close

Вход

Забыли?

вход по аккаунту

JP2006081191

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 JP2006081191
PROBLEM TO BE SOLVED: To provide an acoustic reproduction device and an acoustic
reproduction method capable of correcting distortion due to an actual listening space by
correcting characteristics of an actual listening space with respect to a virtual sound source
generated from a head related transfer function. According to a characteristic of an actual
listening environment characteristic function storing an actual listening characteristic function
for correcting a virtual sound source according to a characteristic to an actual listening space
provided at the time of listening, And a real listening space characteristic correction unit which
reads out the real listening space characteristic function stored in the database of the listening
environment characteristic function and corrects the virtual sound source based on the real
listening space characteristic function. This removes the cause for each distortion and provides
the best sound. [Selected figure] Figure 1
Sound reproduction apparatus and sound reproduction method
[0001]
The present invention relates to an audio reproduction apparatus and an audio reproduction
method, and more particularly to an audio reproduction apparatus and an audio reproduction
method for generating a virtual sound source using a head transfer function (HRTF).
[0002]
In the conventional audio industry, it is intended to reproduce raw sound by forming output
sound on a one-dimensional forward and two-dimensional plane.
09-05-2019
1
Recently, most sound reproducing devices reproduce stereo sound signals by reproducing
monaural sound signals. However, when reproducing a stereo sound signal, the range felt from
the sound signal to be reproduced is limited by the position of the speaker. To this end, studies
have been made to improve the reproduction ability of the speaker in order to feel the freshness
more fully and to generate a virtual signal by signal processing.
[0003]
As a result of such research, a representative system is a surround three-dimensional sound
system using five speakers. This separates out the virtual signals output from the rear speakers.
A method of generating such a virtual signal is a method of introducing a delay due to spatial
movement of the signal, reducing the signal size and transmitting it backward. Most current
sound reproduction devices use stereo sound technology called Dolby Prologic Surround, and if
there is a sound reproduction device capable of reproducing this, the feeling of tension at the
cinema level in ordinary homes You can experience overflowing sounds.
[0004]
Although increasing the number of channels in this way produces the effect of reproducing vivid
sound, it is necessary to add speakers in response to the increased number of channels, which
causes problems such as cost and installation space. Will occur. This problem can be ameliorated
by applying research on how people hear and recognize sounds present in three-dimensional
space. In particular, numerous studies have been conducted on how human hearing recognizes a
three-dimensional acoustic space, and recently virtual sources have been generated and adopted
in application fields.
[0005]
When the concept of such virtual sound source is applied to a sound reproduction apparatus,
that is, providing a sound source using a predetermined number of, for example, two speakers
without using a plurality of speakers to perform sound reproduction in stereo. If possible, this
will provide a great advantage in the implementation of the sound reproduction device. First,
using few speakers provides an economic advantage, and second, it offers the advantage of
09-05-2019
2
minimizing the space occupied by the system.
[0006]
When localizing a virtual sound source using a conventional sound reproducing apparatus, a
head related to or modified from a Head Related Transfer Function (HRTF) measured in an
anechoic chamber. The transfer function was used. However, when using a conventional sound
reproduction device, the effect of the three-dimensional sound reflected at the time of recording
is removed, and the sound actually heard by the listener is not the initial optimum sound but
distorted sound. become. Therefore, there is a problem that it is not possible to provide the
sound that the listener wants. In order to solve this problem, a space transfer function measured
in the optimal listening space is used instead of the head transfer function measured in the
anechoic chamber. However, in this case, the space transfer function used to correct the sound
has much more data to be processed than the head transfer function. As a result, there is a
problem that a relatively high capacity memory is required with another high performance
processor that can calculate the main factors in real time in the circuit.
[0007]
Also, the existing reproduced sound has a problem that it is altered by the speaker and the
listening space used by the actual listener due to the difference between the optimum listening
space intended for recording and the characteristics of the sound reproducing apparatus.
