close

Вход

Забыли?

вход по аккаунту

JP2000134688

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 JP2000134688
[0001]
TECHNICAL FIELD The present invention relates to a microphone array device. In particular, the
present invention relates to an apparatus capable of three-dimensionally arranging microphone
arrays, estimating a sound received at an arbitrary position in space by sound signal processing,
and estimating sounds at many positions with a small number of microphones.
[0002]
2. Description of the Related Art A sound estimation processing technique using a conventional
microphone array device will be described below.
[0003]
In the microphone array device, a plurality of microphones are arranged, and signal processing is
performed using sound signals received by the microphones.
Here, in the microphone array device, the purpose, configuration, application, and effect greatly
differ depending on how the microphones are arranged in the sound field, what kind of sound is
received, and what kind of signal processing is performed. It is. When there are multiple target
signals and noise sources in the sound field, enhancement of high-quality target signals and noise
suppression are central issues in the processing of sound reception by microphones, and
detection of the source position is a TV conference system Useful for various applications such as
03-05-2019
1
visitor reception systems. In order to realize the target signal emphasis, noise suppression, and
sound source position detection processing, it is effective to use a microphone array device.
[0004]
In the prior art, in order to improve the quality of target signal emphasis, noise suppression, and
sound source position detection, the number of microphones constituting the array is increased,
and a large number of data of sound reception signals are collected to execute signal processing.
FIG. 14 shows a microphone array device used for target signal enhancement processing by
conventional synchronous addition. In the microphone array apparatus shown in FIG. 14,
reference numeral 141 denotes actual microphones MIC0 to MICn-1 constituting the
microphone array, delay devices D0 to Dn-1 for adjusting the timing of signals received by each
actual microphone 141, and each actual microphone It is an adder 143 which performs addition
processing of the sound reception signal at 141. The target sound emphasis according to the
related art emphasizes the sound from a specific direction by adding a large number of sound
reception signals which are elements to be added. That is, by increasing the number of actual
microphones 141, the sound signal used for synchronous addition signal processing is increased,
and the strength of the target signal is increased, thereby emphasizing and clearly extracting the
target signal. With regard to noise suppression, noise suppression is performed by performing
synchronous subtraction, and synchronous addition or cross-correlation coefficient calculation is
performed for the sound source position detection process in the assumed direction, and sound
signals are obtained by increasing the number of microphones. It was similar in the point which
improves processing.
[0005]
However, in the microphone array signal processing technology by increasing the number of
microphones, the number of microphones to be prepared increases in order to realize highquality sound reception signal processing, and the size of the microphone array device is
increased. It has the drawback of becoming large. It is also assumed that it may be difficult to
physically arrange the required number of microphones at the required positions in order to
perform the required quality of the received signal estimation.
[0006]
03-05-2019
2
In order to solve the above problems, instead of actually installing and receiving a microphone, a
sound signal that will be received at an assumed position based on a sound signal received from
the actually arranged microphone It is desirable to estimate Furthermore, it is possible to
perform target signal emphasis, noise suppression, a sound source position detection process etc.
as an application form using the sound reception estimation signal.
[0007]
The microphone array device is an effective device capable of estimating the sound reception
signal at any position on the array arrangement with a small number of microphones. Since the
space in which the actual sound propagates is a three-dimensional space, it is preferable that the
microphone array device can perform sound reception signal estimation at an arbitrary position
in the three-dimensional space. In other words, a small number of microphones can be arranged
in a straight line and the estimation error can be suppressed not only for the received signal
estimation at the assumed position on the extension line (one dimension) but also for the signal
from the sound source not on the extension line. Quality received signal estimation is required.
[0008]
In addition, it is desirable to develop better signal processing technology for the signal
processing content itself applied to sound signal estimation, and to improve the quality of target
signal emphasis, noise suppression, and sound source position detection.
[0009]
In view of the problems of the above-described conventional microphone array device, the
present invention realizes a microphone array device in which a small number of microphones
are three-dimensionally arranged, and a small number of microphones can be used An object of
the present invention is to provide a microphone array device that can be estimated.
[0010]
In addition, even if the number of microphone arrays and the arrangement location can not be
made ideal, the interpolation processing or the like can be used to predict and compensate for
the sound reception signal at positions between the plurality of discretely arranged microphone
positions. An object of the present invention is to provide a microphone array device capable of
performing quality received signal estimation.
03-05-2019
3
[0011]
In addition, a microphone array device capable of performing high-quality sound reception signal
estimation by realizing estimation processing superior to sound reception signal estimation at an
arbitrary position in a three-dimensional space than sound reception signal estimation processing
used in the prior art microphone array device. Intended to provide.
[0012]
SUMMARY OF THE INVENTION In order to achieve the above object, a microphone array device
according to the present invention is a microphone array device comprising a plurality of
microphones and a sound reception signal processing unit, wherein the microphones have
respective spatial axes. At least three of the sound reception signal processing units are disposed
on the upper side, and the sound reception signal processing unit is configured to set the
difference between the near points on the time axis of the sound pressure of the sound reception
signal of each microphone; Each spatial axis using the relationship between the difference of the
sound pressure, ie, the inclination, and the difference between the sound pressure's spatial point
on the space axis, ie, the difference between the inclination and the air particle velocity on the
time axis Based on the temporal change of sound pressure of the sound reception signal of each
of the arranged microphones in the direction and the spatial change of air particle velocity, the
sound reception signal at each axial component at any position is estimated and threedimensionally synthesized By before And estimating a sound signal in an arbitrary position in
space.
