close

Вход

Забыли?

вход по аккаунту

?

JP2001326990

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 JP2001326990
[0001]
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an
audio signal processing apparatus and processing method.
[0002]
In the present specification, sound or sound means sound waves in the audible band (about 20
Hz to about 20 kHz) and sound means sound waves in the voice band (about 200 Hz to about 6
kHz) that can be generated by humans. Shall be meant.
[0003]
In general, a non-directional microphone that is less susceptible to the shape of the surrounding
cabinet and wind noise is used as a microphone built in a camera integrated VTR (camera
integrated magnetic recording and reproducing device) such as a video camera.
Then, in order to obtain a stereo sound signal, a plurality of non-directional microphones are
used, and the sound signals from the plurality of non-directional microphones are electrically
processed by the stereo processing unit to have a directivity characteristic. It produces an audio
signal of stereo 2 channels with a sense of reality, and this stereo 2 channel audio signal is
recorded on a magnetic tape together with a video signal.
10-05-2019
1
[0004]
Then, with the intention of zooming the sound at the same time as the image according to the
zooming operation of the zoom lens of the video camera, the directivity characteristic by the
above-mentioned stereo processing unit can be varied according to the zooming of the zoom lens
It has been proposed in the prior art to vary the mix ratio of the sound level of the sound and the
sound level of the sound from the left and right direction.
[0005]
Specifically, since the angle of view of the lens changes depending on the zoom position of the
zoom lens of the video camera, the above-mentioned directivity angle is reduced as the angle of
view decreases at the zoom tele end (telephoto), or The sound level of the forward sound is made
higher than the sound level of the left and right sound in accordance with shooting so that the
distant subject is close to the subject, and conversely, the image is drawn at the zoom wide end
(wide angle). As the angle increased, the directivity angle was increased, and the sound level of
the forward sound was lower than the sound level of the left and right sound.
[0006]
In the following, with reference to FIG. 1, a system of sound zoom processing adapted to the
zoom lens of a conventional video camera will be described.
First, MICR, MICC and MICL indicate non-directional microphones for sound collection, which are
right, center and left microphones, respectively.
Here, as in the microphone arrangement example shown in FIG. 5, the respective microphones
MICR, MICC and MICL are arranged at the apex of an equilateral triangle so as to face in the
forward direction with the photographing screen. The right and left microphones MICR and MICL
are disposed apart from each other by a distance d, and the central microphone MICC is disposed
at a position protruding from the right and left microphones MICR and MICL with respect to the
forward direction with the photographing screen.
[0007]
The acoustic signals 1R, 1C and 1L outputted from the microphones MICR, MICC and MICL,
respectively, are inputted to the amplifiers AMPR, AMPC and AMPL and amplified, and then the
10-05-2019
2
amplified acoustic signals 2R, 2C and 2L become stereo The signal is input to the arithmetic
processing circuit 3 and converted into directional two-channel stereo audio signals 3R and 3L.
[0008]
A zooming control signal 9 from a microcomputer 10 is input to the stereo processing circuit 3,
and the zooming characteristics of the stereo two-channel signals 3R and 3L can be varied by
this signal 9.
The specific circuit configuration of the stereo processing circuit 3 will be described later.
[0009]
Next, the stereo two-channel signals 3R and 3L are gain controlled to a level optimum for poststage processing by the AGC circuit 4. The stereo two-channel signals 4R and 4L whose gains are
controlled are A / D converter (ADC) 5 , And converted to digital signals 5 R and 5 L from analog
signals, and then input to the audio / video recording system signal processing circuit 6.
[0010]
Further, a camera signal from the CCD 2 of the camera head unit including the zoom lens 1 and
the CCD (CCD image pickup device) 2 is input to the camera system signal processing circuit 11
and subjected to signal processing, and a video signal 12 obtained from this is The signal is input
to the audio / video recording system signal processing circuit 6.
[0011]
Here, the zoom position information of the zoom operated by the user from the zoom lens 1 is
input as the zoom position signal 8 to the microcomputer 10, and the microcomputer 10
generates the previous zoom control signal 9 from the zoom position signal 8 There is.
The camera signal output from the CCD 2 is input to the camera system signal processing 11,
where sampling processing, AGC processing, etc. are performed, and further, a built-in A / D
converter (ADC) (not shown) The digital signal is converted to a digital signal, gamma processing,
encoding processing and the like are performed, and input to the above-mentioned audio / video
10-05-2019
3
recording system signal processing 6 as a video signal (digital video signal) 12.
[0012]
In the audio / video recording system signal processing circuit 6, when the camera integrated
type VTR is a camera integrated type digital VTR, the audio signal is subjected to interleaving
processing and framing processing, and the video signal is blocked processing and shuffled. After
ring processing, DCT processing and requantization are performed, framing processing is
performed, multiplexing processing is performed with the above audio signal, and the signal is
input as a recording signal 7 to a recording apparatus for driving a magnetic tape to perform
magnetic processing Recorded on tape.
[0013]
About this recording device, illustration and explanation are omitted.
Further, the detailed circuit configurations of the audio / video recording system signal
processing circuit 6 and the camera system signal processing circuit 11 will not be illustrated nor
described.
[0014]
Next, with reference to FIG. 2, a specific circuit of the stereo processing circuit 3 of FIG. 1 will be
described.
First, the right input audio signal 2R and the left input audio signal 2L are input to the adder 20
and added, and are input to the delay units 21 and 22, respectively, and to the positive terminals
of the adders 25 and 26. Each is input.
[0015]
Next, the acoustic signals 21R and 22L subjected to delay processing by the delay units 21 and
22 are optimized in level by the attenuators 23 and 24, and then the acoustic signal 24L whose
10-05-2019
4
levels are optimized. 24R is input to each negative terminal of the adders 25 and 26 and is
subtracted from the signals 2R and 2L, respectively, to perform matrix operation. Here, each
delay time of the delay units 21 and 22 is set to a time delay corresponding to the speed of
sound movement of the distance d between the left and right microphones MICL and MICR in the
microphone arrangement of FIG. 5.
[0016]
Next, the frequency characteristics of the subtraction output signals 25R and 26L of the adders
25 and 26 are made flat across the entire band by the equalizers 27 and 28, and the output
signals 27R and 28L are input to the level controllers 36 and 37, respectively. The levels of the
output signals 27R and 28L are variably controlled by the zooming control signal 9 input to the
level controllers 36 and 37. Furthermore, the level controlled signals 36R and 37L are input to
one terminals of the adders 29 and 30, respectively.
[0017]
Here, the directivity pattern of the above-mentioned acoustic signals 27R and 28L is shown in
FIG. Generally, the directivity pattern patterns the response in each direction of the acoustic
signal with relative values, the broken line in FIG. 3 is the directivity pattern of the acoustic signal
27R, and the solid line is the directivity pattern of the acoustic signal 28L. The directivity
patterns of the both are symmetrical with respect to the forward direction, and the acoustic
signal 27R has strong directivity in the right direction and the acoustic signal 28L in the left
direction. Therefore, the stereo signal has a sense of presence in the lateral direction Although it
is a directivity pattern, the directivity is not so strong in the forward direction.