[0008]
The present invention was devised to solve the above-mentioned problems, and the object of the
present invention is to correct the characteristics of the actual listening space with respect to the
virtual sound source generated from the head-related transfer function. It is an object of the
present invention to provide an audio reproduction device and an audio reproduction method
capable of correcting distortion due to a listening space.
Another object of the present invention is to provide an audio reproduction apparatus and an
audio reproduction method capable of correcting distortion by the speaker by correcting the
characteristics of the speaker with respect to the virtual sound source generated from the headrelated transfer function.
09-05-2019
3
[0009]
Another object of the present invention is to provide an audio reproduction apparatus and an
audio reproduction method capable of making it possible to make a virtual sound source
generated from a head-related transfer function be recognized as being listened to in an optimal
listening space.
[0010]
In order to solve the above-mentioned problems, in the sound reproducing apparatus according
to the present invention, an audio signal input to an input channel is generated as a virtual sound
source by a head transfer function (HRTF), and an audio signal by the virtual sound source
generated. Is an audio reproduction apparatus output via a speaker, and an actual listening
environment in which an actual listening characteristic function for correcting the virtual sound
source is stored according to the characteristics for the actual listening space provided at the
time of listening It includes a characteristic function database, and an actual listening space
characteristic correction unit that reads out the actual listening space characteristic function
stored in the actual listening environment characteristic function database and corrects the
virtual sound source based thereon.
[0011]
A speaker characteristic function for correcting the virtual sound source is further stored in the
real listening environment characteristic function database of the sound reproducing apparatus
according to the characteristic for the speaker provided at the time of listening, and the real
listening environment characteristic function database And a speaker characteristic correction
unit for correcting the virtual sound source based on the read speaker characteristic function
stored in the storage unit.
[0012]
The sound reproduction apparatus includes: a virtual listening space parameter storage unit in
which a virtual listening space parameter set to be output to an optimal listening space in which
an audio signal from the virtual sound source is expected; The information processing apparatus
may further include a virtual listening space correction unit that reads out the virtual listening
space parameters stored in the space parameter storage unit and corrects the virtual sound
source based thereon.
[0013]
09-05-2019
4
It is preferable that the virtual listening space correction unit corrects only the virtual sound
source corresponding to the audio data input from the front channel among the input channels.
It is preferable that the virtual listening space correction unit corrects only the virtual sound
source corresponding to the audio data input from the rear channel among the input channels.
A sound reproducing apparatus according to the present invention for solving the technical
problems as described above is characterized in that audio data input to an input channel is
generated as a virtual sound source by HRTF and generated. An acoustic reproduction apparatus
in which an acoustic signal from a sound source is output through a speaker, and an actual
listening environment characteristic in which a speaker characteristic function for correcting the
virtual sound source is stored according to the characteristic with respect to the speaker
provided at the time of listening. A speaker characteristic correction unit may be included that
reads a database of functions and the speaker characteristic function stored in the database of
the actual listening environment characteristic function, and corrects the virtual sound source
based on this.
[0014]
Further, in the sound reproducing apparatus according to the present invention, the audio data
input to the input channel is generated as a virtual sound source by a head transfer function
(HRTF), and the generated sound signal by the virtual sound source is output through a speaker
A virtual listening space parameter storage unit for storing a virtual listening space parameter set
to output an audio signal from the virtual sound source to an expected optimal listening space;
And a virtual listening space correction unit for reading out the virtual listening space
parameters stored in the listening space parameter storage unit and correcting the virtual sound
source based thereon.
[0015]
On the other hand, in the sound reproducing method according to the present invention for
solving the above technical problems, the virtual data generated by generating audio data input
to the input channel into a virtual sound source by HRTF is generated. An acoustic reproduction
method in which an acoustic signal from a sound source is output through a speaker, wherein: (a)
an actual listening characteristic function for correcting the virtual sound source according to the
characteristics for the actual listening space provided at the time of listening (B) correcting the
virtual sound source based on a speaker characteristic function for correcting the virtual sound
source according to the characteristics of the speaker provided at the time of listening; (C) based
09-05-2019
5
on a virtual listening space parameter set to output an acoustic signal from the virtual sound
source to an optimal listening space to be predicted; And correcting the sound sources may
further include.