[0013]
According to the above configuration, the inclination on the time axis of the sound pressure
calculated from the temporal change of the sound pressure of the sound signal received by each
microphone and the sound reception signal between the microphones arranged on each axis The
sound signal at any position in space can be estimated by using the relationship between the
calculated air particle velocity and the inclination on the spatial axis.
[0014]
Further, in order to achieve the above object, the microphone array device of the present
invention is a microphone array device including a plurality of microphones and a sound
reception signal processing unit, and at least three microphones are arranged in one direction.
The boundary conditions of sound estimation of each plane constituting at least three layers
three-dimensionally arranged in at least three layers so that the planes do not intersect in units
of planes arranged in at least three lines so that the microphone rows do not intersect
Microphones arranged so as to be obtained, wherein the sound reception signal processing unit
estimates the sound pressure of sound reception signals of the respective microphones between
03-05-2019
4
adjacent points on The relationship between the difference between nearby points on the spatial
axis of velocity, that is, the slope, and the difference between the nearby points on the spatial axis
of sound pressure, that is, the relationship between the slope and air particle velocity on the time
axis Based on the temporal change in sound pressure of the sound reception signal and the
spatial change in air particle velocity at at least three locations aligned in one direction, using the
relationship with the difference between According to the present invention, sound signals at at
least three positions are estimated, and the sound signals are estimated in a direction intersecting
the one direction based on the sound signals at the estimated three positions.
[0015]
With the above configuration, it is possible to obtain the boundary conditions of sound
estimation of each surface constituting the three dimensions from each microphone, and the time
axis of the sound pressure calculated from the temporal change of the sound pressure of the
sound signal received by each microphone Arbitrary three-dimensional space using the
relationship between the inclination above and the inclination of the air particle velocity on the
space axis calculated based on the sound reception signal between the microphones arranged on
each axis. The sound signal of the position of can be estimated.
[0016]
Further, in order to achieve the above object, the microphone array device of the present
invention is a microphone array device including a plurality of directional microphones and a
sound reception signal processing unit, wherein the directional microphones are arranged on
each spatial axis. At least two of the directivity are arranged, and the sound reception signal
processing unit is configured to determine the difference between the near points on the time
axis of the sound pressure of the sound reception signal of each microphone, that is, the near
point on the spatial axis Using the relationship between the difference between the sound
pressure and the inclination, and the difference between the sound pressure near the space axis,
ie, the difference between the inclination and the air particle velocity on the time axis, ie, the
inclination Based on the temporal change in sound pressure of the sound signal of each of the
directional microphones disposed in the axial direction and the spatial change in air particle
velocity, the sound signal at each axial component at an arbitrary position is estimated; Combine
dimensions By, and estimating a sound signal in an arbitrary position on the space.
[0017]
With the above configuration, the inclination on the time axis of the sound pressure calculated
from the temporal change of the sound pressure of the sound signal received by each directional
microphone, and the directivity in which the directivity is matched on each axis and arranged
The sound signal of an arbitrary position in space can be estimated using the inclination on the
03-05-2019
5
spatial axis of the air particle velocity calculated based on the sound reception signal between the
microphones and the correlation between the both.
[0018]
Next, in order to achieve the above object, the microphone array device of the present invention
is a microphone array device including a plurality of directional microphones and a sound
receiving signal processing unit, wherein the directional microphones are arranged in one
direction. A unit of at least two planes arranged at least in two lines so that at least two
directional microphone lines arranged with directivity do not intersect, arranged in at least two
layers three-dimensionally so that the planes do not intersect, and constituting three dimensions
Directional microphones arranged so as to obtain boundary conditions of sound estimation for
each surface, and the sound reception signal processing unit is used to estimate sound in each
direction in three dimensions, the time axis of the sound pressure of the sound reception signal
of each microphone The difference between the neighboring points on the space axis, ie the
difference between the neighboring points on the spatial axis of the air particle velocity, and the
difference between the neighboring points on the spatial axis of the sound pressure, ie The
temporal change in sound pressure and air particle velocity of the sound reception signal of at
least two positions aligned in one direction, using the relationship between the inclination and
the difference between nearby points on the time axis of the air particle velocity and the air
particle velocity Based on the spatial change, sound signals at at least two positions are estimated
along a direction intersecting the one direction, and sound signals in the directions orthogonal to
the one direction are further calculated based on the sound signals at the two estimated
positions. It is characterized by estimating a signal.
[0019]
With the above configuration, it is possible to obtain the boundary conditions of sound
estimation of each surface constituting the three dimensions from each directional microphone,
and the sound calculated from the temporal change of the sound pressure of the sound signal
received by each directional microphone Use of the inclination of pressure on the time axis and
the inclination of air particle velocity on the space axis calculated based on the sound reception
signal between directional microphones arranged on each axis and the correlation between the
both Then, it is possible to estimate the sound signal at any position on the three-dimensional
space.
