JP2001346299A - Sound field correction method and audio unit - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a sound field correction method by which the sound is corrected with a simple configuration even when the number of listening positions is increased when the acoustic characteristic at the prescribed listening positions in a closed space is corrected identical to that of an original sound. SOLUTION: First measuring a sound pressure outputted from a loudspeaker and a response sound pressure with respect to the sound pressure of the loudspeaker measured at prescribed listening positions in a space obtains a transfer function matrix of this space and then a correction filter to realize acoustic correction is calculated from the transfer function matrix, and using the correction filter corrects the acoustic signal outputted to the loudspeaker to provide the corrected signal to the loudspeaker so that the acoustic characteristic at the prescribed listening positions in the space is corrected identical to that of the original sound itself. Thus, the acoustic characteristic at the prescribed listening positions in the closed space is corrected identical to that of the original sound itself without increasing the number of loudspeakers with a simple configuration.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a sound field correction method for correcting an acoustic characteristic at a predetermined viewing position in a closed space so as to be that of an original sound, and an audio apparatus for realizing the acoustic correction. In particular, the present invention relates to a sound field correction method capable of realizing sound correction according to a simple configuration even when the number of viewing positions increases, and an audio device realizing the sound correction.

[0002] In a closed space such as a passenger compartment, a complicated sound field is formed, and the way in which sound is heard differs significantly depending on the viewing position. From now on, it is necessary to provide a correction mechanism for realizing the acoustic characteristics of the original sound at each viewing position, but this correction mechanism is simple even when the number of viewing positions exceeds the number of speakers. It is preferable that the sound correction can be realized according to a simple configuration.

[0003]

2. Description of the Related Art In a vehicle room, the presence of standing waves due to the reflection of sound waves or the interaction of loudspeakers causes the sound pressure level of a specific acoustic mode to decrease depending on the viewing position. A complex sound field is formed.

Conventionally, as disclosed in Japanese Patent Application Laid-Open No. 8-331682, a sound pressure level of a specific frequency component of a received sound at a viewing position is lowered by positioning the viewing position near a node of a standing wave. When sound is played, a speaker is placed near the listening position and the sound of the frequency component whose sound pressure level is lowered is output, so that the sound at the listening position is close to the original sound. Was used.

Further, as disclosed in Japanese Patent Application Laid-Open No. 9-327086, the sound pressure level of the sound collected immediately before the speaker is determined by using a first equalizer provided before the power amplifier connected to the speaker. Is adjusted to be flat over the entire audible frequency band, and then, using a second equalizer provided between the power amplifier and the first equalizer, the sound pressure of the sound collected at the receiving point In some cases, the level is adjusted so as to be flat over the entire audible frequency band so that the sound at the receiving point is close to the original sound.

In Japanese Patent Application Laid-Open No. 10-161667, there is a case where a desired acoustic characteristic is obtained at a measurement point in a vehicle interior acoustic space using an adaptive equalizer. In consideration of an enormous amount of calculation, a method of using an adaptive equalizer that targets only a specific frequency band such as a bass range is described.
The gist of the invention described in JP-A-10-161667 is to improve the adaptive equalizer.

In this adaptive equalizer, in addition to the first speaker that radiates an audio signal in a frequency band to be subjected to adaptive signal processing to a vehicle interior acoustic space, a first speaker that radiates an entire band audio signal to a vehicle interior acoustic space is provided. By adopting a configuration including two speakers and setting a delay time so that audio signals passing through the two speakers reach the measurement point at the same time, desired acoustic characteristics can be obtained at the measurement point in the vehicle interior acoustic space. I have to.

[0008]

However, as disclosed in Japanese Patent Application Laid-Open No. 8-331682, a speaker is arranged in the vicinity of a viewing position and a sound of a frequency component whose sound pressure level is lowered is output. According to the conventional technique of making the sound at the viewing position close to the original sound, there is a problem that a speaker different from the originally prepared speaker must be prepared.

