close

Вход

Забыли?

вход по аккаунту

?

DESCRIPTION JP2008052772

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 JP2008052772
To minimize sound quality deterioration when noise removal is performed on an audio signal on
which intermittent noise is superimposed. A noise generation timing can be properly detected by
detecting a noise generation timing based on a detection signal of a noise detection sensor
provided for a noise generation source. As described above, if noise generation timing can be
properly detected, a section for performing noise removal can be suppressed to the necessary
minimum, and sound quality deterioration can also be suppressed to the minimum. In addition,
since the above noise detection operation is performed only in the section where noise
generation timing is included, the chance of false detection of noise detection is reduced
accordingly, and as a result, the opportunity for sound quality deterioration is reduced
accordingly. Can. [Selected figure] Figure 5
Recording device, noise removal method, noise removal device
[0001]
The present invention is a recording apparatus provided with recording means for recording an
audio signal having a sound collection means as an input source on a required recording
medium, and in particular, a recording apparatus having a noise source for generating
intermittent noise sounds during recording operation. About. The present invention also relates
to a noise removing method and a noise removing device which are applied to, for example, such
a recording device and are suitable for removing noise superimposed on the audio signal.
[0002]
10-04-2019
1
2. Description of the Related Art Conventionally, video camera apparatuses that record captured
video and collected audio on a required recording medium have become widespread. And, as
such a video camera apparatus, in recent years, there has been one which incorporates a hard
disk drive (HDD) and records a photographed video and a collected voice (hereinafter also
referred to as photographed data) in this.
[0003]
As described above, some video camera apparatuses that record shooting data in the HDD
continuously record the shooting data while the magnetic head is loaded on the rotationally
driven magnetic disk. That is, from the recording start instruction to the recording stop
instruction, the shooting data is continuously written to the magnetic disk.
[0004]
However, in this method, since the rotation operation of the magnetic disk is always involved
during the recording operation, an intermittent access method has been proposed as an
improved recording method. In this intermittent access, shooting data is temporarily stored in the
built-in memory after recording starts, and in the meantime, the magnetic head is retracted to a
predetermined parking position and then the rotation of the magnetic disk is stopped. Then,
when the storage amount of imaging data in the built-in memory reaches a predetermined value
or more, rotation drive of the magnetic disk and loading to the recording position of the magnetic
head are performed, and the imaging data stored and read are recorded. After recording for a
predetermined time, the magnetic head is again retracted to the parking position, and the
rotational drive of the magnetic disk is stopped. In this manner, intermittent recording on the
magnetic disk is performed according to the storage amount of the imaging data in the built-in
memory. By this intermittent access method, power consumption can be reduced more than
when recording is performed with the magnetic disk constantly rotated.
[0005]
However, when this method of intermittent access is adopted, when loading the magnetic head
onto the magnetic disk, or when retracting the magnetic head to the parking position, noise due
10-04-2019
2
to physical collision of the device is generated inside the HDD. It becomes something that occurs
in And this noise is picked up by the microphone with which a video camera apparatus is
equipped, and the problem that this will be recorded with imaging data generate | occur |
produces. Therefore, in the case of using the method of intermittent access, it is necessary to
reduce the power consumption, but it is necessary to take measures against such noise.
[0006]
The following patent documents describe a technique for removing noise superimposed on an
audio signal. JP, 2005-203041, A JP, 2002-251823, A JP, 2005-228400, A
[0007]
By the way, when it is made to cope with the noise accompanying the above intermittent access,
it must be taken into consideration that it is of the nature which occurs intermittently and is
different from the stationary noise. Here, the noise removal is a process of modifying the original
signal, and as a result, it is a process that may cause the deterioration of the sound quality. With
regard to stationary noise, there is no choice but to always remove the noise if it is to be
removed, but with regard to intermittent noise, in order to be the minimal modification necessary
for such sound quality deterioration, It is desirable to target only the section where noise occurs.
In consideration of such a thing, in order to remove the noise accompanying the intermittent
access as described above, it is important to first detect the timing at which the noise occurs.
[0008]
Specifically, in order to obtain information on the noise generation timing, it is conceivable to
predict this from, for example, the timing at which the control to the corresponding portion that
is the noise generation source is performed (for example, described in Patent Documents 2 and
3) . However, in the intermittent access as described above, the length of time until the collision
sound occurs with the loading of the magnetic head in response to the recording start
instruction, and the length of time until the collision sound occurs with the parking of the
magnetic head in response to the recording stop instruction. Are known to differ depending on
the difference in recording start / end position and other conditions. Although it is conceivable to
cope with this by providing a certain degree of margin in the noise section to be set according to
the recording start / stop instruction, in consideration of the above-mentioned sound quality
10-04-2019
3
problem, the noise It is desirable to minimize the target section to be removed, and in that sense,
there is a limit to the method of predicting the noise generation section from the timing of the
recording start / stop instruction.
[0009]
Therefore, in the present invention, in view of the above problems, the recording apparatus is
configured as follows. That is, the recording apparatus of the present invention is provided with a
recording means for recording an audio signal having a sound collection means as an input
source on a required recording medium, and a recording having a noise generation source
generating intermittent noise with the recording operation. The apparatus is provided with a
noise detection sensor provided for the noise generation source. Further, analysis of the
detection signal of the noise detection sensor is performed only in a section of the inclusive
section including the timing of occurrence of the noise acquired based on predetermined
information, and the timing of occurrence of the noise is detected based on the analysis result
Noise timing detection means. Further, the noise signal superimposed on the voice signal is
removed by performing predetermined signal processing on the voice signal in the noise section
set based on the generation timing of the noise detected by the noise timing detection means. A
noise removing unit is provided to perform a noise removing process.
[0010]
Further, in the present invention, the noise removal device is configured as follows. That is, the
noise removal apparatus according to the present invention is a noise removal apparatus for
removing intermittent noise superimposed on an audio signal, and includes a noise detection
sensor provided for the noise generation source. Further, analysis of the detection signal of the
noise detection sensor is performed only in a section of the inclusive section including the timing
of occurrence of the noise acquired based on predetermined information, and the timing of
occurrence of the noise is detected based on the analysis result Noise timing detection means.
Further, the noise signal superimposed on the voice signal is removed by performing
predetermined signal processing on the voice signal in the noise section set based on the
generation timing of the noise detected by the noise timing detection means. A noise removing
unit is provided to perform a noise removing process.
[0011]
10-04-2019
4
According to the present invention, the noise generation timing is detected based on the
detection signal from the noise detection sensor provided for the noise generation source.
Therefore, for example, from the control timing for the portion as the noise generation source,
the noise generation interval It is possible to properly detect the noise generation timing, unlike
the case of predicting. Moreover, in the present invention, since the noise detection operation
based on the detection sensor is performed only in the inclusive section in which the noise
generation timing is included, it is not necessary to always perform the noise detection operation.
[0012]
As described above, according to the present invention, the noise generation timing can be
properly detected by detecting the noise generation timing based on the detection signal of the
noise detection sensor provided for the noise generation source. . Since the noise generation
timing can be properly detected in this way, the noise removal section can be minimized, thereby
minimizing the sound quality deterioration that may occur with the noise removal. Can.
[0013]
Further, in the present invention, the noise detection operation is performed only in the inclusive
section in which the noise occurrence timing is included as described above. According to this,
the signal analysis for noise detection is performed accordingly. The processing load can be
reduced, and power consumption can be reduced accordingly. Although noise detection based on
such signal analysis has no possibility of false detection as described above, it is assumed that the
period during which signal analysis for noise detection is performed includes noise generation
timing. By narrowing down to the included interval, the chance of false detection of noise
detection can be reduced accordingly. And if the chance of false detection can be reduced in this
way, the useless noise removal operation based on such false detection will not be performed,
and the chance of sound quality deterioration can be reduced accordingly. That is, sound quality
deterioration can be minimized also in this aspect.
[0014]
The best mode for carrying out the invention (hereinafter referred to as the embodiment) will be
10-04-2019
5
described below. [Arrangement of Recording Apparatus] FIG. 1 is a perspective view showing an
appearance of a video camera apparatus 1 which is an embodiment of the recording apparatus of
the present invention. The video camera apparatus 1 includes a camera lens 1a as illustrated, and
can record a captured image obtained through the camera lens 1a on an internal recording
medium. Also, the video camera apparatus 1 includes an external sound pickup microphone
(microphone) 1 b provided so as to be exposed to the outside of the housing as shown in the
drawing, and the sound pickup by the external sound pickup microphone 1 b is performed The
voiced sound can be recorded on the recording medium along with the photographed video.