[0016]
According to the sound reproduction apparatus and the sound reproduction method of the
present invention, it is possible to obtain an optimum virtual sound source in each of the
listening spaces by correcting the actual listening space.
In addition, by correcting the speaker characteristics, it is possible to obtain an optimal virtual
sound source applied to each speaker.
Furthermore, by making it feel like listening in a virtual listening space, it is possible to feel that
it is listening to sound in an optimal listening space.
[0017]
In addition, since a space transfer function is not used to correct a distorted sound, a large
amount of calculation is not required, and a relatively large memory is not required.
Therefore, when listening to the sound through the virtual sound source, the cause for each
distortion is removed, so a high quality sound can be provided.
[0018]
Hereinafter, preferred embodiments of the present invention will be described in detail based on
the attached drawings.
[0019]
FIG. 1 is a block diagram illustrating an audio reproduction apparatus according to an
09-05-2019
6
embodiment of the present invention, and relates to an audio reproduction apparatus for
correcting the characteristics of an actual listening space.
The sound reproduction apparatus 100 according to the present embodiment includes the head
related transfer function database 110, the head related transfer function application unit 120,
the first combining unit 130, the first band pass filter 140, the second combining unit 180, and
the actual listening environment characteristic. A function database 150, a second band pass
filter 160, an actual listening space characteristic correction unit 170, and a second combining
unit 180 are included.
[0020]
The head related transfer function database 110 stores head related transfer functions measured
in an anechoic chamber.
The head related transfer function according to the present invention refers to a transfer
function in a frequency domain indicating radio waves of sound from the sound source in the
anechoic chamber to the human ear canal. That is, from the structural aspect of the ear, although
the irregular pattern of the pinna distorts the frequency spectrum of the signal reaching the ear
before it enters the ear canal, such distortion is caused by the direction and distance of the
sound, etc. Aspects are changing. Such changes in frequency components play a large role in
recognizing the direction of sound by a person, and the head transfer function indicates the
degree of such frequency distortion. A three-dimensional solid sound can be reproduced using
such a head related transfer function.
[0021]
The head related transfer function application unit 120 stores the head related transfer function
H11 stored in the head related transfer function database 110 for audio data supplied from an
external sound signal providing device (not shown) and input through an input channel. , H12,
H21, H22, H31, H32 are applied. As a result, the left virtual sound source and the right virtual
sound source are generated. Hereinafter, only three input channels are illustrated for
simplification of the drawing, and thereby six head transfer functions are illustrated. However,
the scope of the present invention is not limited to the number of input channels and the number
of head related transfer functions.
09-05-2019
7
[0022]
The head related transfer functions H11, H12, H21, H22, H31, and H32 in the head related
transfer function application unit 120 are the left head related transfer functions H11 applied
when generating a sound source for output to the left speaker 210. , H21, and H31, and headrelated transfer functions H12, H22, and H32 applied to generate a sound source for output to
the right speaker 220.
[0023]
The first combining unit 130 includes a first left combining unit 131 and a first right combining
unit 133.
The first left side combining unit 131 combines the left virtual sound sources output from the
left head transfer functions H11, H21, and H31 to generate a left combined virtual sound source,
and the first right side combining unit 133 generates the right side head The virtual sound
sources on the right side output from the transfer functions H12, H22, H32, H42, and H52 are
synthesized to generate a synthetic virtual sound source on the right side.
[0024]
The first band pass filter 140 receives inputs of the synthesized virtual sound source on the left
and the synthesized virtual sound source on the right output from the first left synthesis unit 131
and the right synthesis unit 133, respectively. Only the correction target region of the input
synthesized virtual sound source passes through the second band pass filter 160. Only the
correction target passes through the second band pass filter 160 among the input synthesized
virtual sound sources on the right side. Therefore, only the correction target area that has passed
through the second band pass filter 160 is output to the actual listening space characteristic
correction unit 170 among the left and right synthesized virtual sound sources. However, the
filtering process by the second band pass filter 160 is an optional matter.
