JP2666058B2 - Sound pickup reproduction control device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial applications]   This invention can be used in a general room
Performing in a large hall etc. (instrumental performance, singing, etc.)
Sound pickup that can create the atmosphere that you are doing
It relates to a raw control device. [Conventional technology]   Playing instruments such as piano and flute in an ordinary room
When performing a song or singing, it is as if the player himself
I feel like playing in a big hall
Can make the performance even more enjoyable
it can. Professional musicians practice in the test room, etc.
Even if you are in a real hall
(Including the tone) if you can play
You can practice effectively according to the actual performance.   Conventionally, with a device that adds the feeling of a hall or a feeling of expansion
Reverberator, karaoke system, surround
Processors and the like have been proposed. These are shown in FIG.
Source (instrument sounds and singing sounds) 12
Suitable for original sound with processor 14 as troll factor 10
Perform signal processing such as adding reverberation, and
Then, the power is supplied to the speakers 20 and 22.   However, these only add artificial reverberation.
And natural reflections in acoustic spaces such as halls
It is completely different from the
I couldn't get the atmosphere right. [Problems to be solved by the invention]   The present invention solves the problems of the conventional technology.
In a real room, such as a normal room or test room
Sound pickup reproduction control device that can give the atmosphere of playing
It is intended to provide. [Means for solving the problem]   This invention is arranged so as to surround the entire circumference of the playing position
At least four or more speaker reproducing means,
Microphone sound pickup means to pick up sound and reflection in acoustic space
The arrival of each reflected sound determined for each virtual sound source position of the sound
To reflected sound data consisting of incoming direction, delay time and amplitude level
Based on the performance position using each of the speaker reproducing means.
Around the acoustic space or a similar model space
To reproduce a large number of reflections at
Of the sound reproducing means and each of the reflected sound data
Each of the speaker reproducing means, which is obtained from the direction data,
Impulse consisting of delay time and gain of reflected sound to be performed
The response characteristic is determined by the reflected sound parameter of each speaker
Parameter storage means for storing each of the
Based on each reflected sound parameter stored in the meter storage means,
Convolution operation on the sound pickup signal of the microphone sound pickup means
By doing so, a large number of
Each of the speakers generates a reflected sound, and
Reflected sound supplied to the corresponding position of the raw means
Generating means, wherein the acoustic space or
When performing in a similar model space,
The state of generation of a large number of reflected sounds generated by the
The feature is that it can be reproduced almost equally including up to
Is what you do.   Note that “reflected sound data” here refers to the acoustic space
Data that constitutes the reflected sound in the
How the reflected sound comes from the virtual sound source distribution
Data such as direction, distance (= delay time) and amplitude level
is there.   The “reflected sound parameter” refers to the reflected sound data.
Place the reflected sound around the playing position in the room.
Each speaker to simulate multiple speakers.
Parameter to generate the reflected sound to be emitted from
Specifically, these are parameters of delay time and gain.
This reflected sound parameter is set to the reflected sound data and the performance position.
It is determined by the relationship between the position of the speaker and the like. [Action]   According to the solution of the present invention, an actual hole or the like
Simulate the reflected sound of the performance based on the reflected sound data
The atmosphere where the performer is playing in the hall etc.
You can taste. [Principle of the invention]   In the present invention, the direction of arrival of the reflected sound as reflected sound data,
Delay time, amplitude level, etc. are used. These reflected sounds
Data is a virtual sound source in an acoustic space such as a hall
It can be obtained from the distribution. Where the virtual sound source
Is viewed from a specific sound receiving point in an acoustic space such as a hall.
Effective reflected sound source. That is, the actual sound source (real
The sound source at the time. ) Is emitted as a direct sound
In addition to directly reaching the sound receiving point, the sound of walls, ceilings, floors, seats, etc.
Reflects at all reflective parts in space and reaches the sound receiving point
I do. In this case, at the sound receiving point, the reflected sound is
From the sound source that is on the extension of the line connecting the reflection points
This can be regarded as a sound
A virtual sound source at a point, that is, a virtual sound source
Can be.   Therefore, the acoustic space at a certain receiving point is
It can be grasped as the distribution of the virtual sound source at the sound point.
Reflected sound of performance from each virtual sound source even in ordinary rooms
Simulates the sound space,
When you experience the atmosphere of playing in that acoustic space
I can.   The position of the virtual sound source is determined by the direction and distance from the sound receiving point.
Simulate the reflected sound from the virtual sound source
The direct sound collected by the microphone to the virtual sound source.
From the direction, a time delay corresponding to the distance, and the reflected sound
What is necessary is just to emit with the volume according to an amplitude level. And this
For each virtual sound source in the acoustic space
If it is done, it will reproduce the state of playing in that acoustic space
Can be   As for how to find the virtual sound source, the hole to be reproduced
Measure and find impulse response in actual acoustic space
Calculated from the method and the shape of the acoustic space such as a hall
There is a method.   Method to obtain by measurement   The former method is called the so-called four-point method.
There are things that can go wrong. These are the four closest points in the acoustic space
Using the time difference of the impulse response of
The coordinates of the virtual sound source are obtained.   The impulse response is simultaneous from the real sound source and the virtual sound source.
Considered to be a pick-up signal at the receiving point when the pulse was emitted
In the early part of the response, reflected sounds are individually identified without overlapping
Since it can be used, the distribution of the virtual sound source is obtained by using this.   The measurement by the four-point method was performed as shown in FIG.
The impulse response of the sound source 26 in the acoustic space 24
The measurement is made at the four touched sound receiving points o, x, y, and z. These sounds
The necessary condition is that the points o, x, y, z are not on one plane,
In order to facilitate the subsequent processing, as shown in FIG.
