close

Вход

Забыли?

вход по аккаунту

?

DESCRIPTION JPH06178399

код для вставкиСкачать
Patent Translate
Powered by EPO and Google
Notice
This translation is machine-generated. It cannot be guaranteed that it is intelligible, accurate,
complete, reliable or fit for specific purposes. Critical decisions, such as commercially relevant or
financial decisions, should not be based on machine-translation output.
DESCRIPTION JPH06178399
[0001]
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a
sound image localization coefficient calculation apparatus of a sound image localization
apparatus for performing realistic sound reproduction in an AV (audio visual) apparatus.
[0002]
2. Description of the Related Art In recent years, in the field of video and audio, sound
reproduction with a large screen and a sense of reality is desired in order to enjoy movies at
home due to the spread of VTRs, and development of hardware corresponding thereto is desired.
.
[0003]
In particular, in the sound of movie software of VTR, as in a movie theater, a Dolby Surround
system in which speakers are disposed side by side or in the rear (or a combination thereof) and
reproduced is widespread.
[0004]
Also, apart from this, a sound image localization device was developed for making it possible to
hear the sound from the side or the rear without putting the speaker on the side or the rear, for a
home TV set etc. ing.
10-04-2019
1
[0005]
Hereinafter, a sound image localization apparatus using two reproduction speakers will be
described.
[0006]
FIG. 9 is a diagram showing the configuration of a conventional sound image localization
apparatus and a system for localizing a sound image to the left rear of the listener.
[0007]
In FIG. 9, 1 indicates a sound image localization apparatus, 2 indicates a signal S (t) (t indicates
continuous time, and indicates that the signal is a time function.
And so on), 3 and 4 are FIR filters, 5 is a speaker for reproduction placed on the left front of the
listener 8, 6 is a speaker for reproduction placed on the right front of the listener 8, and 7 is a
listener HL (t) is an impulse response when an impulse is input from the oscillator 2 to the finite
time impulse response (FIR) filter 3; hR (t) is an FIR response to the FIR filter 4 This is an impulse
response when an impulse is input from the oscillator 2. Hereinafter, hL (t) and hR (t) will be
referred to as sound image localization coefficients in the sound image localization apparatus 1
on the left (L) and right (R).
h1 (t) is the position of the speaker 5 and the left ear of the listener 8 (precisely, the response at
the position of the eardrum when an impulse is input to the speaker 5, but when measurement is
performed, it is performed at the position of the ear canal entrance The head related transfer
function in the same manner hereinafter (hereinafter, referred to as an impulse response to
explain in the time domain).
However, similar results can be obtained even in the frequency domain.
Moreover, what converted the impulse response into a frequency characteristic by Fouriertransforming is called a transfer function. H2 (t) is an impulse response at the position of the
speaker 5 and the right ear of the listener 8, h3 (t) is an impulse response at the position of the
speaker 6 and the left ear of the listener 8, h4 (t) is the speaker 6 The impulse response at the
position of the right ear of the listener 8, h5 (t) is the impulse response at the position of the left
10-04-2019
2
ear of the speaker 7 and the listener 8, h6 (t) is the impulse at the position of the right ear of the
speaker 7 and the listener 8. It is a response.
[0008]
The operation of the conventional sound image localization apparatus and the method of sound
image localization will be described with reference to FIG.
[0009]
In such a configuration, when the signal S (t) from the oscillator 2 is radiated from the target
speaker 7, the sound reaching the ear of the listener 8 is (Equation 1), in the left ear L (t),
[0010]
[Equation 1] L (t) = S (t) * h5 (t) In the right ear R (t),
[0011]
R (t) = S (t) * h6 (t) (where * represents a convolution operation).
It is expressed as).
In practice, the impulse response of the speaker itself is also convoluted, but this is ignored.
Also, it may be considered that the impulse response of the speaker or the like is included in h5
(t) and h6 (t).
[0012]
Also, considering the impulse response and the signal S (t) as discrete digital signals, L (t) → L (n)
R (t) → R (n) h5 (t) → h5 (n) h6 When expressed as (t) → h6 (n) S (t) → S (n), the equations (1)
and (2) are as follows.
[0015] Here, N is the length of the impulse response h5 (n), h6 (n).
10-04-2019
3
[0016]
On the other hand, when the signal S (t) is input from the oscillator 2 to the sound image
localization apparatus 1, the input signal S (t) is branched into two and input to the FIR filter 3
and the FIR filter 4, respectively. After convolution processing is performed in 3 and 4,
respectively, the signal is output from the speakers 5 and 6, respectively.
At that time, the sound reaching the listener 8 is given by the left ear L ′ (t)
[0017] L '(t) = S (t) * hL (t) * h1 (t) + S (t) * hR (t) * h3 (t) In the right ear R' (t), (Equation 6),
[0018] R '(t) = S (t) * hL (t) * h2 (t) + S (t) * hR (t) * h4 (t), and similarly (Equation 7) and (Equation
8) By
[0019] L '(n) = S (n) * hL (n) * h1 (n) + S (n) * hR (n) * h3 (n)
[0020] R '(n) = S (n) * hL (n) * h2 (n) + S (n) * hR (n) * h4 (n)
[0021] Assuming that the sounds can be heard from the same direction if the head related
transfer functions are equal (this premise is generally correct),
[0022] When L (n) = L '(n), by the equations (10) and (11)
[0023] H5 (n) = hL (n) * h1 (n) + hR (n) * h3 (n)
[0024] When R (n) = R '(n), as shown in (Equation 12)
[0025] Since h6 (n) = hL (n) * h2 (n) + hR (n) * h4 (n), the speakers 5 and 6 can be used to hear
10-04-2019
4
the listener 8 from the rear left In order to do so, hL (n) and hR (n) may be determined so as to
satisfy the equations (10) and (12). For example, if Equations (10) and (12) are rewritten in the
frequency domain representation, the convolution operation will be replaced by multiplication,
and then each impulse response will be converted to a transfer function by Fast Fourier
Transform (FFT) . The impulse responses of the FIR filters 3 and 4 can be obtained from the two
(Equation 10) and (Equation 12) because the impulse responses of the FIR filters 3 and 4 are
obtained by measurement. It is also possible to express (Equation 10) and (Equation 12) as a
matrix equation and determine hL (n) and hR (n) by inverse matrix operation.