[0015]
FIG. 2 is a block diagram showing an internal configuration of the video camera device 1. First,
also in FIG. 2, the camera lens 1a shown in FIG. 1 is shown. The camera lens 1a is provided in the
camera block 5 as illustrated. The camera block 5 is a solid-state imaging device portion such as
a CCD (Charge Coupled Device) sensor or a CMOS (Complementary Metal-Oxide Semiconductor)
sensor that detects object light incident through the camera lens 1a, or a solid-state imaging
device portion. It includes a video signal generation unit or the like which transfers a pixel signal
and performs gain adjustment, A / D conversion, etc. to obtain a video signal.
[0016]
The video signal obtained by the camera block 5 is supplied to a video signal processing unit 6
shown in the figure, where predetermined video signal processing such as white balance and Y /
C processing is performed here, and the video signal compression processing unit 7 is Supplied.
The video signal compression processing unit 7 subjects the video signal to compression
processing based on a predetermined video compression method to generate compressed video
data. Then, the compressed video data is output as stream data (video stream Vstrm shown).
[0017]
Also shown in FIG. 2 is the external sound collection microphone 1b shown in FIG. The input
sound signal Ain obtained based on the sound collection operation of the external sound
collection microphone 1b is supplied to the noise detection / removal unit 2 surrounded by a
dashed dotted line in the drawing.
10-04-2019
6
[0018]
The noise detection / removal unit 2 is based on the input audio signal Ain, a detection audio
signal ANin from a noise detection microphone (microphone) 11 described later, and noise
occurrence predicted interval information NcS from a system controller 12 described later. The
detection of the noise generation timing and the removal of the noise based on the noise
generation timing are performed, and the result is output as an output sound signal Aout
illustrated. The internal configuration and operation of the noise detection / removal unit 2 will
be described later.
[0019]
The audio signal processing unit 3 receives the output audio signal Aout, performs
predetermined audio signal processing, and supplies the result to the audio signal compression
processing unit 4. The audio signal compression processing unit 4 performs compression
processing based on a predetermined audio compression method on the output audio signal Aout
supplied from the audio signal processing unit 3 to generate compressed audio data. Then, the
compressed audio data is output as stream data (audio stream Astrm shown).
[0020]
The AV (Audio Visual) stream generation unit 8 receives the video stream Vstrm from the video
signal compression processing unit 7 described above and the audio stream Astrm from the
audio signal compression processing unit 4 and combines them. An AV stream AVstrm is
generated. Then, the AV stream AVstrm is supplied to the buffer memory 9.
[0021]
The buffer memory 9 performs buffering of the AV stream AVstrm by writing / reading the AV
stream AVstrm supplied from the AV stream generation unit 8 to the internal memory according
to an instruction from the system controller 12 described later. .
[0022]
The HDD (hard disk drive) 10 records the AV stream AVstrm supplied from the buffer memory 9
10-04-2019
7
on the internal magnetic disk (magnetic disk 10 B) based on an instruction from the system
controller 12.
[0023]
Here, the internal configuration of the HDD 10 is shown in the following FIG.
As shown in FIG. 3, the HDD 10 is provided with a magnetic disk 10B and a magnetic head 10A
for recording a signal on the magnetic disk 10B.
In addition to this, for example, a drive system for driving the magnetic head 10A to the load /
parking position in the drawing, a drive system for rotating the magnetic disk 10B, and the like
are also provided. The configuration is omitted. Further, in the case of the present embodiment,
the noise detection microphone (microphone) 11 is provided in the HDD 10 (not shown, see FIG.
2), which will be described later.
[0024]
Returning to FIG. 2, the system controller 12 is constituted by a microcomputer provided with,
for example, a central processing unit (CPU), a read only memory (ROM), a random access
memory (RAM), etc., and the video camera apparatus Perform overall control of 1, operation
processing. For example, by controlling each necessary unit according to an operation signal
from an operation input unit (not shown), an operation according to the user's operation input
can be realized, or a storage of data file in HDD 10 or creation / update of management
information Etc.
[0025]
Further, the system controller 12 particularly in the case of the embodiment also executes
control processing for realizing intermittent access at the time of recording of the AV stream
AVstrm (also referred to as shooting data) to the HDD 10. That is, the system controller 12 does
not immediately read out the photographing data from the buffer memory 9 and instruct the
recording start of the photographing data to the HDD 10 immediately after the recording start of
10-04-2019
8
the photographing data, and the storage amount of the photographing data in the buffer memory
9 These instructions are given in response to the threshold being equal to or greater than one.
Then, after the reading of the photographing data from the buffer memory 9 and the start of the
recording by the HDD 10, the photographing from the buffer memory 9 is taken in response to
the accumulation amount of the photographing data in the buffer memory 9 becoming less than
the second threshold. It instructs to stop data reading and recording operation of the HDD 10.
Also thereafter, according to the comparison result of the storage amount of the imaging data in
the buffer memory 9 and the first and second threshold values, the reading start instruction of
imaging data from the buffer memory 9 and the recording start instruction to the HDD 10, buffer
memory An instruction to stop reading out the imaging data from 9 and an instruction to stop
recording on the HDD 10 are repeatedly performed.
[0026]
[HDD Noise] Here, in the video camera device 1 shown in FIG. 2, by adopting such a method of
intermittent access, for example, the magnetic head 10A is used while the magnetic disk 10B is
kept rotating in the HDD 10. Compared to the case where the recording operation is
continuously performed, the power consumption can be reduced accordingly.
[0027]
However, in this intermittent access method, the magnetic head 10A is loaded on the magnetic
disk 10B in the HDD 10 in accordance with the recording start instruction as described above,
but at that time, it is caused by the physical collision of the devices. Noise is generated inside the
HDD 10.
Also, in response to the recording stop instruction, the magnetic head 10A is retracted to the
parking position, but also at this time, noise due to a physical collision of the devices is generated
in the HDD 10.
[0028]
In this manner, noise generated in the HDD 10 during recording operation of shooting data is
collected by the external sound collection microphone 1 b and superimposed on the input sound
signal Ain, and as a result, is recorded in the HDD 10 together with the shooting data. Problems
occur.
10-04-2019
9
[0029]
FIG. 4 is a diagram exemplifying the characteristics of noise (hereinafter, also referred to as HDD
noise) generated in the HDD 10 with intermittent access in this manner, and in FIG. 4A, its time
waveform is also shown in FIG. Shows the spectrogram.
Such HDD noise depends on the individual differences of the HDD 10, but when viewed from the
time waveform (FIG. 4 (a)), it becomes spike noise having a very short duration of about 0.03 sec.
ing. Also, as seen in the spectrogram (FIG. 4 (b)), the power is distributed in a wide frequency
band in the form of impulse, and it can be confirmed that the signal concentrated in time is
distributed in a wide band in frequency. . Although this HDD noise is a signal component with a
very low level, it becomes an auditorily very annoying noise due to these characteristics. Then,
such troublesome noise is generated intermittently at each recording start / stop accompanying
intermittent access, and this is superimposed and recorded on the imaging data (audio signal).
[0030]
Here, in order to prevent the recording of such HDD noise, it is conceivable to remove it from the
input audio signal Ain, but at that time, the HDD noise is generated intermittently. It must be
taken into account that it is different from stationary noise. That is, the noise removal is a
process of modifying the original signal, and as a result, it is a process that may cause the
deterioration of the sound quality. With regard to stationary noise, there is no choice but to
always remove the noise if it is to be removed, but with regard to intermittent noise, in order to
be the minimal modification necessary for such sound quality deterioration, It is desirable to
target only the section where noise occurs.
[0031]
In consideration of such a thing, in order to remove HDD noise as described above, it is important
to first detect the timing at which the noise occurs.
[0032]
As a specific method for that purpose, for example, it is conceivable to predict this from the
timing at which control to the noise source is performed as in the prior art.
