JP6977376B2 - 3D sound field control device and method - Google Patents

Description

The present invention relates to an apparatus and a method for generating a sound field in a closed region in a three-dimensional space.

Boundary sound field control is a technique for generating a desired sound field in a three-dimensional space. In this boundary sound field control, the same sound field as the sound field in the closed region of the original sound field is generated in the closed region of the reproduced sound field different from the original sound field by the following procedure.
(1) In the original sound field, the sound generated at a plurality of control points on the boundary surface surrounding the closed region is recorded.
(2) In the reproduced sound field, assuming the same closed area as the closed area of the original sound field, a plurality of speakers that emit sound from the outside of the closed area toward the closed area are arranged.
(3) In the reproduced sound field, the transfer function from a plurality of speakers to a plurality of control points on the boundary surface of the closed region is obtained, and the inverse function thereof is obtained.
(4) A plurality of sound signals capable of generating the same sound as the sound generated at the plurality of control points in the closed region of the original sound field at the plurality of control points on the boundary surface of the closed region of the reproduced sound field are reversed. Calculated using a function and supplied to multiple speakers.
As a technical document relating to this boundary sound field control, there is, for example, Patent Document 1.

Japanese Unexamined Patent Publication No. 2012-10011

By the way, in the above-mentioned boundary sound field control, the frequency of the controllable sound depends on the control point interval d. Specifically, in boundary sound field control, it is desirable that the control point interval d be 1/4 or less of the shortest wavelength of the sound to be controlled. When the speed of sound is 340.5 m / s (the speed of sound at a temperature of 15 ° C.), the wavelength of the sound having a frequency of 100 Hz is 3.4 m. Therefore, in order to properly control the boundary sound field of 100 Hz sound, it is necessary to set the control point interval d to 85 cm or less. Similarly, in order to properly control the boundary sound field of 1 kHz sound, it is necessary to set the control point interval d to 8.5 cm or less. Therefore, in order to control the boundary sound field of sound in a wide frequency range, it is necessary to arrange a large number of control points on one boundary surface at short control point intervals determined by the upper limit frequency of the frequency range, and the boundary sound field. There is a problem that the amount of calculation for control increases remarkably.

The present invention has been made in view of the above circumstances, and provides a technical means for realizing boundary sound field control capable of controlling sound in a wide frequency range without causing a significant increase in the amount of calculation. With the goal.

INDUSTRIAL APPLICABILITY The present invention includes a sound field control means for executing boundary sound field control for generating a desired sound at a plurality of control points on a boundary surface indicating a boundary of a closed region in a reproduced sound field, and the sound field control means is described. Provided is a three-dimensional sound field control device characterized in that the interval between control points is controlled by switching the boundary surface according to the frequency of the sound.

According to the present invention, the sound field control means controls the interval between control points by switching the boundary surface according to the frequency of the sound, so that it is not necessary to arrange a large number of control points on one boundary surface, which is wide. Boundary sound field control over the frequency range can be performed. Therefore, according to the present invention, it is possible to control the boundary sound field of sound in a wide frequency range without causing a significant increase in the amount of calculation.

It is a figure which shows the structure of the 3D sound field control system using the 3D sound field control apparatus which is 1st Embodiment of this invention. It is a flowchart which shows the operation of the signal conversion part of the 3D sound field control device. It is a block diagram which shows the structure of a part of the signal conversion part of the 3D sound field control apparatus which is 2nd Embodiment of this invention. It is a figure which shows the processing content of the signal conversion part. It is a figure explaining the control filter which stores in the control filter storage part of the 3D sound field control apparatus which is 3rd Embodiment of this invention.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

<First Embodiment>
FIG. 1 is a diagram showing a configuration of a three-dimensional sound field control system using the three-dimensional sound field control device 100 according to the first embodiment of the present invention. The three-dimensional sound field control device 100 according to the present embodiment is a device that executes control for generating the same sound field as the sound field generated in the original sound field 1 in the reproduced sound field 2.

As shown in FIG. 1, in the original sound field 1, a spherical low-frequency boundary surface BLo surrounding a closed region and a spherical high-frequency boundary surface included inside the low-frequency boundary surface BLo are included in the original sound field 1. BHo is provided. Here, the low-frequency boundary surface BLo is a boundary surface provided for sound field control in the low frequency range below a predetermined cutoff frequency, and the high-frequency boundary surface BHo is above the same cutoff frequency. It is a boundary surface provided for sound field control in the high frequency range. These low-frequency boundary surface BLo and high-frequency boundary surface BHo are virtual boundary surfaces without substance.