[0026] Using hL (n) and hR (n) determined in this way, the signals S (n) and hL (n) from the
speaker 5 and the signals S (n) and hR (n) from the speaker 6 By radiating the folded-in, the
listener 8 can feel that the sound is being emitted from the rear without actually sounding the
rear target speaker 7.
[0027] Even if there are three or more reproduction speakers, the sound image can be localized
at an arbitrary position by performing the same process as described above.
[0028] It is the FIR filters 3 and 4 that actually perform the convolution operation. A block
diagram of the basic configuration of the FIR filters 3 and 4 is shown in FIG. In FIG. 10, delay
elements 3-2 for delaying a signal by τ time are connected in series to an input terminal 3-1 for
inputting a signal, and both ends of the delay elements 3-2 are called tap coefficients. A
multiplier 3-3 for multiplying the input signal is connected. The other ends of the multipliers 3-3
are connected to an adder 3-4 for adding a plurality of input signals, and the adder 3-4 is
connected to an output terminal 3-5 for outputting the added signal. ing.
[0029] H (n) (n: 0 to N-1) of the multiplier 3-3 is an impulse response having a certain
characteristic set as a tap coefficient. Usually, such an FIR filter uses a dedicated LSI such as a
DSP (Digital Signal Processor) that performs multiplication and addition at high speed. As shown
in the figure, the impulse response h (n) is set as a tap coefficient in the multiplier 3-3, and a
delay time corresponding to the sampling frequency when converting an analog signal to a
digital signal is set in the delay element 3-2 By repeating multiplication and addition and delay
on the input signal, the convolution operation as shown in Equations 3 and 4 is executed.
[0030] Therefore, by inputting a signal to this FIR filter, the characteristic of the impulse
response h (n) is convoluted into the input signal and output. The above is the case of a digital
signal, so in practice it is necessary to have an A / D converter to convert an analog signal to a
10-04-2019
5
digital signal before this FIR filter and a D / A converter to convert a digital signal to an analog
signal after this. 9, is omitted in FIG.
[0031] Next, a conventional sound image localization coefficient calculation device for a sound
image localization device when two reproduction speakers are used will be described.
[0032] FIG. 11 is a block diagram showing the configuration of a conventional sound image
localization coefficient calculation apparatus.
[0033] In FIG. 11, 9-1, 9-2, 9-3 and 9-4 are reproduction system characteristic input terminals,
10-1 and 10-2 are target characteristic input terminals, 11 is a matrix operator, 12 is a tap
setter. , 14, 15, 15-1, 15-2, 15-3 and 15-4 are FIR filters, 16 is an impulse generator, 17, 18 are
adders, 19 and 20 are subtracters, 21 is a feedback controller, Reference numerals 22-1 and 222 denote sound image localization coefficient output terminals. H1 (n), h2 (n), h3 (n), h4 (n), h5
(n), h6 (n), hL (n) and hR (n) are the same as h1 shown in FIG. (t), h2 (t), h3 (t), h4 (t), h5 (t), h6
(t), hL (t), hR (t) are converted into discrete representations.
[0034] The operation of the conventional sound image localization coefficient calculation
apparatus configured as described above will be described with reference to FIG.
[0035] Reproduction System Impulse Responses h1 (n), h2 (n), h3 (n), h4 (n) which are the
characteristics of the reproduction system from the reproduction system characteristic input
terminals 9-1, 9-2, 9-3, 9-4 Are input, and target impulse responses h5 (n) and h6 (n) which are
target characteristics are input from the target characteristic input terminals 10-1 and 10-2,
respectively, and all input signals are input to the matrix calculator 11. Be done.
[0036] In the matrix operator 11, when the impulse responses h1 (n), h2 (n), h3 (n) and h4 (n)
are reproduction system characteristics, the impulse responses h5 (n) and h6 (n) are target
characteristics. Sound image localization coefficients of such a sound image localization
apparatus, that is, hL (n) and hR (n) candidates h′L (n) and h′R (n) satisfying the above
(Equation 10) and (Equation 12) The reproduction system impulse responses h1 (n), h2 (n), h3
(n), h4 (n), h'L (n) and h'R (n) are output to the tap setting unit 12.
[0037] The tap setter 12 applies the impulse response h'L (n) to the FIR filter 13, the impulse
10-04-2019
6
response h'R (n) to the FIR filter 14, the impulse response h1 (n) to the FIR filter 15-1, and the
FIR filter The impulse response h2 (n) is set to 15-2, the impulse response h3 (n) to the FIR filter
15-3, and the impulse response h4 (n) to the FIR filter 15-4 as tap coefficients.
