁴|＜＜|bx²L. Thus, distortion is reduced by a low value x, which is the case by listening to the audio chain at normal loudness, where the pressure of the human ear changes up to p_Δ± 1 Pa. If the electroacoustic transducer has a smaller membrane surface, the pressure on the membrane will be greater at the same distance and the same loudness. This will increase the adiabatic distortion of the electroacoustic transducer. Adjusting a nonlinear element to reduce a quadratic nonlinearity of an electroacoustic transducer toThree times less is sufficient to perceive a significant enhancement of the sound.

It is well known that SET (single-ended triode) tube amplifiers have a non-linearity of more than 1% at rated power and are not audible to the human ear. Jean Hiraga written an article with much attention and review entitled "Amplifier music-Amplifier Harmonic Distortion Spectroscopy Study" which describes the nonlinear Harmonic structure of various amplifiers and subjectively evaluates their sounds. In addition to not hearing the SET tube amplifier nonlinearities, their nonlinearities cover details of the sound we are no longer hearing. If we assume that the human ear has similar non-linearity and we do not hear it, we do not hear it even if the non-linearity is located in a part of the audio chain. Frequency sine wave f, as is well known₁And increased by 2, 3, 4, 5, 6 times f₁F of (a)₂，f₃，f₄，f₅，f₆Of a same frequency sine wave, wherein the amplitude is: f. of₁At 0db, f₂At-40 db, f₃At-50 db, f₄At-60 db, f₅At-70 db and f₆At-80 db, the sound will be the same in the human ear (fig. 2 b). The hyperbolic function 1/(1-x) (fig. 1a) has a non-linearity with a harmonic distortion structure such that each component is smaller than the constant value of the previous component (fig. 1 b). If the harmonic structure of the human ear is significantly disturbed, we will hear it as a change in sound. The psychoacoustic characteristics of the human ear will be approximated by a fifth order polynomial function:

x-a x²-b x³-c x⁴-d x⁵ [33]

where a, b, c and d are real positive numbers and x is the relative pressure of the human ear. To determine the values of a, b, c and d, we add non-linearity to the audio signal until the distortion of the harmonic structure of the human ear we hear is reached. To determine the coefficient a, a non-linear x + a x is used²It is given by an approximation to the characteristics of the human ear:

x-(2a²+b)x³-(a³+3ab+c)x⁴-(3a²b+4ac+d)x⁵-.. [34]

wherein the component x is removed²And disturbs the harmonic structure of the human ear. To determine the coefficient b, a non-linear x + b x is used³It is given by an approximation to the characteristics of the human ear:

x-a x²-(2ab+c)x⁴-(3b²+d)x⁵-.. [35]

wherein the component x is removed³And disturbs the harmonic structure of the human ear. To determine the coefficient c, a non-linear x + c x is used⁴It gives, using an approximation of the characteristics of the human ear:

x-a x²-b x³-(2ac+d)x⁵-.. [36]

wherein the component x is removed⁴And disturbs the harmonic structure of the human ear. To determine the coefficient d, a non-linear x + d x is used⁵It gives, using an approximation of the characteristics of the human ear:

x-a x²-b x³-c x⁴-.. [37]

wherein the component x is removed⁵And disturbs the harmonic structure of the human ear. Obtaining a component by hearing testAndtolerance of each component ± 30%. The approximate function of the psychoacoustic features of the human ear is:

by applying the Lagrange-Biirmann formula, the following approximate inverse function of the ear is obtained:

due to the psychoacoustic properties of the human earX of the approximation function of⁵The coefficients of the members are small and negligible, as are larger members. To hear enough detail, x, which approximates the psychoacoustic properties of the human ear, is needed²，x³And x⁴The non-linearity introduced by the member is reduced by at least a factor of two. The inverse function of the approximation of the psychoacoustic characteristics of the human ear can be hyperbolicAndthe derivation is given here by a-0.00372, b-0.06061, c-0.002484 and d-0.01313 (fig. 5 a). The inverse function of the approximation of the psychoacoustic properties of the human ear using hyperbolas is:

when developing a taylor series, the first five components are obtained:

to see how the non-linearity of the human ear is reduced, the approximate psychoacoustic features in the human ear In, use Replace x and get the first five components:

due to the fact thatAndapproximating the psychoacoustic properties x of the human ear²，x³And x⁴The non-linearity introduced by the member is reduced by at least a factor of two.