[0018]
Next, the addition output signal 20A of the adder 20 of FIG. 2 is subjected to delay processing by
the delay unit 31, and the level of the delay signal 31A is optimized by the attenuator 32, and the
optimized signal 32A of that level is obtained. Are input to the negative terminal of the adder 33.
Further, the central acoustic input signal 2C is input to the + type terminal of the adder 33, the
output signal 32A from the attenuator 32 is subtracted from the input signal 2C, and the
subtraction output signal 33C is input to the equalizer 34. After the entire frequency band is
flattened, the output signal 34C is input to the level controller 35, and the level of the output
10-05-2019
5
signal 34C is variably controlled by the control signal 39.
[0019]
The output signal 35C of the level controller 35 is input to the other terminal of each of the
adders 29 and 30 described above, and is added to the output signals 36R and 37L of the level
controllers 36 and 37, respectively. And the left addition output signal 3L is output.
[0020]
Here, the delay time of the delay unit 31 is equivalent to the distance of a perpendicular line
which is a straight line connecting the right and left microphones MICR and MICL from the
central microphone MICC in the microphone arrangement example of FIG. By setting the time
delay when moving the sound velocity, the output signal 34C of the equalizer 34 described above
becomes a directivity pattern having primary sound pressure inclination characteristics in the
forward direction as in the directivity pattern shown in FIG.
[0021]
Here, the level controllers 36 and 37 and the level controller 35 are controlled in level by the
zooming control signal 9 and the control signal 39, respectively, but the control signal 39 is the
zooming control signal 9 and the coefficient inverter 38 Since the control is performed with the
reverse characteristics, as shown in the example of the zooming control in FIG. 6, the control
characteristics of the both are maximized (MAX) by the outputs of the level controllers 36 and 37
when the lens zoom position is at the wide end. The output of the level controller 35 is controlled
to be minimum (MIN), and the zoom position is at the tele end, the outputs of the level
controllers 36 and 37 are at minimum (MIN) and the output of the level controller 35 is
maximum (MAX) It is controlled to become
[0022]
Therefore, the output signals 3R and 3L of the adders 29 and 30 for mixing the respective output
signals show a wide-angle horizontal directivity pattern as shown in FIG. 3 when the zoom
position is at the wide end, and when the zoom position is at the tele end. A forward-directed
narrow-angle directivity pattern as shown in FIG. 4 is shown, and in the middle position, a
directivity pattern in which both are mixed is shown.
The above is an example of sound zooming according to the conventional method, and the
10-05-2019
6
zooming characteristic (control example) is fixed as shown in FIG.
[0023]
However, the above-described conventional sound zoom processing system is not necessarily
recorded as sound or voice matched to the shooting screen, and for example, in distant view
photography such as landscape, noise noise in the front increases due to zooming of sound. It is
better to always collect the surrounding sound in a well-balanced manner regardless of the
zooming of the sound.
Also, when shooting a concert or a concert, if the sound generated from a certain instrument or
the voice of a person is emphasized by the zooming operation of the zoom lens of the video
camera and the sense of reality is lowered, the sound may be heard unnaturally .
Therefore, in this case as well, it is more natural to always collect the entire sound regardless of
the zooming position of the zoom lens of the video camera to give a sense of reality.
[0024]
In addition, as a shooting scene often used in general homes by video cameras, there is shooting
of a child at a relatively short distance. In this case, the voice of the child can be recorded clearly
at the zoom tele end, but the sound in the left and right direction is emphasized at the zoom wide
end, and the voice of the child may not be recorded clearly. Therefore, even at the zoom wide
end, if the voice of the child located in the forward direction of the screen is recorded with
emphasis to a certain extent, it will match the purpose of photographing.
[0025]
As described above, there are cases where it is preferable to emphasize the sound and voice
coming from the front of the screen depending on the shooting scene, and cases where it is not
preferable, and the conventional recording method could not cope with this. In addition, there is
also a problem that it is difficult for a general user to make judgment even if it is attempted to
switch between lens zooming and sound zooming in a coordinated manner or not depending on
10-05-2019
7
the shooting scene.
[0026]
SUMMARY OF THE INVENTION In view of the foregoing, the present invention provides an
acoustic signal processing apparatus capable of realizing optimum acoustic zooming according to
an imaging mode of a camera and a zooming position of a zoom lens of the camera. It is intended
to propose a treatment method.
[0027]
According to a first aspect of the present invention, there is provided a plurality of microphones,
and based on the acoustic signals from the plurality of microphones, forward directional acoustic
signals having directivity in the forward direction and directional in the lateral direction Acoustic
signal generating means for generating a left-right direction acoustic signal having flexibility, a
first voice band pass filter for extracting a forward direction voice band signal from the forward
direction acoustic signal from the acoustic signal producing means, and the first First subtracting
means for subtracting a forward voice band signal extracted from a voice band pass filter from a
forward voice signal to generate a forward voice band removal signal; a forward voice band
signal and a forward voice band Control signal generation means for generating a control signal
indicating a control coefficient based on the removal signal, a control mode according to a
shooting mode of the camera and a zoom position of the zoom lens of the camera A forward
direction sound is obtained by multiplying map signal generation means for generating map
signals indicating map values, control signals indicating control coefficients from the control
signal generation means, and map signals indicating control characteristic map values from the
map signal generation means. A level control signal generation multiplying means for generating
a first level control signal for controlling the level of the signal and a first level control signal
from the level control signal generating multiplying means control the level of the forward
direction acoustic signal And an audio signal processing apparatus comprising:
[0028]
According to the first aspect of the present invention, the acoustic signal generation means
comprises a plurality of microphones, and based on the acoustic signals from the plurality of
microphones, the forward acoustic signal having directivity in the forward direction and the
directivity in the lateral direction A first voice band pass filter extracts a forward voice band
signal from the forward voice signal from the voice signal generating means, and a first
subtraction means generates The forward voice band signal extracted from the voice band pass
filter is subtracted from the forward voice signal to generate a forward voice band removal
signal, and the control signal generation means generates the forward voice band signal and the
forward voice band. Based on the removal signal, a control signal indicating a control coefficient
10-05-2019
8
is generated, and the map signal generation means generates a control characteristic map
according to the photographing mode of the camera and the zooming position of the zoom lens
of the camera. The level control signal generating multiplying means multiplies the control signal
indicating the control coefficient from the control signal generating means and the map signal
indicating the control characteristic map value from the map signal generating means, A first
level control signal for controlling the level of the directional acoustic signal is generated, and the
first level control means generates the level of the forward directional acoustic signal by means
of the first level control signal from the level control signal generating multiplying means.
Control.
[0029]
According to a second aspect of the present invention, there is provided an audio signal
processing apparatus according to the first aspect of the present invention, comprising: a second
voice band pass filter for extracting left and right direction voice band signals from left and right
direction sound signals from the sound signal generating means; And second subtraction means
for subtracting the left and right direction voice band signal extracted from the two voice band
pass filters from the left and right direction audio signal to generate a left and right direction
voice band removal signal, and the control signal generation means The acoustic signal
processing apparatus is configured to generate a control signal indicating a control coefficient
based on the forward direction voice band signal, the forward direction voice band removal
signal, and the left and right direction voice band signal and the left and right direction voice
band removal signal.
