JP2003299180A - Method for driving ultrasonic loud speaker and loud speaker system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a new and useful loud speaker system which preprocesses a sound signal, before an ultrasonic carrier is modulated with the sound signal and is supplied to an ultrasonic loud speaker. <P>SOLUTION: A Fourier transformation portion 121 transforms the sound signal into a signal in a frequency region. A weighting portion 122 multiplies each frequency component with a weighting value. A zero setting and inverse Fourier transformation portion 124 removes either a group of weighted positive frequency components or group of weighted negative frequency components, and returns the signal into a signal in a time region. When an ultrasonic signal is modulated with this preprocessed sound signal, an ultrasonic signal of a single sideband (SSB) can be obtained. Weighting computation can be performed by comparatively simple processing, and its processing load is comparatively light, and a favorable result can be obtained even if the band width 25 of a directive ultrasonic transducer 40 is comparatively narrow. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an ultrasonic loudspeaker driving method, a preprocessing device and a loudspeaker system for preprocessing a signal before supplying it to a transducer of the loudspeaker. In particular, the present invention relates to a directional ultrasonic loudspeaker that produces a directional beam of ultrasonic vibrations modulated by an audio-frequency signal.

[0002]

Acoustic signals generated by conventional loudspeakers are usually emitted in almost all directions. That is,
Vibrations from the loudspeaker are not output only in the direction of the narrow angle range. Outputting sound waves in the middle and low frequency bands with high directivity can be achieved only by arranging the individual loudspeakers of the loudspeaker array in a complicated manner.

By the way, when a strong modulated ultrasonic signal is transmitted in the air (for example, reference [3] Masahide Yoneyam)
a etc. “The Audio Spotlight: An Application of No
nlinear Interaction of Sound Wave to a New Type of
Loudspeaker Design ”J. Acoust. Soc. AM. 73 (5), M
ay 1983. [4] F. Joseph Pompei “The Use of Airborn
e Ultrasonic for Generating Audible Sound Beams ”
J. Audio Eng. Soc., Vol. 47, No. 9, 1999, Septembe
r. [5] Mark F. Hamilton “Nonlinear Effects in Sou
nd Beams ”Handbooks of Acoustics, Edited by Malco
lm J. Crocker, Chapter 20, pp221-pp228, John Wiley
& Sons Inc., 1998), due to non-linearity in the air,
It is widely known that the envelope of a modulated ultrasound signal can be demodulated automatically. In addition, since the ultrasonic waves can be easily concentrated as a thin beam, the reproduction of the modulated ultrasonic signal can be limited to this thin beam. If the modulating signal is an acoustic signal, a thin acoustic beam is formed. This is the basic mechanism of directional loudspeakers.

[0004]

In reality, the demodulated signal is not directly proportional to the envelope of the amplitude modulation signal (hereinafter referred to as AM signal). Using a certain simplified hypothesis based on the non-linear theory of acoustics presented by Berktay, in an AM signal, when the primary wave of pressure p ₁ (t) was transmitted, after being modulated in air, It has been demonstrated that a secondary wave of pressure p ₂ (t) is generated.
The pressures p ₁ (t) and p ₂ (t) are expressed by the following equations (1) and (2).

[0005] _{p 1 (t) = P 1} E (t) sin (ω c t) (1)

[0006]

[Equation 1]

Here, P ₁ is the sound source sound pressure (peak value) of the primary wave, E (t) is the envelope of the modulation signal,
ω _c is the carrier frequency, β is the nonlinear coefficient of air, ρ ₀ is the air density, c ₀ is the small signal wave propagation velocity, A is the cross-sectional area of the beam, and z is the axial distance. Yes, α is the absorption coefficient of the sound wave (at the frequency ω _c ), and τ = t−z / c ₀ is the delay time. The relationship between the envelope E (t) of the modulation signal and the acoustic signal f (t) is expressed by equation (3)
It is represented by.

E (t) = 1 + mf (t) (3) Here, m represents the modulation factor of AM. The main function of the pre-processing is to reduce the non-linear effect of Eq. (2) in E (t), that is, to pre-process f (t) before modulation so that the output of the secondary wave becomes f (t). It is directly proportional. this is,
By generating the following E ′ (t) in which E (t) is modified,
Can be realized.

