JPWO2012172618

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 JPWO2012172618
An array microphone apparatus and the like that maintain high sound quality and realize highspeed convergence in sensitivity correction. The present array microphone device is an array
microphone device 1 having a microphone array 11 composed of a plurality of microphone units
11a to 11d, and a signal input for inputting a signal from the microphone units 11a to 11c to be
corrected as a correction target signal Unit 21; a reference signal input unit 22 for inputting a
reference signal; a gain variable unit 23 for amplifying or attenuating the correction target signal
so that the level of the correction target signal and the reference signal are substantially equal;
And a gain control unit 24 that controls the gain when amplifying or attenuating. When the
elapsed time from activation of the array microphone device 1 is less than a predetermined time,
the gain control unit 24 changes the gain by the first change amount per unit time, and the
elapsed time is equal to or longer than the predetermined time And a low speed gain update unit
211 that changes the gain by a second change amount per unit time smaller than the first change
amount.
Array microphone device and gain control method
[0001]
The present invention relates to an array microphone device, and more particularly to an array
microphone device having a sensitivity correction function.
[0002]
Conventionally, an array microphone device is known as one of directional microphones.
03-05-2019
1
The array myrophone device is a device in which a plurality of microphones are installed, and by
processing the input voice, spatial information of sound that can not be obtained by one
microphone can be acquired, and directivity control is performed. It is possible to estimate the
direction of arrival of noise. In recent years, the array microphone device has come to be used for
directivity control for remote sound collection and high sound-to-noise ratio (signal-to-noise
ratio) in a vehicle interior hands-free communication system. .
[0003]
As such an array microphone device, one having a function (sensitivity correction function) of
automatically correcting the sensitivity of each microphone unit constituting the microphone
array to the same characteristic is known (for example, see Patent Document 1). In this array
microphone device, sensitivity correction means for performing sensitivity correction is provided
at the front stage of the directivity synthesis means for performing directivity synthesis. Then,
even if the sensitivity of each microphone unit and the sensitivity of the circuits such as
microphone amplifiers accompanying it (these are collectively referred to simply as the
sensitivity of the microphone unit) are not identical due to changes over time or manufacturing
reasons, etc. The directional combining method works normally on the premise that all levels are
equal.
[0004]
Japanese Patent Application Laid-Open No. 7-131886
[0005]
However, in the conventional array microphone device, when the update amount of gain per time
when amplifying or attenuating the signal from each microphone unit is large, the sound
fluctuation or distortion may be caused by hearing. .
In addition, since the sensitivity correction function is always operated to cope with changes over
time, the amount of gain update per time usually affects the sound quality and distortion factor
of the sound collected by the array microphone device. It is necessary to make it a very small
amount. Furthermore, when the amount of gain update is large, if the amount of gain update per
03-05-2019
2
unit time is small, it takes a long time to converge the gain.
[0006]
These circumstances lead to performance deterioration due to directivity synthesis failure at the
time of activation of the array microphone device, and an increase in manufacturing cost due to
an increase in inspection time in an inspection process at the time of manufacturing the array
microphone device.
[0007]
The present invention has been made in view of the above circumstances, and it is an object of
the present invention to provide an array microphone device and a gain control method for
maintaining high sound quality and realizing high-speed convergence in sensitivity correction.
[0008]
An array microphone apparatus according to the present invention is an array microphone
apparatus having a microphone array configured of a plurality of microphone units, and a
correction is performed to input a signal from a microphone unit to be corrected among the
plurality of microphone units as a correction target signal. A target signal input unit, a reference
signal input unit for inputting a reference signal, and a gain variable unit for amplifying or
attenuating the correction target signal such that the level of the correction target signal is
substantially equal to the level of the reference signal; And a gain control unit that controls a gain
when amplifying or attenuating the correction target signal, and the gain control unit is
configured to perform per unit time when an elapsed time from activation of the array
microphone device is less than a predetermined time. A first gain update unit that changes the
gain by a first change amount; and the elapsed time If serial is longer than a predetermined time,
the second change amount per unit smaller time than the first change amount, and a second gain
update unit for changing the gain.