[0030]
According to a third aspect of the present invention, in the acoustic signal processing device
according to the first aspect, the level control signal generating multiplying means is configured
to generate a second level control signal for controlling the level of the left and right direction
acoustic signal. According to another aspect of the present invention, there is provided an
acoustic signal processing apparatus comprising: second level control means for controlling the
level of the left and right direction acoustic signal according to a second level control signal from
the level control signal generating multiplying means.
[0031]
A fourth aspect of the present invention is the acoustic signal processing apparatus according to
the second aspect, wherein the level control signal generating multiplication means is configured
to generate a second level control signal for controlling the level of the left and right direction
acoustic signal. According to another aspect of the present invention, there is provided an
acoustic signal processing apparatus comprising: second level control means for controlling the
10-05-2019
9
level of the left and right direction acoustic signal according to a second level control signal from
the level control signal generating multiplying means.
[0032]
According to a fifth aspect of the present invention, in the acoustic signal processing device
according to any one of the first to fourth aspects, the map signal indicating the control
characteristic map value from the map signal generating means has a plurality of patterns
according to the photographing mode of the camera. And the control characteristic is changed
according to the zooming position of the zoom lens of the camera.
[0033]
According to a sixth aspect of the present invention, in the acoustic signal processing apparatus
according to the fifth aspect, the first level control signal is a control signal for making the level
of the forward direction audio signal constant at the wide end of the zoom lens. It is a signal
processing device.
[0034]
According to a seventh aspect of the present invention, a forward direction acoustic signal having
directivity in the forward direction and a left and right direction acoustic signal having directivity
in the left and right direction are generated based on the acoustic signals from the plurality of
microphones. Extract the forward voice band signal and subtract the forward voice band signal
from the forward voice signal to generate a forward voice band removal signal, based on the
forward voice band signal and the forward voice band removal signal Control signal indicating
the control coefficient, and a map signal indicating the control characteristic map value
corresponding to the photographing mode of the camera and the zoom position of the zoom lens
of the camera, and the control signal indicating the control coefficient and the control
characteristic map The map signal indicating the value is multiplied to generate a first level
control signal for controlling the level of the forward direction acoustic signal, and the first level
control signal causes the forward direction sound to be generated. An acoustic signal processing
method to control the level of the signal.
[0035]
An eighth aspect of the present invention is the acoustic signal processing method according to
the seventh aspect of the present invention, comprising: extracting left and right direction voice
band signals from left and right direction acoustic signals; subtracting left and right direction
voice band signals from left and right direction acoustic signals; A control signal indicating a
control coefficient is generated based on the forward direction voice band signal and the forward
10-05-2019
10
direction voice band removal signal and the left and right direction voice band signal and the left
direction voice band removal signal by generating the left direction voice band removal signal.
Acoustic signal processing method.
[0036]
According to a ninth aspect of the present invention, in the acoustic signal processing method
according to the seventh aspect, the second level control signal for controlling the level of the left
and right direction acoustic signal is generated, and the second level control signal is generated.
Is an acoustic signal processing method in which the level of
[0037]
According to a tenth aspect of the present invention, in the acoustic signal processing method
according to the eighth aspect, the second level control signal for controlling the level of the left
and right direction acoustic signal is generated, and the second level control signal Is an acoustic
signal processing method in which the level of
[0038]
An eleventh aspect of the present invention is the acoustic signal processing method according to
any of the seventh to tenth aspects of the present invention, wherein the map signal indicating
the control characteristic map value is switched to a plurality of patterns according to the
photographing mode of the camera. The acoustic signal processing method is such that the
control characteristic is changed according to the zooming position of the zoom lens.
[0039]
According to a twelfth aspect of the present invention, in the acoustic signal processing method
according to the eleventh aspect, the first level control signal is a control signal for making the
level of the forward direction audio signal constant at the wide end of the zoom lens. It is a signal
processing method.
[0040]
DETAILED DESCRIPTION OF THE INVENTION Embodiments of the present invention will be
described in detail below with reference to the drawings.
10-05-2019
11
First, with reference to FIG. 7, a system of sound zoom processing in accordance with zooming of
a zoom lens of a video camera will be described.
First, MICR, MICC and MICL indicate non-directional microphones for sound collection, which are
right, center and left microphones, respectively.
Here, as in the microphone arrangement example shown in FIG. 5, the respective microphones
MICR, MICC and MICL are arranged at the apex of an equilateral triangle so as to face in the
forward direction with the photographing screen.
The right and left microphones MICR and MICL are disposed apart from each other by a distance
d, and the central microphone MICC is disposed at a position protruding from the right and left
microphones MICR and MICL with respect to the forward direction with the photographing
screen.
[0041]
The acoustic signals 1R, 1C and 1L respectively output from the microphones MICR, MICC and
MICL are respectively input to the amplifiers AMPR, AMPC and AMPL and amplified, and then
adaptive zoom processing means 50 as acoustic signals 2R, 2C and 2L. Is input to
The adaptive zoom processing means 50 will be described in detail later with reference to FIG.
[0042]
While the zooming control signal 9 similar to that of the conventional example of FIG. 1 is input
from the microcomputer 10 to the adaptive zoom processing means 50, the imaging mode
information signal 52 is newly input.
Based on these signals 9 and 52, the adaptive zoom processing R and L channel signals 50R and
50L generated by the adaptive zoom processing means 50 are gain controlled by the AGC circuit
4 to a level optimum for subsequent processing, The gain-controlled stereo two-channel signals
10-05-2019
12
4R and 4L are input to an A / D converter (ADC) 5 and converted from analog signals to digital
signals 5R and 5L, and then an audio / video recording system signal processing circuit 6 Is
input to
[0043]
Further, a camera signal from the CCD 2 of the camera head unit including the zoom lens 1 and
the CCD (CCD image pickup device) 2 is input to the camera system signal processing circuit 11
and subjected to signal processing, and a video signal 12 obtained from this is The signal is input
to the audio / video recording system signal processing circuit 6.
[0044]
Here, the zoom position signal 8 is input from the zoom lens 1 to the microcomputer 10
described above to generate the zooming control signal 9, and the photographing mode signal 53
is input from the newly provided photographing mode setting means 51. .
For example, in the case of a video camera, the shooting mode setting means 51 blurs a far
background with a shallow camera depth setting menu when the user shoots the video camera,
for example, and clearly shoots a nearby person. The mode is selected from a portrait mode that
can be used and a landscape mode that is suitable for shooting from close to far by increasing
the depth of field.
[0045]
Therefore, the photographing mode information signal 52 inputted to the adaptive zoom
processing means 50 is information which the microcomputer 10 judges from the photographing
mode signal 53 and which mode it is in now.
In practice, the aperture value of the lens and the camera system signal processing 11 are also
changed in accordance with the switching of these camera mode settings, but the details are out
of the scope of the present invention and the illustration and description thereof will be omitted.