The other three sound receiving points x, y,
Arrange so that z forms rectangular coordinates. Distance from origin o
The separation is equal to d.   Take a simple experiment with one reflector installed in an anechoic chamber as an example.
I will tell.   Microphone MIC at each sound receiving point o, x, y, z_o, MIC_x, MIC
_y, MIC_z, Are as shown in FIG. this is,
Direct sound from each microphone MIC_o, MIC_x, MIC_y, MIC_zEach
Time to_o, tx_o, ty_o, tz_oIncident on, reflected sound from reflector
Is to₁, tx₁, ty₁, tz₁This indicates that the light has entered.   FIG. 6 schematically shows the process of the reflected sound.
You. The sound source 26 is in the direction of the y-axis when viewed from the sound receiving points o, x, y, and z
So the direct sound is first the microphone MIC_yIncident on the
Icrophone MIC_o, MIC_x, MIC_zAt almost the same time. this
Therefore, as shown in FIG. ty_o<To_o≒ tx_o≒ tz_o Holds for direct sound, and for reflected sound, ty₁<To₁≒ tz₁<Tx₁ Holds.   Distance ro from each sound receiving point o, x, y, z to the virtual sound source 26 '₁, r
x₁, ry₁, rz₁Is given by the following equation, where v is the sound speed. ro₁＝ v ・ to₁ rx₁= Vtx₁ ry₁= V · ty₁ rz₁= V.tz₁   Set the coordinates of any virtual sound source to (x_n, Y_n, Z_n), That virtual sound source
From the distance to each receiving point o, x, y, z_n, rx_n, r
y_n, rz_nAs the center of each sound receiving point o, x, y, z
The equation of a sphere with on the surface is Xn^Two+ Yn^Two+ Zn^Two= Ro_n ^Two (Xnd)²+ Yn^Two+ Zn^Two= Rx_n ^Two Xn^Two+ (Yn-d)²+ Zn^Two= Ry_n ^Two Xn^Two+ Yn^Two+ (Zn-d)²= Rz_n ^Two Becomes Solving this equation gives Becomes   As described above, the position of the virtual sound source corresponding to each reflected sound
The target can be determined.   In general, the impulse response is not as simple as in FIG.
It has a complex shape with many reflections. specific
Reflected sound creates peaks and impulse responses at each receiving point
To select from, use the cross-correlation of a short section. sand
In other words, microphone MIC_oSection with the output of
The section where the number of relations becomes large_x, MIC
_y, MIC_zThe output time of the reflected sound to
_n, tx_n, ty_n, tz_nTo determine.   The virtual sound source distribution of a certain hall by the above-described four-point method
7 to 9 show an example of the measurement. FIG. 7 shows X-
FIG. 8 shows a projection on a Y plane (horizontal plane), and FIG.
FIG. 9 is a projection view on the XZ plane. O in the figure
The loudness represents the level of the reflected sound, such as the microphone
Lohon MIC_oIs representatively measured.   How to calculate virtual sound source by calculation   As a method to obtain virtual sound sources by calculation instead of measurement
Is a mirror image method. This is as shown in FIG.
The sound is emitted from the real sound source 27, and the sound is received at the receiving point 28.
When receiving, the reflected sound on the wall 24 is placed in the virtual image position
This is considered to be virtually emitted from a certain sound source 30, and
The virtual sound source 30 according to the wall shape of the acoustic sky
Things.   An example of finding the virtual sound source distribution of a certain hall by the mirror image method
Are shown in FIGS. 11 and 12. FIG. 11 shows an XY plane (horizontal).
FIG. 12 is a projection on the YZ plane.
You. In the case of the mirror image method, the amplitude level
Set each according to the distance to the source.   Measured or calculated as described above
The performance from each virtual sound source is based on the virtual sound source distribution data.
When simulating the reflected sound of a performance in a room,
Microphones to collect performances.
Sound, and the collected signal is sent to an appropriate speaker (virtual sound source).
Appropriate time delay (corresponding to direction) (distance to virtual sound source)
And appropriate volume (corresponding to the amplitude level of the reflected sound)
Simulates the reflected sound of the performance by emitting
be able to.   In this case, the performance position in the room and each speaker
Depending on the position, the direction, distance, and level of the reflected sound
Position of the speaker relative to the playing position
(Direction and distance if necessary)
How much sound and delay time from the speaker in the direction reflected sound
Is calculated.   Also, speakers should ideally be in the direction of all virtual sound sources.
Need to be placed in But to do that,
At least the upper hemisphere of the room around the playing position without leakage
It means that speakers must be placed,
is there. Economically, the limit is about 4 to 10 pieces.
Loudspeakers around the room, and
The virtual sound source included in each area is determined by defining the allocation area of the peaker
Simulate the reflected sound of the
Try to rate. According to this method, adjacent switches
Any reflections from the virtual sound source in the middle of the peaker
Strictly because it is emitted on behalf of one simulation
In other words, it is necessary to accurately simulate the direction of a virtual sound source.
It does not mean that if the number of speakers is large enough,
There is no problem in practical use, and the ability to determine the direction of human hearing is limited
Given that, this is enough.   Alternatively, a virtual sound source in the middle between adjacent speakers
If you need to accurately simulate the direction,
The volume distribution between these speakers makes it feasible
is there.   Thus, the reflection from the virtual sound source in the middle of the speaker
Simulate the sound on behalf of any one speaker
And simulation by volume distribution between speakers.
Volume to be emitted from each speaker when
The delay time is described below.   When simulating on behalf of one speaker   FIG. 13 shows eight speakers SP1 to SP around the playing position 34.