The low-frequency boundary surface BLo is provided with virtual M (M is an integer of 2 or more) control points PL _m (m = 0 to M-1) at an almost uniform density, and these controls are provided. A microphone is placed at the point. The high-frequency boundary surface BHo is also provided with virtual M control points PH _m (m = 0 to M-1) at almost uniform densities, and microphones are arranged at these control points. .. Here, the surface area of the high-frequency boundary surface BHo is smaller than the surface area of the low-frequency boundary surface BLo, and the number of control points arranged on each boundary surface is the same. Therefore, the control point spacing in the high frequency boundary surface BHo is shorter than the control point spacing in the low frequency boundary surface BLo.

A recording device 11 is provided in the original sound field 1. The recording device 11 has an M channel sound signal output from each microphone at _{the control point PL m} (m = 0 to M-1) on the low frequency boundary surface BLo, and control on the high frequency boundary surface BHo. The sound signal of the M channel output from each microphone at the point PH _m (m = 0 to M-1) is converted into a digital format and stored.

The reproduced sound field 2 has a low-frequency boundary surface BLr having the same shape and size as the low-frequency boundary surface BLo of the original sound field 1, and a high-frequency boundary surface BHo having the same shape and size as the original sound field 1. The boundary surface BHr for high frequencies is provided. Here, the relative position of the high frequency boundary surface BHr with respect to the low frequency boundary surface BLr is the same as the relative position of the high frequency boundary surface BHo with respect to the low frequency boundary surface BLo. The listener's head is located in a closed area surrounded by the high frequency interface BHr.

The low-frequency boundary surface BLr is provided with virtual M control points QL _m (m = 0 to M-1) having a substantially uniform density. Further, the boundary surface BHr for high frequencies is also provided with virtual M control points QH _m (m = 0 to M-1) having a substantially uniform density. Here, each relative position of the control point QL _m (m = 0 to M-1) on the low frequency boundary surface BLr is the control point PL _m (m = 0 to M-1) on the low frequency boundary surface BLo. Same as each relative position. Further, each relative position of the control point QH _m (m = 0 to M-1) on the high frequency boundary surface BHr is each of the control points PH _m (m = 0 to M-1) on the high frequency boundary surface BHo. Same as relative position.

In the reproduction sound field 2, outside the low frequency boundary surface BLr, N sounds are emitted toward the closed region in the low frequency boundary surface BLr and the closed region in the high frequency boundary surface BHr (N is _{Speakers SP n} (n = 0 to N-1) (an integer of 2 or more) are arranged. Further, an _{amplifier An} (n = 0 to N-1) is connected to _{these speakers SP n} (n = 0 to N-1). An N-channel digital sound signal is input from the three-dimensional sound field control device 100 to the D / A conversion unit 21. The D / A conversion unit 21 converts the N-channel digital sound signal into an N-channel analog sound signal _{and supplies it to the amplifier An} (n = 0 to N-1). Amplifier _{A n (n = 0~N-1} ) , based on the analog sound signal supplied to each drive each speaker _{SP n (n = 0~N-1} ).

The three-dimensional sound field control device 100 executes low-frequency boundary sound field control and high-frequency boundary sound field control by using the speaker SPn (n = 0 to N-1) arranged in the reproduced sound field 2. It is a device to do. Here, in the low-frequency boundary sound field control, the same sound field as the low-frequency component of the sound field in the low-frequency boundary surface BLo generated in the original sound field 1 is placed in the low-frequency boundary surface BLr of the reproduced sound field 2. appear. Further, in the high-frequency boundary sound field control, the same sound field as the high-frequency component of the sound field in the high-frequency boundary surface BHo generated in the original sound field 1 is generated in the high-frequency boundary surface BHr of the reproduced sound field 2. do.

As shown in FIG. 1, the three-dimensional sound field control device 100 includes an operation display unit 101, a sound signal acquisition unit 102, a control filter storage unit 103, and a signal conversion unit 104.

The operation display unit 101 is composed of an operator such as a keyboard and a mouse, and a display unit such as a liquid crystal display. The sound signal acquisition unit 102 is a device that reads digital sound signals of two M channels from the recording device 11.

The control filter storage unit 103 has a low frequency control filter matrix GL (k) = GL _{m, n} (k) (m = 0 to M-1, n = 0 to N-1), which is a function of the angular frequency k. And the high frequency control filter matrix GH (k) = GH _{m, n} (k) (m = 0 to M-1, n = 0 to N-1) are stored.