E ′ (t) = [1 + ∫∫f (τ) dτ ² ] ^1/2 (4) This seemingly simple pretreatment is very difficult to implement. The main difficulty lies in the calculation of the square root. Since the square root calculation is a non-linear calculation, it significantly increases the signal bandwidth. A wide signal bandwidth imposes very severe requirements on the circuit bandwidth, especially the ultrasonic transducer. It is very difficult to manufacture an ultrasonic transducer having a wide bandwidth and high power efficiency at the same time. Further, the calculation of double integration is difficult to carry out because the analog integration is easily saturated.

In order to avoid the direct use of the equation (4) and at the same time reduce the non-linear interference between different frequency components due to the square operation of the equation (2), for example, the reference document
[3] Masahide Yoneyama etc. “The Audio Spotlight:
An Application of Nonlinear Interaction of Sound Wa
ve to a New Type of Loudspeaker Design ”J. Acous
Although t. Soc. AM. 73 (5), May 1983 recommends using a small modulation factor m, if the modulation factor m is decreased, the output of the reproduced acoustic signal decreases, and the entire system is reduced. As the output efficiency is reduced. For example, reference [4] F. Joseph Pompei “The Use of Airborne Ultras
onic for Generating Audible Sound Beams ”J. Audio
Eng. Soc., Vol. 47, No. 9, 1999, September,
It is recommended to follow equation (4) faithfully, but this presents the problems described above. As a compromise, refer to [7] Tomoo Kamakura, Tasahide Yoneyama and Kazuo.
Ikegaya “Studies for the Realization of Parametri
c Loudspeaker ”J. Acoust.Soc. Japan, No. 6 Vol. 4
There is also a method of using a single sideband (hereinafter referred to as SSB) signal without a square root as a modulation signal described in 1, 1985. This method drastically reduces the required frequency band compared to the equation (4), and the direct double sideband (hereinafter, D
It is called SB. ) Although it has an effect of reducing internal self-interference compared to the signal, that is, the formula (3), it has a drawback of degrading sound quality.

In order to bring the SSB signal closer to an ideal envelope with a square root, it is possible to perform preprocessing to improve the sound quality, but preprocessing involves an increase in processing cost.

Therefore, the present invention has been made in view of the above-mentioned circumstances, and an object of the present invention is to provide an acoustic signal before modulating an ultrasonic carrier and supplying it to an ultrasonic transducer of a loudspeaker. The present invention provides a novel and useful ultrasonic loudspeaker driving method, a preprocessing device, and a loudspeaker system for preprocessing.

[0013]

The invention proposes to process the acoustic signal in the frequency domain by weighting each frequency component by a predetermined weighting factor. Further, in the present invention, the ultrasonic signal is pre-processed by removing the positive frequency component group or the negative frequency component group before the acoustic signal is reconverted (returned) to the time domain signal. When modulated by an acoustic signal, a single sideband (hereinafter referred to as SSB) ultrasonic signal can be obtained.

The weighting calculation proposed here can be performed by a relatively simple processing operation, which requires less processing in implementing the present invention. As will be explained below, this weighting calculation provides a practical, accurate and stable pre-processing with a low computational load on the computer and without generating a modulated ultrasound signal which requires a wide band transducer. Is enough for

In order to solve the above-mentioned problems, an ultrasonic loudspeaker driving method according to the present invention pre-processes an acoustic signal to generate a pre-processed acoustic signal, and uses the pre-processed acoustic signal to perform an ultrasonic wave. A method of modulating a signal and using a modulated ultrasonic signal to drive a transducer of a loudspeaker, the method comprising: transforming an acoustic signal into a frequency domain signal; and weighting different frequency components of the transformed signal. Weighting with the corresponding weight values stored in the value table, removing either the positive frequency component group or the negative frequency component group of the weighted frequency components, and the remaining frequency components Transforming from the domain to the time domain, pre-processing the acoustic signal to generate a pre-processed acoustic signal.

In this ultrasonic loudspeaker driving method, the step of weighting the frequency components and the step of removing the positive or negative frequency component group are not meant to be ordered, and these two steps are They may be performed in any order or at the same time.

The preprocessing device according to the present invention is used for a loudspeaker, and includes a first conversion means for converting an acoustic signal into a signal in the frequency domain, and a signal converted by the first conversion means. Weighting means for weighting different frequency components with corresponding weight values stored in the weight value table, and removing means for removing one of the positive frequency component group and the negative frequency component group of the weighted frequency components. And second conversion means for converting the remaining frequency components from the frequency domain to the time domain to generate a pre-processed signal for modulating the ultrasonic waves.