Further, according to the prior art method of using two equalizers to make the sound at the receiving point close to the original sound as disclosed in Japanese Patent Application Laid-Open No. 9-327086, Since two equalizers must be prepared for each sound point, there is a problem that when the number of sound receiving points increases, an enormous number of equalizers must be prepared.

[0010] Further, according to the prior art method of using an adaptive equalizer that targets only a specific frequency band such as a bass range as described in JP-A-10-161667, the speaker originally prepared is There is a problem that another speaker must be prepared.

As described above, according to the conventional technique, a configuration is employed in which the acoustic characteristics at each viewing position in the space are corrected to those of the original sound. When the number of viewing positions increases, the configuration is simplified according to a simple configuration. There is a problem that sound correction cannot be realized.

The present invention has been made in view of the above circumstances, and is directed to a case in which a configuration for correcting acoustic characteristics at a predetermined viewing position in a closed space to that of an original sound is adopted.
It is an object of the present invention to provide a new sound field correction method capable of realizing sound correction according to a simple configuration even when the number of viewing positions increases, and to provide a new audio device realizing the sound correction.

[0013]

In order to achieve this object, in the sound field correction method of the present invention, first, the sound pressure output from a speaker and the sound pressure measured at a predetermined viewing position in space are measured. The transfer function matrix of this space is obtained by measuring the response sound pressure to the sound pressure, and subsequently, a correction filter for realizing that the acoustic characteristic at a predetermined viewing position becomes that of the original sound from the transfer function matrix. Then, the sound signal output to the speaker is corrected by using the correction filter, and the corrected sound signal is output to the speaker so that the sound characteristics at a predetermined viewing position in the space become that of the original sound. To be corrected.

As shown in FIG. 1 (a), the incident sound pressure from the loudspeaker into the space is F, the loudspeaker and the viewing position (listener)
If the sound characteristic of the incident sound pressure F is flat in a closed space such as the interior of a vehicle, if the spatial characteristic between H and the listening sound (sound pressure response) at the viewing position is P, Due to the spatial characteristic H between the speaker and the listening position, a large peak or dip occurs on the acoustic characteristic of the received sound P. Therefore, an acoustic characteristic different from the acoustic characteristic F of the original sound is provided to the listener, and as a result, the sound quality of the received sound P is greatly deteriorated.

Therefore, in the present invention, as shown in FIG. 1B, the incident sound pressure is changed from F to F ^* by using a correction filter G in consideration of the spatial characteristic H between the speaker and the viewing position. By making a change so that the acoustic characteristics of the received sound P ^* become flat, the same acoustic characteristics P ^* as the acoustic characteristics F of the original sound are provided to the listener.

In order to obtain the correction filter G, in the present invention, first, using a microphone, a sound pressure output from a speaker and a response sound pressure to the sound pressure measured at a viewing position in a space are used. Is measured to obtain a space transfer function matrix (including amplitude and phase).

At this time, a ratio value of two measured sound pressures measured on the time axis is calculated, and each matrix element of the transfer function matrix may be obtained by Fourier transforming the ratio. Each matrix element of the transfer function matrix may be obtained by Fourier-transforming the sound pressure and calculating a ratio value on the frequency.

In the present invention, subsequently, a correction filter for realizing that the acoustic characteristic at the viewing position becomes that of the original sound is calculated from the obtained transfer function matrix.

When a sound source is placed in a closed space, the sound pressure incident from the sound source into the space is {F} [Pa], the transfer function matrix of the space is [H], and the sound pressure response at the viewing position is If ｛P｝ [Pa], then ｛P｝ _{m * 1} = [H] _{m * n} ｛F｝ _{n * 1} (1) where {P} is m rows and 1 column, [H] is m rows and n columns, {F} is n rows and 1 column, n is the number of sound sources, and m is the number of viewing positions.

Here, when the number m of viewing positions and the number n of sound sources match (m = n), the inverse matrix can be calculated by making the transfer function matrix H square, and [G] _{n * n} = [H] _{n * n} ^-1 where [G] can calculate the correction filter G according to n rows and n columns.

