JP2016152566

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 JP2016152566
An object of the present invention is to provide an apparatus which is suitable for reducing the
occurrence of dip, and which is further suitable for reducing the processing load of the phase
control signal for performing time alignment processing. A phase control signal generation
device for generating a phase control signal for each frequency band of an audio signal
converted to a frequency domain, setting change means capable of setting and changing a
propagation delay time for each predetermined frequency band And difference value acquiring
means for acquiring the difference value before and after the setting change of the propagation
delay time whose setting has been changed, and updating the phase control amount of the
frequency band for which the propagation delay time has been changed based on the acquired
difference value. An update unit and a phase control signal generation unit configured to
generate a phase control signal for each frequency band by performing smoothing processing of
the phase control amount in the frequency domain using the phase control amount after the
update. [Selected figure] Figure 2
Phase control signal generation apparatus, phase control signal generation method and phase
control signal generation program
[0001]
The present invention relates to a phase control signal generation apparatus, a phase control
signal generation method, and a phase control signal generation program for generating a phase
control signal for performing time alignment processing for adjusting propagation delay times of
a plurality of frequency bands.
[0002]
07-05-2019
1
Generally, speakers are installed at a plurality of positions in a vehicle cabin.
For example, the left front speaker and the right front speaker are installed at symmetrical
positions with respect to the center line of the cabin space. However, these speakers are not
symmetrical with respect to the listener's listening position (driver's seat, passenger's seat, rear
seat, etc.). Therefore, due to the difference in distance between the listening position of the
listener and each of the plurality of speakers (difference in time for the reproduced sound
emitted from each speaker to arrive), a bias of sound image localization due to the Hearth effect
occurs.
[0003]
For example, Patent Document 1 describes an apparatus capable of improving the bias of sound
image localization. The device described in Patent Document 1 suppresses the bias of sound
image localization by performing time adjustment (that is, time alignment processing) so that the
reproduced sound emitted from each speaker simultaneously reaches the listening position of the
listener. More specifically, the apparatus described in Patent Document 1 divides an audio signal
into a high band and a low band by a band dividing circuit, and then reproduces the reproduced
sound emitted from the low band and high band speakers. By adjusting the time of each, the
disturbance of the frequency characteristic due to the deviation of the sound image localization
or the phase interference is corrected over the entire band.
[0004]
However, in the device described in Patent Document 1, there is a problem that the linearity of
the transfer characteristic at the listening position of the listener is deteriorated due to the loss of
signal and double addition in the band division circuit. Further, in the device described in Patent
Document 1, there is also a problem that a dip occurs in the frequency characteristic near the
crossover frequency when the signals divided by the band division circuit are mixed.
[0005]
Therefore, Patent Document 2 proposes an apparatus for performing time alignment processing,
which can improve the linearity of the transfer characteristic at the listening position of the
07-05-2019
2
listener and can suppress the occurrence of dip at the mixing time.
[0006]
In the device described in Patent Document 2, a digital filter is used to improve the linearity of
the transfer characteristic at the listening position of the listener.
More specifically, in the device described in Patent Document 2, an FIR (Finite Impulse Response)
filter is used. The FIR filter described in Patent Document 2 is a high-order filter having a steep
cutoff frequency in order to suppress the occurrence of dips, and has a configuration that
requires a large number of delay circuits and multipliers. In particular, in the case of a
configuration having an FIR filter having a linear phase characteristic with constant phase in all
frequency bands, a large number of delay units and multipliers are required, resulting in an
increase in processing load. Further, in the configuration described in Patent Document 2, as the
number of frequency bands to be divided is increased (as the time-adjusted frequency band is
finely divided), the necessary delay circuits and multipliers are further increased, and the
processing load is further increased. There is also the problem of increasing.
[0007]
Therefore, Patent Document 3 describes an apparatus for performing time alignment processing
for adjusting propagation delay times of a plurality of frequency bands, which is configured to
reduce the occurrence of dips, but is suitable for suppressing the processing load. There is.
Specifically, the device described in Patent Document 3 generates a phase control signal for
phase control (phase rotation and phase offset) for each frequency band in an audio signal, and
performs smoothing processing on the generated phase control signal. By connecting the phase
changes smoothly between the frequency bands, it is possible to reduce the occurrence of dips
while having a configuration in which a large number of FIR filters are unnecessary (that is, a
configuration in which the processing load is reduced). .