_{Here, the low-frequency control filters GL m and n} (k), which are elements of the low-frequency control filter matrix GL (k), are _{impulses from the speaker SP n to the control point QL m} on the low-frequency boundary surface BLr. It is an inverse function of _{the transfer functions HL m and n} (k) obtained by applying the Fourier transform to the response (signal in the time domain). _{Further, the high frequency control filters GH m and n} (k), which are elements of the high frequency control filter matrix GH (k), provide an impulse response _{from the speaker SP n to the control point QH m} on the high frequency boundary surface BHr. It is an inverse function of _{the transfer functions HH m, n} (k) obtained by applying the Fourier transform to (the signal in the time domain).

The signal conversion unit 104 is composed of, for example, a DSP (Digital Signal Processor). The signal conversion unit 104 is a sound field control means for executing boundary sound field control for generating a desired sound at a plurality of control points on the boundary surface in the reproduced sound field 2, and is a boundary according to the frequency of the sound. It functions as a sound field control means for controlling the interval between control points by switching surfaces.

Specifically, the signal conversion unit 104 executes low-frequency boundary sound field control and high-frequency boundary sound field control. Here, in the low frequency boundary sound field control, the _{sound signal supplied to the speaker SP n} (n = 0 to N-1) is the control point PL _m (the control point PL m () on the low frequency boundary surface BLo of the original sound field 1. Reproduces the same sound as the low frequency component of the sound generated in m = 0 to M-1) below the cutoff frequency. Control point QL _m (m = 0 to M-) on the low frequency boundary surface BLr of the sound field 2. Generate the sound signal generated in 1). Further, in the high frequency boundary sound field control, the _{sound signal supplied to the speaker SP n} (n = 0 to N-1) is the control point PH _m (m) on the high frequency boundary surface BHo of the original sound field 1. = 0 to M-1) Reproduces the same high-frequency component of the sound generated in the above cutoff frequency or higher. Control point QHm (m = 0 to M-1) on the high-frequency boundary surface BHr of the sound field 2. Generates the sound signal generated in.

In FIG. 1, in the box showing the signal transforming unit 104, HPF (High Pass Filter; high pass filter) 41H, LPF (Low Pass Filter; low pass filter) 41L, and FT (Fourier Transform; Fourier transform) ) Units 42H and 42L, matrix multiplication units 43H and 43L, IFT (Inverse Fourier Transform) units 44H and 44L, and addition units 45 are shown. These are functions realized by the signal conversion unit 104, which is a DSP, executing a microprogram.

In the signal conversion unit 104, the LPF 41L, the FT unit 42L, the matrix multiplication unit 43L, and the IFT unit 44L are means for realizing low-frequency boundary sound field control. Further, the HPF 41H, the FT unit 42H, the matrix multiplication unit 43H, and the IFT unit 44H are means for realizing high-frequency boundary sound field control. Then, the addition unit 45 adds the sound signal obtained by the low-frequency boundary sound field control and the sound signal obtained by the high-frequency boundary sound field control, and adds the speaker SPn (n = 0 to N−) of the reproduction sound field 2. It is a means for generating a sound signal for driving 1). The details of these means will be clarified in the operation description of the present embodiment in order to avoid duplication of description.

FIG. 2 is a flowchart showing the operation of the signal conversion unit 104 in the present embodiment. Hereinafter, the operation of the present embodiment will be described with reference to FIG. 2.

When a predetermined instruction is given via the operation display unit 101, the signal conversion unit 104 starts parallel execution of the low-frequency boundary sound field control process S10L and the high-frequency boundary sound field control process S10H shown in FIG. Various modes can be considered for this parallel execution. In a preferred embodiment, the signal conversion unit 104 has at least two arithmetic processing means, and the low-frequency boundary sound field control processing S10L and the high-frequency boundary sound field control processing are performed by utilizing two of the arithmetic processing means. S10H is executed in parallel. In another preferred embodiment, the signal conversion unit 104 alternately executes the low-frequency boundary sound field control process S10L and the high-frequency boundary sound field control process S10H by utilizing a common arithmetic processing means by time division control.

The processing contents of the low-frequency boundary sound field control processing S10L are as follows. First, in step S11L, the LPF 41L transmits the digital sound signal of the M channel picked up at _{the control point PL m (m = 0 to M-1) on the low frequency boundary surface BLo via the sound signal acquisition unit 102.} get. Next, in step S12L, the LPF 41L performs low-pass processing for removing components in the band above the cutoff frequency on the digital sound signal of the M channel acquired by the sound signal acquisition unit 102, and outputs the signal. Next, in step S13L, the FT unit 42L divides each of the digital sound signals of the M channel output from the LPF 41L into frames having a predetermined time length, performs Fourier transform in frame units, and performs a Fourier transform coefficient sequence XL _{m of the M channel.} (K) Outputs (m = 0 to M-1).