[0046]
10-05-2019
13
Next, with reference to FIG. 8, a specific circuit of the adaptive zoom processing means 50 of FIG.
7 will be described.
First, the right input audio signal 2R and the left input audio signal 2L are input to the adder 20
and added, and are input to the delay units 21 and 22, respectively, and to the positive terminals
of the adders 25 and 26. Each is input.
[0047]
Next, the acoustic signals 21R and 22L subjected to delay processing by the delay units 21 and
22 are optimized in level by the attenuators 23 and 24, and then the acoustic signal 24L whose
levels are optimized. 24R is input to each negative terminal of the adders 25 and 26 and is
subtracted from the signals 2R and 2L, respectively, to perform matrix operation.
Here, each delay time of the delay units 21 and 22 is set to a time delay corresponding to the
speed of sound movement of the distance d between the left and right microphones MICL and
MICR in the microphone arrangement of FIG. 5.
[0048]
Next, the frequency characteristics of the subtraction output signals 25R and 26L of the adders
25 and 26 are made flat across the entire band by the equalizers 27 and 28, and the output
signals 27R and 28L are input to the adaptive zoom circuit 60.
[0049]
Here, the directivity pattern of the above-mentioned acoustic signals 27R and 28L is shown in
FIG.
Generally, the directivity pattern patterns the response in each direction of the acoustic signal
with relative values, the broken line in FIG. 3 is the directivity pattern of the acoustic signal 27R,
and the solid line is the directivity pattern of the acoustic signal 28L.
10-05-2019
14
The directivity patterns of the both are symmetrical with respect to the forward direction, and
the acoustic signal 27R has strong directivity in the right direction and the acoustic signal 28L in
the left direction. Therefore, the stereo signal has a sense of presence in the lateral direction
Although it is a directivity pattern, the directivity is not so strong in the forward direction.
[0050]
Next, after the addition output signal 20A of the adder 20 of FIG. 8 is subjected to delay
processing by the delay unit 31, the level of the delayed signal 31A is optimized by the
attenuator 32, and the optimized signal 32A of that level is obtained. Are input to the negative
terminal of the adder 33.
Also, the central acoustic input signal 2C is input to the positive terminal of the adder 33, the
output signal 32A from the attenuator 32 is subtracted from the input signal 2C, and the
subtraction output signal 33C is input to the equalizer 34. The output signal 34 C is input to the
adaptive zoom circuit 60 after the frequency characteristic is flattened in the entire band.
As described above, the signals 27R, 28L, and 34C are a right direction signal, a left direction
signal, and a front direction signal, respectively, in which the imaging direction is the front
direction.
Further, the zooming control signal 9 and the photographing mode information signal 52 are
input to the adaptive zoom circuit 60, subjected to adaptive zoom processing to be described
later, and output as a right output signal 50R and a left output signal 50L.
[0051]
Here, the delay time of the delay unit 31 is equivalent to the distance of a perpendicular line
which is a straight line connecting the right and left microphones MICR and MICL from the
central microphone MICC in the microphone arrangement example of FIG. By setting the time
delay when moving the sound velocity, the output signal 34C of the equalizer 34 described above
becomes a directivity pattern having primary sound pressure inclination characteristics in the
forward direction as in the directivity pattern shown in FIG.
10-05-2019
15
[0052]
Next, the circuit configuration of the adaptive zoom circuit 60 of FIG. 8 will be described with
reference to FIG.
First, the input forward direction signal 34 C is input to the band pass filter 8 and only the voice
signal band is passed, and the obtained signal 78 B of the voice band is output from the control
signal generation circuit 82 and the adder 81. It is input to the negative side terminal.
Also, forward direction signal 34 C is input to delay device 9 and delayed by a delay time
corresponding to the delay time when signal 34 C is delayed by band pass filter 78, and then
delayed signal 79 D is added to the + of adder 81. The signal 78D is input to the side terminal,
the previous signal 78B is subtracted from the signal 79D, and the subtraction signal 81A is also
input to the control signal generation circuit 82.
[0053]
Here, the frequency characteristic of the signal 78B in the audio signal band has a band pass
filter characteristic shown as a characteristic curve 90 in FIG. 10A, and the pass band is set to,
for example, about 300 Hz to 5 kHz. Further, the frequency characteristic of the subtraction
signal 81A is a frequency characteristic obtained by adding the characteristic curves 91 and 92
shown in FIG. 10B, and a band pass filter obtained by subtracting the characteristic curve 90 of
FIG. 10A from the acoustic signal band. It has removal characteristics. Further, the abovementioned zooming control signal 9 and the photographing mode information signal 52 are
inputted to the control signal generation circuit 82.
[0054]
Then, level control signals 83 and 84 are output from the control signal generation circuit 82.
The level control signal 83 is input to the level controller 71 that changes the level of the input
signal 27R and the level controller 72 that changes the level of the input signal 28L to control
each level. The level controlled signals 71R and 72L from the level controllers 71 and 72 are
input to one terminal of the adders 74 and 75, respectively. Further, the level control signal 84 is
10-05-2019
16
input to the level controller 73 that changes the level of the input signal 34C to control the level.
The level-controlled signal 73C from the level controller 73 is input to the adders 74 and 75,
added to the previous signals 71R and 72L, respectively, and the right and left signals 50R and
50L are output, respectively.
[0055]
Next, the control signal generation circuit 82 of FIG. 9 will be described in more detail with
reference to FIG. As described above, input signal 78B in FIG. 11 is forward voice signal
{hereinafter referred to as CV (Center Voice) signal}}, input signal 81A is forward voice removed
signal {hereinafter referred to as CO (Center Other) signal and abbreviated Yes. The forward
direction speech signal and the forward direction speech removal signal 78B, 81A are input to
the absolute value processing units 100, 101, and the positive and negative signals 78B, 81A are
absolute valued to be respectively positive signals 100A, 101A. It is converted. These signals
100A and 101A are input to and detected by envelope detection processors 102 and 103, and
envelope signals 102D and 103D of the respective signal waveforms are output. The envelope
signals 102D and 103D are respectively input to the amplifiers 104 and 105 and amplified, and
the amplified signals 104A and 105A are output.
[0056]
At this time, the gains of both the amplifiers 104 and 105 may be changed to balance the levels
of the signals 104A and 105A. This is because in the case of general environmental sound, the
frequency characteristic is often 1 / f characteristic (f is frequency), and when the levels of the
forward direction voice signal and the forward direction voice removed signals 78B and 81A are
compared, This is because there are many signal components in the low frequency region in the
entire band, and the level of the forward direction voice signal 78B often exceeds the level of the
forward direction voice removed signal 81A.
[0057]
The amplified signal 104A is input to the positive terminal of the adder 106, and the amplified
signal 105A is input to the negative terminal of the adder 106, and the signal 105A is subtracted
from the signal 104A. The subtraction signal 106A from the adder 106 is input to the code
detector 107, and the magnitude of the output signals 104A and 105A of the amplifiers 104 and
10-05-2019
17
105 is detected by detecting whether the sign of the signal 106A is positive or negative. The
relationship is determined, and a flag signal 107F is output. That is, when the detection code of
the code detector 107 is positive, it is determined that the level of the signal 104A> the level of
the signal 105A, and when it is negative, it is determined that the level of the signal 104A <the
level of the signal 105A; And the level of the signal 104A = the level of the signal 105A.