[0038] When the setting of the tap coefficient is completed, the impulse generator 16 outputs an
impulse, and the outputted impulse is branched into two and input to the FIR filters 13 and 14,
respectively, and the input impulse is the FIR filter 13, At 14, the convolution processing with the
tap coefficients h'L (n) and h'R (n) is performed and output. The output from the FIR filter 13 is
equivalent to h'L (n), and the output from the FIR filter 14 is equivalent to h'R (n).
[0039] Each of these output signals is branched into two, and the output h'L (n) from the FIR
filter 13 is input to each of the FIR filters 15-1 and 15-2, and the tap coefficients h1 (n) and h2 (
n) and the convolution process is performed respectively and output. Further, the output from
the FIR filter 14 is input to the FIR filters 15-3 and 15-4 for h'R (n), and the convolution process
is performed on the tap coefficients h3 (n) and h4 (n), respectively. It is output.
[0040] The outputs of the FIR filters 15-1 and 15-3 are input and added to the adder 17 and
then output. The outputs of the FIR filters 15-2 and 15-4 are input and added to the adder 18,
and then output. The difference between h5 (n) and the output of the adder 17 is output to the
subtractor 19, and the difference between the h5 (n) and the output of the adder 17 is output.
The output of the unit 18 is input, and the difference between h6 (n) and the output of the adder
18 is output.
[0041] The outputs of the subtractors 19 and 20 are input to the feedback controller 21. In this
feedback controller 21, if the absolute value of the input from the subtractors 19 and 20 is larger
than a preset positive value, the feedback controller 21 preliminarily The instruction signal is
output to the matrix operator 11 so as to perform matrix operation after delaying h5 (n) and h6
(n) for the set time, and perform the matrix operation again as described above. repeat.
[0042] However, when the absolute value of the input from the subtractors 19 and 20 becomes
smaller than the preset positive value in the feedback controller 21, the operation is stopped, and
h'L (n And h′R (n) are output as hL (n) and hR (n), respectively, and the matrix operator 11
outputs hL (n) according to the instruction of the feedback controller 21. n) and hR (n) are output
from the output terminals 22-1 and 22-2, respectively.
10-04-2019
7
[0043] A similar sound image localization coefficient can be obtained by using an operator that
obtains coefficients by performing Fourier transform of the impulse response and performing
operations in the frequency domain instead of the matrix operator.
[0044] Also, the impulse response h'L (n) and h'R (n) are not used in the FIR filters 15-1, 15-2,
15-3, 15-4 without using the impulse generator 16 and the FIR filters 13, 14. You may enter
[0045] As described above, the sound image localization coefficient of the sound image
localization apparatus when two speakers for reproduction are used can be obtained from the
reproduction system characteristic and the target characteristic.
[0046] In addition, a sound image localization coefficient is similarly calculated | required also
when three or more speakers are used.
[0047] By setting the sound image localization coefficient obtained in this way to the sound
image localization apparatus and reproducing it, the sound image can be localized at a position
where the speaker does not actually exist, and from the viewpoint of the listener By making the
sound image localized at a wide position, it becomes possible to reproduce the sound with a
sense of spread and presence.
[0048] SUMMARY OF THE INVENTION However, in the above-mentioned configuration, the tone
color of the sound image localized by the sound image localization apparatus and the tone color
of the sound reproduced from the reproduction speaker without using the sound image
localization apparatus for the listener. There is a difference between them, and as a result, this is
one of the causes of the sound quality deterioration of the localized sound image by the sound
image localization apparatus.
[0049] The present invention takes into consideration such conventional problems, and for the
listener, it is between the timbre of the sound image localized by the sound image localization
apparatus and the timbre of the sound reproduced from the reproduction speaker without using
the sound image localization apparatus. It is an object of the present invention to provide a sound
image localization coefficient calculation device that calculates a sound image localization
coefficient so that there is no difference between the two.
10-04-2019
8
[0050] SUMMARY OF THE INVENTION In the present invention, reproduction system
characteristic input means for inputting reproduction system impulse responses at the left and
right ears of a listener to a plurality of reproduction speakers respectively as input signals, and a
position where sound image is to be localized Target characteristic input means for inputting, as
input signals, target impulse responses at the left and right ears of the listener to the target
speaker actually placed, and sounds at the left and right ears of the listener when sound is output
from the target speaker Peak sound pressure difference detecting means for detecting a pressure
difference, peak time difference detecting means for detecting a difference in arrival time of
sounds reaching the left and right ears of the listener at that time, and one of the plurality of
reproduction speakers Each reproduction system impulse response in the listener's left and right
ears when an impulse is emitted from the reproduction speaker is inputted as an input signal.
Peak time difference adjustment means for adjusting and outputting delay time so that the
difference between arrival times of impulses in the signal becomes equal to the time difference
detected by the peak time difference detection means, and an output signal from the peak time
difference adjustment means Peak sound pressure difference adjusting means for adjusting and
outputting the amplitude so that the sound pressure difference in the input signal becomes equal
to the sound pressure difference detected by the peak sound pressure difference detecting
means; the reproduction system impulse response and the peak sound pressure difference And
an operation means for receiving an output signal from the adjustment means and calculating a
coefficient of the sound image localization apparatus from the input signal, wherein the
operation means outputs an output signal when an impulse is inputted to the sound image
localization apparatus. Reproduction system impulse responses at the left and right ears of the
listener when reproduced by the plurality of reproduction speakers are output signals from the
peak sound pressure difference adjusting means and And outputs the coefficients of properly
made such sound image localization apparatus.