Next, in step S14L, the matrix multiplication unit 43L is a low-frequency control filter matrix GL (k) stored in the control filter storage unit 103 and a Fourier transform coefficient sequence XL _m (k) (m) output from the FT unit 42L. = 0 to M-1), the matrix multiplication process shown in the following equation is executed, and the N-channel Fourier transform coefficient sequence YL _n (k) (n = 0 to N-1) is output.

Here, the low frequency control filters GL _{m, n} (k) are inverse functions of the transfer function _{from the speaker SP n} to the control point QL _{m. Therefore, the Fourier transform coefficient sequence YL n} (k) obtained by multiplying the Fourier transform coefficient sequence XL _m (k) by the low frequency control filters GL _{m, n} (k) is obtained by the Fourier transform at the control point QL _m . It indicates a sound signal for the _{speaker SP n} capable of generating a sound corresponding to the coefficient sequence XL _{m (k).}

Next, in step S15L, the IFT unit _{44L performs an inverse Fourier transform on each of the Fourier transform coefficient sequences YL n} (k) (n = 0 to N-1) of the N channels, and lowers the N channels by one frame each. Outputs a regional digital sound signal.
The above is the processing content of the low frequency boundary sound field control processing S10L.

The processing contents of the high-frequency boundary sound field control processing S10H are as follows. First, in step S11H, the HPF 41H transmits the digital sound signal of the M channel picked up at _{the control point PH m (m = 0 to M-1) on the high frequency boundary surface BHo via the sound signal acquisition unit 102.} get. Next, in step S12H, the HPF 41H performs high frequency pass processing for removing components in the band below the cutoff frequency on the digital sound signal of the M channel acquired by the sound signal acquisition unit 102, and outputs the signal. Next, in step S13H, the FT unit 42H divides each of the digital sound signals of the M channel output from the HPF 41H into frames having a predetermined time length, performs a Fourier transform in frame units, and performs a Fourier transform coefficient sequence XH _{m of the M channel.} (K) Outputs (m = 0 to M-1).

_{Next, in step S14H, the matrix multiplication unit 43H has a Fourier transform coefficient sequence XH m} (k) (m) output from the high-frequency control filter matrix GH (k) stored in the control filter storage unit 103 and the FT unit 42H. = 0 to M-1), the matrix multiplication process shown in the following equation is executed, and the N-channel Fourier transform coefficient sequence YH _n (k) (n = 0 to N-1) is output.

Here, the high frequency control filters GH _{m, n} (k) are inverse functions of the transfer function _{from the speaker SP n} to the control point QH _{m. Therefore, the Fourier transform coefficient sequence YH n} (k) obtained by multiplying the Fourier transform coefficient sequence XH _m (k) by the high frequency control filters GH _{m, n} (k) is obtained by the Fourier transform at the control point QH _m . It shows a sound signal for a _{speaker SP n} capable of generating a sound corresponding to a coefficient sequence XH _{m (k).}

Next, in step S15H, the IFT unit 44H _{performs an inverse Fourier transform on each of the Fourier transform coefficient sequences YH n} (k) (n = 0 to N-1) of the N channels, and the height of the N channels for each frame is high. Outputs a regional digital sound signal.
The above is the processing content of the high frequency boundary sound field control processing S10H.

When the low-frequency boundary sound field control process S10L and the high-frequency boundary sound field control process S10H for one frame are completed, the processing of the signal conversion unit 104 proceeds to step S20. In step S20, the addition unit 45 adds the N-channel low-frequency digital sound signal obtained from the IFT unit 44L and the N-channel high-frequency digital sound signal obtained from the IFT unit 44H between the same channels, and N The digital sound signal of the channel is supplied to the D / A conversion unit 21.

When the process of step S20 is completed, the signal conversion unit 104 determines whether or not the low-frequency boundary sound field control process S10L and the high-frequency boundary sound field control process S10H have been completed for all the sound signals stored in the recording device 11. Is determined (step S30). When this determination result is "NO", the signal conversion unit 104 re-executes the low-frequency boundary sound field control process S10L and the high-frequency boundary sound field control process S10H. If the determination result in step S30 is "YES", the processing of the signal conversion unit 104 ends.