[0058]
The flag signal 107F from the sign detector 107 is input to the up / down counter 108, and
when the flag signal 107 indicates positive, the counter 108 operates as an up counter, and when
negative, it operates as a down counter. , 0 is controlled to hold the pre-count value. Therefore,
the output signal 108UD of the counter 108 is a signal that repeats up and down according to
the sign of the flag signal 107F. The output signal 108 UD of the counter 108 is input to the
limiter processor 109 to set MAX (maximum) and MIN (minimum) of the up / down count values.
When the count value is used as a coefficient, MAX (maximum) is set to 1 and MIN (minimum) is
set to 0.
[0059]
The output signal 109LM of the limiter processor 109 is input to the time constant adder 110,
and a time constant is added so that the change can be heard smoothly in terms of hearing. As an
example of the time constant adder 110, a low pass filter can be mentioned.
[0060]
Next, generation of a coefficient by the up / down counter 108 of FIG. 11 will be described with
reference to FIG. FIG. 12 (a) shows arbitrary waveforms of the input signals 78B and 81A of FIG.
When these input signals 78B and 81A are processed by the absolute value processing units 100
and 101, respectively, signals 100A and 101A of waveforms shown by the solid line in FIG. 12B
are obtained. When the signals 100A and 101A are input to the envelope detection processors
102 and 103 and detected, envelope signals 102D and 103D shown by broken lines in FIG. 12B
are obtained.
[0061]
10-05-2019
18
Referring to FIG. 12C, a signal 104A obtained by amplifying the signal 102D by the amplifier
104 and a signal 105A obtained by amplifying the signal 103D by the amplifier 105 are input to
the adder 106, and the signal 104A is obtained. And the subtraction output signal 106A is input
to the code detector 107 to obtain the flag signal 107F. When the level of signal 104A> the level
of signal 105A, flag 107F indicates positive, and when the level of signal 104A <level of signal
105A, flag 107F indicates negative, and the level of signal 104A = level of signal 105A. When the
level is on, the flag 107F indicates 0. That is, as shown in FIG. 12C, the code is detected as the
flag signal 107F in each period.
[0062]
Referring to FIG. 12D, the up / down counter 108 counts up when the sign of the flag 107F from
the sign detector 107 is positive, and counts down when the sign is negative, and when the sign
is zero. Holds the prestep value. At this time, by making the step value set in the up / down
counter 108 different on the up side and the down side, it is possible to make the attack side
faster and the recovery side slower as in the attack / recovery time constant. Further, if the step
value to be set is increased, the up and down speed is increased quickly, and if the step value is
decreased, the up and down speed is increased, so that the tracking speed can be changed.
Further, in FIG. 12D, the step value is limited by the limiter 109, the upper limit is limited to the
factor 1, and the lower limit is limited to the factor 0.
[0063]
Referring to FIG. 12E, the time constant adder 110 applies low pass filter processing to the
output signal 109LM of the limiter 109 to smooth the level difference of the output signal
108UD of the up / down counter 108, and thereby the signal 110T. I am trying to get
[0064]
As described above, the coefficient signal 110T generated in the time constant adder 110 of FIG.
11 is input to one input terminal of each of the multipliers 112 and 113. The coefficient signal
110T has an input signal level of When the forward voice signal 78B> forward voice removal
signal 81A, the coefficient is large, MAX is 1 and when the signal level is forward voice signal
78B <forward voice removal signal 81A, the coefficient is small , MIN is 0, and between the
coefficients 1 and 0 is always determined by the step value set by the up / down counter 108
and the cut-off frequency of the low-pass filter by the time constant adder 110 according to the
magnitude relationship between signal levels It is changing at tracking speed.
10-05-2019
19
[0065]
Furthermore, the left and right directivity signal 111S output from the map generation circuit
111 and the front directivity signal 111C are input to the other input terminals of the multipliers
112 and 113, respectively, and the coefficient signal 110T is input to the respective signals 111S
and 111C. And the respective multiplied outputs are outputted as the horizontal direction level
control signal 83 and the forward direction level control signal 84.
[0066]
Next, mapping by the map generation circuit 111 of FIG. 11 will be described.
The map generation circuit 111 outputs mapping values of the forward directional signal 111C
and the left and right directional signal 111S with respect to the lens zoom position based on the
zooming control signal 9 and the photographing mode information signal 52.
In the map generation circuit 111, for example, the relationship between the lens zoom position
and each mapping value is described in a table made up of ROM, and the lens zoom position is
judged from the input zooming control signal 9 to cope with it. This can be realized by
appropriately reading the mapping value from the ROM table.
[0067]
Next, the mapping example will be described with reference to FIG.
As shown in FIG. 13, the photographing mode information signal 52 input to the map generation
circuit 111 of FIG. 11 has three modes, ie, a normal mode, a portrait mode and a landscape mode,
for example. ), (B) and (c) show examples of mapping corresponding to each mode.
[0068]
10-05-2019
20
FIG. 13A shows an example of mapping in the normal mode, that is, in the normal camera
shooting mode. In FIG. 13A, the horizontal axis indicates the lens zoom position change from the
wide end to the tele end according to the zooming control signal, and the vertical axis indicates
the mapping value of the corresponding front directional signal 111C and the horizontal
directional signal 111S by MAX. It is set as 1 and it represents in the range which sets it to -1 by
MIN. The horizontal and vertical axes in FIGS. 13 (b) and 13 (c) are the same as the horizontal
and vertical axes in FIG. 13 (a). For example, if the bit word length of the ROM table is expressed
as 16 bits and expressed in 2's complement representation, the value is 7fff hex (corresponding
to 1) to 8000 hex (corresponding to -1) (hex is a hexadecimal number), and the resolution is It is
65535 steps.
[0069]
First, the mapping value of the forward directional signal 111C indicates an upward slope to the
right, and is larger at the tele end and smaller at the wide end. The larger the inclination, the
larger the zooming effect. There is also a portion where the inclination is zero at the wide end,
which will be described later. Also, the mapping value of the left-right pointing signal 111S
shows a downward slope to the right, and similarly, the larger the slope, the larger the zooming
effect. When the mapping value is 0, the mapping values of the forward directional signal 111C
and the left and right directional signal 111S are the same.
[0070]
FIG. 13B shows an example of mapping in the portrait mode in which the camera shooting mode
is set. In the portrait mode, the depth of field is reduced by the lens aperture to blur the distant
view and capture the close view sharply. Etc. suitable for relatively short distance shooting.
Therefore, there are many cases where a person at a short distance emits a voice as a shooting
scene in this case, and in this shooting mode, the zooming control signal of the horizontal axis is
used to increase the zooming amount of the voice according to the zooming of the zoom lens.
With respect to the change from the wide end to the tele end, the zoom value of the vertical axis
is made larger than that in the normal camera photographing mode of FIG. 13A to increase the
zooming effect.