As a first concrete example of the weighting calculation, the calculation of the square root of the equation (4) can be avoided by using a single sideband signal having an appropriate AM modulation factor. In this case, in the frequency domain, the double integral of equation (4) can be expressed as division of the frequency component divided by a factor proportional to the square of the corresponding frequency.

As a second specific example of the weighting calculation, the frequency components can be weighted by corresponding numerical values capable of correcting the frequency dependence of the ultrasonic transducer of the loudspeaker.

Although these two weighting calculations are logically independent, each frequency component is 1 / ω2C.
It is preferable to use these in combination so that weighting calculation is performed with a weighting value proportional to (ω + ω _c ). Note that ω is the frequency, and C (ω + ω _c ) is the measured value of the frequency response of the converter at the frequency ω above the fundamental resonance frequency ω _{c of the} loudspeaker.

The pre-processing is preferably performed digitally, but the pre-processing device is preferably able to input signals in any combination of analog and / or digital signals. The input means of the preprocessor is an analog / analog converter that converts any analog signal contained in the input signal into a digital signal.
Equipped with a digital converter. Similarly, the preprocessor preferably comprises a digital-to-analog converter for producing an analog signal for use in modulation.

[0022]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An ultrasonic loudspeaker driving method, a preprocessing device and a loudspeaker system according to the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a specific configuration of a loudspeaker system to which the present invention is applied. This loudspeaker system is, as shown in FIG. 1, a pre-processing device 10 for pre-processing at least one input audio signal.
And a modulator 20 that amplitude-modulates the ultrasonic carrier generated by the oscillator 30 by a pre-processed acoustic signal, for example, according to Expression (1), and a directional ultrasonic transducer 40 that includes, for example, an ultrasonic loudspeaker. Prepare In addition, it is preferable that the pre-processing apparatus 10 can input either or both of an analog acoustic signal and a digital acoustic signal.

FIG. 2 is a block diagram showing a specific structure of the preprocessing apparatus 10. As shown in FIG. 2, the preprocessing device 10 includes an input unit 11 and a preprocessing unit 12. The input unit 11 includes a filter and an analog / digital converter (hereinafter referred to as a filter and an A / D converter) 111 that converts an input analog audio signal into a digital audio signal, and a digital audio signal from, for example, a compact disc or MP3. And a decoder 112 for decoding The filter and A / D converter 111 filters the input analog acoustic signal, converts it into a digital acoustic signal, and supplies the digital acoustic signal to the preprocessing unit. On the other hand, the preprocessing unit 112
For example, the digital audio signal input from the compact disc player or the compressed digital audio signal distributed in the so-called MP3 format is decoded, and the obtained digital audio signal is supplied to the preprocessing unit 112. Therefore, in any case of the input signal, the signal supplied from the input unit 11 to the preprocessing unit 12 is a digital acoustic signal.

As shown in FIG. 2, the preprocessing unit 12 transforms the digital acoustic signal supplied from the input unit 11 into a frequency domain signal and a Fourier transform unit 121, and a frequency domain signal weighting unit 122. And a zero-setting and inverse Fourier transform unit 124 that removes either the positive frequency component group or the negative frequency component group of the weighted frequency components and converts the weighted frequency component into a time domain signal, and the time domain signal A digital / analog converter and a filter (hereinafter, referred to as a D / A converter and a filter) 125 for converting and filtering an analog signal are provided. The Fourier transform unit 121 includes, for example, a fast Fourier transformer (hereinafter, referred to as FFT unit 121), and converts the digital acoustic signal f (t) in the time domain into the signal F in the frequency domain.
Convert to (ω).

The weighting unit 122 is an FFT conversion unit 121.
The weight value table 123 is added to each frequency component supplied from
The different weight values corresponding to each stored in are applied. Here, the setting of the weight value of the weight value table 123 will be described with reference to FIGS. 3 and 4. The weight value table 123 is composed of, for example, a ROM or a RAM, and it is preferable that the weight values calculated in advance be stored in the memory.

FIG. 3 is a graph showing the gain characteristic as a frequency function of the directional ultrasonic transducer 40, which is composed of an ultrasonic loudspeaker, for example. As shown in FIG. 3, this gain characteristic has a frequency pass band 25, and the center frequency of the band is usually approximately the resonance frequency ω _c . The center frequency of the frequency pass band 25 is preferably the frequency of the ultrasonic wave generated by the oscillator 30.