Next, the effect of this embodiment will be described. In boundary sound field control, there is only one boundary surface. If the boundary sound field control is performed in the high frequency range and the low frequency range using this one boundary surface, a large number of control points are provided on the boundary surface, and the boundary sound field control using all the control points in the high frequency range is performed. In the low frequency range, for example, every other control point is used, and the boundary sound field is controlled in a state where the control points are long. In this case, the size of the control filter matrix is different between the high frequency band and the low frequency band because the number of control points used for the boundary sound field control is different. Therefore, different operations are performed in the high frequency range and the low frequency range, which causes a problem that the boundary sound field control becomes complicated. Further, if there is only one boundary surface, there is a problem that the listener cannot move in the reproduced sound field and control the boundary sound field for the listener when the listener deviates from the boundary surface.

On the other hand, in the three-dimensional sound field control device 100 according to the present embodiment, the signal conversion unit 104 controls the interval between control points by switching the boundary surface according to the frequency of the sound. Specifically, the signal conversion unit 104 controls the low-frequency boundary sound field corresponding to the boundary surface BLr (= BLo) having a long interval between control points in the band below the cutoff frequency, and performs the cutoff. For the band above the frequency, high-frequency boundary sound field control corresponding to the boundary surface BHr (= BHo) in which the intervals between control points are short is performed. Therefore, according to the present embodiment, the boundary sound field control can be realized over a wide frequency band without increasing the number of control points as a whole. Further, in the present embodiment, the number of control points arranged on the boundary surface BLr (= BLo) and the number of control points arranged on the boundary surface BHr (= BHo) are the same. Therefore, the sizes of the low frequency control filter matrix GL (k) and the high frequency control filter matrix GH (k) are the same. Therefore, in the present embodiment, the boundary sound field can be controlled by the same calculation in the low frequency range and the high frequency band. Further, in the present embodiment, the low-frequency boundary sound field control process S10L and the high-frequency boundary sound field control process S10H of FIG. 2 can be synchronously executed in parallel without causing unnecessary waiting time. The addition process (step S20) can be executed. Therefore, according to the present embodiment, the boundary sound field control in the low frequency range and the boundary sound field control in the high frequency range can be efficiently executed. Further, according to the present embodiment, even if the listener moves in the reproduced sound field and deviates from the high frequency boundary surface, if the listener is within the low frequency boundary surface, the boundary in the low frequency range is provided for the listener. Sound field control can be performed.

<Second Embodiment>
FIG. 3 is a block diagram showing a part of the configuration of the signal conversion unit 104A of the three-dimensional sound field control device according to the second embodiment of the present invention. The signal conversion unit 104A has a configuration in which a crossfade unit 46 including a multiplication unit 46L and 46H is added to the signal conversion unit 104 of the first embodiment.

In the first embodiment, in order to properly control the boundary sound field, the cutoff frequency indicating the upper limit of the pass band of LPF41L and the cutoff frequency indicating the lower limit of the pass band of HPF41H exactly match. It is desirable to be there. However, it is difficult to match both cutoff frequencies exactly. Then, for example, when the cutoff frequency fc1 indicating the upper limit of the pass band of LPF41L is lower than the cutoff frequency fc2 indicating the lower limit of the pass band of HPF41H, the sound in the band from the frequency fc1 to the frequency fc2 is reproduced in the reproduction sound field 2. There is a problem that is not done. Therefore, in the second embodiment of the present invention, this defect is solved as follows.

First, in the present embodiment, as shown in FIG. 4A, the cutoff frequency fc1 indicating the upper limit of the pass band of the LPF 41L is set higher than the cutoff frequency fc2 indicating the lower limit of the pass band of the HPF 41H, and the low frequency boundary is formed. A part of the upper limit side of the pass band of the sound field control process S10L and a part of the lower limit side of the pass band of the high frequency boundary sound field control process S10H are overlapped.

Further, in the present embodiment, in the frequency domain where the pass band of the low frequency boundary sound field control process S10L and the pass band of the high frequency boundary sound field control process S10H overlap, the output signal of the low frequency boundary sound field control process S10L ( A crossfade is performed between the signal in the frequency domain and the output signal (signal in the frequency domain) of the high frequency boundary sound field control process S10H. Specifically, in the present embodiment, the Fourier transform coefficient sequence YL _n (k) (n = 0 to N-1) output from the matrix multiplication unit 43L is shown in FIG. 4 (b) by the multiplication unit 46L. The coefficient α (k) is multiplied and supplied to the IFT unit 44L. Further, in the present embodiment, the coefficient β (b) shown in FIG. 4 (b) is obtained by the multiplication unit 46H with respect to the _{Fourier transform coefficient sequence YH n (k) (n = 0 to N-1) output from the matrix multiplication unit 43H.} k) is multiplied and supplied to the IFT unit 44H.