10-04-2019
10
However, in the above-mentioned intermittent access, the length of time until the collision sound
occurs with loading of the magnetic head 10A according to the recording start instruction, and
the length of time until the collision sound with parking of the magnetic head 10B occurs
according to the recording stop instruction Are known to differ depending on the difference in
recording start / end position and other conditions. Although it is conceivable to cope with this
by providing a certain degree of margin in the noise section to be set according to the recording
start / stop instruction, it is removed in consideration of the above-mentioned sound quality
problems. It is desirable to minimize the interval in which the noise reduction interval is
performed, and in that sense, it can be said that there is a limit to the method of predicting the
noise occurrence interval from the timing of the recording start / stop instruction.
[0033]
[Noise Timing Detection According to the Embodiment] Therefore, in the video camera device 1
according to the present embodiment, the noise detection microphone 11 is provided in the HDD
10 as shown in FIG. The generation timing of the HDD noise is detected based on the detection
audio signal ANin obtained based on the sound collecting operation.
[0034]
In FIG. 2, the detection audio signal ANin from the noise detection microphone 11 is supplied to
the noise timing detection / noise interval setting unit 24 in the noise detection / removal unit 2.
The noise timing detection / noise interval setting unit 24 detects the noise occurrence timing
and sets the noise interval based on the detected noise occurrence timing as described later
based on the detection audio signal ANin thus supplied. Do.
[0035]
Here, as described above, HDD noise in the present embodiment is noise generated intermittently
according to start / stop of recording operation in HDD 10, and the generation timing thereof is
from system controller 12 to HDD 10 It can be predicted to some extent from the timing at which
the recording start instruction and the recording stop instruction are performed.
[0036]
10-04-2019
11
Therefore, in the present embodiment, the operation by the noise timing detection / noise section
setting unit 24 is performed only in the section where HDD noise is expected to be generated as
described above, and the useless detection operation is performed. To reduce the
[0037]
In the video camera device 1 shown in FIG. 2, the system controller 12 instructs the noise timing
detection / noise interval setting unit 24 to provide information on the timing thus predicted.
That is, the system controller 12 sets the predetermined section information as the noise
occurrence expected section information NcS in accordance with the recording start instruction
and the recording stop instruction to the HDD 10 along with the control process for the
intermittent access described above. The timing detection and noise section setting unit 24 is
supplied.
[0038]
The noise timing detection / noise interval setting unit 24 detects the detection audio signal ANin
supplied from the noise detection microphone 11 only in the interval indicated by the noise
occurrence prediction interval information NcS supplied from the system controller 12 in this
manner. Signal analysis is performed, and the occurrence timing of the HDD noise is detected
based on the result.
Then, based on the detected noise occurrence timing, a noise interval indicating an interval in
which HDD noise occurs is set, and noise interval information NS indicating the start timing and
the end timing is output.
[0039]
FIG. 5 is a diagram schematically showing the operation of detecting the noise generation timing
and the operation of setting the noise section by the noise timing detection / noise section setting
unit 24. As shown in FIG. 5A shows an example of the detection audio signal ANin including the
HDD noise component, and FIG. 5B shows a noise section to be set according to the detection
10-04-2019
12
audio signal ANin shown in FIG. 5A. An example is shown.
[0040]
First, as shown in FIG. 5A, the noise timing detection / noise interval setting unit 24 determines
the first threshold th-s1 and the second threshold th- predetermined for the amplitude value of
the detection audio signal ANin. s2 is set. The noise timing detection / noise interval setting unit
24 determines that the amplitude value of the input detection audio signal ANin is a value
outside a predetermined range defined by the first threshold th-s1 and the second threshold ths2. It is determined whether or not HDD noise has occurred by determining whether or not it is.
That is, the timing at which the amplitude value of the detection audio signal ANin is out of the
predetermined range by the first threshold th-s1 and the second threshold th-s2 is detected as
the occurrence timing of the HDD noise. At this time, the noise timing detection / noise interval
setting unit 24 detects the amplitude value of the detection audio signal ANin, the threshold
value th-s1, and the threshold value th only in the noise occurrence prediction interval indicated
by the noise occurrence prediction interval information NcS. A comparison with -s2 is to be
made.
[0041]
When the occurrence timing of the HDD noise is detected as a result of the determination as
described above, a predetermined section including the noise occurrence timing is set as a noise
section as shown in FIG. 5B. For example, in this case, as shown in the drawing, a predetermined
section extending before and after the noise generation timing is set as the noise section. Then,
the information of the start point and the end point of the noise section set in this way is output
as the noise section information NcS.
[0042]
Note that for confirmation, the noise timing detection / noise interval setting unit 24 determines
that the amplitude value of the detection audio signal ANin is based on the threshold th-s1 and
the threshold th-s2 in the noise occurrence prediction interval. If the HDD noise is not detected
out of the range, the noise section information NS is not output. Further, when setting of a noise
section is performed on the assumption that HDD noise has been detected, thereafter, within this
set noise section, signal analysis on the detection audio signal ANin (in this case, the first
10-04-2019
13
threshold th-s1 of the amplitude value) The comparison with the second threshold th-s2 is not
performed.
[0043]
Further, in the above example, the amplitude value of the detection audio signal ANin is simply
compared with the predetermined threshold in detecting the noise generation timing, but instead,
for example, the average power within the predetermined time is calculated It is also possible to
detect the noise occurrence timing based on the comparison result of the average power and a
predetermined threshold.
[0044]
[Noise Impairment Determination] In the video camera device 1 of the present embodiment, the
configuration described later based on the information (NS) of the noise section detected and set
by the noise timing detection / noise section setting unit 24 as described above. Although noise
is removed from the input audio signal Ain, as mentioned in the explanation of FIG. 4 above, the
signal levels of the HDD noise in the present embodiment are compared. In some cases, it may be
possible that the sound may be buried in the external sound (background sound) and be erased.
[0045]
6 to 9 schematically show the relationship between such external sound and HDD noise.
6 and 7 show the case where the background sound has a relatively small volume, respectively,
and in FIG. 6 (a) the temporal waveform of the background sound is shown, and (b) the HDD
noise is shown. Further, (c) shows the time waveform of the input audio signal Ain in which the
HDD noise and the background sound are added.
Further, in FIG. 7, the frequency spectra of the background sound (dotted line) and the HDD
noise (solid line) are shown in comparison.
[0046]
First, as can be seen by comparing FIGS. 6A and 6B, the HDD noise has an amplitude of about ±
10-04-2019
14
300 with respect to the 16-bit full scale (± 32767). Based on this, HDD noise is less likely to be
buried in the input audio signal Ain shown in FIG. 6C, for example, when recording under a
relatively small volume background sound as shown in FIG. 6A. . Also, when this is compared
with the frequency spectrum shown in FIG. 7, it can be seen that the spectrum of the HDD noise
is distributed at a power somewhat higher than the spectrum of the background sound. From the
results of FIGS. 6 and 7, it can be seen that the human auditory sense is more likely to perceive
HDD noise under relatively low volume background sound, and it is effective to perform noise
removal in such a case.
[0047]
8 and 9 show the case where the background sound is relatively loud, and FIG. 8 (a) shows the
background sound, FIG. 8 (b) shows the HDD noise, and FIG. 8 (c) shows the same. The time
waveforms of the input audio signal Ain are respectively shown, and FIG. 9 shows the frequency
spectrum of the background sound (broken line) and the HDD noise (solid line) in comparison.
Under relatively loud background sound as shown in FIG. 8A, HDD noise is likely to be buried in
the input audio signal Ain shown in FIG. 8C, and the frequency spectrum shown in FIG. 9 is
viewed. Also, it can be seen that the spectrum of the HDD noise is distributed with power
substantially smaller than that of the background sound. As described above, under loud
background sound, human hearing is hard to perceive HDD noise, and the advantage of
performing noise removal in such a situation is extremely small. Rather, in this case, since the
level of the input audio signal Ain for noise removal is very large, there is a high possibility that
the sound quality deterioration due to the noise removal is more easily heard.
[0048]
From the above, in the video camera device 1 of the present embodiment, the background sound
is set to a relatively small volume, and the noise removal is performed only when the HDD noise
is not buried, and a relatively large volume of background is generated. It is intended to prevent
unnecessary noise removal from being performed when HDD noise is buried under the sound.
[0049]
As a configuration for this, the noise detection / removal unit 2 shown in FIG. 2 is provided with
a delay circuit 21, a noise embedding determination unit 25, a removal operation control unit 27,
a delay circuit 28, and a selection control unit 29.
