close

Вход

Забыли?

вход по аккаунту

JP2018050222

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 JP2018050222
Abstract: The present invention provides a filter generation device, a filter generation method,
and a program for generating a filter for a high resolution digital audio signal. A filter generation
device according to the present embodiment is a filter generation device that generates a filter
for performing an out-of-head localization process on a high resolution digital audio signal, and
includes left and right speakers 5L and 5R. And a filter generation unit 351 that generates a filter
according to the transfer characteristic from the left and right speakers 5L, 5R to the left and
right microphones 2L, 2R. A predetermined frequency not exceeding the Nyquist frequency of
the collected signal is taken as a first frequency (24 kHz). The filter generation unit 351 sets the
amplitude component of the low frequency band BH of the filter according to the frequency
amplitude characteristic of the collected signal, and combines the amplitude component of the
high frequency band BH of the filter with the amplitude component of the low frequency band
BL. Generate to [Selected figure] Figure 1
Filter generation device, filter generation method, and program
[0001]
The present invention relates to a filter generation device, a filter generation method, and a
program.
[0002]
Conventionally, as a method of localizing a sound image outside the head, Patent Document 1
discloses a method of using a head related transfer function (HRTF) of a listener (see, for
10-05-2019
1
example, Patent Document 1).
When headphones or earphones are used, the inverse characteristic of the head transfer function
from the virtual sound source to both ears of the listener and the ear canal transfer function
ECTF (Ear Canal Transfer function) is convoluted into the reproduction signal. By doing this, it is
possible to cancel the characteristics of the headphones or the earphones and reproduce the
sound field as if the sound is being heard from the virtual sound source direction despite the
sound coming from the vicinity of the ear. Measurement of the ear canal transfer function ECTF
is performed by inserting a measurement microphone into the ear canal of the listener or by
using a dummy head.
[0003]
JP 2002-209300 A
[0004]
However, while the ideal acoustic image localization is in a state where the ear canal is open, in
actual measurement, the ear canal is in a closed state because measurement is performed with
headphones or earphones worn.
As a result, resonance occurs in the ear canal and a peak or dip occurs at a specific frequency. As
a result of convoluting the inverse characteristic of the ear canal transfer function (also referred
to as an ear canal correction function) into the reproduction signal, the sound quality on hearing
may deteriorate. In addition, even in the case of three-dimensional sound field reproduction by a
speaker using a head transfer function, resonance may occur due to the influence of the
reflection of the measurement space or the like, resulting in deterioration of the sound quality. It
is difficult to detect a peak or dip generated by resonance from the waveform of the reproduction
signal. Therefore, there is a possibility that the sound field can not be properly reproduced.
[0005]
The present invention has been made in view of the above, and it is an object of the present
invention to provide a filter generation device, a filter generation method, and a program for
generating a filter for a high resolution digital audio signal.
10-05-2019
2
[0006]
A filter generation device according to an aspect of the present invention is a filter generation
device that generates a filter for performing an out-of-head localization process on a high
resolution digital audio signal, and includes left and right speakers, and left and right speakers.
The right and left microphones are attachable to the ear, pick up the measurement signals output
from the left and right speakers, and obtain the collected signals, and obtain the left and right
from the left and right speakers based on the collected signals. And a filter generation unit for
generating a filter according to the transfer characteristic to the microphone, wherein the sound
collection signal is a signal of a predetermined sampling frequency, and a first frequency not
exceeding the Nyquist frequency of the sound collection signal is selected. The transfer
characteristic is an amplitude component of a low frequency band including a frequency below
the first frequency, and a high frequency including a frequency higher than the first frequency.
And the amplitude component of the low frequency band of the filter is set according to the
frequency amplitude characteristic of the collected signal, and the amplitude component of the
high frequency band of the filter is included. To be connected to the amplitude component of the
low frequency band.
[0007]
A filter generation method according to an aspect of the present invention is a filter generation
method for generating a filter for performing an out-of-head localization process on a high
resolution digital audio signal, and outputs measurement signals from left and right speakers.
Collecting the measurement signal using the left and right microphones attachable to the left and
right ears of the listener to obtain a sound collection signal, and based on the sound collection
signal, the left and right speakers Generating a filter according to the transfer characteristic from
the microphone to the left and right microphones, wherein the sound collection signal is a signal
of a predetermined sampling frequency, and the predetermined frequency not exceeding the
Nyquist frequency of the sound collection signal A first frequency, and the transfer characteristic
may be an amplitude component of a low frequency band including a frequency lower than the
first frequency, and a high frequency including a frequency higher than the first frequency. And,
in the step of generating the filter, the amplitude component of the low frequency band of the
filter is set according to the frequency amplitude characteristic of the collected signal, and the
high frequency band of the filter is generated. Is generated so as to be joined to the amplitude
component of the low frequency band.
[0008]
A filter generation method according to an aspect of the present invention is a program that
causes a computer to execute the above-described filter generation method.
10-05-2019
3
[0009]
According to the present invention, it is possible to provide a filter generation device, a filter
generation method, and a program for generating a filter for a high resolution digital audio
signal.
[0010]
FIG. 1 is a block diagram showing a sound field reproduction device according to a first
embodiment.
It is a figure which shows an isometric curve.
3 is a flowchart showing a sound field reproduction method according to Embodiment 1;
It is a figure for demonstrating sweep operation ¦ movement and the change of a frequency
characteristic.
It is a figure for demonstrating sweep operation ¦ movement and the change of a frequency
characteristic.
It is a figure for demonstrating sweep operation ¦ movement and the change of a frequency
characteristic. It is a figure for demonstrating sweep operation ¦ movement and the change of a
frequency characteristic. FIG. 7 is a block diagram showing a sound field reproduction device
according to a second embodiment. FIG. 18 is a block diagram showing an out-of-head
localization processing device according to a third embodiment. FIG. 17 is a diagram showing a
configuration of a filter generation device that generates a filter according to transfer
characteristics in a third embodiment. It is a figure which shows the transfer characteristic Hls in
a frequency domain. It is a figure which shows the frequency characteristic of the speaker
corresponding to HR signal. It is a figure which shows the transfer characteristic Hls calculated ¦
required also by simulation. It is a figure for demonstrating the process which carries out the
level adjustment of the component of a high frequency band. It is a figure for demonstrating the
process which smooths an amplitude value. FIG. 17 is a block diagram showing an out-of-head
localization processing device that performs an auditory sense test in a third embodiment. FIG.
16 is a block diagram showing an out-of-head localization processing apparatus according to a
10-05-2019
4
fourth embodiment. It is a figure which shows the frequency characteristic of HPF and LPF. FIG.
13 is a block diagram showing an out-of-head localization processing apparatus according to a
third modification;
[0011]
The outline of the sound field reproduction apparatus according to the present embodiment will
be described. The sound field reproducing apparatus according to the present embodiment
measures head transfer characteristics (also referred to as head transfer functions) or ear canal
transfer characteristics (also referred to as ear canal transfer functions) of an individual, and uses
these characteristics to perform out-of-head localization etc. It realizes sound field reproduction.
Specifically, in the sound field reproducing apparatus, the sound quality deterioration is
improved by eliminating the influence of the reflection caused in space during the measurement
of the transfer characteristic and the influence of the resonance generated due to the occlusion
of the ear canal.
[0012]
In the present embodiment, sound field processing such as localization outside the head is
realized using head-to-head transmission characteristics from the speaker to the listener's ear, or
external ear-canal transmission characteristics in a state of wearing headphones or earphones. In
space transfer characteristics such as head transfer characteristics or ear canal transfer
characteristics, resonance may occur due to the influence of reflection or the like in the
measurement space, and peaks or dips may occur in the high region. Further, also in
measurement of the ear canal transmission characteristics, resonance may occur by measuring in
a state where the ear canal is closed, and a peak or dip may occur in a high region. This is an
individual characteristic that differs among individuals, which can only be perceived by the
listener, and it is difficult to make correction automatically.
[0013]
Therefore, a frequency sweep signal (sweep signal) whose frequency gradually changes is used.
While listening to the frequency sweep signal, the listener operates a button or the like at a
portion where the volume has felt greatly changed. By doing this, it is possible to specify the
position (frequency) of the peak and the dip. A filter such as a notch filter or peaking filter is
10-05-2019
5
applied around the peak or dip frequency. By doing this, it is possible to remove unnecessary
resonance and correct it to a flat frequency characteristic.
[0014]
After the position (frequency) of the peak or dip is specified by the above operation, the
frequency around it may be repeatedly swept. Further, the listener can adjust the peak level of
the filter so that the sound volume is constant, thereby enabling more detailed correction.
[0015]
In order to prevent a sudden change in volume, it is preferable to apply AGC (automatic gain
control) in the out-of-head localization process.
[0016]
Embodiment 1
A sound field reproduction apparatus 100 according to the present embodiment is shown in FIG.