[0009]
With this configuration, at the time of start-up operation of the array microphone device, the
sensitivity of the microphone unit to be corrected can be adjusted at high speed to the reference
sensitivity.
Therefore, it is possible to reduce the performance deterioration due to directivity synthesis
03-05-2019
3
failure, shorten the inspection time in the inspection process at the time of manufacturing, and
reduce the manufacturing cost.
[0010]
Further, at the time of the normal operation of the start transition of the array microphone
device, the sensitivity of the microphone unit to be corrected can be adjusted to the reference
sensitivity at a low speed.
Therefore, the sound quality of the sound collected by the array microphone device is not
affected by the sound fluctuation and the distortion factor is not deteriorated.
[0011]
That is, both the maintenance of the sound quality and the high-speed convergence of the
sensitivity correction can be achieved.
[0012]
Further, a gain control method of the present invention is a gain control method in an array
microphone apparatus having a microphone array constituted by a plurality of microphone units,
and a correction target signal input unit is used to correct a correction target among the plurality
of microphone units. A correction target signal input step of inputting a signal from the
microphone unit as a correction target signal, a reference signal input step of inputting a
reference signal by the reference signal input unit, a level of the correction target signal, and a
level of the reference signal And a gain control step of controlling a gain when amplifying or
attenuating the correction target signal so that the correction target signal is substantially equal,
and the gain control step includes the gain control step. The elapsed time from activation of the
array microphone device is less than a predetermined time If the elapsed time is equal to or
longer than the predetermined time, the first change amount per unit time may be smaller than
the first change amount. And a second gain update step of changing the gain by a second amount
of change.
[0013]
By this method, at the time of start-up operation of the array microphone device, the sensitivity
03-05-2019
4
of the microphone unit to be corrected can be adjusted at high speed to the reference sensitivity.
Therefore, it is possible to reduce the performance deterioration due to directivity synthesis
failure, shorten the inspection time in the inspection process at the time of manufacturing, and
reduce the manufacturing cost.
[0014]
Further, at the time of the normal operation of the start transition of the array microphone
device, the sensitivity of the microphone unit to be corrected can be adjusted to the reference
sensitivity at a low speed.
Therefore, the sound quality of the sound collected by the array microphone device is not
affected by the sound fluctuation and the distortion factor is not deteriorated.
[0015]
That is, both the maintenance of the sound quality and the high-speed convergence of the
sensitivity correction can be achieved.
[0016]
According to the present invention, high-speed convergence in sensitivity correction can be
realized while maintaining sound quality.
[0017]
Block diagram showing a configuration example of an array microphone device in the first
embodiment of the present invention. Block diagram showing a configuration example of a
sensitivity correction unit in the first embodiment of the present invention. Determination unit in
the first embodiment of the present invention, Flow chart showing operation example of high
speed gain update unit and low speed gain update unit (A) Diagram showing gain update
example at the time of high speed gain update in the first embodiment of the present invention,
(B) First embodiment of the present invention 12A shows an example of gain update at the time
of slow gain update in the embodiment, and FIG. 12C shows a diagram showing gain update at
the time of conventional gain update. FIG. 12A shows directivity when sensitivity correction is
performed in the first embodiment of the present invention. The figure which shows an example
03-05-2019
5
of the frequency characteristic in which directivity synthesis was carried out by the synthetic ¦
combination part, (B) When sensitivity correction is not implemented in 1st Embodiment of this
invention, an example of the frequency characteristic in which directivity synthesis was carried
out by the directivity synthetic ¦ combination part The Block diagram showing a configuration
example of the sensitivity correction unit in the second embodiment of the present invention.
Flow chart showing an operation example of the determination unit, the high speed gain update
unit, and the low speed gain update unit in the second embodiment of the present invention
[0018]
Hereinafter, embodiments of the present invention will be described in detail with reference to
the drawings.
[0019]
First Embodiment FIG. 1 is a block diagram showing a configuration example of an array
microphone device according to a first embodiment of the present invention.