[0025]
09-05-2019
8
The actual listening environment characteristic function database 150 stores the characteristic
functions of the actual listening environment. The actual listening environment characteristic
function means a characteristic function calculated by measuring an impulse signal generated by
a speaker by the operation of the listener 1000 at the listening position of the listener 1000.
Thereby, the characteristics of the speakers 210 and 220 are considered in the actual listening
environment characteristic function, and the characteristics of the actual listening space 200 are
considered. That is, the characteristic of the listening environment is a concept in which all the
characteristics of the listening space and the characteristics of the speaker are taken into
consideration. The characteristics of the actual listening space 200 are defined by the size of the
place (for example, a room, a living room) where the sound reproducing apparatus 100 is placed,
and the horizontal / vertical pattern. Such a characteristic function of the actual listening
environment can be used from now on as long as it is initially measured once as long as the
position and location of the sound reproducing apparatus 100 are not changed. Also, the actual
listening environment characteristic function can be measured via an external input device such
as a remote control.
[0026]
The second band pass filter 160 extracts only the part for the initial reflection sound by the real
listening environment characteristic function stored in the real listening environment
characteristic function database 150. The actual listening environment characteristic function is
classified as a portion for direct transmission sound and reflection sound, and a portion for
reflection sound is classified as a portion for direct reflection sound, initial reflection sound, and
back reflection sound. The second band pass filter 160 according to the present invention
extracts only the portion for the initial reflected sound. This is because the effect of the early
reflections on the actual listening space 200 is the largest, so only the portion corresponding to
this is extracted.
[0027]
The real listening space characteristic correction unit 170 corrects the correction target region of
the synthesized left / right virtual sound source from the first band pass filter 140 so as to
correspond to the real listening space 200, and passes the second band pass filter 160. The
correction is made based on the portion of the actual listening environment characteristic
function for the initial reflection sound. The correction is for eliminating the characteristic of the
actual listening space 200 so that the listener 1000 can always listen to the sound in the
optimum listening space from the actual listening space characteristic correction unit 170.
09-05-2019
9
[0028]
The second combining unit 180 includes a second left combining unit 181 and a second right
combining unit 183. The second left side synthesis unit 181 corrects the correction target area of
the left synthesized synthesized virtual sound source corrected from the real listening space
characteristic correction unit 170 and the remaining area of the left synthesized sound source
that can not pass through the second band pass filter 160. Synthesize An acoustic signal from the
synthesized final left virtual source is provided to the listener 1000 via the left speaker 210.
[0029]
The second right side synthesis unit 183 corrects the correction target area of the synthesized
virtual sound source on the right corrected from the real listening space characteristic correction
unit 170 and the remaining area of the synthesized virtual sound source on the right that can not
pass through the second band pass filter 160 Synthesize The acoustic signal from the
synthesized right final virtual source is provided to the listener 1000 via the right speaker 220.
By this, in the final virtual sound source according to the present embodiment, it is possible to
correct the characteristic with respect to the actual listening space 200 and provide the listener
1000 with the sound in which the characteristic of the actual listening space is reflected.
[0030]
FIG. 2 is a block diagram showing a sound reproducing apparatus according to another
embodiment of the present invention, and relates to the sound reproducing apparatus for
correcting the characteristics of the speakers 210 and 220. As shown in FIG. The sound
reproduction device 300 according to the present embodiment includes a head related transfer
function database 310, a head related transfer function application unit 320, a first combining
unit 330, a band pass filter 340, a real listening environment characteristic function database
350, and a low pass filter. And 360, a speaker characteristic correction unit 370 and a second
combining unit 380.
[0031]
The head transfer function database 310, the head transfer function application unit 320, the
09-05-2019
10
first synthesis unit 330, and the actual listening environment characteristic function database
350 according to the embodiment shown in FIG. The description is the same as the description
for the database 110, the head-related transfer function application unit 120, the first synthesis
unit 130, and the actual listening environment characteristic function database 150, except for
the explanation thereof. Describe.