[0008]
Patent Document 1: Japanese Patent Application Laid-Open No. 7-162985 Patent Document 2:
International Publication 2009/095965 Patent Document 2: Japanese Patent Application LaidOpen No. 2015-12366
07-05-2019
3
[0009]
However, there is a constant need to reduce the processing load for this type of device.
Therefore, while the object of the present invention is a configuration suitable for reducing the
occurrence of dip, a phase control suitable for further suppressing the generation processing
load of the phase control signal for performing time alignment processing. A signal generation
device, a phase control signal generation method, and a phase control signal generation program.
[0010]
The phase control signal generation apparatus according to an embodiment of the present
invention is an apparatus for generating a phase control signal for each frequency band for an
audio signal converted to the frequency domain, and setting and changing the propagation delay
time for each predetermined frequency band Setting change means capable of setting the
difference value acquiring means for acquiring the difference value before and after the setting
change of the propagation delay time whose setting is changed, and the frequency for which the
propagation delay time is changed based on the acquired difference value Phase control signal
generation means for generating a phase control signal for each frequency band by performing
update processing for updating the phase control amount of the band, and performing smoothing
processing of the phase control amount in the frequency domain using the phase control amount
after update Equipped with
[0011]
Further, in the phase control signal generation device according to an embodiment of the present
invention, a weight coefficient holding unit that holds a weight coefficient for each frequency
band, and a weight corresponding to a frequency band whose propagation delay time has been
changed by the setting changing unit. It may be configured to include a weighting factor
acquisition unit that acquires a coefficient from the weighting coefficient holding unit.
In this configuration, the updating means performs phase control of the frequency band for
which the propagation delay time has been changed by the setting changing means based on the
weighting factor acquired by the weighting factor acquisition means and the difference value
07-05-2019
4
acquired by the difference value acquiring means. The amount may be updated.
[0012]
In one embodiment of the present invention, the weighting factor acquired by the weighting
factor acquisition means has, for example, a first value near the center frequency of the
frequency band whose propagation delay time has been changed by the setting change means, A
coefficient having a second value smaller than the first value in a frequency band adjacent to the
frequency band.
[0013]
In one embodiment of the present invention, the weighting factor acquired by the weighting
factor acquiring means is, for example, the propagation delay time set and changed by the setting
changing means so that the phase control amount does not change in adjacent frequency bands.
From the frequency band to the adjacent frequency band, the first value is a value attenuated to a
second value using a rectangular attenuation curve.
[0014]
Also, in an embodiment of the present invention, the frequency band of the phase-controlled
object has, for example, a logarithmic wide bandwidth from low to high.
[0015]
Furthermore, the phase control signal generation device according to an embodiment of the
present invention may be configured to include filter coefficient holding means for holding a
plurality of filter coefficients having different filter orders and cutoff frequencies for each
frequency band.
In this configuration, the phase control signal generation unit may perform the smoothing
process on the phase control amount using different filter coefficients for each frequency band.
[0016]
In one embodiment of the present invention, the filter order and the cutoff frequency may be set
based on the number of frequency spectrum signals in the frequency band.
07-05-2019
5
[0017]
In one embodiment of the present invention, the filter coefficient may be set to a value such that
the smoothing amount by the phase control signal generation means increases as the frequency
band is higher.
[0018]
A phase control signal generation method according to an embodiment of the present invention
is a method of generating a phase control signal for each frequency band for an audio signal
converted to the frequency domain, and propagation delay time for each predetermined
frequency band The propagation delay time setting is changed based on the setting change step
which can be changed in setting, the difference value acquiring step for acquiring the difference
value before and after the setting change of the propagation delay time whose setting is changed,
and the acquired difference value. Phase control signal generation for each frequency band by
performing an update step of updating the phase control amount of the frequency band and
smoothing processing of the phase control amount in the frequency domain using the phase
control amount after the update And a step.
[0019]
Further, a phase control signal generation program according to an embodiment of the present
invention is for causing a computer to execute the above phase control signal generation method.
[0020]
According to one embodiment of the present invention, a phase control signal suitable for
reducing the generation processing load for phase control signals for performing time alignment
processing while having a configuration suitable for reducing the occurrence of dips. A
generator, a phase control signal generation method, and a phase control signal generation
program are provided.