10-04-2019
15
First, the delay circuit 21 receives an input sound signal Ain from the external sound pickup
microphone 1b, applies a delay of a predetermined time length to this, and outputs it. The input
audio signal Ain to which the delay by the delay circuit 21 is given is called an input audio signal
Ain-1. In this case, the input sound signal Ain input to the delay circuit 21 from the external
sound pickup microphone 1b is hereinafter referred to as the input sound signal Ain-0 in order
to distinguish it from the input sound signal Ain after such delay. Also called.
[0050]
Here, the delay time set in the delay circuit 21 is set at least by the noise timing detection / noise
interval setting unit 24 as the input audio signal Ain-1 supplied to the noise embedding
determination unit 25 described below. It may be set in such a manner that a signal portion that
is earlier in timing than the start point of the noise section is supplied. That is, when the noise
section is set, the delay may be performed so that at least a signal portion before the noise
section is supplied to the noise embedding determination unit 25.
[0051]
The noise embedding determination unit 25 analyzes the signal of the input audio signal Ain-1
from the delay circuit 21 and determines whether the HDD noise is embedded in the background
sound (that is, a signal component other than the HDD noise in the input audio signal Ain). judge.
Specifically, the noise embedding determination unit 25 calculates the power for each
predetermined section of the time waveform of the input audio signal Ain-1, and if the power is
larger than a predetermined threshold th-p determined in advance. The determination result that
the HDD noise is not perceived by human hearing, that is, the HDD noise is buried (masked) is
output as the buried determination result information M. If the calculated power is not larger
than the threshold value th-p, a determination result that HDD noise is perceived, that is, HDD
noise is not buried is output as the buried determination result information M. At this time, the
power of the time waveform of the input audio signal Ain-1 can be calculated by the sum of
squares of the values of the input audio signal Ain-1 at each time, for example, as shown in the
following equation. Σt = 0 to T-1 {Ain (t)} <2>
[0052]
10-04-2019
16
Note that the noise burial determination can also be determined based on the result of the power
calculation of the frequency spectrum, as well as based on the result of the power calculation of
the time waveform. That is, the power of the frequency spectrum of the input audio signal Ain-1
is calculated, and when the power is larger than a predetermined threshold value, the
determination result that the HDD noise is buried is output as the buried determination result
information M, and the power is predetermined If it is smaller than the threshold, the
determination result that the HDD noise is not buried is output as the buried determination result
information M. The power of the frequency spectrum can be calculated, for example, by the sum
of squares of the frequency spectrum of the input speech signal Ain-1 at each time as shown in
the following equation. Σ f = 0 to fs / 2 {Ain (f)} <2>
[0053]
Also, for example, when the spectrum of noise to be processed has a characteristic frequency
distribution in a predetermined band, the power may be calculated only from that band. The
power of a predetermined band (for example, frequencies a to b) in the frequency spectrum can
be calculated, for example, as follows. Ff = fa to fb {Ain (f)} <2>
[0054]
By the way, depending on the configuration of the noise embedding determination unit 25
described above, the above-described signal analysis is always performed on the input sound
signal Ain-1, and the embedding determination result information M is always output. However,
the burial determination result information M is information indicating whether the HDD noise is
buried in the background sound in the first place, and therefore it can be considered that there is
no need to always output this.
[0055]
Therefore, as the noise embedding determination unit 25 of the embodiment, the noise interval
information NS from the noise timing detection / noise interval setting unit 24 described above is
input, and only in the interval indicated by the noise interval information NS, The signal analysis
as described above and the noise embedding determination based on the result are performed.
Specifically, only in the section indicated by the noise section information NS, the power
calculation of the time waveform of the input audio signal Ain-1 as described above, and the
10-04-2019
17
comparison with the threshold th-p for the power, and the comparison thereof It is arranged to
output the burial judgment result information M based on the result.
[0056]
The burial determination result information M output from the noise burial determination unit
25 is supplied to the removal operation control unit 27 as shown in the figure, and is branched
to be separated into the selection control unit 29 via the delay circuit 28. Is also supplied for
[0057]
First, the removal operation control unit 27 outputs an ON / OFF control signal for turning on /
off the operation of the noise removal signal generation unit 26, which will be described later,
based on the embedding determination result information M.
Specifically, when the burial determination result information M indicates the determination
result that the HDD noise is not buried, a signal instructing ON as the ON / OFF control signal is
output, and the HDD noise is buried. In the case where the determination result is indicated, a
signal instructing OFF is output as the ON / OFF control signal.
[0058]
Although described later, the noise removal signal generation unit 26 receives the input sound
signal Ain (Ain-2) and performs predetermined signal processing to generate a noise removal
signal from which HDD noise is removed. When the HDD noise is buried in the background
sound by supplying the ON / OFF control signal according to the content of the burial
determination result information M to the noise removal signal generation unit 26 as described
above. The generation operation of the noise removal signal can be performed only when the
generation operation of the noise removal signal is not performed and the HDD noise is not
buried in the background sound.
[0059]
On the other hand, the selection control unit 29 inputs the burial determination result
information M obtained through the delay circuit 28 as described above, and based on the burial
determination result information M, is input to the signal insertion unit 30 shown in the figure. A
10-04-2019
18
selection control signal SLC for selectively selecting one of the two input signals is output.
[0060]
As will be described later, the signal insertion unit 30 receives an input audio signal Ain-3
obtained through the delay circuit 22 and the delay circuit 23 illustrated together with the delay
circuit 21 described above, and generates a noise removal signal. The noise removal signal from
the unit 26 is input.
Then, whether to output the input audio signal Ain-3 as it is or to output the noise removal signal
instead of the input audio signal Ain-3 is selected according to the selection control signal SLC. .
[0061]
The selection control unit 29 instructs the selection of the input audio signal Ain-3 when the
embedding determination result information M indicates the determination result that the HDD
noise is buried in the background sound. Output On the other hand, when the embedding
determination result information M indicates the determination result that the HDD noise is not
buried in the background sound, the selection control signal SLC instructing selection of the
noise removal signal is output. That is, when such a selection control signal SLC is supplied to the
signal insertion unit 30, the input audio signal Ain-3 is output as it is when the HDD noise is
buried in the background sound (that is, the noise removal) Can be output only when the HDD
noise is not buried in the background sound) (that is, the audio signal from which the noise is
removed). .
[0062]
As described above, depending on the burial determination result information M output from the
noise burial determination unit 25, when the HDD noise is not buried in the background sound,
that is, when it is determined that the noise removal is necessary, the input audio signal A noise
removal process for Ain can be performed, and an audio signal subjected to the noise removal
can be selectively output. On the other hand, when HDD noise is buried in the background sound,
that is, when noise removal is not required, noise removal processing is not performed on the
input audio signal Ain, and noise removal is performed. It is possible to selectively output an
10-04-2019
19
unvoiced audio signal.
[0063]
The delay times to be set to the above-described delay circuit 22, delay circuit 23, and delay
circuit 28 will be described later.
[0064]
[Noise Removal] Next, a noise removal method performed by the video camera device 1
according to the present embodiment will be described.
First, when the HDD noise of the embodiment is reconsidered prior to the description of the noise
removal method of the embodiment, the duration of the HDD noise can be roughly viewed from
the time waveform (FIG. 4A). The spike noise is very short, about 0.03 sec. Also, as seen in the
spectrogram (FIG. 4 (b)), the power is distributed in a wide frequency band in the form of
impulse, and the signal concentrated in time is distributed in a wide band in frequency. .
[0065]
As described above, since the generation time of the HDD noise is a very short period, for
example, in the noise removal method using an adaptive filter as conventionally performed, the
convergence of the filter is not in time, and the result is It is not unthinkable that the possibility
of not being able to remove the noise as
[0066]
Also, since HDD noise is characterized by its power distribution over a relatively wide band as
described above, noise is removed, for example, when applying a method of removing this by
adding signals of opposite phases. Can be difficult.
[0067]
Therefore, in the present embodiment, a noise removal method that can properly remove such
HDD noise in consideration of the characteristics of the HDD noise is adopted.
10-04-2019
20
FIG. 10 is a diagram schematically showing the basic concept of the noise removal method
according to the embodiment, and FIG. 10 (a) shows a noise section of the input sound signal Ain
and a time waveform including before and after it. FIG. 10 (b) shows a time waveform including
the noise section and the preceding and following sections of the voice output signal Aout after
noise removal.