P8 is arranged. Here, the acoustic space is adjacent
Line between the center position of the speaker and the playing position 34
Then, it is divided into eight regions d1 to d8 on the horizontal plane. Each area d1 ~
P is the reflected sound at d8_MnThen, at performance position 34,
Shooting sound P_MnSound P of each speaker SP1 to SP8 required to obtain_Mn
(M = 1 to 8) is represented by the following equation.  However NM (M = 1 to 8): Number of virtual sound sources in each area d1 to d8 (=
Number of reflections) U: Unit function t: time τ_n: Delay time of reflected sound   Simulate by volume distribution between adjacent speakers
If you want to   As shown in FIG. 14, for example, four
Speakers SP1 to SP4 are arranged, performance position 38 and each speaker SP
Divide into four quadrants n, m, l, and k by the line connecting 1 to SP4.
Virtual sound sources in the left and right quadrants of each of the players SP1 to SP4
Simulate their reflected sound. That is, the speaker SP
4, simulate the reflected sound in quadrant n with the volume ratio of SP1,
Simulate the reflected sound in quadrant m with the volume ratio of speakers SP1 and SP2.
Rate and the reflection in quadrant l with the volume ratio of speakers SP2 and SP3
Simulates sound and quadrants with the volume ratio of speakers SP3 and SP4
Simulate the reflected sound in k. Simulate each reflected sound
Playback sound P of each speaker SP1 to SP4 necessary for_Ms(M
= 1 to 4) are as follows.   However, P_n, P_m, Pl, Pk: Reflected sound level τ_n, Τ_m, τl, τk: Delay time of reflected sound θ_n, θ_m, τl, τk: Reflected sound on XY plane (horizontal plane)
Direction angle θ₁, θ_Two, θ_Three, θ_Four: Speakers SP1 to SP4 on XY plane
Direction angle τ_Mn, τ_Mm, τ_Ml, τ_Mk: Delay time of each speaker playback sound. ±
Is a correction based on the distance between the performers' ears.
Then we assume the case of 15cm. Nl, Ml, Ll, Kl: Number of virtual sound sources in each quadrant n, m, l, k t: time U: Unit function   In the above formula, the virtual sound in the middle of adjacent speakers
To simulate the direction of the reflected sound from the source
When the signal distribution between them is the COS function shown in Fig. 15,
The linear function shown in Fig. 16 or Fig. 17
Speaker function such as log function shown in Fig. Or speaker characteristics
Use the one that can approximate the reflection direction the most according to
I do.   By the signal distribution described above, the speaker arrangement shown in FIG.
To simulate reflected sounds from all directions
Can be   When performing in an actual hall or the like, the performer
Listening to reflected sound while playing at that position
Therefore, both the sound source and the receiving point
If you set the virtual sound source distribution on the
Using reflected sound parameters of each speaker based on source distribution
By generating reflected sounds, the performer in a normal room
Performing in an atmosphere on a stage such as a hall
it can. Also, not limited to this, the sound source and the receiving point can be variously changed.
The reflected sound parameters based on the virtual sound source distribution
By doing so, you can enjoy performances of various tastes.   In the room, between the speaker and the playing position
Since there is a distance and a time difference occurs, it is emitted from a virtual sound source
This time delay is used to more accurately simulate reflections.
The playback sound from each speaker, taking this into account
To do.   FIG. 18 shows a virtual sound source of a hall using the four-point method.
Of reflected sound data (direction of arrival, distance, amplitude level)
Based on this, the reflected sound was stained with the speaker arrangement shown in FIG.
Should be played from each speaker SP1 to SP4.
Signal P_Mn(M = 1 to 4) was obtained from the above equation (2).
Is the thing   This is the inverse of the signal output from each speaker SP1 to SP4.
Shows the sound structure and impulse in each speaker direction.
You may think of it as a response. Impulse response of neighboring speakers
The answers are interrelated, that is, these speaker methods
The reflected sound located in the distance is correctly
The amplitude level of both impulse responses so that
The delay has been calculated in advance.   Source signals (continuous signals such as record playback signals)
When generating reflected sound by using
For these sample values, these impulse responses are
(For gain and delay time)
Generated (based on the time the sample value was obtained
Count the delay time that generates the reflected sound, and add
Determine the level of each reflected sound with the level multiplied by the gain of
You. ), These reflections obtained for each sample value
If the columns are added to each other at each point in time,
Reflections in the direction are generated and these are
The player at the playing position 38 (Fig. 14)
Has its own virtual sound source distribution
You can enjoy the atmosphere of playing in the hall
You.   Reflected sound production based on reflected sound parameters of impulse response
As the synthesis processing, a method based on a convolution operation described later, etc.
Can be used.   In the present invention, the number of microphones is limited to one.
And a plurality can be used.   Also, pick up sound with the microphone as much as possible in the room.
Desirably, it is not affected by the properties. for that reason
Should have as little reflection as possible in the room itself.
The state can be obtained by performing appropriate sound absorption processing.
You.   Also, in the present invention, the sound picked up by the microphone is
Howling to process and play back through the speakers again
Care must be taken to avoid phenomena. So
For speaker and microphone arrangement for
This will be described in an embodiment described later.   By the way, the reflected sound at the hall etc. is actually about 2-3 seconds
And usually all this
If the sound data is generated by
If you have to store it in memory,
In some cases, the memory capacity becomes enormous. Or
However, the early reflection part (several hundred mm
Seconds), the sound is suddenly interrupted,
The atmosphere of a hall or the like cannot be sufficiently provided.   Therefore, simple filters such as comb filters and all-pass filters
Reverberation composed of repeating (feedback) loops
Use the additional device to generate the middle to late part of the reflected sound
A long reflection sound with a small memory capacity.