The array microphone device 1 shown in FIG. 1 is configured to include a microphone array 11,
an analog-to-digital converter (hereinafter referred to as an AD converter) 13, and a digital signal
processor.
[0020]
The microphone array 11 includes a first microphone unit 11a, a second microphone unit 11b,
an (N-1) th microphone unit 11c, and an Nth microphone unit 11d.
The microphone units 11a to 11d are linearly arranged.
Also, the number of microphone units is not limited to this.
Although each of the microphone units 11a to 11d normally has the same acoustic
characteristics, the level of the output signal may differ depending on the external environment,
03-05-2019
6
the usage period, and the like.
[0021]
The AD conversion unit 13 is a first AD conversion unit 13a to which the first microphone unit
11a is connected, a second AD conversion unit 13b to which the second microphone unit 11b is
connected, and an (N-1) th microphone It includes an (N-1) th AD converter 13c to which the unit
11c is connected, and an Nth AD converter 13d to which the Nth microphone unit 11d is
connected.
The AD conversion unit 13 (each AD conversion unit 13a to 13d) converts an analog signal from
the microphone array 11 into a digital signal.
[0022]
The digital signal processing unit 14 is connected to the AD converters 13a to 13d, and performs
various digital signal processing on digital signals from the AD converters 13a to 13d. The digital
signal processing unit 14 further includes a sensitivity correction unit 15 and a directivity
synthesis unit 16. The digital signal processing unit 14 incorporates a known CPU, a ROM, a
RAM, etc., and the built-in CPU executes a program stored in the ROM to perform a sensitivity
correction function or directivity by the sensitivity correction unit 15. The directivity synthesis
function by the synthesis unit 16 is realized.
[0023]
The sensitivity correction unit 15 includes a first sensitivity correction unit 15a corresponding to
the first microphone unit 11a, a second sensitivity correction unit 15b corresponding to the
second microphone unit 11b, and an (N-1) th microphone unit 11c. , And an (N-1) th sensitivity
correction unit 15c corresponding to. The details of the sensitivity correction unit 15 (the
sensitivity correction units 15a to 15c) will be described later.
[0024]
03-05-2019
7
The directivity synthesis unit 16 synthesizes the digital signals (output signals of the sensitivity
correction units 15a to 15c) whose sensitivity has been corrected by the sensitivity correction
unit 15 so as to direct directivity in a specific direction. For example, the directivity synthesis unit
16 includes an adaptive filter and an adder (not shown), and can change the operation coefficient
to determine the directivity.
[0025]
Next, an operation example of the array microphone device 1 will be described.
[0026]
The sound waves collected by the microphone array 11 are converted into digital signals through
the microphone units 11a to 11d and the AD converters 13a to 13d.
The converted digital signal is subjected to the processing of the sensitivity correction units 15 a
to 15 c and the processing of the directivity synthesis unit 16 in the digital signal processing unit
14. For example, the sampling frequency of the digital signal is 16 kHz, and the digital signal
processing unit 14 also operates at this frequency.
[0027]
Each sensitivity correction unit 15a to each sensitivity correction unit 15c receives the output
signal of the Nth microphone unit 13d as a reference signal, and the level of the output signal of
each sensitivity correction unit 15a to 15c is from the Nth microphone unit 13d. The sensitivity
correction of each of the microphone units 11a to 11c is performed so as to be approximately
equal to the signal level. As a result, even when the sensitivity of each of the microphone units
11a to 11c varies, the levels of the input signals to the directivity synthesis unit 16 become
equal. Therefore, the directivity characteristic by the directivity synthesis unit 16 is correctly
obtained.
[0028]
Next, details of the sensitivity correction unit 15 will be described.
03-05-2019
8
[0029]
FIG. 2 is a block diagram showing a configuration example of the sensitivity correction unit 15
(the sensitivity correction units 15a to 15c).
As shown in FIG. 2, the sensitivity correction unit 15 includes a signal input unit 21, a reference
signal input unit 22, a gain variable unit 23, a gain control unit 24, and a signal output unit 25.