[0068]
In the present embodiment, in consideration of the characteristics of the HDD noise as described
above, the audio signal in the noise section is interpolated using the waveform information of the
audio signal around the noise generation timing.
That is, as shown in FIG. 10B, a noise removal signal for removing noise from the peripheral
waveform of the noise generation timing is generated, and the noise removal signal generated in
this way is inserted into the noise section. Thus, the signal of the noise section is interpolated.
[0069]
In order to realize such a noise removal method, the video camera device 1 according to the
embodiment includes the noise removal signal generation unit 26 and the signal insertion unit
30 shown in FIG. As shown in FIG. 2, the noise removal signal generation unit 26 receives the
input audio signal Ain-2 obtained through the delay circuit 21 and the delay circuit 22, and the
noise interval from the noise timing detection / noise interval setting unit 24. Information NS is
input. The noise removal signal generation unit 26 generates a noise removal signal for
interpolating the noise interval indicated by the noise interval information NS in the input audio
signal Ain-2, and supplies this to the signal insertion unit 30.
[0070]
Further, as described above, the noise removal signal generation unit 26 is supplied with the ON
/ OFF control signal from the removal operation control unit 27, and the noise removal signal
generation operation is turned on based on the ON / OFF control signal. / OFF is to be controlled.
The internal configuration and operation of the noise removal signal generation unit 26 will be
10-04-2019
21
described later.
[0071]
In the signal insertion unit 30, together with the noise removal signal from the noise removal
signal generation unit 26, an input audio signal Ain-3 obtained through the delay circuit 21, the
delay circuit 22 and the delay circuit 23, noise timing detection / noise The noise section
information NS from the section setting unit 24 is input. The signal insertion unit 30 outputs the
input audio signal Ain-3 as it is, or in the noise section indicated by the noise section information
NS in the input audio signal Ain-3, in place of the input audio signal Ain-3. It is possible to select
whether to output the noise removal signal. The selection output of the signal insertion unit 30 is
supplied to the audio signal processing unit 3 as an audio output signal Aout as illustrated.
[0072]
Such selection operation of the signal insertion unit 30 is controlled by the selection control
signal SLC from the selection control unit 29 as described above. That is, when the selection of
the input audio signal Ain-3 is instructed by the selection control signal SLC, the signal insertion
unit 30 selectively outputs the input audio signal Ain-3 as it is. When the selection of the noise
removal signal is instructed by the selection control signal SLC, the noise removal signal is output
instead of the signal of the noise section in the input audio signal Ain-3. In other words, the noise
removal signal is inserted into the noise section in the input audio signal Ain-3.
[0073]
Here, the generation method of the noise removal signal of this embodiment will be described.
FIG. 11 shows an internal configuration of the noise removal signal generation unit 26 shown in
FIG. As shown in FIG. 11, in the noise removal signal generation unit 26, a delay circuit 31, a first
half pitch calculation unit 32, a second half pitch calculation unit 33, a first half signal generation
unit 34, a second half signal generation unit 35, a cross fading processing unit 36 are provided.
As shown in FIG. 2 described earlier, the noise removal signal generation unit 26 receives the
noise interval information NS from the noise timing detection / noise interval setting unit 24 and
the ON / OFF control signal from the removal operation control unit 27. Supplied. Although
illustration is omitted, the noise section information NS is supplied to each unit in the noise
removal signal generation unit 26. The above-mentioned ON / OFF control signal is also supplied
10-04-2019
22
to each part in the noise removal signal generation unit 26, and each part is configured to be
turned ON / OFF in accordance with this ON / OFF control signal.
[0074]
In FIG. 11, first, an input voice signal Ain-2 obtained through the delay circuit 21 and the delay
circuit 22 shown in FIG. 2 is input to the first half pitch calculation unit 32 and the second half
pitch calculation unit 33. Ru. The first half pitch calculation unit 32 calculates, for the input
speech signal Ain-2, a pitch period which is considered to be most correlated before and after the
start point of the noise section indicated by the noise section information NS, and the pitch
thereof First half pitch information Pf indicating a section specified by the period is obtained.
Further, the latter half pitch calculation unit 33 calculates, for the input speech signal Ain-2, a
pitch period that is considered to be most correlated before and after the end point of the noise
section indicated by the noise section information NS, Second-half pitch information P-r
indicating a section specified by the pitch period is obtained.
[0075]
For example, in the first half pitch calculation unit 32 and the second half pitch calculation unit
33, the pitch period can be calculated as the number T of samples that minimizes the average
distortion as defined below. (1 / T) * tt = 0 to T-1 {Ain (t)-Ain (t + T)} <2> or (1 / T) * tt = 0 to T-1
| Ain (t)-Ain ( t + T) | As described above, in the first half pitch calculating section 32 and the
second half pitch calculating section 33, the pitch period which makes the waveform correlation
highest is calculated.
[0076]
In addition, at the time of calculation of the pitch period in this case as described above, the part
in the noise section is also the target of the pitch calculation by making the target before and
after the noise section start point and end point as a reference. This is because, in the
embodiment, the noise section is set wider than the section in which HDD noise actually exists.
Furthermore, HDD noise is a signal with an extremely small level, and a signal dominant in pitch
calculation. It is because it does not become.
[0077]
10-04-2019
23
The first half pitch information Pf obtained by the first half pitch calculation unit 32 is supplied
to the first half signal generation unit 34.
Further, the latter half pitch information Pr obtained by the latter half pitch calculation unit 33 is
supplied to the latter half signal generation unit 35.
[0078]
An input audio signal Ain-2d obtained by delaying the input audio signal Ain-2 by the delay
circuit 31 illustrated is input to the first half signal generation unit 34 and the second half signal
generation unit 35. The delay time to be set in the delay circuit 31 will be described later.
[0079]
The first half signal generation unit 34 generates the first half signal based on the input voice
signal Ain-2d and the first half pitch information Pf as described in FIG. 12 below. Further, the
second half signal generation unit 35 generates a second half signal based on the input voice
signal Ain-2d and the second half pitch information Pr as described in FIG. 13 below.
[0080]
12 and 13 schematically show the process of generating the first half signal in the first half
signal generator 34 and the process of generating the second half signal in the second half signal
generator 35, respectively. First, in FIG. 12, assuming that the noise section for the input speech
signal is set as shown in FIG. 12A, for example, the first half pitch calculator 32 may set the pitch
period as shown in FIG. Is calculated. As understood from the above description, such a pitch
cycle is calculated with the cycle having the highest waveform correlation before and after the
start point of the noise section.
[0081]
10-04-2019
24
Then, the first half signal generation unit 34 performs predetermined signal processing on the
input speech signal based on the first half pitch information Pf indicating two sections (pitch
sections) specified by such a pitch cycle. Generate That is, the first-half signal generation unit 34
first generates a signal of two pitch sections indicated by the first-half pitch information Pf in the
input speech signal (Ain-2d) as shown in FIG. Weighting is performed by multiplying each weight
window data. As shown in the figure, the weighting window data in this case is set such that the
start point of the noise section is “1”, and the window data gradually decreases toward “0”
toward the end point of each pitch section Be done.
[0082]
Then, the first half signal generation unit 34 obtains a weighted addition signal for one pitch
period by adding the audio signals weighted by the window data in this manner as shown in FIG.
12 (d). Then, as shown in FIG. 12 (e), the weighted addition signal for one pitch cycle is repeated
a predetermined number of times in the direction from the start point of the noise section to the
end point of the noise section. Get a signal. As illustrated, in this case, it is assumed that the
above-mentioned weighted addition signal is repeated by the maximum number within the range
where it falls within the noise section to generate the above-mentioned repeated weighted
addition signal.
[0083]
Here, each weighted addition signal is obtained by weighting and adding signals originally having
high waveform correlation, and the repeated weighted addition signal obtained by repeating the
same is an iterative method in which the continuity of the input speech signal is maintained.
Therefore, the repetitively weighted addition signal generated in this way can be an audio signal
which is extremely natural in hearing. The first half signal generation unit 34 outputs this
repetitively weighted addition signal as the first half signal.
[0084]
Further, FIG. 13 shows a generation process of the second half signal in the second half signal
generation unit 35. Also in this case, assuming that the noise section for the input speech signal
is set as shown in FIG. 13A, for example, the second half pitch calculator 33 calculates the pitch
period as shown in FIG. 13B. Be done. Such a pitch period calculated by the second-half pitch
10-04-2019
25
calculation unit 33 is also the period in which the waveform correlation is the highest before and
after the end point of the noise section.