[0071]
10-05-2019
21
FIG. 13C shows an example of mapping in the landscape mode in which the camera shooting
mode is set. In the landscape mode, the depth of field is made deeper by the lens aperture, and in
order to shoot sharply from a distant view to a near view Is suitable. Therefore, in this case, the
mapping value is made smaller than that in the normal camera shooting mode of FIG. 13A so as
to give a sense of reality suitable for the landscape even at the tele end, thereby reducing the
zooming effect. ing.
[0072]
By the way, in each of the above camera shooting modes, the mapping value of the forward
directional signal 111C is set to zero at the wide end side, for example, when shooting a person
against a landscape or shooting a child at a short distance. This is because it is possible to
suppress the reduction of the mapping value because it is better for the purpose of
photographing to keep the voice of the person clear while leaving the forward directional signal
to some extent even on the wide end side.
[0073]
Therefore, also in this example, the mapping value can be freely set, and the zooming value can
always be optimized for the camera shooting mode.
Further, although the mapping values to be stored in the ROM table are made to change linearly
with respect to the zooming control signal, they may be made to have a logarithmic change so
that the change in human hearing can be heard more smoothly.
[0074]
As described above with reference to FIG. 11, the forward direction signal 111C and the left and
right direction signal 111S generated by the map generation circuit 111 are multiplied by the
coefficient signal 110T by the multiplier 113 and the multiplier 112, respectively, to obtain the
forward direction level. The control signal 84 and the horizontal direction level control signal 83
are output. An example of generation of the forward direction level control signal 84 and the left
and right direction level control signal 83 will be described with reference to FIG.
[0075]
10-05-2019
22
In this example, the mapping value generated by the map generation circuit 111 is compared
with the level of the CV signal (forward speech signal as input signal 78B in FIG. 11) and the CO
signal (forward speech cancellation signal as input signal 81A). The zooming is adaptively
controlled by multiplying the coefficient signal 110 T generated by the above.
[0076]
FIG. 14A shows a generation example (one example) of the forward direction level control signal
84 and the left and right direction level control signal 83.
When a voice band signal is input from the forward direction in the portrait mode of FIG. 13B,
the control signal is the coefficient signal 110T, that is, the control signal multiplied by the
coefficient 1 by the signal 111C and the signal 111S of FIG. Is outputted, so that the voice of the
person being photographed is zoomed as much as possible in accordance with the zooming of
the image.
[0077]
FIG. 14B shows a generation example (two examples) of the forward direction level control signal
84 and the left and right direction level control signal 83. In this example, many signals other
than the voice band are input from the front direction, and at this time, it is considered that other
than a person is photographed as a photographing scene, and the signal 111C and the signal
111S of FIG. Since the coefficient of the coefficient signal 110T is a value of 1 or less, the slope
of the output level is reduced, and a control signal which relatively reduces the effect of zooming
the sound is output. Further, although not shown, when the values of coefficient 0 are multiplied
by the signal 111C and the signal 111S of FIG. 11, the zooming effect is lost.
[0078]
FIG. 14C shows a generation example (three examples) of the forward direction level control
signal 84 and the left and right direction level control signal 83. In this example, the voice from
the forward direction becomes an environmental sound other than voice when the person stops
emitting sound during zooming from the wide end to the tele end, or when the photographer
stops shooting a person. It is illustrated that the coefficient of the coefficient signal 110T
10-05-2019
23
gradually decreases and the zooming effect diminishes.
[0079]
FIG. 14D shows a generation example (four examples) of the forward direction level control
signal 84 and the left and right direction level control signal 83. In this example, when the
person emits a voice during zooming from the wide end to the tele end, or when the
photographer starts shooting a person, environmental sound other than the voice from the front
direction becomes a voice. It is illustrated that the coefficient of the coefficient signal 110T
gradually increases and the zooming effect increases.
[0080]
The tracking speed of coefficients during zooming in FIGS. 14C and 14D changes the step value
set by the up / down counter 108 in FIG. 11 and the cutoff frequency of the time constant adder
110, for example, as described above. Can be optimized.
[0081]
Next, referring to FIG. 15, another block diagram of the adaptive zoom circuit 60 in FIG. 8 will be
described.
In FIG. 15, parts corresponding to those in FIG. 9 are assigned the same reference numerals and
redundant explanation will be omitted, and parts different from FIG. 9 will be described. First, the
input signals 27R and 28L are respectively input to the level controllers 71 and 72 described
above. The input signal 27R is input to the negative terminal of the adder 120, the input signal
28L is input to the positive terminal of the adder 120, and the input signal 27R is subtracted
from the input signal 28L. The subtraction output 120 A from the adder 120 is input to the band
pass filter 121 and the delay unit 122.
[0082]
The band pass filter 121 and the delay unit 122 perform the same processing as the band pass
filter 78 and the delay unit 79 described above. The output signal 121 B of the band pass filter
10-05-2019
24
121 is input to the control signal generation circuit 124 and the negative side terminal of the
adder 123. The output signal 122D of the delay unit 122 is input to the positive terminal of the
adder 123, the output 121B is subtracted from the output signal 122D, and the subtraction
output 123A is input to the control signal generation circuit 124.
[0083]
Here, in the output signal 121B from the band pass filter 121, the overlapping portion of the left
and right directional pattern shown in FIG. 3 is removed by the adder 120, and there is no
directivity in the front and rear direction. It has a directivity pattern as shown, and in addition,
only the voice band is passed by the band pass filter 121. Further, the subtraction signal 123A
from the adder 123 is an audio band removal signal having the same left-right directivity as the
output signal 121B from the band pass filter 121B.
[0084]
In the control signal generation circuit 124, based on the signals 121B and 123A, the signals
78B and 81A similar to those described in FIG. 9, the zooming control signal 9 and the
photographing mode information signal 52, the horizontal direction level control signal 125 and
the forward direction A level control signal 126 is generated.
[0085]
Next, the control signal generation circuit 124 of FIG. 15 will be described with reference to FIG.
As described above, the input signal 78B of the control signal generation circuit 124 is a forward
direction voice signal {hereinafter, abbreviated as CV (Center Voice) signal}, and the input signal
81A is a forward direction voice removed signal {hereinafter, CO (Center Other) The input signal
121B is a left / right direction voice signal {hereinafter referred to as SV (Side Voice) signal}}, and
the input signal 123A is a left / right direction voice removed signal {hereinafter referred to as
SO (Side Other) signal and abbreviated Yes. These signals are input to the absolute value
processing units 130 to 133, and signals that extend from positive to negative are absolute
valued to positive signals to obtain absolute value signals 130A to 133A, respectively. These
absolute value signals 130A to 133A are input to envelope detectors 134 to 137, respectively,
and the envelope of the signal waveform is detected. The envelope-detected signals 134D to
137D are respectively input to the amplifiers 138 to 141 and appropriately amplified, and the
10-05-2019
25
respective amplified signals 138A to 141A are input to the level comparison and determination
processing means 142.