[0030]
The signal input unit 21 inputs a digital signal from any one of the AD converters 13a to 13c
corresponding to one of the microphone units 11a to 11c to be corrected as a correction target
signal. For example, in the case of the sensitivity correction unit 15a, a digital signal from the AD
conversion unit 13a corresponding to the microphone unit 11a to be corrected is input.
[0031]
The reference signal input unit 22 inputs the digital signal from the N-th AD conversion unit 13 d
corresponding to the N-th microphone unit 11 d as a reference signal serving as a reference.
That is, signals from microphone units other than the microphone units 11 a to 11 c to be
corrected among the plurality of microphone units are input as the reference signal. The
reference signal may be input from the outside of the array microphone device 1 instead of the
signal included in the microphone unit.
[0032]
The gain varying unit 23 amplifies or attenuates the level of the signal from any one of the AD
converters 13 a to 13 c corresponding to any one of the microphone units 11 a to 11 c to be
corrected. At this time, the gain varying unit 23 amplifies or attenuates the correction target
signal so that the level of the correction target signal and the level of the reference signal become
substantially equal as a result of amplification or attenuation. For example, the signal to be
corrected is amplified or attenuated so that the level of the signal to be corrected from the
03-05-2019
9
microphone unit 11a and the level of the reference signal from the microphone unit 11d are
substantially equal.
[0033]
The gain control unit 24 controls the amount of amplification or the amount of attenuation (gain)
when the variable gain unit 23 amplifies or attenuates the signal to be corrected. Details of the
gain control unit 24 will be described later.
[0034]
The signal output unit 25 is an output terminal or the like that outputs the signal amplified or
attenuated by the gain variable unit 23, that is, the output signal of the gain variable unit 23 to
the directivity combining unit 16.
[0035]
Next, details of the gain control unit 24 will be described.
[0036]
As shown in FIG. 2, the gain control unit 24 includes a band pass filter 26, an absolute value
calculation unit 27, a subtraction unit 28, a determination unit 29, a high speed gain update unit
210, and a low speed gain update unit 211.
[0037]
The band pass filter 26 is a first band pass filter 26 a that limits the pass band of the output
signal of the gain variable unit 23, a second band pass filter 26 b that limits the pass band of the
reference signal from the reference signal input unit 22, Equipped with
These passbands are the same passband.
[0038]
The absolute value calculator 27 calculates the absolute value of the signal from the first band
03-05-2019
10
pass filter 26b and the first absolute value calculator 27a that calculates the absolute value of the
signal from the first band pass filter 26a. The absolute value calculator 27b of
[0039]
The subtraction unit 28 subtracts the calculation result of the second absolute value calculation
unit 27 b from the calculation result of the first absolute value calculation unit 27 a.
[0040]
The determination unit 29 selects the high-speed gain update unit 210 or the low-speed gain
update unit 211 in order to update the gain.
For example, the determination unit 29 determines whether or not the elapsed time from
activation of the array microphone device 1 is less than a predetermined time (for example, 1
second), and when it is less than the predetermined time, the high speed gain update unit 210 is
selected. If the time is longer than a predetermined time, the low speed gain update unit 211 is
selected.
[0041]
The fast gain update unit 210 updates the gain with a relatively large gain update amount (first
change amount) per unit time.
[0042]
The low-speed gain update unit 211 updates the gain with a relatively small gain update amount
(second change amount smaller than the first change amount) per unit time.
[0043]
Thus, since the sensitivity correction unit 15 is configured, the gain control unit 24 controls the
correction target signal from any one of the microphone units 11 a to 11 c to be corrected, which
is input to the signal input unit 21. The signal is amplified or attenuated according to the gain
from the signal output unit 25.
[0044]
03-05-2019
11
Further, with regard to the output signal of the gain varying unit 23, the magnitude of the
correction target signal subjected to the band limitation is calculated as an absolute value by the
first band pass filter 26a and the first absolute value calculating unit 27a.
[0045]
On the other hand, regarding the reference signal from the microphone unit 11d input from the
reference signal input unit 22, the magnitude of the reference signal whose band is limited by
the second band pass filter 26b and the second absolute value calculation unit 27b. Is calculated
as an absolute value.