[0020]
Next, in the microphone array device, the relationship between the inclination on the time axis of
the sound pressure of the sound reception signal and the inclination on the spatial axis of the air
particle velocity is expressed by the equation (Equation 2) Is preferred.
03-05-2019
6
[0022]
Here, x, y and z are space axis components, t is a time component, v is an air particle velocity, p is
a sound pressure, and b is a coefficient.
Next, the microphone array device is adapted to estimate the sound signal at an arbitrary
position in the space, and to provide the sound pressure of the sound signal in the space
direction and the fluctuation of the air particle velocity in another space axis direction. It is
preferable to perform the sound signal estimation process for each spatial axis direction by
treating the sound pressure of the sound signal and the influence of the fluctuation of the air
particle velocity as negligible.
[0023]
According to the above configuration, sound signal processing can be performed as a plane wave
independent of sound signals on each space axis, and sound pressure and air particle velocity can
be efficiently estimated for each space axis.
[0024]
Next, it is preferable that the sound receiving signal processing unit includes a parameter input
unit that receives an input of a parameter for adjusting the content of signal processing.
According to the above configuration, the user can adjust and designate the signal processing
content for the microphone array device.
[0025]
Next, it is preferable that the distance between the arranged adjacent microphones is within a
distance satisfying the sampling theorem on the spatial axis at the frequency of the sound
reception signal.
03-05-2019
7
According to the above configuration, high quality signal processing can be performed in a
necessary frequency range by satisfying the sampling theorem.
[0026]
Next, it is preferable that the microphone array device includes a microphone distance
adjustment unit that variably adjusts the distance between the arranged microphones.
According to the above configuration, the distance between the microphones can be variably
adjusted by an instruction from the outside, automatic adjustment, or the like, and the sampling
theorem can be satisfied in a necessary frequency range.
[0027]
Next, the sound reception signal processing unit includes a microphone position interpolation
processing unit that virtually variably adjusts the distance between the arranged microphones by
performing position interpolation processing on the signals received by the microphones. Is
preferred.
[0028]
According to the above configuration, spatial position interpolation of the sound reception signal
position such as one-dimensional interpolation is performed without moving each microphone
interval itself, so that the sampling theorem can be obtained among a plurality of sound
reception signals without moving each microphone interval itself. Can be adjusted to meet
[0029]
Next, it is preferable that the sound reception signal processing unit includes a sampling
frequency adjustment unit that adjusts a sampling frequency of sound reception processing in
the microphone.
According to the above configuration, by adjusting the sampling frequency, the frequency of the
sound signal and the distance between the microphones can be adjusted to satisfy the sampling
03-05-2019
8
theorem.
[0030]
Next, it is preferable that the sound reception signal processing unit includes a band processing
unit that performs frequency division processing of band division of the sound reception signal
by the microphone and band synthesis.
According to the above configuration, it is possible to adjust the visual signal band, shift the
frequency of the signal received by the microphone, and obtain the same effect as adjusting the
sampling frequency of the signal received by the microphone.
[0031]
Next, it is preferable that the parameter given to the parameter input unit be a sound signal
emphasizing direction parameter specifying a specific direction for emphasizing sound signal
estimation, and the estimation of the sound signal from the sound source in the specific direction
be emphasized. .
[0032]
According to the above configuration, it is possible to give the direction in which the sound signal
estimation is to be emphasized as a parameter, the target signal can be emphasized, and highquality recording can be performed.
Next, it is preferable that the parameter given to the parameter input unit be a sound signal
attenuation direction parameter specifying a specific direction in which the sound signal
estimation is to be reduced, and the sound signal from the sound source in the specific direction
be removed.
[0033]
According to the above configuration, it is possible to give a noise source and the like, as a
03-05-2019
9
parameter, a direction in which it is desired to attenuate the mixed noise, noise can be
suppressed, and high quality recording can be performed.
Next, based on the sound signals estimated at a plurality of arbitrary positions in the sound field,
the microphone array device detects a position at which the cross-correlation is the largest using
a cross correlation function between the estimated sound signals; It is preferable to estimate the
sound source position.
[0034]
With the above configuration, the sound source position can be estimated, and then, by setting
the target signal emphasis direction parameter as the target sound emphasis direction, or by
matching the directivity of the directional microphone with the sound source direction, etc.
Sound processing can be performed.
[0035]
Next, in the microphone array device, the sound reception signal processing unit includes a
sound power detection unit, and the sound power detection unit checks the power of the sound
signal estimated in a certain direction, and whether or not the sound source is in the direction It
is preferable to detect the
[0036]
With the above configuration, it can be detected whether or not there is a sound source in the
assumed direction.
Further, in order to achieve the above object, the present invention is a microphone array device
including a plurality of microphones and a sound reception signal processing unit, and a plurality
of the microphones are arranged in three orthogonal axis directions of a predetermined space,
The sound reception signal processing unit connected to the microphone is characterized in that
a sound signal at an arbitrary position outside the microphone arrangement space is estimated
based on the relationship between the installation position of the microphone and the sound
reception signal.