[0086]
At this time, the gains of the respective amplifiers 138 to 141 are changed to balance each signal
level in the acoustic band. This is because, when the levels of the forward direction voice signal
78B and the forward direction voice removed signal 81A are compared, in the case of a general
environmental sound, the frequency characteristic exhibits a 1 / f characteristic, the voice
occupying almost all low frequency regions This is in order to absorb the level difference due to
the difference between the processing of the forward direction audio signal 78B and the forward
direction audio removal signal 81A in that the signal level of the band often exceeds the voice
band elimination signal level.
[0087]
Next, the output signals 138A to 141A of the amplifiers 138 to 141 are input to the level
comparison and determination processing means 142, and the magnitude comparison of the
respective signal levels is performed to obtain the comparison result signal 142F. The
comparison result signal 142F is input to the coefficient table selection circuit 143. The level
comparison / determination processing means 142 will be described in detail later.
[0088]
The coefficient table selection circuit 143 selects the timely coefficient from the coefficient table
configured by the ROM or the like based on the comparison result signal 142F, and outputs the
forward direction coefficient signal 143C and the left and right direction coefficient signal 143S.
Input to one of 112 terminals. The mapping signals 111C and 111S are input to the other
terminals of the multipliers 113 and 112 from the map generation circuit 111 described above
with reference to FIG. 11, and are multiplied by the forward coefficient signal 143C and the
horizontal coefficient signal 143S, respectively. The multiplication output signals 113C and 112S
are input to time constant adders 145 and 144, respectively.
[0089]
10-05-2019
26
The time constant adders 145 and 144 respectively add time constants to the multiplication
output signals 113C and 112S in order to audibly smooth discontinuous portions of signals
generated at the time of switching the coefficient from the coefficient table selection circuit 143
described above. Do. As the time constant adders 145 and 144, for example, a low pass filter
whose cutoff frequency is set low is used. After the above processing, the forward direction level
control signal 126 and the left and right direction level control signal 125 are output.
[0090]
Next, the operation of the level comparison / determination processing means 142 of FIG. 16 will
be described with reference to the flowchart of FIG. First, for the explanation of the signals 138A
to 141A input to the level comparison / determination processing means 142 of FIG. 16, these
signals 138A to 141A are respectively referred to as signals A, B, C, and D in FIG. A counter, B
counter, C counter and D counter are provided.
[0091]
First, when the signals to be compared are input simultaneously at a certain timing, the counters
corresponding to the input signals are reset to zero in steps 150 to 153. Next, it is judged at step
154 whether signals A and B are signal A> signal B or not, if YES, the B counter is incremented
by 1 at step 157, and if NO, the A counter is incremented at step 158. The value is incremented
by one and then proceeds to step 163 respectively.
[0092]
Next, it is judged at step 155 whether signals B and C satisfy signal B> signal C. If YES, C counter
is incremented by 1 at step 159, and if NO, B counter is incremented at step 160. The value is
incremented by one, and then the process proceeds to step 164. Next, it is judged at step 156
whether signals C and D satisfy signal C> signal D. If YES, D counter is incremented by 1 at step
161, and if NO, C counter is incremented at step 162. The value is incremented by one, and then
the process proceeds to step 163.
10-05-2019
27
[0093]
Further, it is judged whether the signals A and C are signal A> signal C or not in step 163, and if
YES, the C counter is incremented by 1 in step 165, and if NO, in step 166. The A counter is
incremented by one, and then the process proceeds to step 169. Next, it is determined in step
164 whether signals B and D satisfy signal B> signal D. If YES, the D counter is incremented by 1
in step 167, and if NO in step 168 , B counters are incremented by 1, and then the process
proceeds to step 172.
[0094]
Further, it is judged at step 169 whether signals A and D satisfy signal A> signal D. If YES, the D
counter is incremented by 1 at step 170, and if NO, at step 171. The A counter is incremented by
one, and each step proceeds to step 172.
[0095]
Then, in step 172, the counter values of the A, B, C and D counters are read, and the counter
values are sorted in ascending order, and then the process proceeds to step 173.
That is, in step 172, the counter values are sorted in the order of 0 → 1 → 2 → 3. In step 173,
the order of levels is determined in the order of the signals corresponding to the counter having
the counter value, and the determination result is output as the signal F (corresponding to the
signal 142F in FIG. 16). These flows are repeatedly executed each time a new signal is input, and
the level determination of the signals A, B, C, D is always performed, and the signal F is output in
the order of increasing levels.
[0096]
Next, with reference to FIG. 18, a specific example of the coefficient table possessed by the
coefficient table selection circuit 143 of FIG. 16 will be described. Here, for each of the
aforementioned CV, CO, SV and SO signals, each combination of forward coefficient and left /
right coefficient corresponding to all combinations of the level order obtained by the signal 142F,
that is, 24 combinations. I have a coefficient table. The coefficients can be determined
independently with values between 0 and 1, and these coefficients are optimized by experiments
10-05-2019
28
and written to ROM etc. The following points are the points at optimization .
[0097]
1. In the case where the level order of the CV signal is relatively high, this is a case where the
subject in the forward direction as a shooting scene is a person, and the forward direction
coefficient is increased to increase the zooming effect. 2. When the level order of both the CV
signal and the SV signal is high, the sound is input to the microphone over the entire
surroundings as a shooting scene, that is, the sound of the target subject ahead in a relatively
large number of people such as a party is recorded In this case, the forward coefficient and the
lateral coefficient are both raised to increase the zooming effect. 3. If the level order of the SV
signal is relatively high, it is considered that the sound from the side direction is not the sound of
the target subject as the shooting scene, and the coefficient in the left and right direction is
relatively large. Lower. 4. If the level order of both the CO signal and the SO signal is high, it is
assumed that an environmental sound other than voice is input to the microphone from the
entire surroundings as a shooting scene, that is, a shooting scene in which no person is relatively
around the scene. In this case, since it is not necessary to perform zooming in this case, both the
forward direction coefficient and the left and right direction coefficient are lowered to reduce the
zooming effect.
[0098]
Therefore, the coefficient table selection circuit 143 is characterized in that the zooming effect
can be set adaptively by setting the respective coefficients independently according to the
shooting scene under these conditions.
[0099]
Next, with reference to FIG. 19, another system diagram of the sound zoom processing will be
described.
FIG. 19 performs adaptive zoom processing by the MS (mid-side) microphone system on the
system diagram of FIG. 7, and after analog-digital conversion of the adaptive zoom processing by
the A / D converter 5, DSP (digital signal It shows an example of a system in which digital
processing is performed by a processor) or a digital LSI.
10-05-2019
29
[0100]
The acoustic signals 1M and 1S from the M (mid) microphone MICM and S (side) microphone
MICS are preamplified by the amplifiers AMPM and AMPS, and the amplified signals 2M and 2S
are input to the AGC circuit 4 and Is controlled to the most suitable level. The signals 4M and 4S
from the AGC circuit 4 are input to the A / D converters 5 and 5 to be analog-digital converted,
and the digital signals 5M and 5S are input to the adaptive zoom processing means 180.
[0101]
The adaptive zoom processing unit 180 performs adaptive zoom processing described later on
the digital signals 5M and 5S, and outputs digital signals 180R and 180L. The digital signals
180R and 180L, that is, stereo two-channel acoustic signals are input to the audio / video
recording system signal processing circuit 6.