[0046]
Next, an example of gain update by the high-speed gain update unit 210 and the low-speed gain
update unit 211 will be described.
[0047]
FIG. 3 is a flowchart showing an operation example of the determination unit 29, the high speed
gain update unit 210, and the low speed gain update unit 211 according to the present
embodiment.
[0048]
Here, the sensitivity difference (difference in gain) between any of the microphone units 11a to
11c to be corrected and the reference microphone unit 11d is ± 4 dB at the maximum, and
amplification of ± 4 dB is required by the gain varying unit 23 It is assumed that
In addition, the gain is converged in one second after start-up, and the sound quality is not
affected by a minute gain update amount after one second after start-up.
Note that one second is an example.
[0049]
03-05-2019
12
When the array microphone device 1 is activated by turning on the power, in the digital signal
processing unit 14, a time elapsed counter (not shown) performs an update operation every
sampling cycle.
That is, clocking is started (step S101).
Subsequently, the determination unit 29 determines whether the time elapsed counter is less
than one second (step S102).
When the time elapsed counter is 1 second or more, the determination unit 29 selects the highspeed gain update unit 210 (No in step S102).
[0050]
Subsequently, the high speed gain update unit 210 determines whether the subtraction result of
the subtraction unit 28 is positive (step S103).
If the subtraction result by the subtraction unit 28 is positive, the level of the correction target
signal from the gain variable unit 23 is larger than the level of the reference signal, so the highspeed gain update unit 210 performs update to reduce the gain Step S104).
At this time, the high-speed gain update unit 210 sets the gain update amount per one sampling
period to, for example, -0.00025 dB (corresponding to -4 dB in 16000 samples, which is the
number of samples in one second).
[0051]
Also, when the subtraction result by the subtraction unit 28 is negative, the level of the
correction target signal from the gain variable unit 23 is smaller than the level of the reference
signal, so the high speed gain update unit 210 performs update to increase the gain. (Step S105).
03-05-2019
13
At this time, the high speed gain update unit 210 sets the gain update amount per one sampling
period to, for example, +0.00025 dB (corresponding to +4 dB in 16000 samples which is the
number of samples in one second).
[0052]
Then, the high speed gain update unit 210 updates the gain with the gain update amount, and
sets the gain in the gain variable unit 23 (step S106).
[0053]
In this manner, the sensitivity correction is being performed (during gain update) by correcting
the sensitivity difference corresponding to the result subtracted by the subtraction unit 28 at
high speed after startup by a predetermined time (for example, 1 second) by the high-speed gain
update unit 210 The directivity deterioration period can be shortened, and the waiting time in
the inspection process can be shortened to the above predetermined time (for example, 1
second).
[0054]
On the other hand, when the time elapsed counter is 1 second or more, the determination unit 29
selects the low speed gain update unit 211 (Yes in step S102).
[0055]
Subsequently, the high speed gain update unit 210 determines whether the subtraction result by
the subtraction unit 28 is positive (step S107).
If the subtraction result by the subtraction unit 28 is positive, the low-speed gain update unit
211 performs update to reduce the gain because the level of the correction target signal from the
gain variable unit 23 is larger than the level of the reference signal (step S108). ).
At this time, the gain update amount per one sampling period is set to, for example, -0.0000025
dB (corresponding to -0.04 dB in 16000 samples which is the number of samples in one second).
03-05-2019
14
[0056]
When the subtraction result by the subtraction unit 28 is negative, the low speed gain update
unit 211 performs update to increase the gain because the level of the correction target signal
from the gain variable unit 23 is smaller than the level of the reference signal. Step S109).
At this time, the gain update amount per one sampling period is set to, for example, +0.0000025
dB (corresponding to +0.04 dB in 16000 samples which is the number of samples in one second).
[0057]
Then, the low speed gain update unit 211 updates the gain with the gain update amount, and
sets the gain in the gain variable unit 23 (step S106).