[0037]
03-05-2019
10
With the above configuration, it is possible to estimate a sound signal at an arbitrary position
outside the space where the microphones are arranged.
Next, it is preferable that the microphones are mutually connected and supported on a
predetermined spatial axis.
[0038]
It is preferable that the support material be rigid and have a thickness less than 1/2, preferably
less than 1/4 of the wavelength of the maximum frequency of the reception signal, and be hard
to vibrate under the influence of sound.
With the above configuration, microphones to be actually arranged can be arranged at
predetermined position intervals, and vibration due to sound can be suppressed, and a
microphone array device capable of reducing noise to the sound reception signal can be
provided.
[0039]
DESCRIPTION OF THE PREFERRED EMBODIMENTS The microphone array device of the present
invention will be described with reference to the drawings.
First, the basic principle of the sound reception signal estimation process of the microphone
array device of the present invention will be described.
[0040]
Sound is a vibrational wave of air particles as a medium, and the change in pressure of air
generated by sound waves, that is, "sound pressure p", and the time derivative of the change in
position of air particles (displacement) by time, that is, "air particle velocity Two wave equations
shown by the following (Equation 3) are established between v ′ ′.
03-05-2019
11
[0042]
Here, t is time, x, y and z are orthogonal coordinate axes defining a three-dimensional space, K is
bulk modulus (ratio of pressure to expansion),, is density of air as a medium (unit volume Mass)
is shown.
Here, the sound pressure p is a scalar quantity, and the particle velocity v is a vector.
Further, ▽ on the left side of (Equation 3) indicates a type of partial differential operation, and in
the case of orthogonal coordinates (x, y, z), (Equation 4) is expressed.
[0044]
In equation (4), xI, yI and zI represent vectors of unit lengths in the x, y and z axis directions,
respectively.
The right side of (Equation 3) indicates a partial differential operation at time t.
Consider converting the two wave equations shown in (Equation 3) into difference equations of
the type that is handled in actual calculation.
(Equation 3) can be converted to (Equation 5) to (Equation 8).
[0049]
Here, a and b indicate constant coefficients.
Further, tk indicates a sampling time, and xi, yj, zg indicate estimated positions on the x, y, z axes,
and in this case, they are equally spaced.
03-05-2019
12
Also, vx, vy, vz, indicate the x, y, z axis components of the particle velocity.
[0050] Now, as an example of the three-dimensional arrangement of the microphones of the
microphone array device of the present invention, it is assumed that three microphones are
arranged at equal intervals in the x, y and z axis directions. It is a microphone array in which 27
microphones in total of 3 × 3 × 3 are arranged, and the arrangement of microphones is (x0, x1,
x2) as x-coordinates and (y0, y1, y2) as y-coordinates (z0) as z coordinates , Z1, z2). FIG. 1 is a
diagram in which the microphones on the xy plane in which the value of z is z1 in the
microphone array device are highlighted.
[0051] First, in the microphone array device of this three-dimensional arrangement, it is assumed
that the direction of the sound source is one and the direction of the sound source is known. In
order to simplify the explanation, the received signal estimation of a certain point on the x axis is
performed for convenience. A method for estimating the sound pressure and the air particle
velocity in the x-axis direction using the above (Equation 5), (Equation 6) and (Equation 8) in the
reception signal estimation in the x-axis direction of FIG. 1 will be shown below. The y-axis
direction can also be estimated by the same process.
[0052] Now, in the microphone array device shown in FIG. 1, there is a problem that the
equation (8) can not be used as it is because the particle velocity vz in the z-axis direction can not
be obtained. Then, the z-axis component of the air particle velocity is removed from (Equation 8)
to create (Equation 9).
[0054] Here, b 'is a coefficient depending on the sound source direction θ with reference to the
xy plane as shown in (Equation 10).
[0056] Thus, when there is only one sound source and the sound source direction is known,
(Equation 9) can be used for the received signal estimation process, and as shown in (Equation
10), depending on the sound source direction θ. It turns out that it is sufficient to change the
coefficient b '. However, in order to estimate signals from a plurality of sound sources in
unknown directions, an estimation method that does not depend on the sound source direction
θ is required. Therefore, an estimation method that does not depend on the sound source
direction θ will be considered below.
03-05-2019
13
[0057] In general, the distance that the sound source travels in a short time 1 / Fs is not large, so
assuming that the sound source direction θ does not change significantly, the following
(Equation 11) is established. Here, Fs is a sampling frequency.
[0059] Here, if the following (Equation 12) is used, the right side of (Equation 9) can be
estimated from the right side of (Equation 11).
[0061] The coefficient cq in equation (12) is calculated using the following equation (13).
[0063] Similarly, as shown in (Equation 14) using the coefficient cq, the left side of (Equation 9)
can be estimated from the left side of (Equation 11).
[0065] Next, an example of sound reception signal estimation at an arbitrary point will be shown
by the processing of the above equation. As shown in FIG. 1, the microphones are actually
arranged, and the sound reception signal at a point where the microphones are not actually
arranged is estimated based on the sound reception signal obtained from the sound source. (X3,
y0, z1) is selected as a point at which the microphone is not disposed, and first, the sound
pressure p (x3, y0, z1, tk) at time tk at that point is estimated.