You. Reflection in actual halls etc. is sufficient in the initial part
Although it can be clearly recognized as a reflected sound, the later the reflected sound
Are appearing in large numbers, so they become increasingly chaotic,
One becomes unclear. In other words, after the chaos
Is the reflected sound aggregate diffused and statistical?
To reproduce it, you only need to control the amount,
Simple repetition of comb filters and all-pass filters
A reed-like reverberator is sufficient.   FIG. 1 shows the sound pickup reproduction of the present invention utilizing the above principle.
FIG. 3 is a conceptual diagram of a control device. Here, the reproduction
At the hole (for example, Carnegie Hall)
As "reflected sound data", based on the virtual sound source distribution,
Direction of arrival of reflected sound, delay time (corresponding to distance) and amplitude
Calculate the level and simulate these reflected sound data and
Simulate the reflected sound from the speaker arrangement used
Delay time of the reflected sound to be output from each speaker in order to
The amplitude level is used as a “reflected sound parameter” for each speaker.
Control parameters for these reflected sound parameters.
Processor 42 as processor 42, for each virtual sound source
Source (the sound picked up by the microphone) 44 reflected sound
To be reproduced on each speaker to simulate
Generates sound signals and the reflections generated for each of these speakers
Sound signals are transmitted to the speakers 56, 58, 60, via the amplifiers 48, 50, 52, 54.
62 to reflect sound from each virtual sound source.
Is emulating. [Example 1]   FIG. 19 shows an embodiment of the present invention.
The room 80 is subjected to a sound absorbing process to have a dead characteristic. room
At the four corners of 80, speakers 56, 58,
60 and 62 are arranged. Also, there are five
Microphones 81, 82, 83, 84, 85 are arranged. My
The microphone 81 is positioned facing the instrument 90 and the microphone 82
~ 85 are located in the middle of the speakers 56, 58, 60, 62
Have been. By the way, the microphone 81
The microphones 82, 83, 84, and 85 are
Which direction the vessel is facing)
I do. The number of microphones is about 1 to 9
Degree is practical.   When the player 88 plays the instrument 90, the sound is
Microphones 81-85, and these collected signals are
Mixing with the microphone mixing circuit 92 with built-in head-up
Be singed. Then, the processor 46 sets the reflected sound parameter.
The reflected sound is generated based on the data. These generated anti
The shooting sound was a 4-channel amplifier 72 (amplifiers 48, 50, 5 in FIG. 1).
2,54), each speaker 56,58,60,6
By being emitted from 2, the player 88 is as if
Enjoy the atmosphere of playing in the hall yourself
Can be. The reflected sound parameter indicates that the player 88
In the position, by the operation of the remote controller 76,
Adjustable.   A specific example of the processor 46 in FIG. 19 is shown in FIG.
You. In FIG. 20, sound was picked up by microphones 81 to 85
The performance signal is mixed by the microphone mixing circuit 100.
The level is adjusted by the input volume 102. And b
-Pass filter (to prevent aliasing during A / D conversion) and
A / D converter 106 via sample and hold circuit 104
A / D conversion is performed. And furthermore, the frequency characteristics of the reflected sound
Digital filter for each channel.
, 108, 110, 112, 114.   Output from digital filters 108, 110, 112, 114
The performance signals are reflected sound generation circuits 116, 118, 12 of each channel.
0,122. In the reflected sound generation circuits 116, 118, 120, 122
The memory 126 is controlled by a command from the microcomputer 124.
Parameter for each channel stored in the
Time data and gain data)
Each time, a reflected sound signal of the performance signal is generated. Generate
These reflected sound signals are converted by the D / A converter 124
D / A conversion is performed in a division multiplex manner. Output signal of D / A converter 124
Are distributed in a time-division manner for each channel,
Sample and hold circuits and low-pass filters 126, 128,
Via 130 and 132 respectively
Will be returned. Output volume 134, 136, 138, 140 and
And channel power amplifiers 48, 50, 52, 54
, 56, 58, 60, and 62, respectively. This allows
From each channel speaker 56, 58, 60, 62, each corresponding
The reflected sound of the performance from the virtual sound source in the
Sound space such as a hall specified by the distribution of the thought sound source is reproduced
It is.   The memory (ROM) 126 has various acoustic spaces such as halls.
Reflected sound parameters and digital filters 108, 110, 11
2,114 frequency response parameters for each channel
Stored on the remote control
Through the remote control sensor interface 142.
Either of these
The parameters of the rule are read.   The read frequency characteristic parameters are temporarily transferred to RAM.
Sent to the RAM based on the parameters stored in this RAM.
The frequency characteristics of the digital filters 108, 110, 112, 114 are controlled.
It is. The frequency characteristics parameters stored in RAM
Can be adjusted to your preference by operating the remote control 76
It is.   Also, the reflected sound parameter of each channel read out
(Delay time digital and gain data)
Provided in the reflected sound generation circuits 116, 118, 120, 122
Once stored in RAM (parameter memory 160 in FIG. 23 described later).
Transferred and stored in the RAM based on the reflected sound parameters.
And the reflected sound generation circuits 116, 118, 120, 122
Each time, a reflected sound of the performance signal is generated. Stored in RAM
Reflected sound parameters can be used to operate the wireless remote control 76
More fine-tuning is possible, which allows you to reduce the reverberation
Can be changed according to   By the way, the reflected sound generation circuits 116, 118, 120, 122
Of delayed input signals (performance signals)
It is possible to generate a reflected sound signal by (convolution operation)
it can. About reflected sound generation by this convolution operation
This will be described below.   The reflection sound generation by the convolution operation is shown in FIG.