乗算部４６Ｌは、このフーリエ変換係数列の各フーリエ変換係数に各々の角周波数ｋに対応付けられた係数α（ｋ_０）、α（ｋ_１）、…、α（ｋ_Ｌ−１）を乗算することにより次のフーリエ変換係数列ＹＬ’ｎ（ｋ）を算出してＩＦＴ部４４Ｌに供給するのである。

フーリエ変換係数列ＹＨ_ｎ（ｋ）（ｎ＝０〜Ｎ−１）についても同様である。 More specifically, it is as follows. It is assumed that one Fourier transform coefficient sequence YL _n (k) consists of L Fourier transform coefficients (L is an integer of 2 or more) as follows, for example.

The multiplication unit 46L multiplies each Fourier transform coefficient of this Fourier transform coefficient sequence by the coefficients α (k ₀ ), α (k ₁ ), ..., α (k _L-1 ) associated with each angular frequency k. By doing so, the following Fourier transform coefficient sequence YL'n (k) is calculated and supplied to the IFT unit 44L.

The same applies to the Fourier transform coefficient sequence YH _n (k) (n = 0 to N-1).

Here, in the band of the cutoff frequency fc2 or less, α (k) = 1 and β (k) = 0. Therefore, in this band, only the Fourier transform coefficient sequence YL _n (k) (n = 0 to N-1) is supplied to the IFT unit 44L as it is.

On the other hand, in the band having a cutoff frequency of fc1 or higher, α (k) = 0 and β (k) = 1. Therefore, in this band, only the Fourier transform coefficient sequence YHn (k) (n = 0 to N-1) is supplied to the IFT unit 44H as it is.

In the band having a cutoff frequency of fc2 or more and a cutoff frequency of fc1 or less, α (k) gradually decreases from 1 to 0 and β (k) gradually decreases from 0 as the frequency increases. It gradually increases toward 1. Therefore, in this band, as the angular frequency k increases, the Fourier transform coefficient sequence α (k) YL _n (k) after coefficient multiplication gradually decreases, and the Fourier transform coefficient sequence β (k) YHn after coefficient multiplication gradually decreases. A crossfade is performed in which (k) gradually increases.

Therefore, according to the present embodiment, it is possible to solve the problem that the sound in the band from the frequency fc1 to the frequency fc2 is not reproduced in the reproduced sound field 2. Further, in the present embodiment, since the size of the control filter matrix used by the matrix multiplication unit 43L and 43H for multiplication is the same, the multiplication process of the coefficient α (k) by the multiplication unit 46L and the coefficient β (k) by the multiplication unit 46H. ) Can be executed in synchronization with the multiplication process. Therefore, there is an advantage that the crossfade processing can be easily controlled by using the multiplication units 46L and 46H, and the boundary sound field control can be efficiently executed.

<Third Embodiment>
FIG. 5 is a diagram illustrating a control filter stored in the control filter storage unit 103 of the three-dimensional sound field control device according to the third embodiment of the present invention. In the second embodiment, the crossfade unit 46 is added to the signal conversion unit 104 of the first embodiment. On the other hand, in the present embodiment, such a crossfade unit 46 is not added, and instead, the control filter matrix stored in the control filter storage unit 103 is processed in advance, as in the second embodiment. Effect.

As shown in FIG. 5, in the present embodiment, the same coefficients as those in the second embodiment _{are applied to the low frequency control filters GL m, n} (k) which are elements of the low frequency control filter matrix GL (k). α (k) is multiplied, and the low-frequency control filters GL'm _{, n} (k), which are the multiplication results, are stored in the control filter storage unit 103 instead of the low-frequency control filters GL _{m, n (k).} .. Further, in the present embodiment, the same coefficient β (k) as in the second embodiment _{is applied to the high frequency control filter GH m, n} (k) which is an element of the high frequency control filter matrix GH (k). The multiplication is performed, and the high-frequency control filter GH'm _{, n} (k), which is the multiplication result, is stored in the control filter storage unit 103 instead of the high-frequency control filter GH _{m, n (k).}

このフィルタ係数列の各フィルタ係数に係数α（ｋ_０）、α（ｋ_１）、…、α（ｋ_Ｌ−１）を乗算し、次のような低域用制御フィルタＧＬ’_ｍ，ｎ（ｋ）を算出して制御フィルタ記憶部１０３に格納するのである。

高域用制御フィルタＧＨ_ｍ，ｎ（ｋ）についても同様である。 Specifically, it is as follows. It is assumed that one low frequency control filter GL _{m, n} (k) is, for example, the following filter coefficient sequence.