FIG. 4 is a diagram showing a method of selecting a weight value. The broken line 35 indicates the weight value stored in advance in the weight value table 123, and this weight value is used in the weighting unit 122 to execute the double integration before obtaining the square root of the equation (4). To be done. In most cases, the weight value shown by dashed line 35 will be approximately -12 dB / octave.

The weight values actually stored in the weight value table 123 and used in the weighting section 122 are
More preferably, it is selected to be 1 / ω ² C (ω + ω _c ). Here, C (ω + ω _c ) is the measured value of the frequency response of the directional ultrasonic transducer 40 at the frequency ω + ω _{c, as shown} in FIG. This weight value may be zero at frequencies outside the frequency pass band 25 of the directional ultrasonic transducer 40, ie, the ultrasonic loudspeaker, and as described below, the zero-setting and inverse Fourier transform unit. It is preferably zero at all frequencies removed by 124. Note that the weight value may be changed according to the converter gain shown in FIG. 3 at other frequencies C (ω + ω _c ), for example, may be the same. Therefore, the weighting unit 122 not only executes the double integration, but also has an effect of correcting the fluctuation of the frequency gain of the directional ultrasonic transducer 40.

Therefore, the output G of the weighting unit 122
(ω) is expressed by the following equation (5).

[0030]

[Equation 2]

The output of the weighting unit 122 is supplied to the zero setting and inverse Fourier transform unit 124, and the zero setting and inverse Fourier transform unit 124 outputs either the negative frequency component or the positive frequency component of G (ω). Is suppressed to generate a desired baseband, and its sideband is subjected to inverse Fourier transform, preferably fast inverse Fourier transform (hereinafter referred to as IFFT), to be converted into a signal in the time domain. For example, G
A negative frequency component of (ω) is expressed as G (ω−), and a complex signal g (t) obtained by IFFT transforming G (ω−) is represented by Expression (6).

G (t) = g _I (t) + jg _Q (t) = IFFT (G (ω−)) (6) where g _I (t) and g _Q (t) are baseband SSB signals. 2 is shown as outputs I and Q of the zero setting and inverse Fourier transform unit 124 in FIG.

The complex signal g (t) from the zero setting and inverse Fourier transform unit 124 is converted into a D / A converter and filter 125.
And the D / A converter and filter 125 converts the real part g _I (t) and the imaginary part g _Q (t) of the complex signal g (t) into an analog signal and filters the analog signal. Two corresponding outputs of 10 are provided to the modulator 20.

The modulator 20 supplies the ultrasonic carrier supplied from the oscillator 30 and the ultrasonic carrier whose phase is shifted by 90 degrees to the real part g _I (t) of the complex signal g (t).
And the imaginary part g _Q (t) are respectively amplitude-modulated, and one sideband is suppressed and output. The band-passed SSB signal h (t) output from the modulator 20 is expressed by Expression (7).

[0035] h (t) = (1 + mg I (t)) cos (ω c t) + (1 + mg Q (t)) sin (ω c t) (7) The SSB signal h (t), the output amplifier ( After being amplified (not shown), it is used to drive the directional ultrasonic transducer 40. The directional ultrasonic transducer 40 is composed of, for example, an ultrasonic loudspeaker, and emits amplitude-modulated ultrasonic waves in a narrow angle direction.

FIG. 5A shows an ideal square-rooted DSB modulation signal obtained from equation (4) when the acoustic signal contains only a single acoustic signal ω _0.
is a diagram showing a spectrum of (nal). This square root DSB
The bandwidth of the spectrum of the modulated signal is a wide bandwidth in which each component exists at intervals of an integral multiple of ω ₀ above and below the carrier frequency.

FIG. 5B shows a signal when the present invention is applied.
It is a figure which shows the spectrum of No. In this spectrum,
In addition to carrier, single frequency component ω_c−ω₀(Okay
Is ω _c+ Ω₀) Only exists.