[0032]
The low pass filter 360 according to the present embodiment extracts only the sound directly
transmitted based on the actual listening environment characteristic function stored in the actual
listening environment characteristic function database 350.
This is because the direct transmitted sound has the largest effect on the speaker, so only that
portion is extracted. The band pass filter 340 receives the input of the synthesized virtual sound
source on the left and the synthesized virtual sound source on the right output from the first left
synthesis unit 331 and the first right synthesis unit 333 respectively. Only the correction target
region of the input synthesized virtual sound source passes through the band pass filter 340.
Only the correction target area of the input synthesized virtual sound source passes through the
band pass filter 340. Therefore, of the left / right synthesized virtual sound source, only the
correction target area that has passed through the band pass filter 340 is output to the speaker
characteristic correction unit 370. However, the filtering process by the band pass filter 340 is
an optional matter.
[0033]
The speaker characteristic correction unit 370 corrects the correction target region of the
synthesized left / right virtual sound source output from the band pass filter 340 so as to
correspond to the actual listening space 200, and passes the band pass filter 340. Correction is
performed based on the portion of the function directly transmitted sound. A flat response
characteristic can be obtained from the speaker characteristic correction unit 370 by such
correction. This is to correct distortion of the sound reproduced through the left / right speakers
210 and 220 according to the characteristics of the actual listening environment that the user
listens to. In order to perform such correction, the speaker characteristic correction unit 370
includes four correction filters (S11, S12, S21, and S22). Of the four correction filters, the first
correction filter (S11) and the second correction filter (S12), which are two correction filters,
correct the correction target area in the synthetic virtual sound source on the left output from
09-05-2019
11
the first left synthesis section 331 The third correction filter (S21) and the fourth correction filter
(S22), which are the other two correction filters among the four correction filters, are corrected
among the combined virtual sound sources output from the first right combining unit 133.
Correct the target area. The number of correction filters (S11, S12, S21, S22) is four of the
transmission paths existing between the two left / right speakers 210, 220 and two ears
provided by a person. Therefore, correction filters (S11, S12, S21, S22) are provided to
correspond to the respective transmission paths.
[0034]
For example, among the synthesized virtual sound sources on the left side output from the band
pass filter 340, the correction target area is input to the two correction filters (S11 and S12) to
be corrected, and the synthesized virtual sound source on the right side output from the band
pass filter 340 The correction target area is further input to the two correction filters (S21, 22)
and corrected. The second combining unit 180 includes a second left combining unit 181 and a
second right combining unit 183.
[0035]
The second left synthesis unit 181 receives an input of the virtual sound source corrected by the
first correction filter (S11) and the third correction filter (S21). Further, the remaining area of the
left synthetic virtual sound source excluding the correction target area is input to the second left
synthesis unit 181. The second left synthesis unit 181 synthesizes the input sounds to generate a
final virtual sound source on the left side, and outputs an acoustic signal from the final virtual
sound source to the outside through the left speaker 210.
[0036]
The second right side synthesis unit 183 receives an input of the virtual sound source corrected
by the second correction filter (S12) and the fourth correction filter (S22). Furthermore, the
remaining area excluding the correction target area among the synthesized virtual sound sources
on the right side is input to the combining unit 183 on the second right side. The second right
side synthesis unit 183 synthesizes the inputted sounds to generate a final virtual sound source
on the right side, and outputs an acoustic signal according to that to the outside via the right side
speaker 220.
09-05-2019
12
[0037]
Thereby, the characteristics of the speaker owned by the listener 1000 are corrected for the final
virtual sound source according to the present embodiment, and the listener 1000 can hear the
sound excluding the speaker characteristics owned by the listener 1000. . FIG. 3 is a block
diagram showing an audio reproduction apparatus according to a further embodiment of the
present invention, which performs correction to make it possible to recognize that the user can
listen in the optimal listening space, but performs audio reproduction for correcting the previous
channel. It relates to the device.
[0038]
The sound reproduction device 400 according to the present embodiment includes a head
related transfer function database 410, a head related transfer function application unit 420, a
combining unit 430, a virtual listening space parameter storage unit 440, and a virtual listening
space correction unit 450. The description of the head related transfer function database 410
and the head related transfer function application unit 420 of the embodiment according to FIG.