[0021]
It is a block diagram showing composition of an acoustic system concerning one embodiment of
the present invention.
07-05-2019
6
It is a figure which shows the flowchart of the phase control signal generation process performed
in the information processing terminal with which the acoustic system which concerns on one
Embodiment of this invention is equipped.
It is a figure which shows the delay time adjustment screen displayed on the display screen of an
information processing terminal in one Embodiment of this invention.
It is a figure which shows the weighting coefficient referred at the time of execution of the phase
control signal generation process which concerns on one Embodiment of this invention.
It is a figure which shows the filter coefficient referred at the time of execution of the phase
control signal generation process which concerns on one Embodiment of this invention.
It is a figure which shows the phase control signal after the phase smoothing process in, when
predetermined propagation delay time is set in one Embodiment of this invention.
It is a figure which shows an output audio signal in case predetermined ¦ prescribed propagation
delay time is set about the white noise to which the input audio signal was band-limited in one
Embodiment of this invention.
FIG. 8 (a) shows a phase control signal before and after phase smoothing processing in the prior
art (Patent Document 3) and FIG. 8 (b) shows a phase control signal before and after phase
smoothing processing in the present embodiment . FIG. 9 is a frequency characteristic of an
output audio signal when a frequency flat impulse signal is input according to an embodiment of
the present invention, showing a frequency characteristic of the output audio signal when phase
smoothing processing is not performed (FIG. Fig. 9 (b) shows the frequency characteristics of the
output audio signal when a) and phase smoothing processing are performed. FIG. 10 is a diagram
showing an output audio signal in Patent Document 3 (FIG. 10 (a)) and an output audio signal in
the present embodiment for white noise in which the input audio signal is limited to low
frequencies in one embodiment It is a figure (FIG.10 (b)).
[0022]
07-05-2019
7
Hereinafter, embodiments of the present invention will be described with reference to the
drawings. In the following, a sound system will be described by way of example as an
embodiment of the present invention.
[0023]
[Configuration of Sound System 1] FIG. 1 is a block diagram showing the configuration of a
sound system 1 according to an embodiment of the present invention. An audio system 1
according to the present embodiment includes an audio device 10 and an information processing
terminal 20.
[0024]
The audio device 10 is, for example, an on-vehicle audio device mounted on a vehicle, and as
shown in FIG. 1, a short-term Fourier transform (STFT) unit 11, a phase control unit 12 and a
short time inverse A Fourier transform (ISTFT: Inverse Short-Term Fourier Transform) unit 13 is
provided. The audio device 10 adjusts the propagation delay time of the audio signal output to
each speaker (not shown) disposed in the vehicle compartment by cooperating with the
information processing terminal 20 (that is, performs time alignment processing) . As a result, an
audio signal whose propagation delay time has been adjusted (corrected) is output from each
speaker via the power amplifier. Therefore, the user can listen to music and the like in an
environment in which the bias of sound image localization due to the hearth effect is suppressed.
[0025]
An audio signal (for example, an audio signal such as a CD (Compact Disc) or a DVD (Digital
Versatile Disc)) obtained by decoding an encoded signal of a reversible compression format or an
irreversible compression format from the sound source unit (not shown) Is input. The STFT unit
11 performs weighting with the overlap processing and window function on the input audio
signal, then performs conversion from the time domain to the frequency domain by the STFT,
and performs phase control on the real and imaginary frequency spectrum signals. Output to 12.
[0026]
07-05-2019
8
In the present embodiment, an audio signal with a sampling frequency of 96 kHz is input to the
STFT unit 11. The STFT unit 11 has a Fourier transform length of 16,384 samples, an overlap
length of 12,288 samples, and a window function of Hanning. The STFT unit 11 acquires a
16,384-point frequency spectrum signal by performing STFT while time-shifting each 4,096
samples. Here, 8,193 points of frequency spectrum signals up to the Nyquist frequency of 0 Hz
to 48 kHz among 16,384 points are output to the phase control unit 12.
[0027]
The phase control unit 12 controls the phase of the frequency spectrum signal input from the
STFT unit 11 for each frequency band based on the phase control signal (details will be described
later) for each frequency band input from the information processing terminal 20 (phase rotation
And phase offset). The phase control unit 12 outputs to the ISTFT unit 13 a frequency spectrum
signal whose phase is controlled for each frequency band.