According to the invention, a device for implementing the method comprises at least one non-linear element 4 in the audio chain, having the function of adding to the audio chain a non-linearity correcting the non-linearity of at least one electroacoustic transducer and/or up to p for the human ear_ΔThe pressure of the human ear varies by a nonlinear approximation of the psychoacoustic properties of the human ear.

Fig. 7 schematically shows an apparatus 19 for implementing a general method of adding non-linearities in an audio signal according to the invention. The input audio signal 1 is routed into a non-isolated portion of the audio signal 1 and at least one isolated audio signal 1; the isolated audio signal 1 is processed by using a non-linear element 4 in at least one isolated non-linear audio signal 7 and a non-isolated part of the audio signal 1 is combined/merged with the at least one isolated non-linear audio signal 7 in an adder 8 into a processed output audio signal 9. Generating the nonlinear branch includes: an optional filter 2 before the non-linear element 4, an optional amplifier/attenuator 3 before the non-linear element 4, a non-linear element 4, an optional amplifier/attenuator 5 after the non-linear element 4 and an optional filter 6 after the non-linear element 4. The non-linear element 4 will have a quadratic function-x²Or hyperbolic functionAnd

a method of audio signal processing in an audio chain using the apparatus 19 shown in fig. 7, which method corrects for non-linearities of an electroacoustic transducer in the audio chain, also taking into account non-linear psychoacoustic properties of the human ear, the method comprising the steps of: splitting an input audio signal 1 into a non-isolated part of the audio signal 1 and at least one isolated audio signal 1; modifying at least one isolated audio signal 1 in the non-linear element 4 by adding a non-linearity; optionally amplifying/attenuating the at least one isolated audio signal in an amplifier/attenuator 3 before the non-linear element 4 and optionally amplifying/attenuating the at least one isolated audio signal in an amplifier/attenuator 5 after the non-linear element 4, optionally filtering the at least one isolated audio signal in a filter 2 before the non-linear element 4 and optionally filtering the at least one isolated audio signal in a filter 6 after the linear element 4 and obtaining at least one isolated non-linear audio signal 7; and combining the non-isolated part of the audio signal 1 and the at least one isolated nonlinear audio signal 7 in an adder 8 into an output audio signal 9.

Fig. 8 schematically shows an embodiment of the non-linear square element 4. The nonlinear element 4 is preceded by an amplifier/attenuator 3 having a positive value a, the nonlinear element 4 having a quadratic function-x²And having an amplifier/attenuator 5 after the non-linear element 4, said amplifier/attenuator 5 having a positive value b. The secondary non-linear element 4 is derived from a signal multiplier 10, which signal multiplier 10 multiplies the output signal after the amplifier/attenuator 3 with itself and changes its sign in a signal inverter 11. The overall transfer function of the circuit of FIG. 8 is- (ax)²b＝-a²bx². By adjusting the values a and b, it is possible to control how much quadratic non-linearity is added to the linear part of the signal.

Fig. 9 schematically shows an embodiment of the non-linear hyperbolic element 4. The amplifier/attenuator 3 before the non-linear element 4 has a positive value a, the non-linear element 4 has a hyperbolic functionAnd the non-linear element 4 is followed byThe amplifier/attenuator 5 has a positive value b. The hyperbolic non-linear element 4 is derived from a signal inverter 11, a source 12 of a constant 1 value, a signal adder 13, a signal sealer 14 and a signal multiplier 10. The output at the signal adder 13 is 1-x, where the signal further enters the signal sealer 14, which signal sealer 14 splits the signal x ÷ (1-x), where the signal multiplier 10 multiplies x and obtainsThe overall transfer function of the circuit of figure 9 is,

by adjusting the values a and b, any function can be obtainedWhere c and d are any positive values.

FIG. 10 shows schematically the derivation of the nonlinear hyperbolic element 4, the amplifier/attenuator 3 preceding the nonlinear element 4 and having a positive value a, the nonlinear element 4 having a hyperbolic functionThe amplifier/attenuator 5 after the well and the non-linear element 4 has a positive value b. The hyperbolic non-linear element 4 is derived from a source 12 of a value of constant 1, a signal adder 13, a signal sealer 14, a signal multiplier 10 and a signal inverter 11. The output at the signal adder 13 isWherein the signal further enters a signal sealer 14, the signal sealer 14 splits the signal by x ÷ (1+ x), the signal multiplier 10 multiplies by x and obtainsThe overall transfer function of the circuit of FIG. 10 is

By adjusting the values a and b, any function can be obtainedWhere c and d are any positive values.