Performance based on the reflected sound parameter string of each channel
Create signals with various time delays and amplitude levels from signals
And superimpose them. That is, one
Describing a channel and using it
The reflected sound parameter sequence to be calculated is calculated as shown in FIG.
Delay time τ based on signal (direct sound)_i(I = 1,2,
…, N) and gain (amplitude level) g_iCombination of parameters
As shown in FIG. 22, the
Using a delay memory 163 having a multi-tap,
Time τ_iThe delayed signal from each tap corresponding to
And gain g by the amplitude adjusters 152-1 through 152-n._iTo
Each of them is added and combined by the adder 153. This
From adder 153, Is output.   Of the reflected sound generation circuit 116 (same for 118, 120, 122) in FIG.
A specific example is shown in FIG.   Note that the configuration of the delay memory 163 is analog
For charge signals, a charge transfer element such as BBD or CCD is used.
Shift register or RA for digital signals
Use digital memory, etc., programmed with M
However, in the following embodiment,
Is large and the parameters (delay time and gain)
A case where a RAM that can be easily set and changed will be described.   In FIG. 23, the parameter memory (RAM) 160
By operating the earless remote control 76, the memory ROM 126 (20th
Of the reflected sound parameters read from
Channel (for the reflected sound generation circuit 116,
) Is stored in each address. Record
The following table shows the estimated reflected sound parameters.   In this table, τ₀Is one sampling period of the input signal
It shows. Therefore, the delay time data τ₁/ τ₀,
τ_Two/ τ₀, ... (integer value) is the delay time τ₁, τ_TwoCorresponding to, ...
Sample position (ie, how many samples
).   Data memory (delay memory) 163 is composed of RAM
The digital filters 108, 110, 112, 114 (20th
Digitized performance signal output from
Is written and stored in the parameter memory.
Delay time data τ₁/ τ₀, τ₁/ τ₀The delay data at the position corresponding to
Will be issued.   The counter 164 writes data in the data memory 163.
This indicates the current address. One sample of the input signal
It is incremented every cycle.   The counter 165 indicates the read address of the parameter memory.
Specify from 0 to 0 within one sample period of the input signal.
Count up to n and delay time data and gain
Reads each parameter of data.   Multiplexer 166 is an adder to parameter memory 160.
Write dress command from parameter controller 162
Address or read address from counter 165
It switches to either.   The subtracter 167 calculates the current address and the parameter from the counter 164.
Data obtained by subtracting the delay time data from the data memory 160.
Output as an address command of the data memory 163.
You. The data memory 163 reads the parameter memory 160
When the address is 0 (ie, delay time data, gain
(When both data are read)
At this time, the output of the subtracter 167 (ie,
(Current address from the counter 164) is the write address
It joins the data memory 163 as a command. Also, data memo
163 indicates that the read address of the parameter memory 160 is 0 or more.
Otherwise, the mode is switched to the read mode.
The output of the subtracter 167 (that is, when the current address is delayed)
Address that is separated by a distance equivalent to the data between
The command is added to the data memory 163 as a command.   Multiplier 168 is a delayed signal read from data memory 163
The pair read from the parameter memory 160 at that time.
The corresponding gain data is added.   The accumulator 169 outputs the delayed signal output from the multiplier 168.
Is accumulated by register 175 and adder 170 (convolution operation)
Then, the reflected sound signal shown in the above equation (3) is created.
It is. Reflected sound signal created by accumulator 169
Is then D / A converted by the D / A converter 124 (FIG. 20) and output
Is done. The AND circuit 171 generates a reflected sound signal.
Signal C3 interrupts the accumulated data until then.
Then, the accumulated value is reset to 0.   The timing controller 172 operates each of the above circuits.
To create each of the timing signals C1 to C5
It is.   Next, the operation of the apparatus shown in FIG. 23 will be described. (1) Setting of reflected sound parameters   First, reproduced by the operation of the wireless remote control 76
Select the hole to try. This allows the memory 12
From 6 (Fig. 20), the reflected sound parameter of the corresponding hole
τ₁~ Τ_n, g₁~ G_nIs read out and the parameter memory 160
Is written to. The written reflected sound parameter τ₁~
τ_n, g₁~ G_nIs adjusted by operating the wireless remote control 76
Is possible.   When performing this writing and adjustment,
Memory 160 is switched to write mode and multiplexer 1
66 is switched to the parameter address 166 side. (2) Creation of reverberation signal   After setting the parameters, set the parameter memory 16
0 is switched to the read mode, and multiplexer 166 is
To the data memory 163,
Phonon signal) to create a reflected sound signal.
Now.   Creating a reflected sound signal requires one sampling period of the input signal.
As one unit, the input signal to the data memory
At each delay set from the write data memory 163
Interval τ₁~ Τ_nRead delay signal corresponding to
Weights (g₁~ G_n) Accumulation
And create a reflected sound signal. 24th for each journey
This will be described with reference to the time chart in FIG.   Write input signal to data memory 163   Clock C4 goes low at the rise of clock C1.
Thus, the counter 165 is cleared. Therefore, the parameter
The address 0 is specified in the data memory 160, and the delay time data
"0" is read out for both the data and the gain data. And
At the next rising edge of clock C5, clock C2 also rises
The memory 163 is in a write state. At this time, the parameter
The delay time data from the memory 160 is “0” as described above.
Therefore, the output of the subtracter 167 is the output of the counter 164.