[0028]
The ISTFT unit 13 converts the frequency spectrum signal input from the phase control unit 12
from the frequency domain to a time domain signal by the ISTFT, and performs weighting and
overlap addition using a window function on the converted signal. The audio signal obtained
after the overlap addition is a signal whose propagation delay has been corrected according to
the setting change (details will be described later) performed using the information processing
terminal 20, and the circuit from the ISTFT unit 13 to the subsequent stage Output to).
[0029]
The information processing terminal 20 can be brought into a vehicle compartment, for example,
a smartphone, a feature phone, a personal handy phone system (PHS), a tablet terminal, a
notebook PC, a personal digital assistant (PDA), a portable navigation device (PND), a portable
game machine, etc. It is a portable terminal. As shown in FIG. 2, the information processing
terminal 20 includes a setting change receiving unit 21, a weighting factor selecting unit 22, a
normalized delay time generating unit 23, a phase control calculating unit 24, and a phase
smoothing unit 25.
07-05-2019
9
[0030]
[Phase Control Signal Generation Processing] FIG. 2 is a view showing a flowchart of phase
control signal generation processing executed in the information processing terminal 20
according to one embodiment of the present invention. When the information processing
terminal 20 receives a predetermined operation by the user, the display screen displays a delay
time adjustment screen for adjusting the propagation delay time.
[0031]
An example of the delay time adjustment screen displayed on the display screen of the
information processing terminal 20 is shown in FIG. As shown in FIG. 3, the delay time
adjustment screen is an operation screen in which the vertical axis is delay time (Time (unit:
msec)) and the horizontal axis is frequency (Frequency (unit: Hz)), and an audio signal The
propagation delay time for each frequency band is arranged on the horizontal axis and
graphically shown as a bar graph screen. Human auditory characteristics are logarithmic with
respect to frequency. Therefore, the frequency on the horizontal axis is displayed logarithmically
in accordance with human auditory characteristics. Further, the frequency of the horizontal axis
is obtained by dividing the range of 30 Hz to 40 kHz into 32 for each 1⁄3 octave. Therefore, in
the present embodiment, the frequency band of the phase-controlled target has a logarithmically
wide bandwidth from the low band to the high band. By displaying the delay time adjustment
screen illustrated in FIG. 3, the phase control signal generation process shown in FIG. 2 is started.
[0032]
[S11 in FIG. 2 (Change in setting of propagation delay time)] The user wants to change the
setting of propagation delay time by performing, for example, a touch operation on the delay
time adjustment screen displayed on the display screen of the information processing terminal
20 A frequency band can be specified and its delay amount can be input. The propagation delay
time that can be changed is, for example, in the range of ± 5 msec. In addition, on the delay time
adjustment screen, predetermined user auxiliary information may be displayed which assists the
setting change operation of the propagation delay time. As an example of the user auxiliary
information displayed on the delay time adjustment screen, information of the speaker name
07-05-2019
10
currently being changed, its reproducible frequency, frequency information such as an
instrument, vocal etc. For example, an icon of a musical instrument that plays a high-pitched
sound may be superimposed on a high-frequency portion of a bar graph or the like.
[0033]
When the setting change receiving unit 21 receives the setting change operation, the setting
change receiving unit 21 outputs the information of the frequency band designated by the
operation to the weighting coefficient selecting unit 22 and normalizes the information of the
propagation delay time designated by the operation. It is output to the time generation unit 23.
[0034]
The setting change of the propagation delay time is not limited to the manual operation on the
delay time adjustment screen, and may be automatically performed.
As an example, a microphone is installed at the listening position of the listener (driver's seat,
passenger's seat, rear seat, etc.). In this case, the acoustic characteristics of the in-vehicle space
are measured using the microphone installed at the listening position of the listener, and the
setting change value of the propagation delay time for each frequency band is automatically
calculated based on the measured result. The setting change accepting unit 21 outputs the
information of the calculated frequency band to the weighting coefficient selecting unit 22 and
outputs the information of the calculated propagation delay time to the normalized delay time
generating unit 23.