FIG. 1 is a block diagram of the sound field reproduction apparatus 100. As shown in FIG. The
sound field reproduction device 100 reproduces the sound field with respect to the listener U
wearing the headphones 19. Therefore, the sound field reproduction device 100 includes a
sweep signal generation unit 11, a music signal reproduction unit 12, an out-of-head localization
processing unit 13, an AGC (Auto Gain Control) processing unit 14, and a variable filter unit
(filter unit) 15. A filter coefficient calculation unit 16, a setting storage unit 17, an input unit 18,
and a headphone (output unit) 19 are provided. The AGC processing unit 14 can be omitted.
Further, the AGC processing unit 14 can perform AGC processing only at the time of sweep signal
reproduction or music signal reproduction.
[0017]
The sound field reproducing apparatus 100 according to the present embodiment is an
information processing apparatus such as a personal computer, processing means such as a
processor, storage means such as a memory or a hard disk, liquid crystal monitor such as an
10-05-2019
6
organic EL display or plasma display And an input unit such as a touch panel, a button, a
keyboard, and a mouse, and an output unit connected to a speaker or a headphone. Alternatively,
the sound field reproduction device 100 may be a smart phone or a tablet PC. In addition, the
sound field reproduction apparatus 100 incorporates processing means such as a processor and
storage means such as a memory in speakers and headphones as output means, and display
means such as a liquid crystal monitor and inputs such as a touch panel in the headphones etc. It
is good also as composition which can connect with means.
[0018]
The sweep signal generator 11 generates a frequency sweep signal whose frequency changes.
The sweep signal generation unit 11 outputs, as a frequency sweep signal, a sine wave which
sweeps a preset sweep range gradually. The frequency sweep signal is, for example, a pure tone,
and is a signal whose center frequency gradually changes. The sweep signal generation unit 11
outputs the frequency sweep signal to the out-of-head localization processing unit 13. The
frequency sweep signal is subjected to processing described later and output from the
headphone 19. The frequency sweep signal increases in frequency at a constant speed. In
addition, the frequency sweep signal may be continuously increased in frequency or may be
gradually increased in frequency. Alternatively, the frequency may be gradually lowered. The
frequency sweep signal may be a stereo signal.
[0019]
The music signal reproduction unit 12 reproduces music signals recorded in advance on a
memory and a disc. The music signal reproduction unit 12 may not be provided inside the sound
field reproduction apparatus 100, and a music signal from an external sound source may be
input to the out-of-head localization processing unit 13. For example, the music signal may be a
stereo signal output from an external CD player or the like. The music signal is finally output
from the headphones 19 after being subjected to a filtering process described later. The music
signal is, for example, a stereo reproduction signal obtained by digitizing a performance by an
instrument or a natural voice, and is output from the right and left units of the headphone 19.
The music signal may be a signal obtained by digitizing a sound that a person can hear by
hearing, such as a conversation, a cry of an animal, a ripple, etc., as well as a signal obtained by
digitizing a performance by an instrument or a natural voice.
[0020]
10-05-2019
7
During music reproduction, the music signal reproduction unit 12 outputs the music signal to the
out-of-head localization processing unit 13 and the sweep signal generation unit 11 does not
generate a frequency sweep signal. On the other hand, when measuring the filter coefficient for
sound quality adjustment, the sweep signal generation unit 11 outputs the frequency sweep
signal to the out-of-head localization processing unit 13 and the music signal reproduction unit
12 does not reproduce the music signal. That is, a frequency sweep signal is output from the
headphones 19 to perform measurement for eliminating resonance due to personal
characteristics such as the shape of the ear canal. Thus, either the music signal or the frequency
sweep signal is input to the out-of-head localization processing unit 13. The process of
outputting a frequency sweep signal will be mainly described below to measure the filter
coefficient.
[0021]
The out-of-head localization processing unit 13 performs convolution processing on the
frequency sweep signal using the ear canal transmission characteristic. Specifically, an inverse
characteristic (also referred to as an ear canal correction function) of an ear canal transmission
characteristic measured in advance is convoluted into a frequency sweep signal. The out-of-head
localization processing unit 13 outputs the frequency sweep signal subjected to the convolution
processing to the AGC processing unit 14. Also, as described later, the out-of-head localization
processing unit 13 convolves the music signal with the inverse characteristic of the ear canal
transmission characteristic.
[0022]
The AGC processing unit 14 performs processing to keep the signal level (roundness level)
representing the auditory strength of the sound of the frequency sweep signal constant. Here,
even if the sound with a high frequency and the sound with a low frequency have the same
sound pressure, a difference occurs in the auditory strength of the sound that human hearing
senses. An equal sensitivity curve (loudness curve) representing this characteristic is shown in
FIG. The horizontal axis of FIG. 2 is frequency (Hz), and the vertical axis is sound pressure level
(dB). Each curve shows the relationship between the sound pressure level and the frequency for
each signal level representing the auditory strength of the sound. For example, when it is desired
to keep the auditory strength of the sound constant at 60 phon, it can be understood from this
figure that the sound pressure level needs to be varied according to the frequency. Therefore,
10-05-2019
8
when performing the out-of-head localization process on the frequency sweep signal, the AGC
processing unit 14 adjusts the gain in accordance with the equal sensitivity curve. The gain in the
AGC processing unit 14 changes in accordance with the volume, that is, the sound pressure level,
or the frequency. When the AGC processing unit 14 performs AGC (automatic gain control)
processing, the gain is adjusted so that the frequency sweep signal has a constant signal level.
Thus, the listener U can listen to the sound at a constant signal level regardless of the frequency
of the sound. The frequency sweep signal AGC-processed by the AGC processing unit 14 is
output to the variable filter unit 15.
[0023]
The variable filter unit 15 reads the filter coefficients calculated by the filter coefficient
calculation unit 16 and sets filters such as notch filters and peaking filters. The variable filter unit
15 filters the frequency sweep signal using the set filter. In the initial state, a filter with flat
characteristics is set in the filter coefficient calculation unit 16. Therefore, the frequency sweep
signal from the AGC processing unit 14 is output to the headphone 19 as it is.
[0024]
Here, the variable filter unit 15 outputs the frequency sweep signal to the headphone 19 as it is.
The headphones 19 output a frequency sweep signal toward the listener U. The headphones 19
are stereo headphones and output frequency sweep signals to the left and right ears of the
listener U respectively. The listener U listens to the frequency sweep signal output from the
headphones 19.
[0025]
The listener U checks whether the volume changes rapidly while listening to the frequency sweep
signal subjected to the out-of-head localization processing. The range of frequencies to be swept
is preset. In the measurement of the ear canal transfer function, resonance occurs in the high
region, so the sweep range for sweeping the frequency sweep signal is 8 kHz to 20 kHz. Of
course, the sweep range is not limited to 8 to 20 kHz. For example, the sweep range may be 5
kHz to 20 kHz. Further, the sweep range is desirably set arbitrarily for each measurement
environment because the frequency at which peak / dip is likely to occur is different depending
on the measurement environment or the like. Of course, the entire reproduction frequency range
10-05-2019
9
of the headphones 19 may be set as the sweep range. Also, the listener U may designate the
sweep range.
[0026]
If the volume changes rapidly while the listener U listens to the frequency sweep signal, the input
unit 18 is operated. The input unit 18 includes, for example, an input device such as a touch
panel, a keyboard, a mouse, a push button, a lever, or a dial. For example, when the listener U
confirms an abrupt change in volume while listening to the frequency sweep signal, the listener U
presses a frequency determination button provided on the input unit 18. Then, the input unit 18
receives a button operation by the listener U and outputs a signal corresponding to the operation
to the setting storage unit 17.
[0027]
The setting storage unit 17 receives the frequency of the current sweep from the sweep signal
generation unit 11. The setting storage unit 17 includes a memory and the like, and stores the
frequency of the frequency sweep signal when the frequency determination button is pressed.
That is, the setting storage unit 17 stores the frequency at which the volume has changed
rapidly. For example, the setting storage unit 17 stores, as a notch frequency, a frequency at
which the volume sharply decreases. Alternatively, the setting storage unit 17 stores, as a peak
frequency, the frequency at which the volume has sharply increased. The setting storage unit 17
stores a frequency at which the volume of the frequency sweep signal changes in response to an
operation from the listener U who listens to the frequency sweep signal output from the
headphones 19.
[0028]
Then, the setting storage unit 17 outputs the stored frequency to the sweep signal generation
unit 11. Then, the sweep signal generation unit 11 generates a frequency sweep signal that
slowly sweeps the vicinity of the input frequency. That is, the sweep signal generation unit 11
slowly changes the frequency of the frequency sweep signal in the vicinity of the notch
frequency or the peak frequency. The listener U listens to the frequency sweep signal. Then, the
input unit 18 is operated so that the volume is constant.
10-05-2019
10
[0029]
For example, the input unit 18 includes a lever, a dial, and the like for adjusting the volume. The
listener U can operate the input unit 18 to adjust the volume of the sound output from the
headphone 19. The setting storage unit 17 stores the volume adjusted by the listener U. When
sweeping the frequency in the vicinity of the frequency stored in the setting storage unit 17, the
input unit 18 receives an operation of volume adjustment by the listener U.