Since the self-interference between the different frequency components of the upper sideband and the lower sideband is removed, the resulting acoustic signal has a higher fidelity (equation (3)) than the conventional DSB signal (equation (3)). Fidelity). It is true that the signal obtained by applying the present invention has less fidelity than an ideal square root DSB modulated signal, since there is still self interference of different frequency components in the remaining sidebands. However, the ideal square root DSB modulated signal is shown in FIG.
As shown in, it is impossible to actually realize because it requires infinite bandwidth. As is clear from the above description, the loudspeaker system to which the present invention is applied provides a better compromise in system complexity and sound fidelity than the above-mentioned conventional systems or methods. is there.

In the above description, the present invention has been described with reference to only one embodiment, but the present invention can be variously modified and changed according to a specific application without departing from the gist of the present invention. It will be apparent to those skilled in the art.

For example, embodiments of the present invention may be modified to include additional signal processing in the frequency / time domain to further improve the fidelity of the final audio output. For example, these signal processings can be appropriately performed in the frequency domain.

Furthermore, the present invention provides a "section" for performing signal processing.
However, these units do not have to be physically separated, and may be realized as a part of software executed by any number of physically separated signal processing units, for example.

[0042]

According to the present invention, by modulating the ultrasonic carrier with the acoustic signal and pre-processing the acoustic signal before supplying it to the ultrasonic transducer of the loudspeaker, compared with the conventional system, The system configuration can be simplified and the sound quality fidelity can be improved.

The weighting calculation in the present invention is as follows.
It can be done with relatively simple processing,
Less processing is required to implement the invention. Further, as described above, this weighting calculation has a low computational load, and can perform practical, accurate, and stable preprocessing without generating a modulated ultrasonic signal that requires a wideband converter. it can.

(References) [1] Tsuneo Tanaka, Mikio I
wasa, Youichi Kimura and Akira Nakamura. “Directi
onal Loudspeaker System ”US Patent No. 4,823,90
8, 1989. [2] Hans-Joachim Raida and Oskar Bschorr. “Direct
ed Radiator with Modulated Ultrasonic Sound ”US
Patent No. 6,016,351, 2000. [3] Masahide Yoneyama etc. “The Audio Spotlight:
An Application of Nonlinear Interaction of Sound W
ave to a New Type of Loudspeaker Design ”J. Acous
t. Soc. AM. 73 (5), May 1983. [4] F. Joseph Pompei “The Use of Airborne Ultraso
nic for Generating Audible Sound Beams ”J. Audio
Eng. Soc., Vol. 47, No. 9, 1999, September. [5] Mark F. Hamilton “Nonlinear Effects in Sound
Beams ”Handbooks of Acoustics, Edited by Malcolm
J. Crocker, Chapter 20, pp221-pp228, JohnWiley & S
ons Inc., 1998. [6] Elwood G. Norris and James J. Croft III, “Met
hod and Device for Developing A Virtual Speaker Di
stant from the Source ”US Patent No. 6,229,889,
2001. [7] Tomoo Kawamura, Tasahide Yoneyama and Kazuo Ik
egaya “Studies for the Realization of Parametric L
oudspeaker ”J. Acoust. Soc. Japan, No. 6Vol. 41,
1985.

[Brief description of drawings]

FIG. 1 is a block diagram showing a specific configuration of a loudspeaker system to which the present invention is applied.

FIG. 2 is a block diagram showing a specific configuration of the pretreatment device shown in FIG.

FIG. 3 is a graph showing an example of frequency response characteristics of the directional ultrasonic transducer shown in FIG.

FIG. 4 is a diagram showing a method of selecting a weight value.

FIG. 5 is a diagram showing a spectrum of an ideal square root DSB modulated signal and a spectrum of a signal reproduced by a loudspeaker system to which the present invention is applied.

[Explanation of symbols]

10 pre-processing device, 11 input unit, 111 filter and A / D converter, 112 decoder, 12 pre-processing unit, 1
21 Fourier transform unit, 122 Weighting unit, 123
Weight value table, 124 Zero setting and inverse Fourier transform unit, 125 D / A converter and filter