[0051] According to the present invention, when the reproduction system characteristic input
means uses a plurality of reproduction speakers, the reproduction system impulse response is
input as an input signal, and the target characteristic input means inputs the target impulse
response as an input signal. The peak sound pressure difference detection means detects the
sound pressure at the left and right ears of the listener when the sound is output from the target
speaker, and the peak time difference detection means reaches the arrival time of the sound
reaching the left and right ears of the listener at that time To detect differences in
[0052] The peak time difference adjustment means inputs, as an input signal, each reproduction
system impulse response in the listener's left and right ears when an impulse is emitted from one
of the reproduction speakers, and the arrival time of the impulse in the input signal The delay
time is adjusted and output so that the difference between the two is equal to the time difference
detected by the peak time difference detection means.
10-04-2019
9
[0053] The peak sound pressure difference adjusting means receives the output signal from the
peak time difference adjusting means, adjusts the amplitude so that the sound pressure
difference in the input signal becomes equal to the sound pressure difference detected by the
peak sound pressure difference detecting means The listener when the computing means inputs
the reproduction system impulse response and the output signal from the peak sound pressure
difference adjusting means and the output signal when the impulse is input to the sound image
localization apparatus is reproduced by a plurality of reproduction speakers The sound image
localization coefficient of the sound image localization apparatus is output such that the
reproduction system impulse response in the left and right ears is equal to the output signal from
the peak sound pressure difference adjusting means.
[0054] Embodiments of the present invention will be described hereinbelow with reference to
the drawings.
[0055] FIG. 1 is a block diagram showing the configuration of a sound image localization
coefficient calculation apparatus according to a first embodiment of the present invention.
[0056] In FIG. 1, reference numeral 23 denotes a peak sound pressure difference detector for
detecting a difference between peak sound pressures of two signals inputted from the target
characteristic input terminals 10-1 and 10-2, and 24 denotes a target characteristic input
terminal 10- Peak time difference detector which detects the time difference at the time of peak
sound pressure of each of two signals inputted from 1, 10-2; 25 are two signals inputted from
reproduction system characteristic input terminals 9-1 to 9-4 Peak time difference adjuster for
adjusting the delay time so that the time difference at the time of the peak sound pressure
becomes equal to the time difference inputted from the peak time difference detector 24; 26 also
from the reproduction system characteristic input terminals 9-1 to 9-4 27-1 and 27-2 are peak
sound pressure difference adjusters for adjusting the amplitude so that the difference between
the peak sound pressures of the two input signals becomes equal to the sound pressure
difference input from the peak sound pressure difference detector 23 Two signals from the
reproduction system characteristic input terminals 9-1 to 9-4 are input and Switch, aL (n) for
outputting either, aR (n) is left, right output signals from the peak sound pressure difference
regulator 26. The other functional elements similar to those in FIGS. 9 and 11 are indicated by
the same numbers and symbols.
[0057] The operation of the sound image localization coefficient calculating apparatus according
10-04-2019
10
to the first embodiment configured as described above will be described with reference to FIG.
[0058] Reproduction System Impulse Responses h1 (n), h2 (n), h3 (n), h4 (n) which are the
characteristics of the reproduction system from the reproduction system characteristic input
terminals 9-1, 9-2, 9-3, 9-4 Are input, and target impulse responses h5 (n) and h6 (n) which are
target characteristics are input from the target characteristic input terminals 10-1 and 10-2. The
input signals are all branched into two, h1 (n) is input to the matrix calculator 11 and the switch
27-1, and h2 (n) is input to the matrix calculator 11 and the switch 27-2, h3 (n) is input to the
matrix calculator 11 and the switch 27-1, h4 (n) is input to the matrix calculator 11 and the
switch 27-2, and h5 (n) is the peak sound pressure difference detector 23 and the peak time
difference The signal h6 (n) is also input to the detector 24 and is also input to the peak sound
pressure difference detector 23 and the peak time difference detector 24.
[0059] The peak sound pressure difference detector 23 obtains the peak sound pressure of each
of h5 (n) and h6 (n) input, and outputs the difference to the peak sound pressure difference
adjuster 26. Further, the peak time difference detector 24 obtains the time of each peak sound
pressure of the input h5 (n) and h6 (n), and outputs the time difference to the peak time
difference adjuster 25.
[0060] The switch 27-1 receives h1 (n) and h3 (n), and the switch 27-2 receives h2 (n) and h4
(n). Here, when it is desired to cause the listener to localize the sound image on the left side, the
switch 27-1 outputs h1 (n) and the switch 27-2 outputs h2 (n), respectively. When it is desired to
localize the sound image, the switch 27-1 outputs h3 (n) and the switch 27-2 outputs h4 (n). The
signals output from the switches 27-1 and 27-2 as described above are input to the peak time
difference adjuster 25.
[0061] The peak time difference adjuster 25 adjusts the delay time so that the time difference at
the peak sound pressure of the two input signals becomes equal to the time difference input from
the peak time difference detector 24, and the two adjusted signals are Output to the peak sound
pressure difference adjuster 26. The peak sound pressure difference adjuster 26 adjusts the
amplitude so that the difference between the peak sound pressures of the two input signals
becomes equal to the sound pressure difference input from the peak sound pressure difference
detector 23, and the two adjusted signals Are output to the matrix calculator 11.
[0062] Here, one of the output signals from the peak sound pressure difference adjuster 26 is
represented as aL (n) and the other as aR (n), and the peak time difference adjuster 25 and the
10-04-2019
11
peak sound pressure difference adjuster 26 adjust aL (n) is the result of adjusting h1 (n) or h3
(n), and aR (n) is the result of adjusting h2 (n) or h4 (n).