[0030]
The setting storage unit 17 stores the volume in association with the frequency. That is, the peak
or dip frequency is associated with the volume adjusted at that frequency. The filter coefficient
calculation unit 16 calculates a filter coefficient based on the frequency and the volume stored in
the setting storage unit 17. The filter coefficient calculation unit 16 calculates the filter
coefficient in real time using the frequency and the volume that have already been determined.
[0031]
The filter coefficients calculated in real time by the filter coefficient calculation unit 16 are set in
the variable filter unit 15. As a result, the characteristic of the variable filter, which is flat in the
initial state, changes. The filter coefficient calculation unit 16 applies a filter coefficient to the
frequency sweep signal subjected to the localization processing outside the head. By doing this,
the peak level of the frequency sweep signal subjected to the out-of-head localization process
fluctuates.
[0032]
Then, the listener U operates the input unit 18 when it is determined that the frequency sweep
signal aurally becomes a constant level by the volume operation. For example, when the listener
U reaches a certain level, the adjustment completion button is pressed. By this, the peak level at
which the volume becomes constant is determined. The setting storage unit 17 stores the filter
coefficient and the volume as level information when the adjustment completion button is
pressed. The filter coefficient calculation unit 16 calculates a final filter coefficient according to
10-05-2019
11
the level information. The filter coefficient calculation unit 16 calculates the filter coefficient
based on the level information when the sound volume is adjusted so that the sound volume
becomes constant in the vicinity of the frequency stored in the setting storage unit 17.
[0033]
The final filter coefficients calculated from the frequency and level information are set in the
variable filter unit 15. Thus, the measurement for eliminating the resonance due to the personal
property is completed. When the measurement is completed, the input to the out-of-head
localization processing unit 13 is switched from the sweep signal to the music signal. By this, the
normal music reproduction mode is set, and sound field reproduction using music signals
becomes possible. That is, out-of-head localization processing in the out-of-head localization
processing unit 13 and filtering processing in the variable filter unit 15 are performed on the
music signal.
[0034]
The out-of-head localization processing unit 13 performs convolution on the music signal using
the ear canal correction function. The variable filter unit 15 filters the music signal subjected to
the convolution processing using the above-described frequency sweep signal, and outputs the
filtered signal to the headphone 19. At the time of music reproduction, AGC processing in AGC
processing unit 14 is not performed. In addition, at the time of measurement using the frequency
sweep signal, the out-of-head localization processing unit 13 may not perform the convolution
process on the frequency sweep signal.
[0035]
Thus, the listener U is made to listen to the frequency sweep signal to specify the peak frequency
or dip frequency. In this way, it is possible to eliminate the resonance according to the personal
characteristics of the listener U. Furthermore, filter coefficients for correcting peaks or dips
caused by the measurement environment and the like are set. Therefore, the sound field can be
properly reproduced.
[0036]
10-05-2019
12
Next, sound quality adjustment in the sound field reproducing method according to the present
embodiment will be described with reference to FIGS. 3 to 7. FIG. 3 is a flowchart showing sound
quality adjustment in the sound field reproduction method. 4 to 7 are graphs showing changes in
frequency characteristics in the adjustment operation. In FIGS. 4 to 7, the horizontal axis is the
frequency, and the vertical axis is the volume heard by the listener U.
[0037]
When the measurement is started, the frequency of the frequency sweep signal is swept (S1) to
check the center frequency of the peak and the dip. Here, the sweep signal generation unit 11
sweeps a sweep range of 8 kHz to 20 kHz as shown in FIG. Here, only the high frequency side of
the listening range to which the listener U can listen is swept. While listening to the frequency
sweep signal, the listener U determines whether or not a sudden volume difference is felt (S2).
That is, it is determined whether the listener U feels a volume difference in the frequency sweep
signal output at a constant level. If the sudden volume difference is not felt (NO at S2), the
frequency is continuously swept.
[0038]
When a sudden volume difference is felt (YES in S2), the listener U presses the frequency
determination button of the input unit 18 (S3). That is, the listener U presses the frequency
determination button at the timing when the volume becomes maximum or minimum. Then, the
setting storage unit 17 stores the frequency at the time when the frequency determination
button is pressed (S4). As shown in FIG. 5, the frequency when the frequency determination
button is pressed is determined as the center frequency of the peak or dip.
[0039]
Next, it repeatedly sweeps slowly around the stored frequency (S5). That is, as shown in FIG. 6,
the vicinity of the stored frequency is set as the level adjustment range. The sweep signal
generator 11 outputs a frequency sweep signal for sweeping the level adjustment range
including the center frequency. The sweep speed in the level adjustment range is slower than the
sweep speed in S1. That is, the sweep signal generation unit 11 sweeps the level adjustment
10-05-2019
13
range more slowly than the sweep in the sweep range. Thus, a part of the sweep range of 8 to 20
kHz is extracted and swept slowly as a level adjustment range.
[0040]
Then, while sweeping the level adjustment range slowly, the listener U operates the volume (S6).
At a frequency at which the dip is present, the volume heard to the listener U decreases.
Therefore, as shown in FIG. 6, the listener U raises the volume so that the volume that can be
heard becomes constant. On the other hand, when there is a peak in the frequency characteristic,
the listener U reduces the volume at the center frequency in order to make the volume constant.
The listener U adjusts the volume level while listening to the frequency sweep signal. By doing
this, it is possible to adjust the volume that the listener U listens to at the center frequency.
[0041]
The setting storage unit 17 stores the operated volume, and the filter coefficient calculation unit
16 calculates the filter coefficient based on the stored volume and frequency (S7). The filter
coefficients calculated in real time are set in the variable filter unit 15 (S8). By doing this, the
characteristics of the filter change. That is, the filter coefficient at the peak frequency or notch
frequency changes. Then, the variable filter unit 15 performs filter processing on the frequency
sweep signal and outputs the result to the headphones 19. That is, the variable filter unit 15
outputs the frequency sweep signal multiplied by the filter coefficient to the headphones 19.
[0042]
The headphones 19 output the filtered frequency sweep signal to the listener U. The listener U
determines whether the frequency sweep signal output from the headphones 19 can be heard at
a constant level (S9). That is, when sweeping in the level adjustment range shown in FIG. 6, the
listener U determines whether the volume becomes constant regardless of the frequency.
[0043]
If it is determined that the frequency sweep signal can not be heard at a constant level (NO at
10-05-2019
14
S9), the processing from step S5 is repeated until the sound can be heard at a constant level. That
is, the listener adjusts the volume while listening to the sweep signal in the level adjustment
range. Therefore, as shown in FIG. 7, the processes of S5 to S9 are repeated until the frequency
sweep signal is heard at a constant level. If it is determined that the frequency sweep signal is
heard at a constant level (YES in S9), the listener U presses the adjustment completion button.
Thereby, the setting storage unit 17 stores the volume and the filter coefficient when the
adjustment completion button is pressed as the level information (S10). By doing this, as shown
in FIG. 7, the sound can be heard at a substantially constant volume regardless of the frequency.
[0044]
As shown in FIG. 7, when the volume becomes constant, the filter coefficient calculation unit 16
calculates the final filter coefficient from the center frequency and the level information (S11).
That is, the setting storage unit 17 stores the level information corresponding to the volume
when the volume becomes constant in the level adjustment range in association with the
frequency. Then, the filter coefficient calculation unit 16 calculates the filter coefficient at the
frequency based on the frequency and the level information stored in the setting storage unit 17.
Thereafter, the final filter coefficient is set to the variable filter unit 15 (S12). Thus, the
measurement of the filter coefficient is completed.
[0045]
At the time of music reproduction, final filter coefficients are set in the variable filter unit 15. In
the case of reproducing the music signal, after the out-of-head localization processing unit 13
performs the out-of-head localization processing on the music signal, the variable filter unit 15
performs the filtering process on the music signal. That is, the filter coefficient included in the
filter set in the variable filter unit 15 is multiplied to the music signal. Then, the headphones 19
output the filtered music signal to the listener U. That is, the headphone 19 outputs the music
signal subjected to the out-of-head localization process and the filter process to the listener U,
whereby the sound field is reproduced.
[0046]
Thus, the filter coefficient is determined by measurement using the sweep signal. Then, by
filtering with the filter including the determined filter coefficient, it is possible to eliminate the
10-05-2019
15
resonance caused by the individual characteristic of the ear canal shape. Therefore, the music
signal subjected to the localization outside the head can be appropriately corrected. Therefore,
even when the headphones 19 are used, the sound field can be appropriately reproduced. In the
above description, although the sound field reproduction apparatus using the headphones 19 is
shown, the same process can be applied to a sound field reproduction apparatus using
earphones.