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Sun Shaobing             Republic of Singapore 117684 Singapore             Le Science Park II The Alpha             # 03-08 Science Park Road 10             Sony Electronics (Singapo             Le) Private Limited Singa             Inside Paul Research Laboratory (72) Mayor Gang, the inventor             Republic of Singapore 117684 Singapore             Le Science Park II The Alpha             # 03-08 Science Park Road 10             Sony Electronics (Singapo             Le) Private Limited Singa             Inside Paul Research Laboratory (72) Inventor Kanzo Okada             Republic of Singapore 117684 Singapore             Le Science Park II The Alpha             # 03-08 Science Park Road 10             Sony Electronics (Singapo             Le) Private Limited Singa             Inside Paul Research Laboratory (72) Inventor Tong Kok Ren             Republic of Singapore 117684 Singapore             Le Science Park II The Alpha             # 03-08 Science Park Road 10             Sony Electronics (Singapo             Le) Private Limited Singa             Inside Paul Research Laboratory F-term (reference) 5D019 AA07 FF01                 5D020 AC11

Claims (15)

[Claims] 1. A transducer for a loudspeaker using a pre-processed acoustic signal to generate a pre-processed acoustic signal, the pre-processed acoustic signal being used to modulate an ultrasonic signal, and the modulated ultrasonic signal being used to modulate the ultrasonic signal. In the method for driving an ultrasonic loudspeaker for driving a sound signal, a step of converting the acoustic signal into a signal in a frequency domain, and different frequency components of the converted signal with corresponding weight values stored in a weight value table. A step of weighting, a step of removing one of the positive frequency component group and the negative frequency component group of the weighted frequency component, and a step of converting the remaining frequency component from the frequency domain to the time domain. A method for driving an ultrasonic loudspeaker, comprising: pre-processing the acoustic signal to generate the pre-processed acoustic signal. 2. The method for driving an ultrasonic loudspeaker according to claim 1, wherein the weight value changes in inverse proportion to the square of the corresponding frequency. 3. ω is a frequency corresponding to the weight value, and ω_c
Is a constant, and C (ω + ω _c) To the above loudspeaker conversion
Frequency ω_cAnd the measured frequency response at
The above weight value is 1 / ω^TwoC (ω + ω_c)
The ultrasonic loudspeaker according to claim 1, wherein
Driving method. 4. The acoustic signal is a digital signal, the pre-processing is digital processing, and the method further comprises the step of generating the acoustic signal from an input signal including an analog signal. 7. A method for driving an ultrasonic loudspeaker according to. 5. The method for driving an ultrasonic loudspeaker according to claim 4, wherein the input signal further includes a digital signal. 6. The pre-processing comprises the step of converting the pre-processed digital signal into an analog signal, the analog signal being filtered prior to being used for the modulation. Item 6. A method for driving an ultrasonic loudspeaker according to Item 4 or 5. 7. The method for driving an ultrasonic loudspeaker according to claim 1, wherein the ultrasonic beam emitted from the ultrasonic loudspeaker is a directional beam. 8. A preprocessing device for use in a loudspeaker, wherein first conversion means for converting an acoustic signal into a frequency domain signal, and different frequency components of the signal converted by the first conversion means are weighted. Weighting means for weighting with the corresponding weight value stored in the value table, removing means for removing either the positive frequency component group or the negative frequency component group of the weighted frequency components, and the remaining A preprocessing device comprising: a second conversion unit that converts a frequency component from a frequency domain to a time domain to generate a preprocessing signal for modulating an ultrasonic wave. 9. The preprocessing apparatus according to claim 8, wherein the weight value of the weight value table changes in inverse proportion to the square of the corresponding frequency. 10. Let ω be a frequency corresponding to a weight value, and ω
_{When c} is a constant and C (ω + ω _c ) is a measurement value of the frequency response at the frequency ω _c of the loudspeaker, the weight value should be proportional to 1 / ω ² C (ω + ω _c ). The pretreatment device according to claim 8, wherein 11. Preprocessing according to any one of claims 8 to 10, further comprising input means supplied with an input signal and having an analog / digital converter for converting an analog component of the input signal into a digital signal. apparatus. 12. The preprocessing device according to claim 11, wherein a digital signal is further supplied to the input means. 13. The digital / analog converter for converting the pre-processed signal of a digital signal into an analog signal, the analog signal being filtered prior to being used for the modulation. Or the pretreatment device according to item 12. 14. A loudspeaker system having the preprocessing device according to claim 8, wherein the preprocessing signal is supplied, and an ultrasonic signal is generated using the preprocessing signal. A loudspeaker system further comprising: a modulator that modulates and generates a modulated ultrasonic signal; and an ultrasonic transducer that generates an ultrasonic beam from the modulated ultrasonic signal. 15. The loudspeaker system of claim 14, wherein the ultrasonic transducer produces a directional beam.