[0085]
The second half signal generation unit 35 also generates a second half signal by performing
predetermined signal processing on the input speech signal based on the second half pitch
information P-r indicating two pitch sections specified by such a pitch period. Do. That is, the
second half signal generation unit 35 applies a weight window as shown in FIG. 13C to the
signals of the two pitch sections indicated by the above second half pitch information Pr in the
input speech signal (Ain-2d). Weighting is performed by multiplying each data. As the weighting
window data in this case, as shown in the figure, the end point of the noise section is “1”, and
the window data decreases gradually toward “0” toward the end point of each pitch section. It
is set.
[0086]
Then, even in the latter half signal generation unit 35, the input speech signal weighted by each
window data in this manner is added as shown in FIG. Get Then, as shown in FIG. 13 (e), the
weighting addition signal for one pitch cycle is repeated a predetermined number of times in the
direction from the end point of the noise section to the start point of the noise section to repeat
the weighting addition signal repeatedly. obtain. Also in this case, the above-mentioned weighted
addition signal is repeated by the maximum number within the range which falls within the noise
section to generate the above-mentioned repeated weighted addition signal.
[0087]
The second half signal generation unit 35 outputs the repetitively weighted addition signal
generated in this manner as a second half signal. The second half signal can be generated by the
same method as in the first half signal as described above, so that an audio signal that is
extremely natural for hearing can be obtained.
[0088]
10-04-2019
26
Description will be returned to FIG. The cross-fade processing unit 36 inputs the first half signal
output from the first half signal generation unit 34 and the second half signal output from the
second half signal generation unit 35, performs cross fade processing on them, and generates a
noise removal signal.
[0089]
FIG. 14 is a view schematically showing the cross fade processing of the cross fade processing
unit 36. As shown in FIG. As shown in FIG. 14A, the cross fade processing unit 36 weights the
first half signal from the first half signal generation unit 34 by multiplying the weight window
data as shown in the figure. Similarly, as shown in FIG. 14B, the second half signal from the
second half signal generation unit 35 is also weighted by multiplying the weight window data as
shown in the figure. The window data for weighting the first half signal and the second half
signal have a length in which the first half signal and the second half signal overlap, as shown in
the figure, and gradually from “1” to “0”. The one that falls is set. Specifically, the window
data for the first half signal is set such that the start point of the second half signal is “1” and
gradually decreases toward “0” toward the end point of the first half signal. Further, as the
window data for the second half signal, that whose end point of the first half signal is “1” is set
to gradually decrease toward “0” toward the start point of the second half signal.
[0090]
Then, the cross fade processing unit 36 adds the first half signal and the second half signal
weighted in this way as shown in FIG. Get an addition signal. The cross fade processing unit 36
outputs the first and second half weighted addition signals (that is, cross fade signals) generated
by such processing as a noise removal signal.
[0091]
As described above, this noise removal signal is input to the signal insertion unit 30 shown in
FIG. 2 (FIG. 11), and this is inserted into the noise section in the input voice signal (Ain-3), so that
the noise is eliminated. Removal is to be done.
[0092]
10-04-2019
27
In the noise detection / removal unit 2 in which noise removal is performed as described above, a
plurality of delay circuits are provided, but here, in each of the delay circuits (21, 22, 23, 28, 31)
Organize the delay time to be set.
First, as the delay time of the delay circuit 21, as described above, noise set by at least the noise
timing detection / noise interval setting unit 24 as the input audio signal Ain-1 supplied to the
noise embedding determination unit 25. It may be set in such a manner that a signal portion that
is earlier in timing than the start point of the section is supplied.
[0093]
Further, as the delay time of the delay circuit 22, after the noise embedding determination by the
noise embedding determination unit 25 is performed, the signal portion before the noise interval
of the input audio signal Ain-2 is supplied to the noise removal signal generation unit 26. It
should just be set to be done.
[0094]
In addition, for the delay circuit 31, after the output of the first half pitch information Pf, the first
half signal generation unit 34 is supplied with a signal portion that is a predetermined minute
before the noise section start point of the input voice signal Ain-2d. The delay time may be set so
that a signal portion that is a predetermined amount of time before the end of the noise section
of the input audio signal Ain-2d is supplied to the second half signal generation unit 35 after the
output of -r.
[0095]
The delay circuit 23 delays the signal insertion unit 30 so that the signal portion before the noise
section of the input audio signal Ain-3 is supplied to the signal insertion unit 30 after the noise
removal signal generation unit 26 generates the noise removal signal. The time may be set.
[0096]
Furthermore, the delay time of the delay circuit 28 is set such that the updated burial
determination result information M is supplied to the selection control unit 30 after the noise
removal signal generation unit 26 generates the noise removal signal. Just do it.
[0097]
10-04-2019
28
By setting such delay time for each delay circuit, an operation of removing noise by interpolating
the noise section according to the noise detection for the input audio signal input continuously,
and noise The operation of performing the noise embedding determination on the noise section
in accordance with the detection and the operation of controlling execution / non-execution of
noise removal in accordance with the embedding determination result can be appropriately
performed.
[0098]
[Summary and Modification of Embodiment] As described above, according to the video camera
device 1 of the present embodiment, detection from the noise detection microphone 11 provided
in the HDD 10 as a noise generation source Since the occurrence timing of the HDD noise is
detected based on the signal, the noise occurrence timing can be properly detected, unlike the
case where the noise occurrence section is predicted from the timing of the recording start / stop
instruction to the HDD 10, for example. As a result, the noise section can set a minimum
necessary section.
In this way, if the noise section can be set to the necessary minimum, it can be minimized even as
a sound quality deterioration portion that may occur with noise removal.
[0099]
Then, in the present embodiment, the signal analysis for detecting the noise generation timing
based on the detection signal of the noise detection microphone 11 as described above is
performed only within a predetermined section in which the noise generation timing is included.
And
Specifically, in accordance with the recording start / stop instruction to the HDD 10, it is
performed only within the range indicated by the noise occurrence predicted section information
NcS output by the system controller 12.
As a result, the noise timing detection / noise interval setting unit 24 does not need to constantly
perform signal analysis for noise timing detection, and processing load and power consumption
can be reduced accordingly.
10-04-2019
29
[0100]
Also, at this time, since the above signal analysis is performed only in the section where HDD
noise is expected to occur, there is no detection failure of HDD noise due to the reduction of the
signal analysis section as described above. It can be
That is, from this, it is possible to achieve both the prevention of HDD noise detection omission
and the reduction of processing load and power consumption by the reduction of the analysis
period.
[0101]
Also, noise timing detection based on signal analysis as described above has no possibility of
false detection, but as described above, the period during which signal analysis for noise timing
detection is performed is within the noise occurrence prediction interval NcS. If it is restricted
only to that, the chances of false detection of the noise timing detection will be reduced
accordingly, and accordingly, the chance of unnecessary noise removal operation based on such
false detection can also be reduced.
Then, if it is possible to reduce the chance of useless noise removal operation based on the false
noise detection as described above, the chance of sound quality deterioration can also be
reduced. That is, according to the present embodiment, the sound quality deterioration can be
minimized even in the aspect of reducing the period of signal analysis for noise detection as
described above.
[0102]
In the embodiment, the noise removal signal is generated only when HDD noise is not buried in
the background sound in the input audio signal Ain by the operations of the noise embedding
determination unit 25, the removal operation control unit 27, and the selection control unit 29.
The generation operation of the noise removal signal by the generation unit 26 is executed, and
the noise removal signal (that is, the audio signal from which the noise is removed) is selected
10-04-2019
30
and output by the signal insertion unit 30. Thus, it is necessary to embed the HDD noise in the
background sound and perform the noise removal by selectively outputting the audio signal from
which the noise is removed only when the HDD noise is not buried in the background sound. In
the case where there is no noise, unnecessary noise removal can be effectively prevented. And if
it is possible to prevent unnecessary noise removal as described above, it is possible to effectively
reduce the opportunity for the sound quality deterioration associated with the noise removal.
[0103]
Also, as described above, only when the HDD noise is not buried in the background sound in the
input audio signal Ain, the noise removal signal generation unit 26 executes the noise removal
signal generation operation, thereby the HDD noise is generated. It is possible to effectively
prevent unnecessary noise removal operation from being performed when it is not necessary to
bury in background noise and perform noise removal, thereby reducing processing load and
power consumption accordingly. be able to.