3 is the same as the description of the head related transfer function database 110 and the head
related transfer function application unit 120 according to the embodiment according to FIG.
Therefore, except for the explanation thereof, the characteristic description only in the present
embodiment will be described below.
[0039]
The virtual listening space parameter storage unit 440 stores parameters of the predicted
optimal listening space. In addition, the parameter of the optimal listening space anticipated is,
for example, meaning the parameter corresponding to the atmospheric absorption degree, the
reflectivity, and the size of the virtual listening space 500, and is set by non-real time analysis
(non real time analysis). The virtual listening space correction unit 450 corrects the virtual sound
source using each parameter stored in the virtual listening space parameter storage unit 440.
That is, correction is performed so as to recognize that the listener 1000 listens to sound in a
virtual listening environment regardless of the listening environment. This is needed because of
the current technical limitations of defining sound images using an anechoic chamber measured
head transfer function. The virtual listening space 500 means an ideal listening space, for
09-05-2019
13
example, a recording space applied at the time the sound was first recorded.
[0040]
Therefore, the virtual listening space correction unit 450 provides each parameter to the left
combining unit 431 and the right combining unit 433 of the combining unit 430, and the left
combining unit 431 and the right combining unit 433 are each configured with the left and right
combined virtual sound sources. The parameters are respectively synthesized to generate a final
left / right virtual sound source. The generated acoustic signals from the left and right virtual
sound sources are output to the outside through the left and right speakers 210 and 220.
[0041]
By this, the listener 1000 can feel as if to hear the sound in the optimal virtual listening space
500 by the final virtual sound source according to the present embodiment. FIG. 4 is a block
diagram showing the sound reproduction apparatus according to the embodiment, and relates to
the sound reproduction apparatus that corrects only the front channel although it makes
corrections to make it recognize that the user is listening in the optimum listening space.
[0042]
The explanation of the head related transfer function database 510 and the head related transfer
function application unit 520 of the embodiment shown in the figure is the description of the
head related transfer function database 110 and the head related transfer function application
unit 120 of the embodiment according to FIG. The virtual listening space parameter storage unit
540 of the embodiment according to FIG. 4 is the same as the explanation for the virtual listening
space parameter storage unit 440 of the embodiment according to FIG. . Hereinafter, the
characteristic description only in the present embodiment will be described.
[0043]
The embodiment in FIG. 4 differs from the embodiment in FIG. 3 in that correction is made so as
to recognize that the user listens in the optimum listening space, but using a method of applying
09-05-2019
14
each parameter only to the front channel It is in. The reason why each parameter is applied only
to the front channel is that, in the case where a virtual sound source is potentialed in front of the
listener 1000 using a head-related transfer function, the listener 1000 correctly recognizes the
directivity of the sound source. Although this is possible, it is because there is no feeling of
expansion of sound (surround effect) when localization is performed by the head related transfer
function. Therefore, in order to compensate such a problem, each parameter is applied only to
the front channel so that the listener 1000 can recognize the sense of expansion of sound to the
virtual sound source localized forward by the head related transfer function.
[0044]
The virtual listening space correction unit 550 according to the present embodiment reads the
virtual listening space parameters stored in the virtual listening space parameter storage unit
540 and applies this to the combining unit 530. The synthesis unit 530 according to the present
embodiment includes a left final synthesis unit 531 and a right final synthesis unit 533.
Furthermore, the synthesis unit 530 includes a left intermediate synthesis unit 535 and a right
intermediate synthesis unit 537.
[0045]
Of the audio data input to the front channel (INPUT 1, INPUT 2), the audio data input to the left
head transfer function (H11, H21) passes through the left head transfer function (H11, H21) and
then the left final synthesis unit It is output to 531. Furthermore, of the audio data input to the
front channel (INPUT 1, INPUT 2), the audio data input to the right head transfer function (H12,
H22) passes through the right head transfer function (H12, H22) and then the right end final It is
output to the combining unit 533.