The counter 164 in the data memory 163
The input signal is written to the address indicated by the output.   Read delay signal from data memory 163   When writing to data memory 163 is completed, the data
The memory 163 is in the read mode. Clock C5 is 1 sun
One writing operation and n reading operations within a pulling cycle.
It rises n + 1 times in total. The counter 165
Count C5, and the count value is
In addition to memory 160, delay time data and gain data
Parameter τ₁~ Τ_n, g₁~ G_NIs read. For example, Coun
When the counter value of data 165 is "1",
From the address 1 of the memory 160 to the delay time data τ₁/ τ₀and
Gain data g₁Is read. Further, address 2
To τ_Two/ τ₀And g_Two, ... from address n to τ_n/ τ₀And g_nBut
Each is read.   Delay time parameter read from parameter memory 160
The meter is calculated by the subtracter 167 and the count value of the counter 164.
From the subtracter 167, the count value of the counter 164, that is,
Only the distance indicated by the delay time data based on the current address
The previous address is output and supported from data memory 163
Signal X stored at the address₁~ X_nIs read
Is done.   Weighting   The delayed signal read from data memory 163 is multiplied by
At the unit 168,
Each corresponding gain data g₁~ G_nIs given.   Accumulation   Whether the count value of the counter 165 is "1"
Data output from the multiplier 168 while the
Data g₁・ X₁~ G_n・ X_nIs obtained by accumulating. This accumulation is performed
The accumulator 169 counts the counter 165
When the clock value is “1”, the clock C3 falls and the previous accumulation
The value is reset. That is, the counter 165 is "1".
, The AND circuit 171 is turned off, and the output of the adder 170 is output.
Is the output g of the multiplier 168₁・ X₁Only value and register 175
Will be retained. Register 17 at the next clock C5 timing
5 is g₁・ X₁Is output, and the next data g_Two・ X_TwoJoin
Is calculated and the value of register 175 is rewritten.
Repeat the addition (accumulation) sequentially, and add n terms Is obtained, this value is output as a reflected sound signal.
You.   By the above operation, the input signal (performance signal) is sampled.
The reflected sound signal is generated for each ring cycle. In addition,
In the above description, one of the reflected sound signals having a plurality of channels is used.
Only shown for one channel, other channels
Can be generated with a similar configuration.   By the way, the parameters from the memory (ROM) 126 (Fig. 20)
The reflected sound pattern of a certain hole in the memory (RAM) 160 (Fig. 23)
After reading the parameters,
Can be adjusted to adjust the reflection characteristics.
Wear. The contents of the adjustment include, for example, the following.   FIG. 25 shows the reflected sound of each channel shown in FIG.
When the parameters of speakers 1 and 2 are delayed
Multiply the value between by a factor to increase the delay time relatively or
It is a contraction. This is the hall to be reproduced
Is equivalent to changing the size of
If (> 1) is multiplied to increase the delay time, the hole will spread.
Feel like, multiply by small coefficient (<1) to shorten delay time
It feels like the hole is narrower.   In this way, the spaciousness of the hall is about 0.0 to 3.0 times
(By increasing the memory capacity, it is possible to
You. ) Can be adjusted.   Fig. 26 shows the gain of the reflected sound parameter sequence (reflected sound amplitude).
(Equivalent to width register)
More live feeling (LIVENESS) can be changed. That is,
If the slope of the in is steep, it will have dead characteristics,
If you do, it will be a live characteristic. This is a large delay time
The higher the size of the register, the bigger the reflected sound
This is achieved by: In addition, the delay of the reflected sound parameter sequence
Periodically change delay time gain or both
You can also. For example, a sinusoidal low-frequency signal
If you shake the data value, the reproduced acoustic space will be audible
The spatial clarity is blurred (DIFFUSION)
It is also possible to obtain a special sound effect.   Here is another example of the arrangement of the speaker and microphone
explain.   Figure 27 layout   FIG. 27 shows an arrangement row of four speakers and four microphones.
It is. That is, the speakers 56, 58, 60, 62
Microphones 82, 83, 84, 85 are located on the ceiling at the four corners of the well
Speakers 56, 58, 60, 62
Have been. Each microphone 82, 83, 84, 85 is
Because it is located at the same distance from the cars 56, 58, 60, 62,
Howling is less likely to occur. Microphone
82, 83, 84, 85 are mounted on the ceiling wall.
Arranged so that the frequency response of the crophone input is flat.
ing.   Figure 28 layout   FIG. 28 shows microphones 82, 83, 84 and 85 placed on the floor of room 80.
These are arranged at the four corners. In this arrangement, the microphone pickup
The sound pressure at the point is the highest over the entire frequency band,
The characteristics are flat. In addition, speakers 56, 58, 60, 62
The effect of direct sound radiation is small, and the howling margin is small.
large.   Fig. 29 layout   FIG. 29 shows four directional microphones 82, 83, 84, 85.
Near the four corners of the ceiling, the direction of the instrument sound source (the center of room 80)
Direction). Microphones 82,83,
84 and 85 are located near the radiation axes of the speakers 56, 58, 60 and 62.
Although it is arranged, it has strong directivity and speakers 56, 58,
60, 62, so the speakers 56, 58, 60, 62
The sound is not picked up and the howling margin is significantly increased. [Example 2]   In the embodiment shown in FIG. 30, sound is collected by the microphones 82, 83, 84, 85.
The signals are individually processed. That is,
The picked-up signals of the microphones 82, 83, 84, 85 are input volume 140
~ 143, low pass filter and sample and hold circuit 1
A / D conversion is performed by A / D converters 148 to 151 via 44 to 147. A
The / D-converted signal of each channel is
Input to the data 108, 110, 112, 114, respectively.