[0035]
[S12 (Selection of Weighting Coefficient) in FIG. 2] The weighting coefficient selecting unit 22
holds the weighting coefficient for each frequency band calculated in advance in a predetermined
memory area. When information on the frequency band designated by the setting change
operation in processing step S11 (setting change of the propagation delay time) is input from
setting change accepting unit 21, weighting factor selection unit 22 selects the frequency band
indicated by the information. The corresponding weighting factor is read out from the
predetermined memory area, and is output to the normalized delay time generation unit 23
together with the information on the frequency band.
07-05-2019
11
[0036]
FIGS. 4A and 4B illustrate weight coefficients held in a predetermined memory area. In each of
FIGS. 4A and 4B, the vertical axis represents a weighting factor, and the horizontal axis
represents the number of phase control signals for each band updated by the phase control unit
12 of the audio apparatus 10. Show. The number of phase control signals is "31" in FIG. 4A and
"2560" in FIG. 4B. As shown in FIGS. 4 (a) and 4 (b), the weighting factor has a value of 1 near
the center frequency of the frequency band and does not affect the adjacent frequency band
(adjacent And the attenuation curve is attenuated to be zero using a rectangular attenuation
curve so that the amount of phase control does not substantially change in the frequency band of
interest. The weighting factor illustrated in FIG. 4A has a central frequency band of 400 Hz and
adjacent frequency bands of 315 Hz and 500 Hz. The weighting factor illustrated in FIG. 4B has a
center frequency band of 31.5 kHz and adjacent frequency bands of 25 kHz and 40 kHz.
[0037]
[S13 (updating of normalized delay time) in FIG. 2] The normalized delay time generation unit 23
operates on the frequency band designated by the setting change operation at processing step
S11 (change of setting of propagation delay time) by the operation. The difference value between
the designated propagation delay time and the current delay time is calculated as the difference
delay time. The normalized delay time generation unit 23 multiplies the calculated differential
delay time by the weighting factor input from the weighting factor selection unit 22, adds the
product to the current delay time, and further, the frequency band In order to control only the
propagation delay time while maintaining the phase of the signal, the resolution of the
propagation delay time is converted to be the reciprocal of the sampling frequency. As a result,
the normalized delay time for each frequency band is updated (more accurately, only the
normalized delay time for the frequency band specified by the setting change operation in
processing step S11 (change in setting of propagation delay time) is current. Changed for the
value).
[0038]
[S14 in FIG. 2 (update of phase control signal)] The phase control calculation unit 24 performs
setting change operation at the updated normalized delay time and the frequency corresponding
to this (processing step S11 (change of setting of propagation delay time)). Update the phase
07-05-2019
12
control amount for each frequency band by multiplying the frequency contained in the frequency
band specified by (more accurately, only the phase control amount for the frequency band
specified by the setting change operation Of the value of V.sub.2) and output to the phase
smoothing unit 25 as an updated phase control signal. Here, phase control is to control the
amount of rotation of the phase of the frequency spectrum signal. Controlling the amount of
phase rotation is equivalent to controlling the propagation delay time in the time domain. In
addition, a phase offset corresponding to the frequency is given to the phase rotation for each
frequency band.
[0039]
In Patent Document 3, when the propagation delay time of a part of frequency bands is changed,
the phase control signal is generated again over the whole frequency band, but in this
embodiment, the phase band of the changed frequency band is changed. Only the phase control
signal is generated (updated) again. Therefore, in the present embodiment, the processing load at
the time of generating the phase control signal can be further suppressed.
[0040]
[S15 (phase smoothing) in FIG. 2] The phase smoothing unit 25 receives the phase input from
the phase control calculation unit 24 for the frequency band designated by the setting change
operation in processing step S11 (change in setting of propagation delay time). The control signal
is smoothed by integration processing using an FIR low pass filter. As a result, the change of the
phase between the frequency bands having different propagation delay times becomes smooth,
and the disturbance of the frequency characteristic (the occurrence of dip) due to the phase
interference between the frequency bands can be suppressed. The phase smoothing unit 25
outputs to the phase control unit 12 a phase control signal in which the disturbance of the
frequency characteristic is suppressed.
[0041]
In Patent Document 3, when the propagation delay time of a part of the frequency band is
changed, phase smoothing processing is performed over the entire frequency band, but in this
embodiment, the phase is changed only in the changed frequency band. Smoothing processing is
performed. Also in this respect, in the present embodiment, the processing load at the time of
07-05-2019
13
phase control signal generation can be further suppressed.