Fig. 11 shows a preferred audio chain embodiment comprising at least one device 19 and a method for audio signal processing in said audio chain. The audio chain comprises: a preamplifier 16 of the input audio signal 15, connected to first means 19 for audio signal processing by using hyperbolic non-linearity; an audio frequency divider 18 connected to the first device 19 (after the first device 19), the audio frequency divider 18 splitting the processed audio signal in the second device 19 into two signal branches by frequency range. At least two second means 19 for audio signal processing using quadratic non-linearity are connected to the audio frequency divider 18 (after the audio frequency divider) and each of the two said second means 19 is also connected to a respective power amplifier 20 and two electro-acoustic transducers 21 are connected to respective power amplifiers 20. The raw input audio signal 15 enters a preamplifier 16 which controls the loudness. The signal from the preamplifier 16 enters a first means 19 for audio signal processing by using hyperbolic non-linearity. The processed signal from first device 19 enters audio frequency divider 18 and audio frequency divider 18 splits the signal into more branches by frequency range. After the audio frequency divider 18 the signal from each branch enters a second correlation means 19 for audio signal processing by using quadratic non-linearity. The processed signal from each second correlation means 19 enters an associated power amplifier 20, the associated power amplifier 20 routing the amplified signal to an associated electroacoustic transducer 21. Each second means 19 for signal processing by using quadratic non-linearity is configured to reduce the quadratic non-linearity of the electroacoustic transducer 21 by at least three times, taking into account the amplification of the power amplifier 20 which affects the required amount of non-linearity. If the amplification is higher, the quadratic non-linearity required on the associated second means 19 is larger. For using hyperboloidThe first means 19 for signal processing with line non-linearity is configured to be in the pressure change region p in consideration of the amplification of the power amplifier 20, the efficiency of the electroacoustic transducer 21 and the distance of the human ear from the electroacoustic transducer_ΔThe non-linearity of the psychoacoustic features of the human ear within ± 1Pa is reduced by at least a factor of two. The hyperbolic non-linearity on the first signal processing means 19 also needs to be larger if the amplification is larger and/or the efficiency of the electroacoustic transducer is higher and/or the distance of the human ear from the electroacoustic transducer is smaller.

A method of audio signal processing in an audio chain as shown in fig. 11 is performed by a device 19, which method corrects for non-linearities of an electroacoustic transducer in the audio chain, also taking into account non-linear psychoacoustic properties of the human ear, the method comprising the steps of: amplifying/attenuating the input signal 15 in an adjustable preamplifier 16; audio signal processing in the first means 19 by applying hyperbolic non-linearities; splitting the audio signal into two branches by frequency range in an audio frequency divider 18; processing the split audio signal in each branch in the second means 19 by applying a quadratic non-linearity; the split audio signals in each branch are power amplified in a power amplifier 20 and the audio signals of each branch are routed to an associated electroacoustic transducer 21.

Another embodiment of the apparatus 19 and the method within the audio chain is shown in fig. 12. The input audio signal 15 enters a preamplifier 16 which controls loudness. The signal from the preamplifier 16 flows to a first means 19 for audio signal processing by using quadratic and hyperbolic non-linearities. The processed signal from the first device 19 flows to the power amplifier 20, the power amplifier 20 delivers the amplified signal to the audio divider 18, and the audio divider 18 splits the signal into more branches by frequency range. After the audio frequency divider 18, the signal from each branch flows to a corresponding electroacoustic transducer 21. The quadratic non-linearity of the electroacoustic transducer 21 is reduced by at least three times by using a quadratic and hyperbolic non-linear signal processing means 19 configured to take into account the amplification of the power amplifier 20 which affects the required amount of quadratic non-linearity. Furthermore, the device 19 is configured to take into account the amplification of the power amplifier 20, the effectiveness of the electroacoustic transducer 21The rate and distance of the human ear from the electroacoustic transducer will be in the pressure change region p_ΔThe non-linearity of the psychoacoustic features of the human ear within ± 1Pa is reduced by at least a factor of two. A larger hyperbolic non-linearity on the device 19 is also required if the amplification is larger and/or the efficiency of the electroacoustic transducer is higher and/or the distance of the human ear from the electroacoustic transducer is smaller. Since the arrangement 19 reduces the quadratic non-linearity of several electroacoustic transducers having different quadratic non-linearities and operating in different frequency ranges, the arrangement applies the filter 2 before the non-linear element 4 and/or the filter 6 after the non-linear element 4, so that it adjusts the quadratic non-linearity for the different frequency ranges. The means 19 are designed to use quadratic and hyperbolic non-linearities by adding them simultaneously to the input audio signal 1 in the adder 8, or as a chain of series-connected means 19.