The reflected sound is generated in the same manner as in the illustrated embodiment. like this
, Each microphone pick-up signal is independent, so indoors
You can also control the direction of the instrument. Sometimes
For sound sources with directivity,
A more natural hall feeling can be reproduced.   In this case, the arrangement of the speakers and microphones
If the arrangement shown in FIG. 29 is used,
Sound from the speaker
Signal processing of characteristics becomes easy. [Example 3]   By the way, the reflected sound by the reflected sound parameter of the present invention
In the generation, it tries to reproduce the reflection sound of the hall etc. faithfully
And the memory capacity according to the duration of the reflected sound is required
Therefore, if you try to obtain a long reverberation time, the memory capacity will increase.
It will be a big one.   The embodiment of FIG. 31 solves this problem and requires less memory.
Long reverberation time is obtained with capacity.
That is, the initial reflected sound is based on the aforementioned reflected sound parameter.
In the reflection sound generation process, each sound is generated individually and almost expanded.
Comb shape for middle to late reflections that are becoming diffuse
Simple loops such as filters and all-pass filters
In the reverberation sound generation process using a reverberation adding device
It is generated.   In FIG. 31, the reflected sound generation circuits 116, 118, 120, 122
The same as those used in the respective embodiments of FIG. 20 and FIG.
The initial reflection from each channel input signal
Generate sound. Also, reverberation sound generation circuits 186, 188, 190, 192
It is composed of a comb filter, an all-pass filter, etc.
Each channel input signal or reflected sound generation circuit 116, 118, 12
0-122 output signals to generate their mid-late reflections
You. These reflected sound generation circuits 116, 118, 120, 122 and reverberation
The output signals of the sound generation circuits 186, 188, 190, 192
And is subjected to the same signal processing as in the previous embodiment.
And supplied to the speakers 56, 58, 60, 62.   The reflected sound characteristics obtained in the embodiment shown in FIG.
Output signal of each channel when input)
An example is shown in FIG. The initial reflected sound part is the reflected sound generation circuit 11
6 (118,120,122), with mid-late reflections reverberation
It is obtained by the sound generation circuit 186 (188, 190, 192).   The memory capacity of the reflected sound generation circuits 116, 118, 120, 122
If it allows, the reflection sound generation circuits 116, 118, 120, 122
186, 188, 190, 19
2 can also be used to cover the late reflection portion.   In addition, the receiving area of the reflected sound generation circuits 116, 118, 120, 122 and
The reverberation sound generation circuits 186, 188, 190, 192
There is no need to cut it, overlap it at the joint
You may make it.   The reflected sound generation circuit 116 (118, 120, 122 and the reverberation sound generation circuit 18
FIG. 33 shows a specific example of 6 (188, 190, 192).   The reflected sound generation circuit 116 is the same as that shown in FIG.
Multi-type delay
Using the memory 163, each tap corresponding to the delay time τi
, And extract the delay signals from the
The gain gi is given by 1 to 152-n, and the adder 1
Synthesize at 53. Thereby, from the adder 153,Is output.   The delay memory 163 has a delay longer than the delay time τn.
A tap corresponding to time τx is provided, and from this tap
Is given a gain gx by the amplitude adjuster 152-X.
The reverberation sound generation circuit 186 inputs the reverberation sound.   The reverberation generation circuit 186 is composed of an all-pass filter.
ing. That is, the reverberation generation circuit 186 converts the input signal
Input to the delay circuit 196 via the adder 194, the delay circuit 196
Output of amplifier 198 (gain 1-g^Two) Via adder 194
Has returned to. The output of the adder 194 is the amplifier 200 (gain
−g) is added to the output of the delay circuit 196 by the adder 202 via
It is. Thus, reverberation signal is output from adder 194.
Is forced.   Reverberation of the reflected sound signal output from the reflected sound generation circuit 116
The reverberation signal output from the sound generation circuit 186 is added to the adder 204.
It is added and output. Delayed from reflected sound generation circuit 116
The reflected sound whose time is τ1 to τn is output, and the reverberation sound is generated.
From the circuit 186, the delay time after τx which is later than τn
Since an echo is output, the two are connected from the adder 204.
A series of long reflected signals is output.   FIG. 34 shows the duration of reflection in the embodiment of FIG.
Enlarge (approximately twice as large as in Fig. 32)
You. ), The impression of expanding the size of the hall (SIZE)
That's what I did. This is the reflected sound generation circuit 116,
Delay time parameters and reverberation at 118,120,122
Parameter of delay time in sound generation circuit 186,188,190,192
Data by a coefficient (≒ 2). [Example 4]   The embodiment of FIG. 35 generates reflected sound for each channel.
Before adding the reverb sound
is there.   That is, the mixed microphone pickup signal is A / D
After the A / D conversion by the converter 106, the reverberation sound generation circuit 210
Reverberation is applied, and then the digital filter 10
Via 8,110,112,114, reflected sound generation circuit 116,118,120,1
22 and the reflected sound is generated for each channel.
You. According to such a configuration, the reverberation
Since there is no need to prepare a generation circuit, the configuration is simplified.
You. 〔The invention's effect〕   As described above, according to the present invention, the performance position
At least four or more speakers should be
Microphone collecting means for placing live means and collecting performance sounds
And the reflected sound data of the hall etc.
From the sound source distribution, etc.
Data on the direction of arrival of the sound and the
Based on the arrangement of the speaker playback means,
The parameters of the number of reflected sounds to be emitted at the step are determined, and
The microphone picks up the sound based on the shooting parameters.
Creates a large number of reflections of the
Loudspeaker playback means.