[0066] The sound pressure p is estimated using (Equation 5), (Equation 6), (Equation 13) and
(Equation 14). Here, xi−xi−1 = yj−yj−1 = (sound speed / sampling frequency). In this case, a
becomes 1 in equation (4).
[0067] First of all, from the sound signal received by each microphone, the velocity of the next
air particle, vx (x0, y0, z1, tk), vx (x1, y0, z1, tk), vy (x0, y0, z1, Calculate tk), vy (x0, y1, z1, tk), vy
(x1, y0, z1, tk), vy (x1, y1, z1, tk).
[0068] The equations (15) and (16) are derived from the equations (5) and (6).
[0070] 【0070】ここで、i=0,1、j=0、g=1である。
03-05-2019
14
[0072] 【0072】ここで、i=0,1、j=0,1、g=1である。 Second, the coefficients
c-1, c0 and c1 are calculated. The equation (17) is derived from the equation (13).
[0074] Next, the air particle velocity vx (x2, y0, z1, tk) at x2 is calculated. The equation (18) is
derived from the equation (14).
[0076] Finally, the sound pressure p (x3, y0, z1, tk) at x3 is calculated. The equation (19) is
derived from the equation (4).
[0078] The sound pressure p and the air particle velocity v at any point on the x-axis can be
estimated by repeating the above first to fourth processes in the same manner in the x-axis
direction. Next, a specific example of a microphone array device to which the basic principle of
the sound reception signal estimation process at an arbitrary position in the above threedimensional space is applied will be shown as an embodiment. The arrangement of the
microphones, the device of the distance between the microphones, the device of the sampling
frequency, and the like will also be described.
[0079] (Embodiment 1) FIG. 2 shows an example in which three microphones are arranged on
each axis as an example in which at least three microphones are arranged on each space axis.
[0080] In this type of microphone array device, the estimation of the sound reception signal at
an arbitrary position S (xs1, ys2, zs3) is performed at each position corresponding to the
component on the space axis of the arbitrary position S in the defined three-dimensional space
The signal is estimated and calculated as a vector sum of three-dimensional components.
[0081] As shown in FIG. 2, upon estimation of the received sound signal of the assumed position
S (xs1, ys2, zs3) in the defined three-dimensional space, first, reception of a position indicating a
component on the spatial axis of the assumed position S Perform sound signal estimation. That is,
first, applying the basic principle of the above-described reception signal estimation processing,
(xs1, 0, 0) on the x axis, (0, ys2, 0) on the y axis, (0, 0, on the z axis The received signal
estimation is performed at each position of zs3). Next, the estimated sound reception signal at the
assumed position S may be obtained by synthetically calculating the vector sum of the estimated
sound reception signals of the respective axis components.
03-05-2019
15
[0082] Here, in the embodiment in which each component in the spatial axis direction is
combined to obtain an estimated sound reception signal, another spatial axis is given to the
fluctuation of the sound pressure of the sound signal and the air particle velocity in one spatial
axis direction. The sound reception signal estimation process can be easily handled by treating
the sound pressure of the sound signal in the direction and the influence of the fluctuation of the
air particle velocity as negligible.
[0083] As described above, in each spatial axis direction, according to the above-described basic
principle of the received signal estimation, the difference between the near points on the time
axis of the sound pressure of the received signal of each microphone, that is, the inclination and
air particle velocity on the space axis. Using the relationship between the difference between the
nearby points, ie, the slope, and the difference between the nearby points on the spatial axis of
the sound pressure, ie, the gradient, and the difference between the nearby points on the time
axis of the air particle velocity, ie, the slope, Based on the temporal change in sound pressure of
the sound reception signal of each of the arranged microphones in each spatial axis direction and
the spatial change in air particle velocity, the sound reception signal in each axial component at
any position is estimated; By combining the dimensions, the sound signal at any position in the
space can be estimated.
[0084] Second Embodiment As shown in FIG. 3, in the microphone array device of the second
embodiment, at least three microphones are arranged in a plane so that at least three
microphones arranged in one direction do not intersect each other. As an example of a
microphone array device of a minimum configuration, as an example of microphone array
devices arranged at least three levels in a three-dimensional manner so that the planes do not
intersect and arranged to obtain boundary conditions for sound estimation of each plane
constituting three dimensions This is an example in which the microphones of
[0085] In this type of microphone array device, estimation of the sound reception signal at an
arbitrary position S (xs1, ys2, zs3) is performed in at least three lines in one direction (for
example, a direction parallel to the x axis) as shown in FIG. These three estimated sound
reception signals are obtained as the next step by obtaining sound reception signals at
predetermined positions (for example, (xs1, y0, z0), (xs1, y1, z0), (xs1, y2, z0)) from In this case,
a sound receiving signal at a predetermined position (e.g., xs1, ys2, z0) in the next axis
component is determined, assuming that it is an estimated sequence in. Repeating this process,
as shown in FIG. 4b, determine at least three in the next axial direction (for example, remaining
(xs1, ys2, z1), (xs1, ys2, z2)), and these three estimated sound receiving signals The final
estimated sound reception signal is obtained from (arbitrary position S (xs1, ys2, zs3)).