[0047]
Although the above description has described the case where there is a dip in the frequency
characteristic, the sound quality can be similarly adjusted even when there is a peak. That is, the
volume may be reduced in S6 so as to reduce the volume at the peak frequency. Thus, the sound
quality can be adjusted so that the volume at the peak frequency is reduced.
[0048]
Also, even when there are two or more peak frequencies and dip frequencies, the volume may be
adjusted for each frequency. That is, sound volume adjustment is performed for each of the
peaks and dips included in the sweep range. Then, the filter coefficient calculation unit 16
obtains the filter coefficient according to the level information when the volume adjustment is
performed and the frequency corresponding thereto. By doing this, an appropriate filter can be
set, so that the sound field can be properly reproduced. Also, the width of the notch filter or
peaking filter may be adjusted.
[0049]
When the frequency presented on the display unit is displayed, the listener U can easily
understand. The listener may also adjust the speed at which the frequency sweep signal
generator 11 sweeps the frequency.
[0050]
Second Embodiment The sound field reproduction apparatus according to the present
10-05-2019
16
embodiment will be described with reference to FIG. FIG. 8 is a block diagram showing a sound
field reproduction device 200 according to the second embodiment. In the present embodiment,
the sound field is reproduced using the speakers 29 instead of the headphones 19. That is, the
speaker 29 is used instead of the headphones 19.
[0051]
The speaker 29 is a speaker having a plurality of channels such as a stereo speaker and a
surround speaker. Furthermore, in the present embodiment, a pseudo surround processing unit
23 is provided instead of the extra-head localization processing unit 13 according to the first
embodiment. The configuration other than the pseudo surround processing unit 23 is the same
as that of the first embodiment, and thus the description thereof is omitted.
[0052]
The sweep signal generated by the sweep signal generation unit 21 and the music signal
reproduced by the music signal reproduction unit 22 are input to the pseudo surround
processing unit 23. In the pseudo surround processing unit 23, head-related transfer
characteristics (also referred to as head-related transfer functions) measured in advance are set.
The pseudo surround processing unit 23 performs convolution processing of head related
transfer characteristics (also referred to as head related transfer functions). The pseudo surround
processing unit 23 outputs the frequency sweep signal subjected to the convolution processing
to the AGC processing unit 24. The AGC processing unit 24, variable filter unit 25, filter
coefficient calculation unit 26, setting storage unit 27, and processing in the input unit 28 are
the AGC processing unit 14, variable filter unit 15, filter coefficient calculation unit 16, and
setting storage of the first embodiment. The same as the unit 17 and the input unit 18.
Therefore, as in the first embodiment, the notch or peak frequency is determined, and the filter
coefficient calculation unit 26 calculates the filter coefficient.
[0053]
Then, a filter having the calculated filter coefficient is set in the variable filter unit 25. The
speaker 29 outputs the frequency sweep signal multiplied by the filter coefficient to the listener
U. The listener adjusts the volume as in the first embodiment while listening to the frequency
sweep signal output from the speaker 29. When the final filter coefficient is calculated, the input
10-05-2019
17
to the pseudo surround processing unit 23 is switched from the frequency sweep signal to the
music signal. Then, the pseudo surround processing unit 23 and the variable filter unit 25
perform processing on the music signal. The music signal that has undergone the processing of
the pseudo surround processing unit 23 and the variable filter unit 25 is output from the speaker
29.
[0054]
In the present embodiment, after the pseudo surround processing unit 23 convolutes the head
transfer characteristic into the music signal, the variable filter unit 15 performs the filtering
process on the music signal. By doing this, the surround sound field output from the speaker 29
can be reproduced. Furthermore, filter coefficients for correcting peaks or dips caused by the
measurement environment and the like are set. Therefore, the sound field can be properly
reproduced.
[0055]
Third Embodiment In this embodiment, the transfer characteristic of an individual is measured,
and a filter corresponding to the transfer characteristic is used to realize sound field
reproduction such as localization outside the head.
[0056]
Furthermore, in the present embodiment, an out-of-head localization process is performed on a
high resolution digital audio signal (hereinafter referred to as HR signal). In the following
description, a signal collected at a sampling frequency of 96 kHz will be described as an HR
signal or an HR signal. On the other hand, a signal (low resolution signal) collected at a sampling
frequency of 48 kHz is set as a non-high resolution signal (non-HR signal). Of course, the
sampling frequency is not limited to the above values.
[0057]
For a non-HR signal with a sampling frequency of 48 kHz, the Nyquist frequency is 24 kHz.
10-05-2019
18
Hereinafter, a band less than 24 kHz will be described as a low frequency band, and a band
greater than 24 kHz will be described as a high frequency band. The first frequency indicating
the boundary between the high frequency band and the low frequency band is the Nyquist
frequency 24 kHz. Of course, the first frequency may be different from 24 kHz. For example, the
first frequency can be changed according to the sampling frequency.
[0058]
First, the configuration of the out-of-head localization processing device 301 for performing the
out-of-head localization processing will be described with reference to FIG. FIG. 9 is a block
diagram showing the configuration of an out-of-head localization processing device 301 which is
an example of a sound field reproduction device. The extra-head localization processing device
301 corresponds to the extra-head localization processing unit 13 in the first embodiment.
[0059]
The out-of-head localization processing device 301 reproduces the sound field with respect to
the listener (user) U wearing the headphones 343. Therefore, the out-of-head localization
processing device 301 performs sound image localization processing on the Lch and Rch stereo
input signals XL and XR. The Lch and Rch stereo input signals XL and XR are audio reproduction
signals output from an audio device or the like compatible with HR signals. Note that the out-ofhead localization processing device 301 is not limited to a physically single device, and part of
the processing may be performed by a different device. For example, part of the processing may
be performed by a personal computer or the like, and the remaining processing may be
performed by a DSP (Digital Signal Processor) or the like incorporated in the headphone 343.
[0060]
The out-of-head localization processing device 301 includes a transfer characteristic processing
unit 310, a filter unit 341, a filter unit 342, and a headphone 343.
[0061]
The transfer characteristic processing unit 310 performs filter processing according to the
transfer characteristic.
10-05-2019
19
The transfer characteristic processing unit 310 includes convolution calculation units 311 to
312 and 321 to 322, and adders 324 and 325. The convolution calculation units 311 to 312 and
321 to 322 perform convolution processing using space acoustic transfer characteristics. The
transfer characteristic processing unit 310 receives stereo input signals XL and XR from an
audio device or the like corresponding to the HR signal. In the transfer characteristic processing
unit 310, space acoustic transfer characteristics are set. The transfer characteristic processing
unit 310 convolutes space acoustic transfer characteristics to the stereo input signals XL and XR
of each channel.
[0062]
The space acoustic transfer characteristics have four transfer characteristics Hls, Hlo, Hro and
Hrs. The four transfer characteristics can be determined using a filter generation device
described later.
[0063]
Then, the convolution unit 311 convolutes the transfer characteristic Hls with the Lch stereo
input signal XL. The convolution operation unit 311 outputs the convolution operation data to
the adder 324. The convolution operation unit 321 convolutes the transfer characteristic Hro
with respect to the Rch stereo input signal XR. The convolution operation unit 321 outputs the
convolution operation data to the adder 324. The adder 324 adds the two convolution operation
data and outputs the result to the filter unit 341.
[0064]
The convolution unit 312 convolutes the transfer characteristic Hlo with the Lch stereo input
signal XL. The convolution operation unit 312 outputs the convolution operation data to the
adder 325. The convolution unit 322 convolves the transfer characteristic Hrs with respect to the
Rch stereo input signal XR. The convolution unit 322 outputs the convolution data to the adder
325. The adder 325 adds the two convolution operation data and outputs the result to the filter
unit 342. Thus, the transfer characteristic processing unit 310 performs convolution processing
using a filter corresponding to the transfer characteristics Hls, Hlo, Hro, and Hrs.
10-05-2019
20
[0065]
In the filter units 341 and 342, an inverse filter for canceling the ear canal transmission
characteristic is set. Then, the inverse filter is convoluted with the reproduction signal subjected
to the processing in the transfer characteristic processing unit 310. A filter unit 341 convolves
an inverse filter on the Lch signal from the adder 324. Similarly, filter section 342 convolves an
inverse filter on the Rch signal from adder 325. The reverse filter cancels the characteristic from
the headphone unit to the microphone when the headphone 343 is attached. That is, when the
microphone is disposed at the entrance of the ear canal, the transfer characteristic between the
entrance of the ear canal of each listener and the reproduction unit of the headphones or the
reproduction unit of the eardrum and the headphones is canceled.
[0066]
The filter unit 341 outputs the corrected Lch signal to the left unit 343L of the headphone 343.
The filter unit 342 outputs the corrected Rch signal to the right unit 343R of the headphone 343.
The listener U wears a headphone 343. The headphones 343 output the Lch signal and the Rch
signal to the listener U. Thereby, the sound image localized outside the head of the listener U can
be reproduced.