By multiplying each filter coefficient of this filter coefficient sequence by the coefficients α (k ₀ ), α (k ₁ ), ..., α (k _L-1 ), the following low-frequency control filter GL'm _{, n} ( k) is calculated and stored in the control filter storage unit 103.

The same applies to the high frequency control filters GH _{m and n (k).}

The operation of this embodiment is the same as that of the first embodiment except for the storage contents of the control filter storage unit 103.

Also in this embodiment, the same effect as that of the second embodiment can be obtained. Further, according to the present embodiment, since the crossfade portion 46 of the second embodiment is unnecessary, the same effect as that of the second embodiment can be obtained with a smaller amount of calculation than that of the second embodiment.

<Other embodiments>
Although the first to third embodiments of the present invention have been described above, other embodiments of the present invention can be considered. For example:

_{(1) In each of the above embodiments, a microphone is attached to the control point PL m} (m = 0 to M-1) of the boundary surface BLo of the original sound field _{and the control point PH m} (m = 0 to M-1) of the boundary surface BHo. The sound signals at each control point were acquired by arranging and using these microphones. However, instead of doing so, an acoustic simulation of the propagation state of the sound to each control point of the original sound field may be executed, and a sound signal indicating the sound at each control point may be acquired by this acoustic simulation.

(2) Each of the above embodiments can also be implemented in the following embodiments.

(2-1) Original sound field designation information that specifies a desired original sound field by a user who uses the reproduced sound field, song designation information regarding a song to be played in the original sound field, and a reproduced sound field configuration regarding the configuration of the reproduced sound field. Information (for example, information on the shape, size, and layout of the wall surrounding the reproduced sound field) and _{speaker arrangement information on the arrangement position of the speaker SP n} (n = 0 to N-1) in the reproduced sound field are stored in a server on the network. Send to.

(2-2) In the server, based on the speaker arrangement information, a boundary surface for low frequencies and a boundary surface for high frequencies of an appropriate size that fits inside _{the speaker SP n (n = 0 to N-1) are set.} Set. In addition, the same number of control points are arranged on the boundary surface for low frequencies and the boundary surface for high frequencies. Then, the server obtains an impulse response from each speaker to each control point based on the reproduced sound field configuration information and speaker arrangement information received from the user, and from this impulse response, a control filter matrix for low frequencies and a high frequency range are obtained. Find the control filter matrix. Further, in the server, in the original sound field indicated by the original sound field designation information received from the user, each control point of the boundary surface for low frequencies and each control point of the boundary surface for high frequencies similar to the reproduced sound field are provided to provide a song. A sound signal indicating the sound generated at each control point when the song indicated by the specified information is played is obtained by simulation. Then, the server has the low-frequency control filter matrix obtained in this way, the high-frequency control filter matrix, the sound signal corresponding to each control point of the low-frequency boundary surface, and the high-frequency boundary. A sound signal corresponding to each control point on the surface is transmitted to a user who uses the reproduced sound field.

(2-3) The user gives each information received from the server to, for example, the three-dimensional sound field control device of the first embodiment, and causes the boundary sound field control in the low frequency range and the high frequency range to be executed.

According to this aspect, even if the user does not have the means for calculating the low frequency control filter matrix and the high frequency control filter matrix, the low frequency and high frequency according to the first embodiment. The boundary sound field control can be executed by the three-dimensional sound field control device.

(3) In the first to third embodiments, two boundary surfaces are used and the frequency band of sound is divided into two to control the boundary sound field. However, three or more boundary surfaces are used to control the sound. The boundary sound field may be controlled by dividing the frequency band of the above into three or more.

(4) In the first to third embodiments, a plurality of boundary surfaces are spherical surfaces, but the plurality of boundary surfaces do not necessarily have to be spherical surfaces, and if there is an inclusion relationship between the plurality of boundary surfaces. , It may be a closed curved surface other than a spherical surface.

(5) In the first to third embodiments, the sound signal obtained by _{the boundary sound field control is supplied to the speaker SP n} (n = 0 to N-1) of the reproduced sound field in real time, but the boundary sound field control is performed. The sound signal obtained by the above may be stored in the storage means, and then the sound signal may be read out from the storage means _{and supplied to the speaker SP n} (n = 0 to N-1).