03-05-2019
16
[0086] As described above, in the microphone array device according to the second embodiment,
for each direction and each row, the sound pressure of the sound reception signal of each
microphone is determined between the near points on the time axis according to the abovedescribed basic principle of the sound reception signal estimation. Difference, ie, the relationship
between the inclination and the difference between adjacent points on the spatial axis of the air
particle velocity, and the difference between the adjacent points on the spatial axis of the sound
pressure, ie, the inclination and the vicinity of the air particle velocity on the time axis
Intersection with one direction based on temporal change of sound pressure of sound reception
signal and spatial change of air particle velocity in at least three positions aligned in one
direction, using the relation with difference between points, that is, inclination. It is possible to
estimate sound signals at at least three positions along the direction, and further to estimate the
sound signals in the direction intersecting the one direction based on the sound signals at the
estimated three positions.
[0087] (Third Embodiment) In the third embodiment, directional microphones are used as
microphones to be used, and upon arrangement of each directional microphone, the direction of
each directivity is used in accordance with each axial direction. . Boundary conditions in one
direction with directivity can be obtained the same effect as originally obtained.
[0088] FIG. 5 shows an example in which two or more directional microphones are used, and at
least two directional microphones are arranged on each spatial axis with directivity as a
minimum configuration.
[0089] In this type of microphone array apparatus, the directivity is matched along each axis,
and the estimation of the sound reception signal at an arbitrary position S (xs1, ys2, zs3) is
performed in the space of an arbitrary position S in a defined three-dimensional space. The sound
reception signal at each position corresponding to the on-axis component is estimated from the
two sound reception signals and calculated as a vector sum of three-dimensional components.
[0090] As in the first embodiment, in the embodiment in which the components in the spatial
axis direction are combined to obtain the estimated sound reception signal, the variation is
applied to the sound pressure of the sound signal and the air particle velocity in one spatial axis
direction. The sound reception signal estimation process can be easily handled by treating the
influence of the sound pressure of the sound signal in the other spatial axis direction and the
fluctuation of the air particle velocity as negligible.
03-05-2019
17
[0091] Thus, the microphone array device of the third embodiment uses at least two directional
microphones in each spatial axis direction to determine the difference between the near points
on the time axis of the sound pressure of the sound reception signal of each microphone, that is,
the inclination And the relationship between the difference between adjacent points on the
spatial axis of air particle velocity, that is, the inclination, and the difference between the adjacent
points on the spatial axis of sound pressure, that is, the difference between the inclination and air
particle velocity on the time axis By using the relationship with the difference or inclination,
based on the temporal change of the sound pressure of the sound reception signal of each of the
directional microphones arranged in each spatial axis direction and the spatial change of the air
particle velocity, The sound signal at any position in the space can be estimated by estimating the
sound reception signal in each axis component and combining it three-dimensionally.
[0092] Fourth Embodiment In the fourth embodiment, directional microphones are used as
microphones to be used. As shown in FIG. 6, at least two directional microphones are arranged in
one direction in arranging each directional microphone. Boundary condition of sound estimation
of each surface constituting at least two layers and three-dimensionally arranged at least in two
layers so that the planes do not intersect in units of planes arranged in at least two rows so that
the microphone rows do not intersect As an example of the microphone array device arranged to
be obtained, this is an example in which eight directional microphones of the minimum
configuration are arranged. As in the third embodiment, the same effect as that obtained from
the beginning of the boundary condition in one direction in which the directivity is matched can
be obtained. In addition, except for the point that it is possible to estimate the received signal
estimation process for one direction and column from the two signals, the received signal
estimation process for an arbitrary position S in the three-dimensional space is the one described
in the second embodiment. Is the same as
[0093] (Fifth Embodiment) In the fifth embodiment, the characteristic of the microphone array
device can be adjusted by devising the distance between the microphones to be arranged, and
the distance between adjacent microphones arranged can be set to Within the interval satisfying
the sampling theorem on the spatial axis in frequency. The certainty of the estimation process
shown in the above-mentioned basic principle of received signal estimation becomes higher as
the microphone interval is narrower. At this time, the maximum value lmax of the distance
between adjacent microphones is expressed by Equation 20 because it is necessary to satisfy the
sampling theorem.
[0095] As described above, it is sufficient that the distance between adjacent microphones is
within the range satisfying (Equation 20) with respect to the maximum frequency of the sound
03-05-2019
18
signal to be received which is assumed. As shown in FIG. 7, the microphone array device of the
fifth embodiment includes a microphone distance adjustment unit 73 that variably adjusts the
distance between the arranged microphones. For example, the moving device is attached to the
support of the microphone by the microphone interval adjustment unit 73, and the microphone
itself is moved by external input instruction or autonomous adjustment, and the microphone
according to the frequency characteristic of the sound output from the sound source Adjust the
interval variably.