[0063] The matrix operator 11 inputs h1 (n), h2 (n), h3 (n), h4 (n), aL (n), aR (n), and impulse
responses h1 (n), h2 (n), The sound image localization coefficient of the sound image localization
apparatus in which aL (n) and aR (n) are target characteristics when h3 (n) and h4 (n) are
reproduction system characteristics, that is, the left side of equation 10 is aL (n) The candidates
h'L (n) and h'R (n) of hL (n) and hR (n) are calculated so as to satisfy the equation obtained by
changing aR (n) on the left side of Equation 12.
[0064] On the other hand, aL (n) and the output of the adder 17 (h1 (n) + h3 (n)) are input to the
subtractor 19, and the difference between aL (n) and the output of the adder 17 is output. AR (n)
and the output of the adder 18 (h2 (n) + h4 (n)) are input to the unit 20, and the difference
between the aR (n) and the output of the adder 18 is output. This is the same as the operation of
the conventional sound image localization coefficient calculation device described using 11, and
the sound image localization coefficients hL (n) and hR (n) are finally output from the output
terminals 22-1 and 22-2, respectively. .
[0065] The peak sound pressure difference detector 23, the peak time difference detector 24, the
peak time difference adjuster 25, and the peak sound pressure difference adjuster 26 shown in
FIG. 1 will be described below.
[0066] FIG. 2 is a schematic diagram showing the configuration and operation of the peak sound
pressure difference detector.
[0067] In FIG. 2, 23-1 and 23-2 are input terminals of the peak sound pressure difference
detector, 23-3 is a divider, and 23-4 is an output terminal.
[0068] The operation of the peak sound pressure difference detector 23 configured as described
above will be described. First, the input signal 1 (h5 (n)) is input from the input terminal 23-1
and the input signal 2 (h6 (n)) is input from the input terminal 23-2, and the peak sound
pressure A of the input signal 1 is input The peak sound pressure B of the signal 2 is detected
and output to the divider 23-3, A / B is performed in the divider 23-3, and the result is output
from the output terminal 23-4. In FIG. 2, the input signal 1 and the input signal 2 are
schematically represented by a graph showing time t on the horizontal axis and peak sound
10-04-2019
12
pressures A and B according to voltage values on the vertical axis. Needless to say, when the
sound pressure is expressed in decibels, A-B is performed using a subtracter instead of the
divider 23-3.
[0069] Thus, the peak sound pressure difference detector 23 detects and outputs the difference
between the peak sound pressures of the two input signals.
[0070] FIG. 3 is a schematic view showing the configuration and operation of the peak time
difference detector.
[0071] In FIG. 3, 24-1 and 24-2 are input terminals of the peak time difference detector, and 243 are output terminals.
[0072] The operation of the peak time difference detector 24 configured as described above will
be described. First, input signal 1 (h5 (n)) is input from input terminal 24-1, input signal 2 (h6
(n)) is input from input terminal 24-2, and the time at the peak sound pressure of input signal 1
The time t2 at the time of the peak sound pressure of t1 and the input signal 2 is detected, and
the difference Δt between t1 and t2 is determined and output from the output terminal 24-3. In
FIG. 3, the input signal 1 and the input signal 2 are schematically represented as in FIG.
[0073] In this way, the peak time difference detector 24 detects and outputs the time difference
at the peak sound pressure of the two input signals.
[0074] FIG. 4 is a schematic view showing the configuration and operation of the peak time
difference adjuster.
[0075] In FIG. 4, reference numerals 25-1, 25-2, 25-3 denote input terminals of the peak time
difference adjuster, and 25-4, 25-5 denote output terminals.
[0076] The operation of the peak time difference adjuster 25 configured as described above will
be described. First, the input signal 1 (h1 (n), h3 (n)) is input from the input terminal 25-1 and
the input signal 2 (h2 (n), h4 (n)) is input from the input terminal 25-2 The time t'1 at the peak
sound pressure of the input signal 1 and the time t'2 at the peak sound pressure of the input
10-04-2019
13
signal 2 are detected, and the input signal 3 (output terminal 24-3 in FIG. 3) is detected from the
input terminal 25-3. The time difference Δt is input as the output Δt) from the input signal, and
the input signal 2 is delayed by (Δt−t′2 + t′1) so that the difference between t′1 and t′2
becomes equal to Δt. The time at the peak sound pressure of the signal 2 is adjusted to be t3,
and the result is output from the output terminal 25-5, and the input signal 1 is output as it is
from the output terminal 25-4.
[0077] In this manner, the peak time difference adjuster 25 adjusts and outputs the time
difference at the peak sound pressure of the two input signals to be equal to the input time
difference.
[0078] FIG. 5 is a schematic view showing the configuration and operation of the peak sound
pressure difference regulator.
[0079] In FIG. 5, 26-1, 26-2, 26-3 are input terminals of the peak sound pressure difference
adjuster, 26-4 are arithmetic units, 26-5 are multipliers, and 26-6, 26-7 are output terminals. is
there.
[0080] The operation of the peak sound pressure difference adjuster 26 configured as described
above will be described. First, input signal 1 (output from output terminal 25-4 in FIG. 4) is input
from input terminal 26-1 and input signal 2 from input terminal 26-2 (output from output
terminal 25-5 in FIG. 4) The peak sound pressure A ′ of the input signal 1 and the peak sound
pressure B ′ of the input signal 2 are detected.