[0042]
For example, consider the case where the entire frequency band ranging from the low band to
the high band is divided into 1/3 octaves. In this case, in the low band of several hundreds Hz or
less in which the number (the number of points) of frequency spectrum signals per frequency
band is relatively small, the amount of smoothing by integration processing becomes relatively
large compared to the high band. Therefore, in the low band, the difference between the phase
control signal before the phase smoothing process and the phase control signal after the phase
smoothing process becomes large.
[0043]
Therefore, in the present embodiment, the filter coefficients of the FIR low pass filter are
calculated in advance for each frequency band and held in a predetermined memory area. The
filter coefficients have different values for the filter order and the cutoff frequency for each
frequency band. The phase smoothing unit 25 reads the filter coefficient from a predetermined
memory area, and performs smoothing processing of the phase control signal by integration
processing.
[0044]
The following shows the filter order and the normalized cutoff frequency with respect to the
number of points P (the number of frequency spectrum signals in the frequency band). -Number
of points P <20 filter order: No setting Normalized cutoff frequency: No setting-20 <number of
points P <40 filter order: 4 normalized cutoff frequency: 1/4-40 <number of points P <80 filter
order : 8 normalized cutoff frequency: 1/8 · 80 ≦ point number P <160 filter order: 16
normalized cutoff frequency: 1/16 · 160 ≦ point number P filter order: 32 normalized cutoff
frequency: 1/1 32
[0045]
07-05-2019
14
In the above example, the smoothing process is not performed in the frequency band where the
number of points P is less than 20, and in the frequency band where the number of points P is
20 or more, the filtering amount increases as the frequency band where the number of points P
increases. The order is long and the normalized cutoff frequency is set small. Further, in the
frequency band where the number of points P is 160 or more, the filter order and the cutoff
frequency are fixed values so as to maintain the smoothing amount. Illustratively, the frequency
band where the number of points P is less than 20 corresponds to the frequency band less than
250 Hz, and the frequency band where the number of points P is 160 or more corresponds to
the frequency band of 2.5 kHz or more.
[0046]
As an example, FIG. 5 (a) shows filter coefficients corresponding to a frequency band where the
number of points P is 40 or more and less than 80, and FIG. 5 (b) shows filter coefficients
corresponding to a frequency band where the number of points P is 160 or more. Indicates In
each of FIGS. 5 (a) and 5 (b), the vertical axis is the amplitude, and the horizontal axis is the
number of samples (filter order).
[0047]
As described above, in the present embodiment, by providing the optimal filter coefficient for
each frequency band, the phase control signal is smoothed by the optimal amount according to
the number of frequency spectrum signals in the frequency band. Thereby, in the low band, the
setting accuracy of the delay time is improved, and in the middle and high bands, the dip due to
the interference between the bands is reduced while the setting accuracy of the delay time is
maintained.
[0048]
[More detailed illustration] In the delay time adjustment screen shown in FIG. 3, phase smoothing
processing when the propagation delay time in the 4 kHz band is set to 0 msec (that is, when the
propagation delay time is not changed at all) The subsequent phase control signal is illustrated in
FIG. 6A, and the phase control signal after phase smoothing processing in the case where the
propagation delay time in the 4 kHz band is set to 1 msec is illustrated in FIG. In each of FIGS. 6
(a) and 6 (b), the vertical axis is phase (Phase (unit: degree)), and the horizontal axis is frequency
07-05-2019
15
(Frequency (unit: Hz)). In addition, for white noise whose input audio signal is band-limited in the
4 kHz band, the output audio signal when the propagation delay time is set to 0 msec and 1 msec
is shown in FIGS. 7A and 7B, respectively. I will illustrate. In each of FIGS. 7 (a) and 7 (b), the
vertical axis represents amplitude (Amplitude), and the horizontal axis represents time (Time
(unit: msec)). The phase control signal has a phase angle limited to ± 180 degrees.
[0049]
6 and 7, it can be seen that the output audio signal is delayed according to the propagation delay
time set on the delay time adjustment screen.
[0050]
The phase control signal before and behind the phase smoothing process in patent document 3 is
illustrated to FIG. 8 (a), and the phase control signal before and behind the phase smoothing
process in this embodiment is illustrated to FIG. 8 (b).