With a limited number of speaker playback means, each reflected sound
Can give the same directionality as the direction of arrival of the reflected sound in the actual sound field,
In the playing position, a reflected sound field is realized all around.
It is. Therefore, the speaker reproducing means and the microphone
The actual hall in the room where the
Enjoy playing musical instruments and singing in an atmosphere similar to
Can practice.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a conceptual diagram of a sound pickup reproduction control device according to the present invention. FIG. 2 is a conceptual diagram of a conventional sound control device. FIG. 3 is a perspective view showing a method of measuring a virtual sound source by the four-point method. FIG. 4 is a perspective view showing the arrangement of the microphones in FIG. FIG. 5 is a waveform diagram showing a measurement result of an impulse response by the microphone of FIG. FIG. 6 is a diagram showing a method of calculating a virtual sound source position based on the measurement results of FIG. 7, 8, and 9 are views showing virtual sound source distributions obtained by the four-point method. FIG. 7 is an XY plane projection view, and FIG.
The figure is a YZ plane projection view, and FIG. 9 is an XZ plane projection view. FIG. 10 is a diagram showing the principle of a virtual sound source measurement method using a mirror image method. 11 and 12 are diagrams showing a virtual sound source distribution obtained by the mirror image method. FIG. 11 is an XY plane projection view, and FIG.
It is a Z plane projection view. FIG. 13 is a plan view showing a reflected sound reproduction state by eight surrounding speakers. FIG. 14 is a plan view showing a reflected sound reproduction state by four surrounding speakers. FIG. 15, FIG. 16, and FIG. 17 show the sound volume distribution between the speakers for simulating the reflected sound in the middle of the adjacent speakers, and FIG. FIG. 16 is based on a linear function, and FIG. 17 is based on a log function. FIG. 18 shows the fourteenth method based on the reflected sound measurement data.
FIG. 9 is a diagram showing a reflected sound parameter string used to create a reflected sound to be supplied to each speaker when the reflected sound is simulated by the speaker arrangement shown in FIG. FIG. 19 is a block diagram showing one embodiment of the present invention. FIG. 20 is a block diagram showing a configuration example of the processor 46 in FIG. FIG. 21 is a diagram showing a reflected sound parameter string used for reflected sound generation in the reflected sound generation circuits 116, 118, 120, and 122 in FIG. FIG. 22 is a circuit diagram showing a reflected sound generation circuit 116 (118, 120, 122) of FIG. 20 configured to generate a reflected sound signal of an input signal by a convolution operation using the reflected sound parameter of FIG. is there. FIG. 23 shows the reflected sound generation circuit 116 (118, 120, 122) of FIG.
It is a block diagram which shows the specific example of. FIG. 24 is a time chart showing the operation of the circuit of FIG. FIG. 25 is a reflected sound parameter sequence showing an example in which the delay time of the reflected sound parameter is multiplied by a coefficient to adjust the sense of the size of the hole. FIG. 26 is a reflected sound parameter sequence showing an example of adjusting the live feeling by changing the inclination of the reflected sound parameter sequence. 27 to 29 are perspective views showing other examples of the arrangement of the speaker microphone. FIG. 30 is a block diagram showing a second embodiment of the present invention. FIG. 31 is a block diagram showing a third embodiment of the present invention. FIG. 32 is a diagram showing a reflected sound signal pattern obtained in the embodiment of FIG. FIG. 33 is a block diagram showing a configuration example of a reflected sound generation circuit and a reverberation sound generation circuit in FIG. FIG. 34 is a diagram showing a reflected sound signal pattern obtained by finely adjusting the sense of the hole width in the embodiment of FIG. 31. FIG. 35 is a block diagram showing a fourth embodiment of the present invention. 56,58,60,62 …… Speakers (speaker playback means), 81,8
2,83,84,85 …… Microphone (microphone collecting means), 88…
... a player (performance position), 116, 118, 120, 122 ... reflected sound generation circuit (reflected sound generation means), 126 ... memory (parameter storage means).

Continuation of front page    (56) References JP-A-59-90897 (JP, A)                 58-125486 (JP, U)                 Shokai Sho 57-43686 (JP, U)                 Tokuko Hei 5-62752 (JP, B2)                 Tokiko Hei 5-62753 (JP, B2)                 Tokiko Sho 61-640 (JP, B2)                 Tokiko Sho 60-13640 (JP, B2)                 7-15280 (JP, Y2)

Claims (1)

(57) [Claims] At least four or more speaker reproducing means arranged so as to surround the entire circumference of the playing position, microphone collecting means for collecting the performance sound, and a sound source corresponding to each virtual sound source position of the reflected sound in the acoustic space Based on the reflected sound data comprising the direction of arrival of each reflected sound, the delay time, and the amplitude level, a number of reflected sounds in the acoustic space or a model space similar to the acoustic space are reproduced around the playing position using the respective speaker reproducing means. An impulse response characteristic comprising a delay time and a gain of a reflected sound to be emitted by each speaker reproducing means, which is obtained from an arrangement position of each speaker reproducing means and each direction of arrival data of the reflected sound data for reproduction. Is stored as a reflected sound parameter of each speaker reproducing unit, and each of the parameters stored in the parameter storing unit is stored. Based on the shooting parameters, a convolution operation is performed on the sound pickup signal of the microphone sound pickup means, thereby generating a large number of reflected sounds to be emitted by each speaker reproduction means, and corresponding to each of the speaker reproduction means. Reflected sound generating means for supplying the reflected sound to each of the positions, and generating a large number of reflected sounds generated for the player when performing in the acoustic space or a model space similar thereto. A sound collection / reproduction control device characterized in that it can be reproduced almost equally including the direction of the reflected sound.