By being corrected by the correction filter G,
The incident sound pressure ｛F ^* ｝ [Pa] incident on the space becomes ｛F ^* ｝ _{n * 1} = [H] _{n * n} ^-1 ｛F｝ _{n * 1} , which is a transfer function matrix H of the space. By being modulated, the sound pressure response ｛P ^* ｝ [Pa] at the viewing position is as follows: ｛P ^* ｝ _{n * 1} = [H] _{n * n} [H] _{n * n} ^-1 ｛F｝ _{n * 1} Becomes

In this equation, "[H] _{n * n} [H] _{n * n}
^As can be seen from the fact that “ ⁻¹ ” is a unit matrix, the sound pressure response ｛P ^* ｝ [Pa] at the viewing position has the same acoustic characteristics as the original sound ｛F｝ [Pa].

Thus, in the present invention, when the number of viewing positions is equal to the number of speakers, processing is performed to calculate the correction filter by calculating the inverse matrix of the transfer function matrix.

When the acoustic characteristics of all the seats in the passenger compartment are to be corrected simultaneously, the number of viewing positions m (for example, if four people are seated, a total of eight viewing positions from the left and right ears are obtained). ) Is larger than the number n of sound sources (for example, four speakers are provided in front, rear, left and right).

At this time, the transfer function matrix H is not square. Therefore, by multiplying the transposed matrix of the transfer function matrix H (1) from the left of equation of both sides, [H] ^{_{m * n T {P} m}} * 1 = [H] _{m * n} ^T (H) _{m * n} {F} _{n * 1} ·· (2) where {P} is m rows and 1 column, {F} is n rows and 1 column [H] is m rows and n columns, [H] ^T is n rows and m columns obtain.

[(H) on the right side of the equation (2)
_{m * n} ^T [H] _{m * n} ”has n rows and n columns, so that an inverse matrix can be calculated. From this, by multiplying this inverse matrix from the left of both sides of equation (2), ([H] _{m * n} ^T [H] _{m * n} ) ⁻¹ [H] _{m * n} ^T ｛P｝ _{m * 1} = [E] ｛F｝ _{n * 1} ·· (3) where [E] is a unit matrix.

From this, the correction filter [G] is calculated as follows: [G] _{n * m} = ([H] _{m * n} ^T [H] _{m * n} ) ⁻¹ [H] _{m * n}
^T where [G] is n rows and m columns, while from equation (3),
The sound pressure at the listening position is ideal sound {P ^*} and the incident sound pressure that is required to become {F ^{*} _{is, {F *} n * 1}} = ( [H] _{m * n} ^T (H) _{m * n} ) ^-1 [H]
_{m * n} ^T ｛P ^{* *} _{m * 1} can be predicted.

Substituting this equation into ｛P ^* ｝ _{m * 1} = [H] _{m * n} ｛F ^* ｝ _{n * 1} derived from equation (1), the matrix of the ideal sound {P ^* } Considering that the matrix is the same as the matrix of the target sound {P ^** }, which is the sound pressure, if {P ^* } on the right side is replaced with {P ^** }, then {P ^* } _{m * 1} = [H] _{m * n} ([H] _{m * n} ^T [H] _{m * n} )
^-1 [H] _{m * n} ^T ｛P ^** ｝ _{m * 1} However, ｛P ^* ｝ and｝ P ^** ｝ are corrected to m rows and 1 column, and the ideal sound ｛P ^* ｝ [Pa] is the target sound Has the same acoustic characteristics as the original sound {F} [Pa].

That is, by distributing an original sound {F} having a size of n rows and 1 column, a target sound {P ^** } having a size of m rows and 1 column is created, and a correction filter [G ] _{N * m} = ([H] _{m * n} ^T [H] _{m * n} ) ^-1 [H] _{m * n}
^When the product multiplied by ^T is incident on the space, it is multiplied by a transfer function matrix of m rows and n columns, so that ｛P ^* ｝ _{m * 1} = [H] _{m * n} ([H] _{m * n} ^T [H] _{m * n} )
^-1 [H] _{m * n} ^T {P ^** } _{m * 1} is obtained, and the sound pressure at the viewing position becomes the ideal sound {P ^* }.