In each of FIG. 8A and FIG. 8B, the vertical axis is a phase (Phase (unit: degree)), and the
horizontal axis is a frequency (Frequency (unit: Hz)). In each of FIGS. 8A and 8B, a thin solid line
indicates a phase control signal before phase smoothing processing, and a thick solid line
indicates a phase control signal after phase smoothing processing.
[0051]
In Patent Document 3, since the phase smoothing process is performed using the same filter
coefficient over the entire frequency band, the amount of smoothing by the integration process
becomes relatively larger as the number of points P decreases. As a result, as shown in FIG. 8A, a
difference occurs between the phase control signals before and after the phase smoothing
process (in particular, a low band of several hundred Hz or less). On the other hand, in the
present embodiment, the optimum filter coefficient is given to each frequency band, whereby the
phase control signal is smoothed by the optimum amount according to the number of frequency
spectrum signals in the frequency band. As a result, as shown in FIG. 8B, even in the low
frequency band of several hundred Hz or less with a small number of points P, almost no
difference occurs in the phase control signal before and after the phase smoothing process. It can
be seen that it is improved.
07-05-2019
16
[0052]
FIG. 9 illustrates frequency characteristics of an output audio signal when a frequency flat
impulse signal is input. FIG. 9A shows the frequency characteristics of the output audio signal
when the phase smoothing process of the present embodiment is not performed, and FIG. 9B
shows the output audio when the phase smoothing process of the present embodiment is
performed. Indicates the frequency characteristics of the signal. In each of FIGS. 9A and 9B, the
vertical axis represents a signal level (Level (unit: dB)), and the horizontal axis represents a
frequency (Frequency (unit: Hz)).
[0053]
As compared with FIGS. 9 (a) and 9 (b), dips are significantly improved in all frequency bands by
performing the phase smoothing process, and while setting accuracy of the propagation delay
time is maintained, between the frequency bands. It can be seen that the interference is reduced.
[0054]
For low frequency white noise in which the input audio signal is band-limited to the 0 Hz to 63
Hz band, the output audio signal in Patent Document 3 is illustrated in FIG. 10A, and the output
audio signal in this embodiment is FIG. To illustrate.
In each of FIGS. 10 (a) and 10 (b), thin solid lines indicate output audio signals when the
propagation delay time is set to 0 msec, and thick solid lines indicate that the propagation delay
time is set to 5 msec. Indicates the output audio signal when In each of FIGS. 10 (a) and 10 (b),
the vertical axis is amplitude (Amplitude), and the horizontal axis is time (Time (unit: msec)).
[0055]
According to the present embodiment, the output audio signal is delayed according to the
propagation delay time which has been changed while maintaining the amplitude (see FIG. 10
(b)), as disclosed in Patent Document 3 (see FIG. 10 (a)). It can be seen that it has been improved.
[0056]
07-05-2019
17
The above is a description of an exemplary embodiment of the present invention.
Embodiments of the present invention are not limited to those described above, and various
modifications are possible within the scope of the technical idea of the present invention. For
example, the contents of the embodiment of the present application also include the contents in
which the embodiment etc. or the obvious embodiment etc. which are clearly illustrated in the
specification are appropriately combined.
[0057]
As an example, in the above-mentioned embodiment, although various elements which constitute
audio system 1 are divided into audio device 10 and information processing terminal 20 and
provided, the present invention is not limited to this. In another embodiment, all of the various
elements constituting the sound system 1 may be provided in the audio device 10, or all may be
provided in the information processing terminal 20.
[0058]
In the above embodiment, the phase smoothing process using the corresponding filter order and
the normalized cutoff frequency is performed only for the frequency band designated by the
setting change operation in the processing step S11 (setting change of the propagation delay
time). Although implemented, the present invention is not limited thereto. In another
embodiment, phase smoothing processing may be performed using the corresponding filter order
and normalized cutoff frequency not only for the frequency band designated by the operation but
also for all frequency bands.
[0059]
DESCRIPTION OF SYMBOLS 1 audio system 10 audio apparatus 11 S TFT unit 12 phase control
unit 13 IS TFT unit 20 information processing terminal 21 setting change reception unit 22
weight coefficient selection unit 23 normalized delay time generation unit 24 phase control
calculation unit 25 phase smoothing unit
07-05-2019
18