[0102]
Further, the zooming control signal 9 and the photographing mode information 52 from the
microcomputer 10 described above are input to the adaptive zoom processing means 180. The
other functional blocks are assigned the same reference numerals as in FIG. 7 and redundant
explanations are omitted.
[0103]
Next, the MS microphone system will be described with reference to FIGS. 20 and 21. In the MS
microphone system, FIG. 23 is obtained by electrically combining signals from an M microphone
having a directivity pattern in the forward direction shown in FIG. 20 and an S microphone
having a directivity pattern in the left and right direction shown in FIG. A stereo signal having
directivity in the directions of the M + S axis and the M-S axis shown in FIG. In general, M
microphones use unidirectional microphones, and S microphones use bi-directional microphones.
The characteristic of the MS microphone system is that the directivity angle formed by the M + S
axis and the MS axis can be easily varied by the level of the M microphone at the time of
combining. When the angle is narrowed and the level of the M microphone is decreased, control
10-05-2019
30
can be performed so that the directivity angle is expanded in the left and right direction. Also,
similarly, the level of the S microphone may be changed to change the directivity angle formed
by the above-mentioned M + S axis and the M-S axis.
[0104]
Next, referring to FIG. 22, a block diagram of the adaptive zoom processing means 180 of FIG. 19
will be described. First, the M input signal 5M having directivity in the forward direction is input
to the level controller 190 and is also input to the band pass filter 78 and the delay unit 79, as
described for the adaptive zoom circuit 60 in FIG. The same processing is performed and input to
the control signal generation circuit 124.
[0105]
In addition, the S input signal 5S having directivity in the left and right direction is input to the
band pass filter 121 and the delay unit 122 as described in the adaptive zoom circuit 60 of FIG.
The signal is input to the circuit 124 and further to the negative terminal of the adder 191 and to
one terminal of the adder 192. In the adder 191, the S input signal 5S is subtracted from the
output signal 190M of the level controller 190, and the MS output signal 180R is output.
Further, the adder 192 adds the output signal 190M of the level controller 190 and the S input
signal 5S, and outputs the M + S output signal 180L.
[0106]
Further, the control signal generation circuit 124 is the same as the control signal generation
circuit of FIG. 16 and thus the redundant description will be omitted, but in the example of FIG. It
is used as the level variable signal of the level controller 190 for changing the pointing angle by
the -S axis, and the horizontal direction level control signal 125 in FIG. 16 is not used. Although
not shown, the level controller 190 may be provided on the S input side, in which case the leftright direction level control signal 125 is used as the level variable signal, and the forward
direction level control signal 126 in FIG. do not do.
[0107]
10-05-2019
31
In addition, although illustration and description are omitted, as in the control signal generation
circuit 82 in FIG. 9, the control signal generation circuit 124 in FIG. 22 is controlled only from
the forward direction audio signal 78B and the forward direction audio removal signal 81A. In
this case, the control signal 84 in FIG. 9 is used as the level variable signal of the level controller
190 in FIG. Therefore, according to the system example shown in FIG. 22, the adaptive zoom
processing can be realized also by the directional angle variable control.
[0108]
According to the sound signal processing apparatus and processing method of the embodiment
of the present invention described above, it is possible to realize optimum sound zooming
according to the imaging mode of the camera and the zooming position of the zoom lens of the
camera, that is, the shooting scene. it can.
[0109]
In addition, a plurality of correlations between the zoom lens zooming position and the acoustic
zooming are provided as mapping coefficients, and by switching them according to the camera
imaging mode information, it is possible to perform more optimal acoustic zooming.
[0110]
Furthermore, according to the acoustic signal processing apparatus of the embodiment of the
present invention, since all digital circuits can be configured, it is easy to incorporate in an LSI,
and the circuit scale is hardly increased by semiconductor miniaturization and high density in the
future. Realization is possible without any problem.
[0111]
Furthermore, when the present invention is applied to a still image recording device such as a
digital camera, when simultaneous sound can also be recorded, sound can be recorded at an
audio level and directivity pattern that matches the lens zooming position.
[0112]
As a camera, a video camera (camera for moving images), a digital camera (camera for still
images), etc. are possible.
[0113]
10-05-2019
32
According to the present invention described above, it is possible to obtain an acoustic signal
processing apparatus and processing method capable of realizing optimum acoustic zooming
according to the imaging mode of the camera and the zooming position of the zoom lens of the
camera. Can.
[0114]
Brief description of the drawings
[0115]
1 is a block diagram showing a conventional sound zoom processing system.
[0116]
2 is a block diagram showing a detailed configuration of the stereo processing unit 3 in the
sound zoom processing system of FIG.
[0117]
3 is a diagram showing a left-right pointing pattern.
[0118]
4 is a diagram showing a forward direction pointing pattern.
[0119]
5 is a diagram showing an example of the microphone arrangement.
[0120]
6 is a diagram showing an example of zooming control.
[0121]
7 is a block diagram showing a sound zoom processing system (acoustic signal processing
apparatus) according to the embodiment of the present invention.
[0122]
10-05-2019
33
8 is a block diagram showing the adaptive zoom processing block 50 in FIG.
[0123]
9 is a block diagram showing the adaptive zoom circuit 60 in FIG.
[0124]
10 is a characteristic curve diagram showing an example of filter characteristics.
[0125]
11 is a block diagram showing a control signal generation circuit 82 in FIG.
[0126]
12 is an explanatory diagram of coefficient generation of the control signal generation circuit of
FIG.
[0127]
13 is a characteristic curve diagram showing an example of mapping in the map generation
circuit 111 in FIG.
[0128]
14 is a characteristic curve diagram showing an example of control signal generation in the
control signal generation circuit of FIG.
[0129]
15 is a block diagram showing another example of the adaptive zoom circuit 60 in FIG.
[0130]
16 is a block diagram showing a control signal generation circuit 124 in FIG.
[0131]
17 is a flowchart showing the operation of the level comparison determination processing means
142 of FIG.
10-05-2019
34
[0132]
18 is a table showing an example of a coefficient table of the coefficient table selection circuit in
FIG.
[0133]
19 is a block diagram showing another example of the sound zoom processing system (acoustic
signal processing apparatus) according to the embodiment of the present invention.
[0134]
FIG. 20M is a diagram showing a microphone directivity pattern.
[0135]
21S is a diagram showing a microphone directivity pattern.
[0136]
22 is a block diagram showing the adaptive zoom processing means 180 in FIG.
[0137]
FIG. 23 is a diagram showing the MS microphone directivity pattern.
[0138]
Explanation of sign
[0139]
Right, center and left microphones as MICR, MICC and MICL non-directional microphones for
sound collection, 50 adaptive processing means, 60 adaptive zoom circuits, 71, 72, 73 level
controllers, 74, 75 adders, 78 bands Pass filter, 79 delay, 81 adder, 82 control signal generation
circuit, 111 map generation means, 112, 113 multiplier.
10-05-2019
35
Документ
Категория
Без категории
Просмотров
0
Размер файла
53 Кб
Теги
jp2001326990
1/--страниц
Пожаловаться на содержимое документа