[0046]
On the other hand, of the audio data input to the rear channel (INPUT 3), the audio data input to
the head related transfer function (H31) on the left passes through the head related transfer
function (H31) on the left as a virtual sound source on the left It is output to the intermediate
synthesis unit 535. Furthermore, of the audio data output from the rear channel (INPUT 3), the
audio data input to the head related transfer function (H32) on the right side passes through the
head related transfer function (H32) on the right side as a virtual sound source on the right side.
09-05-2019
15
The result is output to the right intermediate combining unit 537. Although only the back
channel (INPUT 3) is illustrated in the drawing for simplification of the drawing, the number of
back channels may be two or more.
[0047]
The left / right intermediate synthesis units 535 and 537 synthesize the left and right virtual
sound sources input from the rear channel (INPUT 3), respectively. The left synthesized virtual
sound source synthesized by the left intermediate synthesizing unit 535 is output to the left final
synthesizing unit 531, and the right synthesized virtual sound source synthesized by the right
intermediate synthesizing unit 537 is output to the right final synthesizing unit 533. . The left /
right final combining units 531 and 533 are virtual sound sources output from the left / right
intermediate combining units 535 and 537, virtual sound sources and virtual listening spaces
output directly from head related transfer functions (H11, H12, H21, H22) Synthesize
parameters. That is, the virtual sound source output from the left intermediate synthesis unit 535
is synthesized by the left final synthesis unit 531, and the virtual sound source output from the
right intermediate synthesis unit 537 is synthesized by the right final synthesis unit 537.
[0048]
The sound signal by the left / right final virtual sound source synthesized by the left / right final
synthesis unit 531, 533 is output to the outside through the left / right speakers 210, 220. FIG. 5
is a block diagram showing an audio reproduction apparatus according to a further embodiment
of the present invention, which performs correction to make it recognize that the user is listening
in the optimal listening space, but reproduces only the back channel. It relates to the device.
[0049]
Descriptions of the head related transfer function database 610 and the head related transfer
function application unit 620 according to the embodiment illustrated in the figure are the same
as the descriptions of the head related transfer function database 110 and the head related
transfer function application unit 120 according to the embodiment illustrated in FIG. The virtual
listening space parameter storage unit 640 in the embodiment according to FIG. 5 is the same as
the explanation for the virtual listening space parameter storage unit 440 in the embodiment
according to FIG. . Hereinafter, the characteristic description only in the present embodiment will
09-05-2019
16
be described.
[0050]
The embodiment in FIG. 5 is different from the embodiment in FIG. 3 in that although correction
is made to make it recognize that the user is listening in the optimal listening space, a method of
applying each parameter only to the rear channel is used. It is. The reason for applying each
parameter only to the rear channel is that, usually, when a virtual sound source is localized
behind the listener 1000 using a head-related transfer function, a virtual sound source with a
front potential due to the characteristics of human cognitive ability There is a risk of confusion.
By applying each parameter only to the back channel to eliminate such confusion, it emphasizes
the cognitive ability of the person's back space, so that the listener 1000 can recognize the
virtual sound source localized to the back, Apply each parameter only to the channel.
[0051]
Virtual listening space correction unit 650 according to the present embodiment reads the
virtual listening space parameters stored in virtual listening space parameter storage unit 640
and applies this to combining unit 630. The synthesis unit 630 according to the present
embodiment includes a left final synthesis unit 631 and a right final synthesis unit 633.
Furthermore, the synthesis unit 630 includes a left intermediate synthesis unit 635 and a right
intermediate synthesis unit 637.
[0052]
Of the audio data input to the front channels (INPUT 1, INPUT 2), the audio data input to the left
head transfer function (H11, H21) passes through the left head transfer function (H11, H21) to
the left It is output to the final synthesis unit 631. Furthermore, of the audio data output from
the front channels (INPUT 1, INPUT 2), the audio data input to the right head transfer function
(H12, H22) passes through the right head transfer function (H12, H22). , And output to the right
final synthesis unit 633.