[0067]
It is preferable to measure the transfer characteristics Hls, Hlo, Hro, Hrs according to the actual
listener U. For example, by attaching a microphone to the ear of the listener U and performing
impulse response measurement, it is possible to obtain the transfer characteristics Hls, Hlo, Hro,
and Hrs according to the shape of the auricle of the listener U. As described above, by using the
transfer characteristics Hls, Hlo, Hro, and Hrs acquired by actually attaching a microphone to the
listener's U ear, the external localization processing can be appropriately performed.
[0068]
Here, the case where the HR localization processing is performed on the HR signal will be
10-05-2019
21
described. In order to obtain an HR signal, it is necessary to prepare a microphone corresponding
to the HR signal. Normally, the audible band is said to be 20 Hz to 20 kHz, but in order to
correspond to the HR signal, it is necessary to prepare an HR signal compatible microphone
capable of collecting high frequency sound of 20 kHz or higher. The HR signal compatible
microphone has sensitivity to high frequency bands, and there is a problem in miniaturization.
[0069]
For example, the diameter of the human ear canal entrance is about 7.5 mm, while the actually
available HR signal compatible microphone is about 1.5 cm. It is not usually possible to obtain a
small HR signal compatible microphone that can be worn near the entrance of the human ear
canal. Also, even if there is an HR signal-compatible microphone of a size that can be worn at the
entrance of the human ear canal, it is considered to be very expensive. For this reason, it is
unrealistic to wear an HR signal compatible microphone for each listener U. Therefore, in the
present embodiment, the microphone measures the amplitude value of the transfer characteristic
in the low frequency band, and the filter generation device generates the amplitude value of the
transfer characteristic in the high frequency band.
[0070]
Hereinafter, the configuration of the filter generation device 350 will be described with reference
to FIG. The filter generation device 350 includes left and right speakers 5L and 5R, left and right
microphones 2L and 2R, and a processing unit 351. As shown in FIG. 10, an impulse response is
measured by measuring the impulse sound outputted by the left and right speakers 5L, 5R with
the microphones 2L, 2R. The collected sound signals acquired by the microphones 2L and 2R are
output to the processing unit 351. The processing unit 351 is, for example, an arithmetic
processing unit such as a personal computer. The processing unit 351 functions as a filter
generation unit that generates a filter based on the collected sound signal. Details of processing
in the processing unit 351 will be described later.
[0071]
In FIG. 10, the transfer characteristics measured by the microphones 2L and 2R are shown as
transfer characteristics H'ls, H'lo, H'ro and H'rs. Transfer characteristic H'ls between left speaker
5L and left microphone 2L, transfer characteristic H'lo between left speaker 5L and right
10-05-2019
22
microphone 2R, transfer characteristic H'ro between right speaker 5L and left microphone 2L
The transfer characteristic H'rs between the right speaker 5R and the right microphone 2R is
measured. That is, the transfer characteristic H'ls is acquired by the left microphone 2L collecting
the measurement signal output from the left speaker 5L. The right microphone 2R picks up the
measurement signal output from the left speaker 5L to acquire the transfer characteristic H'lo.
The left microphone 2L picks up the measurement signal output from the right speaker 5R to
obtain the transfer characteristic H'ro. The right microphone 2R picks up the measurement signal
output from the right speaker 5R to acquire the transfer characteristic H'rs.
[0072]
As described above, the impulse response is measured by measuring the impulse sound
outputted by the left and right speakers 5L and 5R with the microphones 2L and 2R. The
processing unit 351 stores the collected sound signal acquired based on the impulse response
measurement in a memory or the like. Thereby, the transfer characteristic H'ls between the left
speaker 5L and the left microphone 2L, the transfer characteristic H'lo between the left speaker
5L and the right microphone 2R, the transfer characteristic between the right speaker 5L and the
left microphone 2L H'ro, transfer characteristics H'rs between the right speaker 5R and the right
microphone 2R are measured. That is, the transfer characteristic H'ls is acquired by the left
microphone 2L collecting the measurement signal output from the left speaker 5L. The right
microphone 2R picks up the measurement signal output from the left speaker 5L to acquire the
transfer characteristic H'lo. The left microphone 2L picks up the measurement signal output from
the right speaker 5R to obtain the transfer characteristic H'ro. The right microphone 2R picks up
the measurement signal output from the right speaker 5R to acquire the transfer characteristic
H'rs.
[0073]
As described above, since microphones compatible with HR signals are difficult to miniaturize,
the microphones 2L and 2R are microphones incompatible with HR signals. That is, the collected
sound signals acquired by the microphones 2L and 2R are non-HR signals. Therefore, the left and
right speakers 5L and 5R can also be made non-HR signal compatible speakers. The sampling
frequency of the collected signal is 48 kHz.
[0074]
10-05-2019
23
Since the collected signal is a non-HR signal, the transfer characteristic does not include the high
frequency band component of 24 kHz or more. On the other hand, the stereo input signals XL
and XR of the actual out-of-head localization process include components of a high frequency
band. Therefore, in the present embodiment, the processing unit 351 calculates the transfer
characteristic of the high frequency band.
[0075]
A method of calculating the transfer characteristic of the high frequency band will be described
with reference to FIG. FIG. 11 shows the transfer characteristic Hls in the frequency domain. The
horizontal axis is frequency (Hz), and the vertical axis is amplitude (dB) of the transfer
characteristic Hls. That is, FIG. 11 shows the frequency amplitude characteristic of the transfer
characteristic Hls. Further, in FIG. 11, the transfer characteristic H'ls by the collected sound
signal is indicated by a solid line. The transfer characteristic H 'ls in the frequency domain can be
obtained by making the collected signal in the time domain discrete Fourier transform.
[0076]
In FIG. 11, a band including frequencies higher than the first frequency (Nyquist frequency = 24
kHz) is shown as a high frequency band BH, and a band including frequencies lower than the first
frequency is shown as a low frequency band BL. Also, a band from the second frequency to the
first frequency is shown as an interpolation band BL1 with 14 kHz as the second frequency. The
second frequency may be a frequency less than the Nyquist frequency, and is not limited to 14
kHz. The second frequency is preferably a frequency of 10 kHz or more.
[0077]
Here, the amplitude value of the transfer characteristic H'ls at the first frequency (24 kHz) is
taken as an amplitude value Yb [dB]. The processing unit 351 obtains a peak of the transfer
characteristic H'ls in the interpolation band BL1. Here, the frequency and amplitude value of the
peak are referred to as frequency fp [Hz] and amplitude value Yp [dB]. When a plurality of peaks
exist in the interpolation band BL1, the frequency and amplitude value of the highest frequency
peak are set as the frequency fp and the amplitude value Yp. When there is no peak in the
interpolation band BL1, the processing unit 351 sets the second frequency and its amplitude
10-05-2019
24
value as the frequency fp and the amplitude value Yp. The frequency fp is 14 kHz or more and
less than 24 kHz. The amplitude value Yp is also referred to as a second amplitude value.
[0078]
Then, for the band of 0 to fp, the processing unit 351 uses the amplitude value of the transfer
characteristic H'ls as it is as the amplitude value of the transfer characteristic Hls. Therefore, the
amplitude value of the transfer characteristic Hls at the frequency fp is the amplitude value Yp.
In addition, for the fp to 48 kHz band, the processing unit 351 calculates the amplitude value
based on the amplitude value Yp. Hereinafter, how to obtain the amplitude value in the band
from fp to 48 kHz will be described.
[0079]
Here, the processing unit 351 calculates six frequency amplitude characteristics (1) to (6). In the
frequency amplitude characteristic (3), the amplitude value (also referred to as a first amplitude
value) of the transfer characteristic Hls at 24 kHz is the amplitude value Yb. The amplitude value
of the transfer characteristic Hls in the high frequency band BH is constant at the amplitude
value Yb. The amplitude value of the transfer characteristic Hls in the fp to 24 kHz band is
obtained by interpolating the amplitude value Yp and the amplitude value Yb. That is, the
amplitude value of the transfer characteristic Hls at fp to 24 kHz is calculated so as to
complement between the amplitude value Yp and the first amplitude value. Here, the amplitude
value of the transfer characteristic Hls at fp to 24 kHz is determined by a known method such as
linear interpolation. Alternatively, the amplitude value of the transfer characteristic Hls at fp to
24 kHz may be the amplitude value of the measured transfer characteristic H'ls.
[0080]
The processing unit 351 changes the amplitude value (also referred to as a first amplitude value)
of the transfer characteristic Hls at 24 kHz from the amplitude value Yb to obtain the remaining
frequency characteristics (1), (2), (4) to (6). Seeking. For example, the first amplitude value is (Yb6) in the frequency amplitude characteristic (1), the first amplitude value is (Yb-3) in the
frequency amplitude characteristic (2), and the first in the frequency amplitude characteristic (4).
The amplitude value is (Yb + 3), the first amplitude value is (Yb + 6) in the frequency amplitude
characteristic (5), and the first amplitude value is (Yb + 9) in the frequency amplitude
10-05-2019
25
characteristic (6). Preferably, the first amplitude value is set in a range not exceeding the
amplitude value Yp. For example, when (Yb + 9) exceeds Yp, the frequency amplitude
characteristic (6) may not be determined.