[0096] When the microphone spacing is reduced to satisfy (Equation 20), it is necessary to adjust
the coefficients with respect to (Equation 5) to (Equation 8) shown in the above described sound
reception signal estimation processing. Assuming that the maximum value of the microphone
spacing is lmax, and the coefficients a and b in (Equation 5) to (Equation 8) are abase and bbase,
the coefficients for the interval l become the values shown in (Equation 21) .
[0098] As described above, the microphone array is moved so that the microphone spacing can
be variably adjusted by moving the microphone itself according to an external input instruction
to the microphone spacing adjusting unit 73 or by autonomous adjustment, so that the equation
(20) is satisfied. The device configuration of the device can be adjusted.
[0099] (Sixth Embodiment) In the sixth embodiment, the sound reception signal estimation
process of the microphone array device according to the present invention performs sampling on
the spatial axis shown in equation 20 with respect to the frequency characteristics of the sound
output from the sound source. A microphone array device that can be adjusted to satisfy the
theorem, which is similar to that of the fifth embodiment by interpolating on a spatial coordinate
axis, instead of the method of actually changing the distance between microphones shown in the
fifth embodiment. Get the effect.
[0100] Here, in order to simplify the explanation, interpolation adjustment in the x-axis direction
is shown. Needless to say, similar interpolation adjustment is possible in the y-axis direction and
the z-axis direction. The sound receiving signal processing unit of the microphone array device
includes a microphone position interpolation processing unit as shown in FIG. The microphone
position interpolation processing unit 81 virtually variably adjusts the interval between the
arranged microphones by performing position interpolation processing on the signal received by
each microphone.
[0101] Let lbase be the original microphone spacing. If calculation is performed by interpolation
03-05-2019
19
as shown in (Equation 22), it is possible to estimate a sound reception signal similar to the case
where the adjacent microphone interval is changed to l.
[0103] As described above, the interpolation processing is performed on the frequency
characteristic of the sound output from the sound source by the microphone position
interpolation processing unit 81 so that the sampling theorem on the spatial axis shown in
(Equation 20) is satisfied. Can be adjusted.
[0104] (Seventh Embodiment) In the seventh embodiment, the sampling frequency is adjusted in
the sound receiving process by the microphone, the oversampling is performed on the frequency
characteristic of the sound output from the sound source, and the sound receiving signal
estimation process at an arbitrary position Improve the certainty of
[0105] In the microphone array device of the seventh embodiment, as shown in FIG. 9, the sound
reception signal processing unit includes a sampling frequency adjustment unit that adjusts the
sampling frequency of sound reception processing by the microphone. The sampling frequency is
changed by the sampling frequency adjustment unit 91 so as to be oversampling.
[0106] The probability of the estimation process shown in the above-mentioned basic principle
of received signal estimation becomes higher as oversampling is performed. At this time, the
minimum value Fsmin of the sampling frequency is Fsmin = (maximum frequency of the signal to
be received * 2) in order to satisfy the sampling theorem. The maximum frequency of this
received signal is determined by the cut-off frequency of the analog low-pass filter in front of the
AD (analog-digital) converter. Therefore, oversampling can be realized by increasing the sampling
frequency of the AD converter while keeping the cutoff frequency of the low pass filter constant.
[0107] Assuming that the coefficients of (Equation 5) to (Equation 8) when the sampling
frequency is Fsmin are abase and bbase, the coefficients when the sampling frequency is Fs have
values shown in the following (Equation 23).
[0109] As described above, by setting the sampling frequency to the oversampling by the
sampling frequency adjustment unit 91, it is possible to improve the certainty of the sound
reception signal estimation process at an arbitrary position.
03-05-2019
20
[0110] Eighth Embodiment In the eighth embodiment, an effect similar to that of sampling
frequency adjustment is obtained by dividing the sound reception processing by the microphone
into bands and shifting the frequency of each signal to a low frequency, and reception at an
arbitrary position is performed. It is intended to improve the certainty of the sound signal
estimation process.
[0111] The microphone array device of the eighth embodiment is shown in FIG. As shown in FIG.
10, the sound reception signal processing unit 72 includes a band processing unit 101 that
performs band division processing of the sound reception signal in the microphone array 71, up
sampling, and low pass filter processing. Relative sampling frequency adjustment is performed
by frequency-shifting the signal subjected to the band division processing by the band processing
unit 101 to the low band in the original band to improve the likelihood of the received signal
estimation processing at an arbitrary position. .
[0112] A tree structure filter bank or a polyphase filter bank can be used as the band division
filter 102 of the band processing unit 101. Here, the band division filter 102 divides the signal
into four bands. Next, upsampling is performed four times by zero filling by the upsampling unit
103. Finally, it passes through a low pass filter 104 whose cutoff frequency is Fc = Fs / 8.
[0113] The above-described frequency shift processing of the band processing unit 101 obtains
the same effect as the sampling frequency adjustment, and improves the likelihood of the sound
reception signal estimation processing at an arbitrary position. (Embodiment 9) In this
embodiment 9, only the estimated sound in a specific direction is emphasized by setting
parameters to the sound reception signal processing unit of the microphone array device, the
target sound is emphasized, and the estimated sound in a specific direction is attenuated. Noise
suppression.