[0081] The sound pressure difference C is input from the input terminal 26-3 as the input signal
3 (output A / B from the output terminal 23-4 in FIG. 2), and A ', B', C are input to the computing
unit 26-4. The arithmetic result A ′ / (B ′ · C) from the arithmetic unit 26-4 is output to the
multiplier 26-5, and the input signal 2 is A ′ / (B ′ · C) in the multiplier 26-5. The peak sound
pressure of the output signal is A ′ / C, which is equal to the sound pressure difference C from
the peak sound pressure A ′ of the input signal 1. The output signal from the multiplier 26-5 is
output from the output terminal 26-7, and the input signal 1 is output as it is from the output
terminal 26-6.
[0082] When the sound image is localized on the left side with respect to the listener, the output
from the output terminal 26-6 is aL (n), and the output from the output terminal 26-7 is aR (n).
10-04-2019
14
When the sound image is localized, the output from the output terminal 26-6 is aR (n), and the
output from the output terminal 26-7 is aL (n).
[0083] Thus, the peak sound pressure difference adjuster 26 adjusts and outputs the difference
between the peak sound pressures of the two input signals to be equal to the input sound
pressure difference C.
[0084] Thus, the target impulse responses h5 (n) and h6 (n), which are target characteristics, are
not input as they are to the matrix calculator 11, but the reproduction system impulse responses
h1 (n), h3 (n) or h2 ( The matrix calculator 11 is inputted with the amplitudes and delay times of
n) and h4 (n) adjusted. In this case, the calculated sound image localization coefficients hL (n)
and hR (n) are set to the sound image localization apparatus, and predetermined reproduction
speakers are connected, and an impulse is input to the sound image localization apparatus In
order to localize the sound image on the left side of the listener, the impulse response in the
listener's left ear is equal to h1 (n) and the impulse response in the listener's right ear changes
the delay time and amplitude of h2 (n). (However, the time delay due to feedback control is
ignored.) Also, to localize the sound image to the right of the listener, the impulse response in the
listener's right ear is equal to h4 (n), and the impulse response in the left ear of the listener has
the delay time and amplitude of h3 (n). (However, as in the case of the above-mentioned left side
localization, the time delay due to feedback control is ignored. ), It has become a sound image
localization coefficient.
[0085] In this manner, sound image localization to the target position is realized by realizing the
difference in arrival time and sound pressure of the person who has reached the listener's left
and right ears when sound is output from the speaker placed at the position where the sound
image is to be localized. If the sound image is localized on the left side of the listener by the
sound image localization apparatus, the localized sound image and the sound reproduced from
the left reproduction speaker without using the sound image localization apparatus. It is possible
to minimize the difference in timbre that occurs between them.
[0086] When the sound image is localized on the right side of the listener by the sound
localization device, the difference between the timbres generated between the localized sound
image and the sound reproduced from the right reproduction speaker without using the sound
localization device is It is possible to minimize it. By this, it is possible to calculate a sound image
localization coefficient that minimizes the sound quality deterioration of the localized sound
image by the sound image localization apparatus.
10-04-2019
15
[0087] In this embodiment, the case where the sound image is localized on the left side of the
listener and the case where it is localized on the right side is distinguished, but always h1 (n), h2
(n) or h3 (n) regardless of the position of the sound image localization. And h 4 (n) may be input
to the peak time difference adjuster 25.
[0088] Further, in this embodiment, although both of the output signals from the peak time
difference adjuster 25 and the peak sound pressure difference adjuster 26 are two, the output
from the adjustment method in the peak time difference adjuster 25 and the peak sound
pressure difference adjuster 26 Since one of the signals is equivalent to the output signal from
the switch 27-1 or the switch 27-2, stop outputting the signal equivalent to the output signal
from the switch 27-1 or the switch 27-2 The output signals from the peak time difference
adjuster 25 and the peak sound pressure difference adjuster 26 are both one, and the output
signal from the switch 27-1 or the switch 27-2 is branched and input to the matrix calculator 11.
good.
[0089] Further, in this embodiment, although the outputs of the switches 27-1 and 27-2 are
input to the peak time difference adjuster 25, only one of the outputs of the switches 27-1 and
27-2 is input, A signal obtained by adjusting the delay time of the input signal itself and the input
signal may be output.
[0090] Further, in this embodiment, the peak sound pressure difference adjuster 26 is connected
after the peak time difference adjuster 25 to adjust the peak sound pressure difference after
adjusting the peak time difference. The peak time difference adjuster 25 may be connected after
the peak sound pressure difference adjuster 26 as an embodiment of the invention to adjust the
peak time difference after adjusting the peak sound pressure difference, and the operation at that
time is the same as described above. To be done.
[0091] Next, a second embodiment of the present invention will be described with reference to
the drawings.
[0092] FIG. 6 is a block diagram showing the configuration of a sound image localization
coefficient calculation apparatus according to a second embodiment of the present invention.
[0093] In FIG. 6, 28 is an amplitude frequency characteristic difference detector for detecting the
10-04-2019
16
difference between the amplitude frequency characteristics of the two signals input from the
target characteristic input terminals 10-1 and 10-2, and 29 is a reproduction system
characteristic input terminal 9- Amplitude frequency characteristic for adjusting the amplitude
on the frequency axis so that the amplitude frequency characteristic difference of the two signals
inputted from 1 to 9-4 becomes equal to the amplitude frequency characteristic difference
inputted from the amplitude frequency characteristic difference detector 28 Difference adjuster
30 receives two signals from reproduction system characteristic input terminals 9-1 and 9-4 and
outputs either of them. BL (n), bR (n) adjust amplitude frequency characteristic difference Output
signal from the control unit 29. The same functional elements as those in FIG. 1 are indicated by
the same numbers and symbols.