A method of audio signal processing in an audio chain as shown in fig. 12, performed by the apparatus 19, which corrects for non-linearities of an electroacoustic transducer in the audio chain, also taking into account non-linear psychoacoustic properties of the human ear, comprises the steps of: amplifying/attenuating the input signal 15 in an adjustable preamplifier 16; audio signal processing in the first means 19 by using quadratic and hyperbolic nonlinearities; amplifying the audio signal in a power amplifier 20; splitting the audio signal into two branches by frequency range in an audio frequency divider 18; and routes the signal of each branch to the associated electroacoustic transducer 21.

According to the method of the invention, the device 19 reduces the non-linearity of the approximate psychoacoustic properties of the human ear by a factor of two and/or reduces the quadratic non-linearity of the electroacoustic transducer by a factor of 3, and the pressure of the human ear changes by a factor of p_Δ＝±1Pa。

Furthermore, according to the method of the present invention, the audio signal may be processed in an analog format or in a digital format.

The invention also relates to a computer program adapted to run on a processor and to perform the method steps according to the invention when executed on a computer device.

FIG. 13 illustrates the use of an analog multiplier 24 as a non-linear elementTo obtain a quadratic characteristic and an embodiment of the means 19 using an analog multiplier/sealer 25 to obtain a hyperbolic characteristic. The input audio signal 1 reaches the inverting input stage 23, which then flows to a different branch having a non-linear element 4. The first branch has an input filter 2 configured as an adjustable first-order high-pass RC filter, an adjustable amplifier/attenuator 3 configured by using an operational amplifier, a resistor and a potentiometer, and a nonlinear element 4 configured as an analog multiplier 24. The second and third signal processing branches are implemented by: conveniently adjustable combined amplifier/attenuator 3 constructed by using operational amplifier, resistor and potentiometer, and by using a circuit having characteristicsThe analog multiplier/scaler 25 of (a) into a single nonlinear element 4. The outputs of the three branches of the non-linear part of the signal 7 enter an adder 8 formed by a resistor network, the adder 8 converting the non-linear output voltage signal 7 and the audio signal after the input stage 23 into a current sum forming the output audio signal 9, wherein the output inverter stage 26 converts them into the output voltage 9 a.

The inverse psychoacoustic characteristics of the human ear can also be approximated by other functions, and the derivation of the nonlinear element 4 can be performed by applying the nonlinearity of electronic components such as diodes, transistors and vacuum tubes. FIG. 6a shows a pass through non-linearity x^1.5Approximation of the non-linearity of the inverse function of the human earNon-linearity x^1.5Current/voltage characteristic I k · U corresponding to a vacuum diode^1.5. The approximation in FIG. 6a is by x + ((a-x)^1.5-a^1.5+1.5·a^0.5x) b, a-5.31423 and b-0.0366175 (solid line), which when expanded in a taylor series, yields the first five components:

to see how the non-linearity of the human ear is reduced, the approximate psychoacoustic features in the human ear In, use Replace x and get the first five components:

due to the fact thatAndapproximate psychoacoustic properties of the human ear x²，x³And x⁴The non-linearity introduced by the member is reduced by at least a factor of two.

Fig. 14 shows the realization of the non-linear element 4 by applying a vacuum diode. The input signal flows to a resistive network connected to a constant voltage-Va which adds a DC component to the input signal flowing to a voltage follower made by an operational amplifier. After the voltage follower, the signal flows to k.U with a current/voltage characteristic I ═ k.U^1.5The vacuum diode 27. The linear component is removed by applying an inverting amplifier 28 and a resistor 29 that converts the output voltage of the inverting amplifier 28 into a current added by the current of the vacuum diode 27. The DC component is removed by applying a constant voltage + Vb and a resistor 30. The sum of the currents of the vacuum diode 27, the resistor 29 and the resistor 30 is converted into an output voltage on an inverting amplifier 31. Transmission characteristic of the whole circuitIs ((a-x)^1.5-b + c x). d, a, b, c and d are positive values.

Application of the invention

Audio signal processing methods and apparatus are used in the audio chain to reduce unwanted non-linearities in the electroacoustic transducer and the human ear. Due to the adjustability of the device to various electro-acoustic transducers and the human ear, the device is widely used in the audio industry.