[0114] FIG. 11 shows a configuration example of the microphone array device of the ninth
embodiment. The microphone array apparatus includes a parameter input unit 111 that receives
an input of a parameter for adjusting the content of signal processing.
[0115] The parameter given to the parameter input unit 111 is a sound signal emphasizing
direction parameter for designating a specific direction for emphasizing sound signal estimation,
and the specification shown in the basic principle as the reception signal estimation processing
of the reception signal estimation processing unit 72 The addition / subtraction processing unit
112 performs the addition processing on the estimation result of the direction of {circle over (1)}
03-05-2019
21
to emphasize the sound signal from the sound source in the specific direction.
[0116] Further, the parameter given to the parameter input unit 111 is a sound signal
attenuation direction parameter specifying a specific direction in which the sound signal
estimation is to be reduced, and an addition / subtraction processing unit as a reception signal
estimation processing of the reception signal estimation processing unit 72 The noise signal from
the specific direction is suppressed by performing subtraction processing for removing the sound
signal from the sound source in the specific direction by 112.
[0117] Tenth Embodiment A tenth embodiment is to detect whether or not there is a sound
source at a plurality of arbitrary positions in a sound field. In detecting the sound source, based
on the estimated sound signal, use the cross correlation function between the estimated sound
signals, or check the power of the sound signal obtained by performing synchronous addition of
the signals estimated in a certain direction, Whether or not to detect.
[0118] When using the cross correlation function between the estimated sound signals, as shown
in FIG. 12, the sound signal estimated in each direction by the cross correlation calculation unit
121 is used as a basis for the sound reception signal estimation of the sound reception signal
estimation processing unit 72. To calculate the cross-correlation between the estimated sound
signals. The sound source position can be estimated by detecting the position at which the crosscorrelation calculated by the sound source position detection unit 122 is the largest.
[0119] Further, as shown in FIG. 13, the sound reception signal processing unit 72 of the
microphone array device includes a sound power detection unit 131 for detecting the presence
of a sound source based on the sound power of the sound signal thus estimated. The sound
power detection unit 131 checks the power of the sound signal obtained by performing
synchronous addition of the signals estimated in the assumed direction, and the sound source
detection unit 132 detects the sound source in the direction when the sound power exceeds a
predetermined value. I will judge.
[0120] Here, as a result of synchronous addition in the x-axis direction, the sound power pow of
px (x1, y1, tk) is calculated using (Equation 24), and when the value is equal to or greater than
the threshold value, the x-axis direction Judge that there is a sound source in
[0122] For example, when the sound source to be detected is a person, the value of the sound
03-05-2019
22
power may be the sound power of the voice emitted by the person. If the detection sound source
is a car, it may be the power of the sound emitted by the engine sound.
[0123] In each of the above-described embodiments, the number of microphones constituting the
microphone array apparatus, the arrangement, and the interval having specific values are given
as an example for the convenience of description and are intended to be limited. Needless to say,
it is not a thing.
[0124] According to the microphone array device of the present invention, it is possible to
estimate received sound signals at more arbitrary positions with a smaller number of
microphones, which contributes to space saving.
[0125] According to the microphone array device of the present invention, the relationship
between the inclination on the time axis of the sound pressure of the sound reception signal of
each microphone and the inclination on the space axis of the air particle velocity, the inclination
on the space axis of the sound pressure and the air Using the relationship between particle
velocity and inclination on the time axis, arbitrary based on the temporal change of sound
pressure of the sound reception signal of each microphone and the spatial change of air particle
velocity in each spatial axis direction. The sound signal at any position in the space can be
estimated by estimating the sound reception signal in each axis component of the position of 3
and combining three-dimensionally.
[0126] Further, according to the microphone array device of the present invention, the boundary
conditions of sound estimation of each surface constituting the three dimensions can be obtained
from each microphone, and the inclination on the time axis of the sound pressure of the sound
reception signal of each microphone And the inclination of the air particle velocity on the space
axis, and the relationship between the inclination of the sound pressure on the space axis and the
inclination of the air particle velocity on the time axis, each of the arranged in each spatial axis
direction Based on the temporal change of sound pressure of the sound reception signal of the
microphone and the spatial change of air particle velocity, the sound reception signal in each
axial component at any position is estimated, and three-dimensional synthesis is performed on
the space. Sound signals at any position can be estimated.
[0127] Furthermore, according to the microphone array device of the present invention, high
quality signal processing can be performed in a necessary frequency range by satisfying the
sampling theorem. In order to satisfy the sampling theorem, the distance between microphones is
03-05-2019
23
variably adjusted, position interpolation processing of the sound reception signal is performed by
each microphone to perform virtual variable adjustment, the sampling frequency is adjusted, and
the frequency of the signal received by the microphone is shifted can do.
[0128] Further, according to the microphone array device of the present invention, it is possible
to perform the addition processing and the subtraction processing of the estimation signal by the
parameter setting given to the parameter input unit, and to perform the target sound emphasis
and the noise suppression.
[0129] Further, according to the microphone array device of the present invention, the sound
source position can be estimated by using the cross correlation function between the estimated
sound signals and detecting the sound power.
03-05-2019
24
1/--страниц
Пожаловаться на содержимое документа