According to the present invention, when the number of viewing positions is larger than the number of speakers, an inverse matrix of the matrix operation of the transfer function matrix and its transposed matrix is obtained, and the transposed matrix and its inverse are calculated. By calculating a matrix operation with a matrix, a correction filter is calculated, and an acoustic signal to be output to a speaker is distributed and input to the correction filter.

In the present invention, the sound signal to be output to the speaker is corrected using the correction filter calculated as described above, and the corrected sound signal is output to the speaker. It realizes that the acoustic characteristics are those of the original sound.

As described above, according to the present invention, even if the number of viewing positions increases, the acoustic characteristics at a predetermined viewing position in a closed space can be reduced according to a simple configuration without increasing the number of speakers. Can be corrected to the original sound.

[0033]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail according to embodiments.

FIG. 2 shows an embodiment of the present invention.

In this embodiment, it is realized that the sound characteristics at a plurality of viewing positions set in the cabin space 2 of the automobile 1 are corrected so as to be those of the original sound.

Here, for convenience of explanation, the cabin space 2 includes:
Speaker 3-1 (LF) provided on the left side of the front
And a speaker 3-2 (R
F) and a speaker 3-3 (L
B) and a speaker 3-4 (R
It is assumed that four speakers B) are provided.

For convenience of explanation, four listeners α,
β, γ, δ are assumed, and from this, the viewing position corresponding to the left ear position of the listener α, the viewing position corresponding to the right ear position of the listener α, and the left ear position of the listener β A viewing position corresponding to the right ear position of the listener β, a viewing position corresponding to the left ear position of the listener γ, a viewing position corresponding to the right ear position of the listener γ, and a listener δ And a viewing position corresponding to the right ear position of the listener δ.

In FIG. 2, reference numeral 10 denotes an optical disk reproducing apparatus for generating a left reproduction signal and a right reproduction signal;
Is a control unit for controlling the optical disk reproducing apparatus 10 and executing processing for realizing the present invention, 12 is a display constituting an operation panel, 13 is an operation key group constituting an operation panel, and 14-p (p = 1 ~
4) an amplifier for amplifying a reproduction signal output from the optical disk reproducing apparatus 10 and outputting the amplified signal to the corresponding speaker 3-p (p = 1 to 4); 15 a digital filter processing for the reproduction signal output from the optical disk reproducing apparatus 10 A digital filter for applying a Fourier transform to the amplifier 14-p, a Fourier transform circuit 16 (which may be of a software configuration) for performing a Fourier transform, a selector circuit 17, an A / D converter 18,
9 is a waveform memory.

In the present invention, as the first stage processing, the filter coefficients of the digital filter 15 for realizing that the acoustic characteristics at the eight viewing positions to those of the reproduced signal are obtained. Subsequently, as a process of the second stage, a process is performed in which the filter coefficients are set in the digital filter 15 so that the acoustic characteristics at the eight viewing positions to those of the reproduced signal are corrected.

Here, when the first-stage processing is executed, as shown in FIG. 2, a pseudo random signal generated by the control unit 11 is replaced by an amplifier instead of the playback signal output from the optical disk playback apparatus 10. The output is output to 14-p. The digital filter 15 does not execute the digital filter processing, and outputs the pseudo random signal to the amplifier 14-p as it is.

When the process of the first stage is executed, the eight viewing positions and the front positions of the four speakers 3-p (A,
B, C, D), and collects sound pressure signals detected by those microphones.

The selector circuit 17, the A / D converter 18, and the waveform memory 19 are provided for collecting the microphone output.

FIG. 3 shows an embodiment of the detailed configuration of the selector circuit 17, the A / D converter 18, and the waveform memory 19.