[0081]
The amplitude value of the transfer characteristic Hls in the high frequency band BH is constant
at the first amplitude value. For example, in the frequency amplitude characteristic (1), the
amplitude value of the transfer characteristic Hls in the high frequency band BH is constant at
the first amplitude value (Yb-6). In the frequency amplitude characteristic (2), the amplitude
value of the transfer characteristic Hls in the high frequency band BH is constant at the first
amplitude value (Yb-3). Also for the frequency amplitude characteristics (4), (5), (6), the
amplitude value of the transfer characteristic Hls in the high frequency band BH is constant at
each first amplitude value.
[0082]
The amplitude value of the transfer characteristic Hls in the fp to 24 kHz band is obtained by
interpolating the amplitude value Yp and the first amplitude value. Here, the amplitude value of
the transfer characteristic Hls at fp to 24 kHz is determined by a known method such as linear
interpolation. Thus, frequency amplitude characteristics (1) to (6) are obtained by setting the first
amplitude value in 3 dB steps. The frequency amplitude characteristics (1) to (6) are candidates
for the transfer characteristic Hls, respectively.
[0083]
As described above, the transfer characteristic Hls can be determined based on the measured
transfer characteristic H'ls. The other transfer characteristics Hlo, Hrs, and Hro can also be
determined based on the transfer characteristics H'lo, H'ro, and H'rs. Then, the transfer
characteristics Hls, Hlo, Hrs, and Hro are respectively subjected to inverse discrete Fourier
transform. A filter is generated by determining four transfer characteristics Hls, Hlo, Hrs, Hro in
the time domain. Here, one filter includes four transfer characteristics Hls, Hlo, Hrs, and Hro in
the time domain.
10-05-2019
26
[0084]
In the above description, frequency amplitude characteristics (1) to (6) are obtained for each of
the transfer characteristics Hls, Hlo, Hrs, and Hro. That is, six types of filters are generated.
Therefore, by carrying out a hearing test, the optimum filters (transfer characteristics Hls, Hlo,
Hrs, Hro) are sought among the plurality of filters. The hearing test will be described later. The
optimal filters (transmission characteristics Hls, Hlo, Hrs, Hro) obtained in the hearing test are set
in the convolution calculation units 311 to 312 and 321 to 322 shown in FIG.
[0085]
Thus, the processing unit 351 sets the amplitude component of the low frequency band BL of the
filter according to the frequency amplitude characteristic of the collected signal, and connects
the amplitude component of the high frequency band BH of the filter to the amplitude
component of the low frequency band It is generated to fit. By doing this, it is possible to
generate a filter corresponding to the HR signal. For the low frequency band BL, the transfer
characteristic of the listener U can be used, so that the outside localization processing can be
appropriately performed. Further, the measurement of the transfer characteristics H'lo, H'ro, H'rs
can be performed by the microphones 2L, 2R and the speakers 5L, 5R which do not support the
HR signal. The microphones 2 </ b> L and 2 </ b> R that do not support HR signals are compact
and can be worn on the left and right ears. Therefore, measurement can be performed simply and
easily.
[0086]
The processing unit 351 sets a first amplitude value at the first frequency of the transfer
characteristic based on the amplitude value Yb of the frequency amplitude characteristic of the
collected signal at the first frequency (24 kHz). This enables appropriate localization processing
outside the head. In addition, since no complicated processing is performed, the filter can be
generated easily.
[0087]
The processing unit 351 sets, as a first amplitude value, a value obtained by correcting the
10-05-2019
27
amplitude value Yb of the frequency amplitude characteristic of the collected signal at the first
frequency. In the above description, the value (for example, Yb-3, Yb + 3 or the like) obtained by
adjusting the level of the amplitude value Yb is set as the first amplitude value. Thereby, the
outside localization processing can be appropriately performed by a simple method. In addition,
since no complicated processing is performed, the filter can be generated easily.
[0088]
In the present embodiment, a value obtained by changing Yb at intervals of 3 dB is set as the first
amplitude value, but the setting of the first amplitude value is not limited to such a method. For
example, the first amplitude value may be set at 2 dB intervals, or may be set at intervals other
than the constant interval. Further, the first amplitude value is set in a range not exceeding the
amplitude value Yp. This enables appropriate localization processing outside the head.
[0089]
In the above description, the amplitude value of the transfer characteristic Hls in the high
frequency band BH is a constant value, but the present embodiment is not limited to this. The
amplitude value of the transfer characteristic Hls in the high frequency band BH may be
gradually decreased or gradually increased at a constant slope. Alternatively, the amplitude value
of the transfer characteristic Hls in the high frequency band BH may be set in accordance with a
preset pattern.
[0090]
In the third embodiment, the frequency amplitude characteristics (1) to (6) are obtained, but the
number of frequency amplitude characteristics to be obtained may be one or more. If the number
of frequency amplitude characteristics is one, the hearing test is not necessary. If the number of
frequency amplitude characteristics is two or more, the audibility test to be described later is
performed.
[0091]
10-05-2019
28
(Modification 1) In Modification 1 of Embodiment 3, the frequency amplitude characteristic of
the high frequency band is calculated according to the simulation. The other configurations and
methods are similar to those of the third embodiment, and therefore the description thereof is
omitted.
[0092]
FIG. 12 is a diagram showing frequency amplitude characteristics of a speaker compatible with
HR signals. The horizontal axis of FIG. 12 is frequency, and the vertical axis is sound pressure
(dB). FIG. 12 shows the frequency characteristics when the HR signal compatible speaker is
disposed at an angle of 10 degrees from the front of the listener U.
[0093]
FIG. 13 is a diagram showing the transfer characteristic Hls obtained from the frequency
characteristic of FIG. Specifically, FIG. 13 shows the result of simulating the transfer
characteristic Hls when the HR signal compatible speaker is installed at the above angle. As such
a simulation, for example, it is described in Simulation technology for reproducing threedimensional sound according to an individual
(http://www.nict.go.jp/publication/shuppan/kihou-journal/ kihou-vol56no1̲2 / 0403. pdf).
[0094]
In the above-mentioned literature, the FDTD (Finite-Difference Time Domain) method is used. In
this method, for example, the head transfer function HRTF of the dummy head with respect to
the speaker disposed at a predetermined angle can be determined. Therefore, it can be
determined by the transfer characteristic Hls from the speaker to the entrance of the ear canal.
The amplitude component of the high frequency band BH is estimated by the simulation based
on the frequency characteristic data of the speaker. Here, the transfer characteristic obtained by
simulation is referred to as a transfer characteristic H "ls.
[0095]
10-05-2019
29
As in the first embodiment, the amplitude value of the low frequency band BL is the amplitude
value of the measured transfer characteristic H'ls. As shown in FIG. 13, the transfer characteristic
H'ls of the low frequency band BL and the transfer characteristic H''ls of the high frequency band
BH are connected to generate the transfer characteristic Hls. That is, the processing unit 351
generates the transfer characteristic Hls so as to connect the amplitude component of the high
frequency band to the amplitude component of the low frequency band.
[0096]
(Modification 2) In the modification 2 of the embodiment 3, the frequency amplitude
characteristic of the high frequency band is measured by the HR signal compatible microphone.
The other configurations and methods are similar to those of the third embodiment, and
therefore the description thereof is omitted.
[0097]
The HR signal compatible microphone is considered to be very expensive as described above, so
it is difficult to wear and measure for each listener U. Therefore, a representative transfer
characteristic is measured with an HR signal compatible microphone attached to a person other
than the listener U or a dummy head. The measurement of the transfer characteristic using the
HR signal compatible microphone is the same as the configuration shown in FIG. 10, and the
microphones 2L and 2R are attached to a person other than the listener U or a dummy head.
[0098]
A transfer characteristic measured by an HR signal compatible microphone mounted on a person
other than the listener U or a dummy head is referred to as a transfer characteristic H "ls. For the
high frequency band BH, the transfer characteristic H′′ls is used, and for the low frequency
band BL, the transfer characteristic H′ls measured with the HR signal incompatible microphone
is used. Then, as in the first modification, the transfer characteristic H'ls of the low frequency
band BL and the transfer characteristic H''ls of the high frequency band BH are connected to
generate the transfer characteristic Hls. That is, the processing unit 351 generates the transfer
characteristic Hls so as to connect the amplitude component of the high frequency band to the
amplitude component of the low frequency band.
10-05-2019
30
[0099]
In the first and second modifications, the transfer characteristic H ′ ls of the low frequency
band BL is obtained by measurement according to the listener U. That is, as shown in FIG. 10, the
transfer characteristic H'ls is measured with the HR signal incompatible microphone attached to
the listener U. Therefore, since the filter according to the listener U can be used, the out-of-head
localization process can be appropriately performed.