[0104]
Further, in the embodiment, the signal analysis for the noise burial determination by the noise
burial determination unit 25 as described above is not always performed, but is performed only
in a section where HDD noise is generated.
Specifically, this is performed only within the noise section set by the noise timing detection /
noise section setting unit 24. In this way, if the period for performing signal analysis for noise
burial determination is narrowed only within the section where noise is generated, the
processing load on the noise burial determination unit 24 can be reduced accordingly, and the
consumption thereof It is also possible to reduce power consumption.
[0105]
Then, even in the noise burial determination, the signal analysis is performed only in the section
where noise is generated at least in this way, so that there is no omission in the determination of
the noise burial determination, and this can be avoided. Therefore, it is possible to achieve both
the prevention of the determination failure and the reduction of the processing load and the
power consumption as described above.
[0106]
10-04-2019
31
In the embodiment, when the noise generation timing is detected by the noise detection
microphone provided for the noise generation source, the control of the noise removal operation
according to the above-described noise embedding determination is performed. However, this
also offers the following advantages.
That is, when the noise detection microphone is used in this manner, the noise is erroneously
detected because the background sound is picked up by the noise detection microphone under a
situation where the background sound is relatively large. Although there is no possibility, the
noise removal control based on the result of the noise burial judgment as described above is
performed even if the noise is erroneously detected under a relatively large background sound
condition. In fact, noise removal can be prevented.
[0107]
Further, in the embodiment, the noise timing detection / noise interval setting unit 24 sets a
predetermined interval based on the detected noise timing as a noise interval, and the input
audio signal Ain for this noise interval in the input audio signal Ain. It is assumed that the noise
removal is performed by inserting the noise removal signal generated from. According to this, as
in the case of HDD noise of this embodiment, for noise whose power generation period is very
short and power is distributed over a relatively wide frequency band, for example, the method
using the conventional adaptive filter or the reverse Noise removal can be performed more
effectively compared to the case where a phase signal addition method or the like is adopted.
[0108]
Further, in the embodiment, the pitch calculation is performed on the input speech signal Ain
before and after the start point of the noise section and the input speech signal Ain before and
after the end point of the noise section, and the noise section is calculated based on the obtained
pitch information P. The first half signal and the second half signal are generated to interpolate,
and the first half signal and the second half signal are cross-faded to generate the noise removal
signal. Can produce a denoised signal that can be made to be quite natural in hearing. That is,
such a noise removal signal can be used to interpolate the noise section quite naturally in terms
of hearing.
10-04-2019
32
[0109]
-Modification of Embodiment-<First Modification> Hereinafter, a modification of the embodiment
will be described. First, in the first modification, the operation as the embodiment described
above is realized by software processing. Here, as an example, a case where the operation as the
embodiment is realized by software processing of a system controller that performs overall
control of the video camera device will be described.
[0110]
FIG. 15 is a block diagram showing an internal configuration of a video camera device 15 as a
first modification. In addition, in this figure, about the part which becomes the same as the part
already demonstrated in FIG. 2, the same code | symbol is attached | subjected and description is
abbreviate | omitted. As shown, in this case, the input audio signal Ain-1 from the delay circuit
21, the input audio signal Ain-2 from the delay circuit 22, the input audio signal Ain-2d from the
delay circuit 31, and the delay circuit 23 Are input to the system controller 16 respectively. The
system controller 16 is also supplied with a detection audio signal ANin from the noise detection
microphone 11. The system controller 16 in this case also performs control processing for
intermittent access in the same manner as the system controller 12 described above.
[0111]
16 to 18 are flowcharts showing processing operations to be executed by the system controller
16 in order to realize the operations according to the embodiment described above. The
processing operations shown in these figures are executed by the system controller 16 based on
a program stored in, for example, an internal ROM.
[0112]
First, FIG. 16 shows the processing operation corresponding to the operation of the noise timing
detection / noise interval setting unit 24 shown in FIG. First, in step S101, generation of a
recording start / stop instruction is waited. That is, generation of a recording start instruction or
a recording stop instruction to the HDD 10, which is performed by control processing for
10-04-2019
33
intermittent access which is performed in parallel with the processing operation shown in this
figure, is awaited.
[0113]
Then, when the recording start / stop instruction is generated, the detection audio signal ANin is
input in step S102. Furthermore, in the subsequent step S103, the signal analysis in the noise
occurrence prediction interval is performed. That is, in response to the timing of the recording
start / stop instruction being detected as described above, a noise occurrence prediction interval
(noise occurrence prediction interval information NcS) based on the timing is set, and the input
detection audio signal ANin And the threshold value th-s1 and the threshold value th-s2
described above.
[0114]
In the following step S104, it is determined whether or not the amplitude value of the detection
audio signal ANin is out of the predetermined range. That is, based on the result of the signal
analysis, it is determined whether or not the amplitude value of the detection audio signal ANin
exceeds a predetermined range of the threshold th-s1 and the threshold th-s2. If, in step S104, a
negative result is obtained because the amplitude value of the detection audio signal ANin does
not exceed the predetermined range, the process proceeds to step S105, and it is determined
whether the noise occurrence prediction interval has ended. Then, if a negative result is obtained
because the noise occurrence prediction section has not ended yet, the process returns to step
S104, and if a positive result is obtained because the noise generation prediction section has
ended, the process proceeds to step S107 and signal analysis End and become "RETURN".
[0115]
If a positive result is obtained in step S104 that the amplitude value of the detection audio signal
ANin exceeds the predetermined range, the noise period (noise period information) is detected
for a predetermined long period including the detection timing in step S106. After setting as NS),
the signal analysis is finished in step S107, and "RETURN" is made.
[0116]
10-04-2019
34
FIG. 17 also shows the processing operation corresponding to the operation of the noise
embedding determination unit 25.
In FIG. 17, in step S201, as the noise detection standby process, the setting of the noise section in
step S106 shown in FIG. 16 is put on standby. Then, when the noise section is set, in step S202,
the input speech signal Ain-1 is input.
[0117]
In the following step S203, the power (P-NS) of the input speech signal Ain-1 in the noise section
is calculated. That is, as the power P-NS, for example, the power of the time waveform of the
input audio signal Ain-1 in the noise section is calculated as in the noise embedding
determination unit 25 described above. The power of the time waveform of such an input audio
signal Ain-1 can be calculated by the sum of squares of the values of the input audio signal Ain-1
at each time as described above.
[0118]
In step S204, determination processing is performed as to whether the calculated power P-NS
exceeds the threshold th-p. If an affirmative result is obtained because the power P-NS exceeds
the threshold value th-p, the process proceeds to step S205, the noise embedding determination
is performed, and "RETURN" is obtained. On the other hand, if a negative result is obtained
because the power P-NS does not exceed the threshold th-p, the process proceeds to step S206,
noise non-embedding determination is performed, and "RETURN" is obtained.
[0119]
Further, FIG. 18 shows processing operations corresponding to the operations of the noise
removal signal generation unit 26, the removal operation control unit 27, the selection control
unit 29, and the signal insertion unit 30. First, in step S301, the noise non-embedding
determination in step S206 is put on standby. When the noise non-embedding determination is
made, the input sound signal Ain-2 is input in step S302.
10-04-2019
35
[0120]
In the following step S303, processing is performed to calculate first half pitch information P-f
and second half pitch information P-r. That is, the pitch period before and after the noise section
start point of the input speech signal Ain-2 and the pitch period before and after the end point
are calculated similarly to the first half pitch calculation unit 32 and the second half pitch
calculation unit 33 described above First half pitch information Pf and second half pitch
information Pr are obtained from the pitch period.
[0121]
In step S304, an input audio signal Ain-2d is input. Then, in step S305, processing for generating
a first half signal is executed based on the first half pitch information Pf. That is, for the input
speech signal Ain-2d, signals of two pitch sections indicated by the first half pitch information Pf
are respectively multiplied by weight window data as shown in FIG. A weighted addition signal is
generated, and the weighted addition signal is repeated a predetermined number of times from
the start point to the end point of the noise section as shown in FIG. 12D to generate a first half
signal.