[0058]
As described above, the gain update amount is limited to about ± 0.04 dB per second by the lowspeed gain update unit 211, so that the sound quality is not degraded even if the gain update is
constantly performed, and the update operation is perceived audibly You can try not to
[0059]
Next, an example of gain update updated by the high speed gain update unit 210 or the low
speed gain update unit 211 will be described.
[0060]
FIG. 4 shows an example of gain update when the microphone units 11a to 11c to be corrected
have 4 dB sensitivity lower than that of the reference microphone unit 11d, that is, 4 dB
amplification is required by the gain variable unit 23. It is a thing.
[0061]
FIG. 4A is a diagram showing an example of gain update by the high-speed gain update unit 210.
As shown in FIG.
03-05-2019
15
Referring to FIG. 4A, it can be understood that the gain converges to +4 dB one second after the
start of the array microphone device 1.
Further, as shown in FIG. 4A, the gain update amount (first change amount) by the high-speed
gain update unit 210 is, for example, the difference between the correction target signal and the
reference signal and the high-speed gain updater update unit 210. The amount of change is
based on a period for performing gain update (for example, one second after start-up), which is 1
dB or more per second.
[0062]
FIG. 4B is a diagram showing an example of the gain update performed by the low speed gain
update unit 211.
Referring to FIG. 4 (b), it can be understood that the gain of +4 dB is maintained after one second
after start-up, and the fluctuation of the gain is suppressed to such an extent that the sound
quality is not affected.
That is, the gain update amount (second change amount) by the low-speed gain update unit 211
is a change amount not perceptually perceived, and is, for example, less than 1 dB per second.
[0063]
FIG. 4 (c) is a diagram showing a conventional gain update.
Referring to FIG. 4C, it is understood that it takes 100 seconds to converge to a gain of +4 dB,
assuming that the amount of gain update per sampling period that does not affect the sound
quality is +0.0000255 dB. it can.
[0064]
03-05-2019
16
Next, the frequency characteristic of directivity based on the presence or absence of sensitivity
correction (gain update) will be described.
[0065]
FIG. 5 shows frequency characteristics of directivity of 0 degrees and 90 degrees when
directivity is synthesized by the directivity synthesis unit 16 of the array microphone device 1 in
two microphone units.
FIG. 5 (a) shows the case where the 4 dB sensitivity correction is performed, and FIG. 5 (b) shows
the case where the 4 dB sensitivity correction is not performed.
[0066]
Referring to FIG. 5A, when the sensitivity correction in the present embodiment is performed, flat
frequency characteristics are obtained as the directivity of 0 degrees, and from the frequency
characteristics of directivity of 0 degrees as the 90 ° flat frequency characteristics. A flat
frequency characteristic which is reduced by about 6 dB is obtained.
Therefore, it can be understood that directivity characteristics as designed can be obtained. On
the other hand, referring to FIG. 5B, when the sensitivity correction in the present embodiment is
not performed, the frequency characteristic of the directivity of 0 degrees and the frequency
characteristic of the directivity of 90 degrees do not become flat. Therefore, it can be understood
that the directivity characteristics as designed can not be obtained.
[0067]
According to the array microphone device 1 of this embodiment, the level of each output signal
(sensitivity-corrected correction target signal) of the sensitivity correction units 15a to 15c is a
signal from the Nth microphone unit 11d (reference signal) At the time of start of the array
microphone array device 1, the gain of the gain variable unit 23 is rapidly increased or decreased
at the time of startup of the array microphone array 1 so as to be approximately equal to As a
result, high-speed convergence of sensitivity correction (gain update) and maintenance of sound
03-05-2019
17
quality can be realized in a compatible manner.
[0068]
Second Embodiment FIG. 6 is a block diagram showing a configuration example of the sensitivity
correction unit 15B (sensitivity correction units 15a to 15c) in a second embodiment of the
present invention. The configuration of the sensitivity correction unit 15B shown in FIG. 6 of the
present embodiment is basically the same as the sensitivity correction unit 15 shown in FIG. 2 of
the first embodiment. In the sensitivity correction unit 15B shown in FIG. 6, components having
the same functions as those of the sensitivity correction unit 15 shown in FIG. The difference
between FIG. 6 and FIG. 2 is that the output signal 31a of the first absolute value calculator 27a
and the output signal 31b of the second absolute value calculator 27b are input to the
determination unit 29. The configuration other than the sensitivity correction unit 15B in the
array microphone device 1 of the present embodiment is the same as that of the array
microphone device 1 of the first embodiment, and thus the description thereof is omitted.