[0094] The operation of the sound image localization coefficient calculating apparatus according
to the second embodiment configured as described above will be described with reference to FIG.
[0095] Reproduction System Impulse Responses h1 (n), h2 (n), h3 (n), h4 (n) which are the
characteristics of the reproduction system from the reproduction system characteristic input
terminals 9-1, 9-2, 9-3, 9-4 Are input, and target impulse responses h5 (n) and h6 (n) which are
target characteristics are input from the target characteristic input terminals 10-1 and 10-2, and
h1 (n) and h4 (n) are two each. Branched, h1 (n), h2 (n), h3 (n), h4 (n) are input to the matrix
operator 11, and branched h1 (n), h4 (n) are input to the switch 30 .
[0096] The above h5 (n) and h6 (n) are input to the amplitude frequency characteristic
difference detector 28, and the amplitude frequency characteristic difference detector 28 obtains
the amplitude frequency characteristics of each of the input h5 (n) and h6 (n). The difference
between the amplitudes on the frequency axis is output to the amplitude frequency characteristic
difference adjuster 29. The switch 30 inputs h1 (n) and h4 (n), and outputs h1 (n) if the listener
wants the sound image to be localized on the left side, and wants the sound image to be localized
on the right for the listener And h 4 (n) is output to the switch 30, and the signal output from the
switch 30 is input to the amplitude frequency characteristic difference adjuster 29.
[0097] The amplitude frequency characteristic difference adjuster 29 receives the output from
the switch 30 and the output from the amplitude frequency characteristic difference detector 28,
and from the amplitude frequency characteristic difference detector 28 from the amplitude
frequency characteristic of the signal input from the switch 30. The signal whose amplitude is
reduced on the frequency axis by the input amplitude frequency characteristic difference and the
signal input from the switch 30 are output to the matrix calculator 11.
10-04-2019
17
[0098] Here, one of the output signals from the amplitude frequency characteristic difference
adjuster 29 is represented as bL (n) and the other as bR (n), and bL (n) is desired to localize the
sound image on the left side to the listener. (n) is equivalent to h1 (n), bR (n) is the result of
adjusting h1 (n) by adjustment with the amplitude frequency characteristic difference adjuster
29, and the sound image is localized to the right with respect to the listener When it is desired to
make bR (n) equal to h4 (n), bL (n) is the result of adjusting h4 (n) by the adjustment of the
amplitude frequency characteristic difference adjuster 29.
[0099] Next, in the matrix operator 11, h1 (n), h2 (n), h3 (n), h4 (n), bL (n), bR (n) are input, and
impulse responses h1 (n), h2 (n) The sound image localization coefficient of the sound image
localization apparatus in which bL (n) and bR (n) are target characteristics when h.sub.h3, h.sub.3
(n) and h.sub.4 (n) are reproduction characteristics, that is, the left side of equation (10) Calculate
the candidates h'L (n) and h'R (n) of hL (n) and hR (n) that satisfy the equation obtained by
changing bR (n) to the left side of bL (n) and (Equation 12) The difference between bL (n) and the
output of the adder 17 is output to the subtracter 19, and the difference between bL (n) and the
output of the adder 17 is output. The output of 18 is input and the difference between bR (n) and
the output of the adder 18 is output, and the other operations are the same as the operation of
the conventional sound image localization coefficient calculation apparatus described using FIG.
Sound image localization coefficients hL (n) and hR (n) are output from the output terminals 22-1
and 22-2, respectively.
[0100] The amplitude frequency characteristic difference detector 28 and the amplitude
frequency characteristic difference adjuster 29 will be described in detail below.
[0101] FIG. 7 is a block diagram showing the configuration of an amplitude frequency
characteristic difference detector.
[0102] In FIG. 7, 28-1 and 28-2 are input terminals of the amplitude frequency characteristic
difference detector, 28-3 and 28-4 are Fourier transformers, 28-5 is a difference extractor, and
28-6 is an output terminal. .
[0103] The operation of the amplitude frequency characteristic difference detector 28
configured as described above will be described.
10-04-2019
18
[0104] First, h5 (n) is input from the input terminal 28-1 to h6 (n) from the input terminal 28-2,
and h5 (n) and h6 (n) are input to the Fourier transformers 28-3 and 28-4, respectively. Is input,
Fourier transformation is performed, and the amplitude frequency characteristics F (h5 (n)) and F
(h6 (n)) as the result are output to the difference extractor 28-5. The difference extractor 28-5
receives F (h5 (n)) and F (h6 (n)) and extracts the difference in amplitude on the frequency axis,
and the result is F (h5 (n)) / F (h6 (n)) is output, and the output from the difference extractor 285 is output from the output terminal 28-6.
[0105] Needless to say, when the amplitude frequency characteristic is expressed in decibels, the
output of the difference extractor 28-5 becomes F (h5 (n))-F (h6 (n)).
[0106] In this manner, the amplitude frequency characteristic difference detector 28 detects and
outputs the difference between the amplitude frequency characteristics of the two input signals.
[0107] FIG. 8 is a block diagram showing the configuration of an amplitude frequency
characteristic difference adjuster.
[0108] In FIG. 8, 29-1 and 29-2 are input terminals of the amplitude frequency characteristic
difference adjuster, 29-3 is a Fourier transformer, 29-4 is a divider, 29-5 is an inverse Fourier
transformer, 29-6, 29-7 is an output terminal.
[0109] The operation of the amplitude frequency characteristic difference adjuster 29 configured
as described above will be described.