[0122]
In step S306, a process for generating a second half signal is performed based on the second half
pitch information Pr. That is, for the input speech signal Ain-2d, signals of two pitch sections
indicated by the latter half pitch information Pr are respectively multiplied by the weight window
data as shown in FIG. While generating a weighted addition signal, the latter half signal is
generated by repeating this weighted addition signal a predetermined number of times from the
end point of the noise section to the start point side as shown in FIG. 13D.
[0123]
Then, in step S307, cross fade processing of the first half signal and the second half signal is
executed. That is, for the first half signal and the second half signal generated as described above,
cross fade processing as shown in FIG. 14 is performed to obtain a cross fade signal (noise
removal signal).
10-04-2019
36
[0124]
In the following step S308, processing is performed to insert a noise removal signal into the
noise section of the input audio signal Ain-3 and output it. That is, the noise removal signal is
selected from the input voice signal Ain-3 to be input and the noise removal signal generated as
described above, and the noise removal signal is selected from the signal portion of the noise
section of the input voice signal Ain-3. Output instead. As a result, a signal from which HDD noise
has been removed can be output as the output sound signal Aout shown in FIG.
[0125]
<Other Modifications> Although the embodiments of the present invention have been described
above, the present invention should not be limited to the embodiments described above. For
example, in the embodiment, as the noise detection sensor provided for the noise generation
source, the microphone as the noise detection microphone 11 is provided, but as the noise
detection sensor, other than such a microphone For example, any sensor that can detect the
occurrence of noise, such as a vibration sensor, can be used.
[0126]
Here, for example, in the case of using the above-described vibration sensor, the following effects
can be expected. That is, when the background sound is large, there is no possibility that the
noise may be erroneously detected if only the microphone is used as the noise detection sensor,
but for example, the noise detection based on the detection signal from such a vibration sensor is
also combined By setting it as (that is, performing noise detection based on the analysis result of
detection signals of both sensors), it is possible to prevent erroneous detection when the
background sound is large as described above.
[0127]
Further, in the embodiment, the information of the noise generation section where noise is
generated is set according to the generation timing of the recording start / stop instruction to the
10-04-2019
37
HDD 10, but the information of the noise generation expected section Can also be acquired based
on information other than the recording start / stop instruction. For example, in response to the
recording start / stop instruction as described above, a control signal for driving the magnetic
head 10A to the load / parking position in the HDD 10 is output to the corresponding drive unit.
It is also possible to obtain information on the expected occurrence interval.
[0128]
Further, the noise removal method described in the embodiment is merely an example, and
various noise removal methods can be applied, such as removing an audio signal in a noise
section by filter processing, for example. Even when any noise removal method is adopted, the
deterioration of the sound quality accompanying the noise removal can not be avoided, so the
noise detection method as exemplified in the embodiment is adopted in such a case as described
above. By minimizing noise reduction intervals (noise intervals) and reducing noise removal
opportunities accompanying reduction of noise false detection opportunities, sound quality
deterioration can be prevented as well.
[0129]
Further, in the embodiment, the case where the signal analysis for noise embedding
determination is performed only in the noise section set based on the noise occurrence timing is
illustrated, but instead, the recording to the HDD 10 as described above is performed. It is also
possible to perform signal analysis for the burial determination only in the noise occurrence
expected section acquired in response to the start / stop instruction. In this way, if signal analysis
for noise burial determination is performed at least only in a section in which noise is generated,
signal analysis without omission of noise burial determination and noise for burial determination
It is possible to achieve both the processing load and the reduction of power consumption.
[0130]
In the embodiment, the recording device of the present invention is exemplified as being capable
of only recording on the recording medium, but it is also possible to configure so as to enable
reproduction as well.
[0131]
Further, the present invention is not limited to such a video camera device 1, but a recording
10-04-2019
38
device in which noise from a noise generation source may be superimposed and recorded on an
input audio signal having a sound collection microphone as an input source. It can be applied
widely widely and suitably.
[0132]
Further, although the embodiment exemplifies the case where the noise removal is performed
before recording the input voice signal from the sound collection microphone, the present
invention is not limited to this, and the case of removing the noise from the voice signal already
recorded. Can also be suitably applied.
In that case, first, for detection of the noise occurrence timing, for example, the timing
information of the recording start / stop instruction performed during recording of the target
audio signal is held, and after recording, acquired based on the held timing information. The
noise occurrence timing detection for the recorded audio signal may be performed in each noise
occurrence predicted interval.
Then, each noise section may be set based on the detected noise generation timing. Further, as
the noise embedding determination, the signal embedding of the recorded audio signal may be
performed only in the noise section set as described above, and the noise embedding
determination may be performed for each noise section. Then, as the noise removal, the signal
processing as described above is performed on the recorded audio signal only for the noise
section determined as the non-embedding noise among the noise sections set as described above.
A removal signal may be generated and inserted as a signal of the noise section.
[0133]
In addition, also as noise removal control performed on the recorded voice signal in this manner,
selection of the voice signal for which noise removal has not been performed and the voice signal
for which noise removal has been performed is performed according to the noise embedding
determination result. There is no change in what is being done.
[0134]
Further, for confirmation, when performing noise removal on the recorded audio signal as
described above, it is not necessary to provide each delay circuit as described above.
10-04-2019
39
[0135]
FIG. 2 is an external perspective view of the recording apparatus of the embodiment.
It is a block diagram shown about an internal configuration of a recording device of an
embodiment.
FIG. 2 is a diagram showing an internal configuration of an HDD provided in the recording
apparatus of the embodiment. It is a figure which shows the characteristic of HDD noise. It is a
figure for demonstrating the method of the noise timing detection / noise area setting as
embodiment. As a diagram schematically showing the relationship between background sound
(external sound) and HDD noise, an audio signal in which the background sound when the
background sound has a relatively small volume, the HDD noise, the HDD noise and the
background sound are added It is the figure which each showed the time waveform of. As a
diagram schematically showing the relationship between background sound (external sound) and
HDD noise, it is a diagram showing the frequency spectra of the background sound and HDD
noise in the case where the background sound has a relatively small volume, in comparison. As a
diagram schematically showing the relationship between background sound (external sound) and
HDD noise, a background sound when the background sound has a relatively large volume, an
audio signal in which the HDD noise, the HDD noise and the background sound are added It is
the figure which each showed the time waveform of. As a diagram schematically showing the
relationship between background sound (external sound) and HDD noise, it is a diagram showing
the frequency spectrum of the background sound and HDD noise in a case where the background
sound is relatively loud, in comparison. It is the figure which showed typically the basic view of
the noise removal method of embodiment. It is the block diagram which showed the internal
configuration of the noise removal signal generation part with which the recording device of an
embodiment is provided. It is a figure showing typically the generation processing of the first half
signal. It is the figure which showed typically the production | generation process of the second
half signal. It is a figure showing typically about cross fade processing. It is a block diagram
shown about an internal configuration of a recording device as a first modification of the
embodiment. Among processing operations to be performed to realize the operation as the
embodiment, it is a flowchart showing a processing operation corresponding to the noise timing
detection / noise interval setting operation. Among the processing operations to be performed to
realize the operation as the embodiment, it is a flowchart showing the processing operation
corresponding to the noise embedding determination operation. Among processing operations to
be performed to realize the operation as the embodiment, it is a flowchart showing processing
operations corresponding to noise removal signal generation, removal operation control,
10-04-2019
40
selection control, and signal insertion operation.
Explanation of sign
[0136]
1, 15 video camera apparatus, 1a camera lens, 1b external sound pickup microphone, 2 noise
detection / removal unit, 3 audio signal processing unit, 4 audio signal compression processing
unit, 5 camera block, 6 video signal processing unit, 7 video Signal compression processing unit,
8 AV stream generation unit, 9 buffer memory, 10 HDD, 10A magnetic head, 10B magnetic disk,
11 noise detection microphone, 12, 16 system controller, 21, 22, 23, 28, 31 delay circuit, 24
noise timing detection / noise interval setting unit 25 noise embedding determination unit 26
noise removal signal generation unit 27 removal operation control unit 29 selection control unit
30 signal insertion unit 32 first half pitch calculation unit 33 second half pitch calculation unit ,
34 first half signal generator, 35 second half signal generator, 36 crossfade processor
10-04-2019
41
Документ
Категория
Без категории
Просмотров
0
Размер файла
62 Кб
Теги
jp2008052772, description
1/--страниц
Пожаловаться на содержимое документа