As shown in FIG.
Receives sound pressure signals detected by microphones arranged in front of four speakers 3-p (p = 1 to 4), selects one of them according to an instruction from the control unit 11, and outputs the selected one. A first secretor circuit 170 to be input and a sound pressure signal detected by microphones arranged at eight viewing positions to input and select and output one of them according to an instruction from the control unit 11. Secretor circuit 171
It is composed of

On the other hand, the A / D converter 18 includes a first A / D converter 180 for A / D converting a sound pressure signal selectively output from the first secretor circuit 170, and a second secretor circuit 17
1. A / D conversion of the selected and output sound pressure signal
And a D converter 181.

On the other hand, the waveform memory 19 includes a first waveform memory 190 for storing a digital signal (sound pressure signal at a speaker position) output from the first A / D converter 180, and a second waveform memory 190 for storing a second signal.
And a second waveform memory 191 for storing a digital signal (sound pressure signal at the viewing position) output from the A / D converter 181.

FIG. 4 shows an embodiment of a processing flow executed by the control unit 11 to realize the first-stage processing.

Next, according to this processing flow, a description will be given of processing for obtaining a filter coefficient for realizing that the acoustic characteristics at the eight viewing positions to those of the reproduced signal.

When the control unit 11 is started to execute the first-stage processing, first, as shown in the processing flow of FIG. In the following step 2, "1" is set to a variable i.

Subsequently, in step 3, control is performed so that only the amplifier 14-p (p = j) indicated by the value of the variable j is operated and the other amplifiers 14-p (p ≠ j) are not operated. Then, control is performed so that a pseudo-random signal (or any other suitable sound signal) may be output only to the speaker 3-j indicated by the value of the variable j.

Subsequently, in step 4, by controlling the first selector circuit 170, the speaker 13 indicated by the variable j
The time series data of the sound pressure signal detected by the microphone disposed at the front position of -j is stored in the first waveform memory 190, and at the same time, the second selector circuit 17
1 to control the viewing position (i =
1), the viewing position when i = 2, the viewing position when i = 3,...) Stores the time-series data of the sound pressure signal detected by the microphone arranged in the second waveform memory 191. To be stored.

Subsequently, in step 5, the two sampled sound pressure signals are subjected to Fourier transform using the Fourier transform circuit 16, and a ratio value on the frequency (= [sound pressure at the viewing position] / [ By calculating the sound pressure at the speaker position]), the element H _ij of the transfer function matrix H of the vehicle interior space 2 is calculated.
Is calculated.

At this time, an element H _ij of the transfer function matrix H may be calculated by calculating a ratio value of the sampled sound pressure signal on the time axis and performing a Fourier transform on the ratio value.

Subsequently, in step 6, the value of the variable i is incremented by one, and in the following step 7, the value of the variable i is set to the maximum value m corresponding to the number of viewing positions (in this embodiment, m = 8), and returns to step 3 when it is determined that it has not been exceeded.

On the other hand, when it is determined in step 7 that the value of the variable i has exceeded the maximum value m, the process proceeds to step 8 where the value of the variable j is incremented by one. The maximum value n whose value corresponds to the number of speakers
(N = 4 in the case of this embodiment) is determined.

When it is determined that the value of the variable j does not exceed the maximum value n by this determination processing, step 2
When it is determined that the transfer function matrix H has been exceeded, it is determined that the calculation of the transfer function matrix H has been completed.
0, and when “m = n”, [G] _{n * n} = [H] _{n * n} ^-1 where [G] calculates a correction filter G according to n rows and n columns, In the case of “m ≠ n”, [G] _{n * m} = ([H] _{m * n} ^T [H] _{m * n} ) ⁻¹ [H] _{m * n}
^T [G] calculates the correction filter G according to n rows and m columns.

The matrix operation executed at this time is to select a frequency, perform a matrix operation at the selected frequency using the value of the transfer function matrix H at the selected frequency, and perform the matrix operation for all frequencies. Execute.