[0069]
FIG. 7 is a flowchart showing an operation example of the determination unit 29, the high speed
gain update unit 210, and the low speed gain update unit 211 of this embodiment.
[0070]
When the array microphone device 1 is activated by power on, the determination unit 29
calculates the level sum of the level of the output signal 31a and the level of the output signal
31b (step S201).
Then, it is determined whether the level sum is equal to or greater than a threshold (step S202).
The threshold value here may be, for example, residual noise (for example, -60 dBFS) of the
microphone units 11a to 11c.
[0071]
When it is determined that the level sum is equal to or more than the threshold, the time lapse
03-05-2019
18
counter in the digital signal processing unit 14 performs update operation. That is, time counting
is started (step S203). The subsequent operation is the same as steps S102 to S109 in FIG. On
the other hand, when the level sum is equal to or less than the threshold, the gain control unit 24
does not update the gain of the gain variable unit 23 (step S204).
[0072]
As described above, in the array microphone device 1 according to the present embodiment,
when the elapsed time after the sum of the level of the correction target signal and the level of
the reference signal exceeds a predetermined value is less than the predetermined time, the high
speed gain update unit The gain is changed by 210 (an example of a first gain update unit), and
the gain is changed by a low speed gain update unit 211 (an example of a second gain update
unit) when the elapsed time is equal to or longer than a predetermined time.
[0073]
Therefore, the high speed gain update unit 210 and the low speed gain update unit 211 operate
only when voices are sufficiently picked up by the microphone units 11a to 11c to be corrected
and the microphone unit 11d to be the reference, so that the minute level is not sufficient. Stable
level determination can be prevented.
[0074]
The sampling frequency, the gain, the operation time of the high-speed gain update unit 210, the
threshold value, and the like are not limited to those described above.
Also, the high-speed gain update unit 210 may determine the gain update amount from the gain
and the allowable correction time.
In addition, the low-speed gain update unit 211 may set the gain update amount as a level
change amount without a sense of discomfort in hearing. In this case, for example, 0.1 dB or less
per second is desirable.
[0075]
03-05-2019
19
Although the invention has been described in detail and with reference to specific embodiments,
it will be apparent to those skilled in the art that various changes and modifications can be made
without departing from the spirit and scope of the invention. This application is based on
Japanese Patent Application No. 2011-134562 filed on Jun. 16, 2011, the contents of which are
incorporated herein by reference.
[0076]
The present invention is useful for an array microphone device or the like that maintains high
sound quality and realizes high-speed convergence in sensitivity correction.
[0077]
DESCRIPTION OF SYMBOLS 1 array microphone apparatus 11 array microphone 11a 1st
microphone unit 11b 2nd microphone unit 11c 1st microphone unit 11d (N-1) th microphone
unit 11d Nth microphone unit 13 analog digital conversion part (AD conversion part) 13a 1st
analog Digital conversion unit 13b Second analog-to-digital conversion unit 13c Nth (N-1)
analog-to-digital conversion unit 13d Nth analog-to-digital conversion unit 14 Digital signal
processing unit 15, 15B Sensitivity correction unit 15a First sensitivity correction unit 15b
Second sensitivity correction unit 15c The (N-1) th sensitivity correction unit 16 directivity
correction unit 21 signal input unit 22 reference signal input unit 23 gain variable unit 24 gain
control unit 25 signal output unit 26 band pass filter 26a first Band pass filter 26b second band
pass filter 27 absolute value calculation unit 27a first absolute value calculation unit 27b second
absolute value calculation unit 28 subtraction unit 29 determination unit 210 high speed gain
update unit 211 low speed gain update unit 31a output signal 31b of first absolute value
calculation unit Output signal of absolute value calculator of 2
03-05-2019
20