[0110] First, a (n) (h1 (n), h4 (n)) is input from the input terminal 29-1 and from the output
terminal 28-6 of the amplitude frequency characteristic difference detector 28 of FIG. 7 from the
input terminal 29-2. The output amplitude frequency characteristic difference F (h5 (n)) / F (h6
(n)) is input, and a (n) input from the input terminal 29-1 is branched into two, one of which is an
output terminal 29-6, and the other is input to the Fourier transformer 29-3, and the flie
transformer 29-3 receives a (n) and performs flie transform, and the result is F (a (n) ) To divider
29-4.
[0111] In this divider 29-4, F (a (n)) and F (h5 (n)) / F (h6 (n)) are input and division is
10-04-2019
19
performed, and the result is F (a (n))・ F (h6 (n)) / F (h5 (n)) is output to the inverse Fourier
transformer 29-5, and in the inverse Fourier transform 29-5, F (a (n)) ・ F (h6 (n)) / F (h5 (n)) is
input, inverse Fourier transform is performed, and the result is F ~ 1 (F (a (n)) · F (h6 (n)) / F (h5
(n)) And the output from the inverse Fourier transformer 29-5 is output from the output terminal
29-6.
[0112] When the amplitude frequency characteristic is expressed in decibels, the input from the
input terminal 29-2 is F (h5 (n))-F (h6 (n)), and the output of the divider 29-4 is F (a (n)) + F (h6
(n))-F (h5 (n)), and the output of the inverse Fourier transformer 29-5 is F.about.1 (F (a (n)) + F
(h6 (n)). Needless to say,))-F (h5 (n))).
[0113] When the sound image is localized to the left with respect to the listener, the output from
the output terminal 29-6 is bL (n) and the output from the output terminal 29-7 is bR (n). When
the sound image is localized, the output from the output terminal 29-6 is bR (n), and the output
from the output terminal 29-7 is bL (n).
[0114] In this manner, the amplitude frequency characteristic difference adjustment unit 29
outputs the input signal after subtracting the amplitude on the frequency axis by the input
amplitude frequency characteristic difference from the amplitude frequency characteristic of the
input signal. Be done. That is, from one input signal, two signals are generated and output such
that the frequency characteristic difference is equal to the input frequency characteristic
difference. However, one of the output signals is equivalent to the input signal.
[0115] Thus, the target impulse responses h5 (n) and h6 (n), which are target characteristics, are
not input as they are to the matrix calculator 11, but the vibration of the reproduction impulse
response h1 (n) or h4 (n) The sound image localization coefficients hL (n) and hR (n) calculated
by inputting the frequency characteristic adjusted to the matrix calculator are set in the sound
image localization apparatus and are predetermined speakers for reproduction. When you
connect impulses and input impulses to the sound image localization device, to localize the sound
image on the left side of the listener, the impulse response in the listener's left ear equals h1 (n)
and the impulse response in the listener's right ear The amplitude frequency characteristic of h1
(n) is changed (however, the time delay due to feedback control is ignored).
[0116] Also, to localize the sound image to the right of the listener, the impulse response in the
listener's right ear is equal to h4 (n), and the impulse response in the listener's right ear changes
the amplitude frequency characteristics of h4 (n). (Similar to left-side localization, the time delay
10-04-2019
20
due to feedback control is ignored. ), It has become a sound image localization coefficient.
[0117] As described above, sound image localization to the target position is realized by realizing
the amplitude frequency characteristic difference of the sound reaching the listener's left and
right ears when the sound is output from the speaker placed at the position where the sound
image is to be localized. Not only is this possible, it is possible to minimize the difference in
timbre between the sound image localized by the sound image localization device and the sound
reproduced from the reproduction speaker without using the sound image localization device. By
this, it is possible to calculate a sound image localization coefficient that minimizes the sound
quality deterioration of the localized sound image by the sound image localization apparatus.
[0118] Although this embodiment distinguishes the case where the sound image is localized on
the left side and the right side of the listener, either h1 (n) or h4 (n) is always used regardless of
the position of the sound image localization. May be input.
[0119] Also, in this embodiment, there are two output signals from the amplitude frequency
characteristic difference adjuster, but one of the output signals is an output signal from the
switch 30 according to the adjustment method with the amplitude frequency characteristic
difference adjuster. Therefore, the output of the signal equivalent to the output signal of the
switch 30 is not output, and the output signal of the switch 30 is branched to the matrix operator
as one output signal from the amplitude frequency characteristic difference regulator. You may
enter it.
[0120] As described above, according to the sound image localization coefficient calculation
apparatus of the present invention, the arrival time difference of the sound reaching the left and
right ears of the listener when the sound is output from the speaker placed at the position where
the sound image is to be localized. Not only is it possible to localize the sound image to the target
position by realizing the sound pressure difference, or by realizing the frequency characteristic
difference of the sound that has reached the listener's left and right ears, the sound image
localization device When the sound image is localized on the left side, it is possible to minimize
the difference in timbre generated between the localized sound image and the sound reproduced
from the left reproduction speaker without using the sound image localization apparatus. It
becomes.
[0121] Also, when the sound image is localized on the right side of the listener by the sound
image localization apparatus, the difference in timbre generated between the localized sound
10-04-2019
21
image and the sound reproduced from the right reproduction speaker without using the sound
image localization apparatus It is possible to minimize the By this, it is possible to calculate a
sound image localization coefficient that minimizes the sound quality deterioration of the
localized sound image by the sound image localization apparatus.
10-04-2019
22
Документ
Категория
Без категории
Просмотров
0
Размер файла
36 Кб
Теги
description, jph06178399
1/--страниц
Пожаловаться на содержимое документа