Subsequently, at step 11, the digital filter 1 for realizing this correction filter according to a known technique.
The filter coefficient of 5 is calculated, and the process ends.

That is, since the frequency characteristics of each matrix element of the correction filter G have been obtained, the filter coefficients of the digital filter 15 for realizing the frequency characteristics are calculated.

The correction filter realized by the digital filter 15 has eight viewing positions set in the cabin space 2 of the automobile 1 as described in [Means for Solving the Problems]. This realizes that the acoustic characteristics of the original sound become those of the original sound.

From this, in the present invention, as the second stage processing, the calculated filter coefficients are set in the digital filter 15 so that the acoustic characteristics at the eight listening positions are output from the optical disc reproducing apparatus 10. A process is performed to correct the signal.

At this time, since the correction filter G has the size of 4 rows and 8 columns, as shown in FIG. 5, the correction filter G is distributed so that the number of left reproduction signals output from the optical disk reproduction device 10 becomes four. Then, the write reproduction signals output from the optical disk reproducing apparatus 10 are distributed so as to be four, and the eight reproduced signals are processed so as to be input to the digital filter 15.

As described in the above section "Means for Solving the Problems", this distribution signal should be the target sound at eight viewing positions.

As shown in FIG. 6, the correction filter G realized by the digital filter 15 performs a matrix operation with a matrix of a target sound created by distributing a reproduction signal output from the optical disk reproducing apparatus 10, as shown in FIG. By outputting the obtained acoustic signal to the four speakers 3-p (p = 1 to 4), it is possible to confirm that the acoustic characteristics at the eight viewing positions to those of the reproduced signal output from the optical disc reproducing apparatus 10 are obtained. It does.

In the embodiment shown in FIG. 2, the number m of the viewing positions set in the cabin space 2 of the car 1 is the speaker 3-p.
Is assumed to be greater than the number n, but when, for example, only is set as the viewing position, when the two numbers are equal, the size of the correction filter G is 4 rows and 4 columns. As shown in FIG. 7, the processing is performed so that the reproduction signal output from the optical disk reproducing apparatus 10 is directly input to the digital filter 15 without being distributed.

That is, in the example of FIG. 7, when the present invention is not followed, the left reproduction signal output from the optical disk reproducing device 10 is output to the amplifiers 14-1 and 14-3, The present invention is applied by mounting a digital filter 15 for realizing a 4-by-4 correction filter G, since a configuration is adopted in which a write reproduction signal to be output is output to the amplifiers 14-2 and 14-4. If so, processing is performed so as to input the digital filter 15 as it is without distributing it.

In this case, the reproduced signal input to the digital filter 15 should be the target sound at the four viewing positions. From this, as shown in FIG. 8, the correction filter G realized by the digital filter 15 converts the acoustic signal obtained by the matrix operation with the matrix of the target sound defined by the reproduced signal output from the optical disc reproducing apparatus 10 into four signals. By outputting to the speaker 3-p (p = 1 to 4), it is possible to realize that the acoustic characteristics at the four viewing positions are the same as those of the reproduced signal output from the optical disk reproducing device 10.

Next, an experiment conducted to verify the effectiveness of the present invention will be described.

In this experiment, for the sake of simplicity, the optical disk reproducing device outputs only one reproduction signal, and the rear room speaker and the rear right speaker are used as speakers, and the vehicle space 2 This was performed by verifying the correction effect obtained at four viewing positions (specific positions are shown in FIG. 10 described later) set at the rear seat positions.

From these experimental conditions, according to the processing described above,
As shown in FIG. 9, a rear left correction filter G1,
Two correction filters, that is, a rear light correction filter G2 are calculated.

FIG. 10 shows a rear left correction filter G1 for realizing flat correction of the acoustic characteristics of the four viewing positions from the first point to the fourth point using two speakers, the rear left and the rear right. Illustrating the experimental data of
FIG. 11 shows an experiment of a rear right correction filter G2 for realizing flat correction of acoustic characteristics of four viewing positions from a first point to a fourth point using two speakers, a rear left and a rear right. The data is illustrated.