[0100]
In the first modification, the transfer characteristic H′′ls of the high frequency band BH is a
simulation result. In the second modification, the transfer characteristic H ′ ′ ls of the high
frequency band BH is measured by an HR signal compatible microphone attached to a person
other than the listener U or a dummy head. It is not necessary to perform measurement with a
very expensive HR signal compatible microphone for each listener U. Therefore, the transfer
characteristic corresponding to the HR signal can be easily obtained.
[0101]
In Modified Examples 1 and 2, as shown in FIG. 14, the amplitude value of the transfer
characteristic may be largely different in the vicinity of the first frequency (24 kHz). In such a
case, it may not be possible to appropriately perform out-of-head localization processing.
Therefore, in the first and second modifications, it is preferable to adjust the level of the transfer
characteristic H ′ ′ ls of the high frequency band BH (see the arrow in FIG. 14). Here, the
amplitude characteristic is translated up and down by adjusting the DC component of the high
frequency band BH.
[0102]
The transfer characteristic H ′ ′ ls of the high frequency band BH is adjusted to match the
transfer characteristic H ′ ls of the low frequency band BL measured according to the listener U.
In this way, it is possible to appropriately perform out-of-head localization processing. Thus, the
amplitude component of the high frequency band BH is level-adjusted so that the amplitude
10-05-2019
31
component of the high frequency band BH is joined to the amplitude component of the low
frequency band BL.
[0103]
Alternatively, a band for smoothing the amplitude value may be provided in the vicinity of 24
kHz. The smoothing process will be described with reference to FIG. In FIG. 15, the band
disposed on the low frequency side of the high frequency band BH is taken as a band B2. A
predetermined frequency included in the high frequency band BH is a frequency fa. The
frequency fa is a frequency higher than 24 kHz. The band B2 is in the range of 24 kHz or more
and the frequency fa or less. In the band B2, the amplitude value is subjected to a smoothing
process.
[0104]
Alternatively, smoothing processing can be performed in band B3 instead of band B2. The band
B3 is a band disposed on the high frequency side of the low frequency band BL. For example, a
predetermined frequency included in the low frequency band BL is a frequency fb. The frequency
fb is lower than 24 kHz. The band B2 is in the range of 24 kHz or less at the frequency fa or
more. In the band B3, the amplitude value is subjected to smoothing processing.
[0105]
Alternatively, instead of the band B2 or the band B3, the smoothing process may be performed in
the band B4 across 24 kHz. The band B4 is in the range of the frequency fb to the frequency fa.
In the band B4, the amplitude value is subjected to a smoothing process. A moving average or a
weighted moving average can be used for the smoothing process. The smoothing process makes
it possible to generate a more appropriate filter in which the low frequency band and the high
frequency band are smoothly continuous.
[0106]
As described above, in the first and second modifications, the transfer characteristic Hls is
10-05-2019
32
generated by connecting the transfer characteristic H'ls of the low frequency band BL and the
transfer characteristic H''ls of the high frequency band BH. Similarly, transfer characteristics Hlo,
Hro and Hrs are generated. Then, with the four transfer characteristics Hls, Hlo, Hro, and Hrs as
one set, the inverse discrete Fourier transform is performed as in the first embodiment. Thereby,
it is possible to generate a filter including four transfer characteristics Hls, Hlo, Hro, Hrs in the
time domain as one set.
[0107]
(Hearing test) A hearing test for determining an optimal filter from a plurality of trial listening
filters will be described. By performing the audibility test, it is possible to reproduce outside the
head with the sound quality according to the preference of the listener U.
[0108]
Here, the plurality of trial listening filters are generated by one or more methods of the third
embodiment and any of the first and second modifications thereof. For example, in the third
embodiment, since the frequency amplitude characteristics of (1) to (6) are obtained, six filters
are generated. Further, in the first modification, a plurality of filters can be generated by
changing the simulation method or changing the frequency characteristics of the HR signal
compatible speaker used for the simulation. In the second modification, a plurality of filters can
be generated by changing the person wearing the HR signal compatible microphone or the
dummy head. Let these filters be filters for audition. Therefore, each of the plurality of trial
listening filters is obtained by the method according to any one of the first embodiment, the first
modification, and the second modification. Each audition filter includes four transfer
characteristics Hls, Hlo, Hro and Hrs.
[0109]
The out-of-head localization processing apparatus 500 for performing the audibility test will be
described with reference to FIG. The out-of-head localization processing apparatus 500 includes
an adjustment signal generation unit 521, a music signal reproduction unit 512, an out-of-head
localization processing unit 302, a filter selection unit 522, a setting storage unit 517, an input
unit 518, and headphones 343. Is equipped.
10-05-2019
33
[0110]
The input unit 518 and the music signal reproduction unit 512 are the same as the input unit 18
and the music signal reproduction unit 12 shown in FIG. 1, respectively, so the detailed
description will be omitted. The headphones 343 are similar to the headphones 343 of FIG. 10
and the headphones 19 of FIG. However, the music signal reproduction unit 512 and the
headphones 343 correspond to the HR signal. That is, the music signal reproduction unit 512
reproduces the HR signal collected at the sampling frequency of 96 kHz as a music signal. The
headphones 519 output the HR signal to the listener U.
[0111]
The adjustment signal generation unit 521 outputs an adjustment signal for determining an
optimum filter to the out-of-head localization processing unit 302. That is, the adjustment signal
generation unit 521 outputs the HR signal collected at the sampling frequency of 96 kHz as an
adjustment signal. Also, the adjustment signal is a stereo signal including Lch and Rch. As the
adjustment signal, it is preferable to use a sound source such as a stringed instrument rich in
harmonics. Specifically, the adjustment signal is generated by collecting performance sounds
such as cello and gamelan (ethnic music from Indonesia) at a sampling frequency of 96 kHz.
[0112]
The out-of-head localization processing unit 302 includes the transfer characteristic processing
unit 310 shown in FIG. 9, a filter unit 341, and a filter unit 342. The stereo adjustment signals
become the stereo input signals Lch and Rch in FIG. Then, the out-of-head localization processing
unit 302 outputs the adjustment signal subjected to out-of-head localization processing to the
headphones 519. The headphones 519 output the adjustment signal subjected to the outside
localization processing to the listener U. Thus, a hearing test is performed.
[0113]
The listener U operates the input unit 518 according to the aural sense of the adjustment signal
subjected to the out-of-head localization processing. That is, the listener U inputs whether or not
10-05-2019
34
the appropriate out-of-head localization processing has been performed by the audition filter.
The input from the input unit 518 is stored in the setting storage unit 517. The filter selection
unit 522 stores a plurality of trial listening filters. When the input of the sense of hearing for one
filter is completed, the filter selection unit 522 switches the filter. That is, the filter selection unit
522 switches the trial listening filter in order and outputs it to the out-of-head localization
processing unit 302. In this way, the auditory test is performed for the number of stored
auditioning filters.
[0114]
The listener U operates the input unit 518 to input an optimal filter showing the best hearing.
Then, the setting storage unit 517 stores an optimal filter. The optimal filter is set as the music
playback filter. At the time of reproduction of the music signal, the out-of-head localization
processing unit 302 performs out-of-head localization processing using the optimal filter
determined in the audibility test. The listener U compares aurally with the adjustment signal
corresponding to the HR signal, and selects a favorite filter as the optimum filter. Then, an out-ofhead localization process is performed using an optimal filter for the music signal corresponding
to the HR signal. In this way, it is possible to appropriately perform out-of-head localization
processing.
[0115]
The listener U can switch between the hearing test and the music reproduction by operating the
input unit 518. When the listener U inputs an instruction to perform a hearing test, the
adjustment signal generation unit generates an adjustment signal. After the hearing test ends,
when the listener U inputs an instruction to perform music reproduction, the music signal
reproduction unit 512 reproduces the music signal. If there is only one set of auditioning filters,
the hearing test may not be performed.
[0116]
The filter units 341 and 342 need to have amplitude components in the high frequency band
also for the inverse filters set. However, the amplitude component of the high frequency band of
the inverse filter that cancels the ear canal transfer characteristic has little influence on the outof-head localization processing. For example, in the band of 14 kHz or more, the effect of
10-05-2019
35
localization outside the head is obtained without canceling the headphone characteristic.
Therefore, the amplitude component of the high frequency band can be set by the method shown
in JP-A-2015-126269. That is, the amplitudes of the high-frequency boundary frequency and the
Nyquist frequency shown in JP-A-2015-126269 may be used as the amplitude of the high
frequency band.
[0117]
Fourth Embodiment In the fourth embodiment, the hearing test is performed by a different
method. Hereinafter, the out-of-head localization processing apparatus for performing the
audibility test will be described with reference to FIG. FIG. 17 mainly describes only the process
of performing the hearing test. That is, in the following description, processing for determining
an optimal filter from a plurality of trial listening filters is described. The out-of-head localization
process using the optimum filter is the same as that of the third embodiment, and therefore the
description thereof is omitted. For example, the input unit, the headphones, and the setting
storage unit are not shown.