(6) In the first to third embodiments, the boundary sound field is controlled in the reproduced sound field by using the sound signal obtained at the control point of the original sound field, but the scope of application of the present invention is limited to this. It's not something. The present invention is applicable to a three-dimensional sound field control device that controls a boundary sound field that generates an arbitrary desired sound at a control point on the boundary surface of the reproduced sound field.

(7) In the first to third embodiments, using a speaker with a physical as reproduction sound field of the speaker _SP n (n = 0~N-1), a speaker _SP n (n = 0~N-1 ) May be realized by a virtual speaker. Alternatively, the entire speaker SPn (n = 0 to N-1) may be realized as a virtual speaker by playing back headphones.

1 ... Original sound field, 2 ... Playback sound field, 11 ... Recording device, BLo, BLr ... Low frequency boundary surface, BHo, BHr ... High frequency boundary surface, PL _m (m = 0 to M- 1), PH _m (m = 0 to M-1), QL _m (m = 0 to M-1), QH _m (m = 0 to M-1) ... Control point, SP _n (n = 0 to 0) N-1) …… Speaker, _An (n = 0 to N-1) …… Amplifier, 21 …… D / A conversion unit, 100 …… Three-dimensional sound field control device, 101 …… Operation display unit, 102 ... Sound signal acquisition unit, 103 ... Control filter storage unit, 104 ... Signal conversion unit, 41L ... LPF, 41H ... HPF, 42L, 42H ... FT unit, 43L, 43H ... Matrix multiplication unit, 44L , 44H ... IFT part, 45 ... Addition part, 46 ... Crossfade part, 46L, 46H ... Multiplication part.

Claims (7)

The first boundary sound field control that generates sound in the first frequency band at a plurality of control points on the first boundary surface surrounding the closed region in the reproduced sound field, and the first that surrounds the closed region in the reproduced sound field. A second boundary sound field control that generates sound in a second frequency band higher than the first frequency band at a plurality of control points on the second boundary surface having a surface area smaller than that of the first boundary surface. 3 dimensional sound, characterized in that have a sound field control means for performing field controller. The sound field control means is between the plurality of speakers arranged outside the closed area surrounded by the first boundary surface and the closed area surrounded by the second boundary surface and the plurality of control points. The three-dimensional sound field control device according to claim 1, wherein a control filter showing the acoustic characteristics of the above is provided for each boundary surface. The number of control points on the first boundary surface and the number of control points on the second boundary surface are the same.
The sound field control means is
The first boundary sound field control using the low frequency control filter matrix showing the acoustic characteristics between the plurality of speakers and each control point on the first boundary surface,
The matrix showing the acoustic characteristics between the plurality of speakers and each control point on the second boundary surface, and using a high frequency control filter matrix having the same size as the low frequency control filter matrix. Second boundary sound field control and
The three-dimensional sound field control device according to claim 2, wherein the two are synchronized and executed in parallel. The first frequency band and the second frequency band are partially overlapped and adjacent to each other, and the output signals of the first and second sound field control are combined with the first frequency band and the second frequency band. The three-dimensional sound field control device according to any one of claims 1 to 3, wherein the frequency bands of the above are crossfaded in overlapping frequency regions. The three-dimensional sound field control device according to any one of claims 1 to 4, wherein the first boundary surface and the second boundary surface are spherical surfaces. A first boundary sound field control that generates sound in the first frequency band at a plurality of control points on the first boundary surface surrounding a closed region in the reproduced sound field.
Sound in a second frequency band higher than the first frequency band is transmitted to a plurality of control points on a second boundary surface having a surface area smaller than that of the first boundary surface surrounding the closed region in the reproduced sound field. The second boundary sound field control to be generated and
A three-dimensional sound field control method characterized by executing. The number of control points on the first boundary surface and the number of control points on the second boundary surface are the same.
Between a plurality of speakers arranged outside the closed area surrounded by the first boundary surface and the closed area surrounded by the second boundary surface and each control point on the first boundary surface. The first boundary sound field control using the low frequency control filter matrix showing the acoustic characteristics, and
The matrix showing the acoustic characteristics between the plurality of speakers and each control point on the second boundary surface, and using a high frequency control filter matrix having the same size as the low frequency control filter matrix. Second boundary sound field control and
The three-dimensional sound field control method according to claim 6, wherein the two are synchronized and executed in parallel.

Images