FIG. 12 shows a case where the rear left speaker and the cabin space 2 when the correction filter G1 is not applied.
Experimental data on the sound pressure F1 incident on the
1 shows experimental data of the sound pressure F1 ^* incident from the rear left loudspeaker to the vehicle interior space 2 in the case of applying No. 1. In the figure, the data indicated by the solid line is the experimental data of the incident sound pressure F1, and the data indicated by the broken line is the experimental data of the incident sound pressure F1 ^* .

FIG. 13 shows experimental data of sound pressure F2 incident from the loudspeaker of the rear light to the cabin space 2 when the correction filter G2 is not applied, and the rear light when the correction filter G2 is applied. The experimental data of the sound pressure F2 ^* incident from the speaker to the cabin space 2 is shown. In the figure, the data indicated by the solid line is the experimental data of the incident sound pressure F2, and the data indicated by the broken line is the experimental data of the incident sound pressure F2 ^* .

FIG. 14 shows experimental data of sound pressure response at the first viewing position at this time, and FIG. 15 shows experimental data of sound pressure response at the second viewing position at this time. FIG. 16 shows experimental data of sound pressure response at the third viewing position at this time, and FIG.
The experimental data of the sound pressure response at the fourth viewing position at this time is shown. Here, the solid line is experimental data of sound pressure before correction, and the broken line is experimental data of sound pressure after correction.

As can be seen from the experimental results shown in FIGS. 14 to 17, the effectiveness of the present invention was verified by applying the correction filter to verify that the acoustic characteristics at the four viewing positions could be corrected flat. .

FIG. 18 shows the experimental data shown in FIG. 14 converted to an actual vehicle and displayed in a 1/3 octave band.
9 shows the experimental data of FIG. 15 converted to an actual vehicle and displayed as a 1/3 octave band, and FIG. 20 shows the experimental data of FIG. 16 converted to an actual vehicle and displayed as a 1/3 octave band. FIG. 21 illustrates the experimental data of FIG. 17 converted into an actual vehicle and displayed in a 1/3 octave band.
Here, the black color is the experimental data of the sound pressure before the correction, and the hatched lines are the experimental data of the sound pressure after the correction.

As shown in FIG. 18, the sensory evaluation of the acoustic characteristic at the first viewing position was increased from 4 points to 0 point by applying the correction filter. Further, as shown in FIG. 19, the sensory evaluation of the acoustic characteristics at the second viewing position increased from two points to zero due to the application of the correction filter. Further, as shown in FIG. 20, the sensory evaluation of the acoustic characteristics at the third viewing position increased from 2 points to 0 point by applying the correction filter. Further, as shown in FIG. 21, the sensory evaluation of the acoustic characteristics at the third viewing position is
The value increased from 3 points to 0 point due to the application of the correction filter.

This sensory evaluation was performed by comparing the original sound in the listening room. First, sound is recorded near the speakers. Next, a comparison was made between the actual sound and the reproduced sound at the listener's seat, and the evaluator rated the sound with the 0-point when he felt the same sound as the original sound and 4 points when he felt it as a different sound. The evaluator took the average value as 10 adult boys in their 20s.

From the sensory evaluation of the acoustic characteristics, it was verified that the acoustic characteristics became flat by using the correction filter. That is, according to the present invention, using two speakers, the acoustic characteristics at four viewing positions from the first point to the fourth point set as a number larger than the number of speakers are corrected to be flat. We were able to verify that it was possible.

The experimental data of the corrected sound pressure shown in FIGS. 18 to 21 may not seem to show a sufficiently flat characteristic at first glance, but from the viewpoint of the human auditory characteristics, the flatness of such a flat sound is not enough. It is sufficient to have the property, and the above-mentioned sensory evaluation can be obtained by realizing this flatness.