[0118]
The out-of-head localization processing apparatus 400 includes an adjustment signal generating
unit 421, an LPF (low pass filter) 411, a down sampling unit 412, an out-of-head localization
processing unit 401, an up sampling unit 413, an LPF 414, and an HPF (high pass filter). 431, a
variable amplifier 432 and an adder 440.
[0119]
The adjustment signal generation unit 421 generates an adjustment signal for determining an
optimal filter, similarly to the adjustment signal generation unit 521.
That is, the adjustment signal generation unit 421 outputs the HR signal collected at the
sampling frequency of 96 kHz as the adjustment signal. The sampling frequency of the
adjustment signal is 96 kHz. The adjustment signal generated by the adjustment signal
generation unit 421 is input to the LPF 411 and the HPF 431.
[0120]
10-05-2019
36
The LPF 411 is a low pass filter with a cutoff frequency of 24 kHz. Therefore, the LPF 411
passes components in the low frequency band and blocks components in the high frequency
band. The downsampling unit 412 downsamples the adjustment signal that has passed through
the LPF 411. Thus, the sampling frequency of the adjustment signal is 48 kHz. The
downsampling unit 412 outputs the downsampled adjustment signal to the out-of-head
localization processing unit 401.
[0121]
The out-of-head localization processing unit 401 corresponds to the out-of-head localization
processing unit 302 shown in FIG. Therefore, the out-of-head localization processing unit 401
includes the transfer characteristic processing unit 310 shown in FIG. 9, the filter unit 341, and
the filter unit 342. The out-of-head localization processing unit 401 performs out-of-head
localization processing on the down-sampled adjustment signal. Since the sampling frequency of
the adjustment signal is downsampled to 48 kHz, the out-of-head localization processing unit
401 can use the transfer characteristics H'ls, H'lo, H'ro, and H'rs. The transfer characteristics H'ls,
H'lo, H'ro and H'rs are measured in the configuration shown in FIG. 10, and are measured by the
microphones not supporting the HR signal.
[0122]
Next, the upsampling unit 413 upsamples the adjustment signal subjected to the outside
localization processing. Thereby, the sampling frequency of the adjustment signal becomes 96
kHz. The LPF 414 is a low pass filter with a cutoff frequency of 24 kHz. Thus, the LPF 414
passes components in the low frequency band and blocks components in the high frequency
band. The adjustment signal that has passed through the LPF 414 is input to the adder 440.
[0123]
The adjustment signal generated by the adjustment signal generation unit 421 is input to the
HPF 431. The HPF 431 is a high pass filter with a cutoff frequency of 24 kHz. Thus, the HPF 431
passes components in the high frequency band and blocks components in the low frequency
band. The adjustment signal that has passed through the HPF 431 is amplified by the variable
10-05-2019
37
amplifier 432 and input to the adder 440. The adder 440 adds the adjustment signal from the
LPF 414 and the adjustment signal from the variable amplifier 432, and outputs the result to
headphones. That is, the components of the low frequency band and the components of the high
frequency band are combined by the adder 440.
[0124]
As described above, an out-of-head localization process using a filter is performed only on the
low frequency band component that has passed through the LPF 412. That is, for the
components in the high frequency band that has passed through the HPF 431, the out-of-head
localization process using the filter is not performed. The component of the high frequency band
and the component of the low frequency band thus processed are synthesized by the adder 440
and output from the headphone.
[0125]
Here, by changing the amplification factor of the variable amplifier 432, an auditory sense test
can be performed. For example, the amplification factor of the variable amplifier 432 is increased
stepwise or continuously. Then, the input is performed at the timing when the listener U has the
best hearing. Thereby, the optimal amplification factor can be determined.
[0126]
FIG. 18 is a view schematically showing frequency characteristics of the HPF 431 in the hearing
test. The horizontal axis in FIG. 18 is the frequency, and the vertical axis is the gain. Further, FIG.
18 shows the frequency characteristics of the LPFs 411 and 414. By stepwise changing the
amplification factor of the variable amplifier 432, the frequency characteristic of the HPF 431 is
as shown in FIG. In other words, the gain of the HPF 431 can be adjusted. The gain of the HPF
431 can be relatively increased or decreased relative to the gains of the LPFs 411 and 414.
[0127]
Therefore, it is possible to change the component of the high frequency band to perform the
10-05-2019
38
hearing test. The components of the high frequency band can be level adjusted to the
components of the low frequency band. In other words, the components of the high frequency
band can be joined to the components of the low frequency band. Thus, the listener U can
determine the optimum filter from the plurality of trial listening filters. Furthermore, an optimal
filter can be generated based on the measurement result with the HR signal incompatible
microphone.
[0128]
(Modification 3) In the modification 3, a hearing test is performed by processing different from
the processing of the fourth embodiment. FIG. 19 is a block diagram showing the configuration
of the out-of-head localization processing apparatus 400 according to the third modification. In
the third modification, the filter storage unit 402 stores a filter obtained by upsampling the
transfer characteristics H'ls, H'lo, H'ro, and H'rs. The out-of-head localization processing unit 401
performs out-of-head localization processing using the up-sampled filter coefficient. Description
of the same contents as the contents of the embodiment described above will be omitted as
appropriate.
[0129]
The adjustment signal generation unit 421 outputs the adjustment signal to the out-of-head
localization processing unit 401 and the HPF 431. Since the adjustment signal is an HR signal,
the sampling frequency is 96 kHz. The out-of-head localization processing unit 401 performs
out-of-head localization processing of the adjustment signal. Here, the adjustment signal
subjected to the localization processing outside the head is output to the adder 440.
[0130]
As in the fourth embodiment, the HPF 431 is a high pass filter having a cutoff frequency of 24
kHz. Thus, the HPF 431 passes components in the high frequency band and blocks components
in the low frequency band. The adjustment signal that has passed through the HPF 431 is
amplified by the variable amplifier 432 and input to the adder 440. The adder 440 adds the
adjustment signal from the LPF 414 and the adjustment signal from the variable amplifier 432,
and outputs the result to headphones. That is, the components of the low frequency band and the
components of the high frequency band are combined by the adder 440.
10-05-2019
39
[0131]
Thus, the out-of-head localization processing unit 401 performs out-of-head localization
processing using the up-sampled filter coefficients. The components of the high frequency band
passed through the HPF 431 are not subjected to the out-of-head localization processing using a
filter. The component of the entire band and the component of the high frequency band thus
processed are synthesized by the adder 440 and output from the headphone.
[0132]
Here, by changing the amplification factor of the variable amplifier 432, a hearing test can be
performed. For example, the amplification factor of the variable amplifier 432 is increased
stepwise or continuously. Then, the input is performed at the timing when the listener U has the
best hearing. Thereby, the optimal amplification factor can be determined. Thereby, the same
effect as that of the fourth embodiment can be obtained.
[0133]
In the third and fourth embodiments, the frequency (first frequency) which is the boundary
between the high frequency band BH and the low frequency band BL is 24 kHz, but it may be a
frequency which does not exceed the Nyquist frequency. Also, using the two methods of the third
and fourth embodiments, one listener U may perform a hearing test to determine an optimal
filter. Further, in the third and fourth embodiments, the localization processing is performed
using headphones, but as in the second embodiment, sound image reproduction may be
performed using a speaker.
[0134]
Some or all of the above signal processing may be performed by a computer program. The
programs described above can be stored and supplied to a computer using various types of nontransitory computer readable media. Non-transitory computer readable media include tangible
storage media of various types. Examples of non-transitory computer readable media are
10-05-2019
40
magnetic recording media (eg flexible disk, magnetic tape, hard disk drive), magneto-optical
recording media (eg magneto-optical disk), CD-ROM (Read Only Memory), CD-R, CD-R / W,
semiconductor memory (for example, mask ROM, PROM (Programmable ROM), EPROM (Erasable
PROM), flash ROM, RAM (Random Access Memory)) are included. Also, the programs may be
supplied to the computer by various types of transitory computer readable media. Examples of
temporary computer readable media include electrical signals, light signals, and electromagnetic
waves. The temporary computer readable medium can provide the program to the computer via
a wired communication path such as electric wire and optical fiber, or a wireless communication
path.
[0135]
As mentioned above, although the invention made by the present inventor was concretely
explained based on an embodiment, the present invention is not limited to the above-mentioned
embodiment, and can be variously changed in the range which does not deviate from the gist.
Needless to say.
[0136]
11, 21, 512 sweep signal generation unit 12, 22 music signal reproduction unit 13 external
localization processing unit 14, 24 AGC processing unit 15, 25 variable filter unit (filter unit) 16,
26 filter coefficient calculation unit 17, 27, 517 Setting storage unit 18, 28, 518 Input unit 19
Headphones (output unit) 23 Pseudo surround processing unit 29 Speaker 100, 200 Sound field
reproduction device 301 Out-of-head localization processing device 343 Headphones 521
Adjustment signal generation unit 522 Filter selection unit
10-05-2019
41
1/--страниц
Пожаловаться на содержимое документа