The present invention has been described with reference to the illustrated embodiments.
The present invention is not limited to this. For example, in the embodiment, a method of measuring a transfer function matrix by outputting a pseudo-random signal to only one speaker 3-p is used. However, transmission is performed while outputting a pseudo-random signal to all speakers 3-p. A method of measuring a function matrix may be used.

[0082]

As described above, according to the present invention,
Even if the number of viewing positions increases, the acoustic characteristics at a predetermined viewing position in a closed space can be corrected to those of the original sound without increasing the number of speakers according to a simple configuration.

Then, even when the number of viewing positions is larger than the number of speakers, the correction can be realized.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram of the present invention.

FIG. 2 is an embodiment of the present invention.

FIG. 3 is an embodiment of the present invention.

FIG. 4 is an embodiment of a processing flow executed by a control unit;

FIG. 5 is an embodiment of the present invention.

FIG. 6 is an explanatory diagram of a correction process.

FIG. 7 is an embodiment of the present invention.

FIG. 8 is an explanatory diagram of a correction process.

FIG. 9 is an explanatory diagram of experimental conditions.

FIG. 10 is an explanatory diagram of experimental data.

FIG. 11 is an explanatory diagram of experimental data.

FIG. 12 is an explanatory diagram of experimental data.

FIG. 13 is an explanatory diagram of experimental data.

FIG. 14 is an explanatory diagram of experimental data.

FIG. 15 is an explanatory diagram of experimental data.

FIG. 16 is an explanatory diagram of experimental data.

FIG. 17 is an explanatory diagram of experimental data.

FIG. 18 is an explanatory diagram of experimental data.

FIG. 19 is an explanatory diagram of experimental data.

FIG. 20 is an explanatory diagram of experimental data.

FIG. 21 is an explanatory diagram of experimental data.

[Description of Signs] 1 Automobile 2 Cabin space 3 Speaker 10 Optical disk playback device 11 Control unit 12 Display 13 Operation key group 14 Amplifier 15 Digital filter 16 Fourier transform circuit 17 Selector circuit 18 A / D converter 19 Waveform memory

 ────────────────────────────────────────────────── ─── Continued on the front page (72) Inventor Takeshi Toi 1-13-27 Kasuga, Bunkyo-ku, Tokyo F-term in the Faculty of Science and Engineering, Chuo University 5D020 AD02 CE02 5D062 BB03 BB07 CC02 CC12 CC16

Claims (5)

[Claims] 1. A sound field correction method for correcting an acoustic characteristic at a predetermined viewing position in a closed space to be that of an original sound, comprising: a sound pressure output from a speaker; and a sound measured at the viewing position. A transfer function matrix of the space is obtained by measuring a response sound pressure with respect to pressure, a correction filter for realizing the correction is calculated from the transfer function matrix, and an acoustic signal output to the speaker using the correction filter And outputting the same to the speaker. 2. The sound field correction method according to claim 1, wherein when the number of viewing positions is equal to the number of speakers, the correction filter is calculated by calculating an inverse matrix of the transfer function matrix. Calculating the sound field correction method. 3. The sound field correction method according to claim 1, wherein, when the number of viewing positions is larger than the number of speakers, an inverse matrix of a matrix operation of the transfer function matrix and its transposed matrix. By calculating a matrix operation of the transposed matrix and the inverse matrix,
A sound field correction method comprising: calculating the correction filter; distributing an acoustic signal to be output to the speaker; and inputting the audio signal to the correction filter. 4. An audio apparatus for outputting an acoustic signal to a speaker disposed in a closed space, wherein a sound pressure output from the speaker and a response sound pressure to the sound pressure measured at a predetermined viewing position in the space. And a correction filter that is calculated from the transfer function matrix of the space obtained from the above, and corrects the sound characteristic at the viewing position to be that of the original sound. The sound signal output to the speaker is input to the correction filter. And an audio signal output from the correction filter in response to the input to the speaker. 5. The audio device according to claim 4, wherein when the number of viewing positions is larger than the number of speakers, an audio signal output to the speakers is distributed and input to the correction filter. An audio device characterized by being configured as described above.