[0053]
09-05-2019
17
On the other hand, of the audio data input to the rear channel (INPUT 3), the audio data input to
the head related transfer function (H31) on the left passes through the head related transfer
function (H31) as a virtual sound source on the left It is output to the left intermediate synthesis
unit 635. Furthermore, of the audio data output from the rear channel (INPUT 3), the audio data
input to the head related transfer function (H32) on the right side passes through the head
related transfer function (H32) on the right side as a virtual sound source on the right side. It is
output to the right intermediate combining unit 637. In the present drawing, only (INPUT 3) is
shown as the rear channel for simplification of the drawing, but the number of rear channels may
be two or more.
[0054]
The left / right intermediate synthesis units 635 and 637 synthesize the left and right virtual
sound sources and virtual listening space parameters input from the rear channel (INPUT 3),
respectively. The left synthesized virtual sound source synthesized by the left intermediate
synthesis unit 635 is output to the left final synthesis unit 631, and the right synthesized virtual
sound source synthesized by the right intermediate synthesis unit 637 is output to the right final
synthesis unit 633. .
[0055]
The left / right final combining units 631 and 633 combine the virtual sound source output from
the left / right intermediate combining units 635 and 637 and the virtual sound source output
directly from the head related transfer function. The sound signal by the left / right final virtual
sound source synthesized by the left / right final synthesis unit 631, 633 is output to the outside
through the left / right speakers 210, 220. FIG. 6 is a flowchart for explaining the sound
reproducing method according to the present invention.
[0056]
Based on FIG. 1, FIG. 2, FIG. 3 and FIG. 6, first, when audio data is input through the input
channel (S700), the input audio data is the head transfer function (H11, H11, It is input to H12,
H21, H22, H31, H32) and applied (S710). The left / right virtual sound sources output from the
left / right head transfer functions (H11, H12, H21, H22, H31, H32) are synthesized according to
09-05-2019
18
the left / right head transfer functions, and are preset virtual It is synthesized including the
listening space parameter. That is, the virtual listening space parameter is applied to correct the
left / right virtual sound sources (S720).
[0057]
Further, the corrected virtual sound source is synthesized according to the preset speaker
characteristic function and the left / right head related transfer functions and corrected
according to the speaker characteristic (S730). Note that the speaker characteristic function
means a characteristic function having only the characteristic for the speaker characteristic.
Therefore, the real listening environment characteristic function presented above may be applied.
On the other hand, the virtual sound source whose speaker characteristics have been corrected is
synthesized according to the real listening space characteristic function and the head transfer
function on the left / right, and the real listening space characteristic is corrected (S740). The
real listening space characteristic function means a characteristic function having a property
with respect to the real listening space characteristic. Thus, the real listening environment
characteristic function presented above may be applied.
[0058]
As described above, the virtual sound source corrected in steps S720, S730, and S740 is output
to the listener 1000 via the left / right speaker 220 (S750). However, the steps S720, S730, and
S740 can be performed in any order. Although the preferred embodiments of the present
invention have been illustrated and described based on the drawings, the scope of protection of
the present invention is not limited to the above-described embodiments, and the invention
described in the claims and equivalents thereof It extends to the thing.
[0059]
FIG. 1 is a block diagram showing an audio reproduction apparatus according to an embodiment
of the present invention, and is a diagram related to an audio reproduction apparatus that
corrects characteristics of an actual listening space. FIG. 17 is a block diagram showing an audio
reproduction device according to another embodiment of the present invention, and is a diagram
related to the audio reproduction device that corrects the characteristics of the speakers 210 and
220. FIG. 6 is a block diagram showing an audio reproduction apparatus according to a further
09-05-2019
19
embodiment of the present invention, which relates to an audio reproduction apparatus that
corrects the front channel but makes corrections so as to recognize that the user is listening in
the optimal listening space. FIG. FIG. 10 is a block diagram showing an audio reproduction
apparatus according to a further embodiment of the present invention, wherein the audio
reproduction apparatus corrects only the front channel while making corrections so as to
recognize that the user is listening in the optimal listening space. FIG. FIG. 6 is a block diagram
showing an audio reproduction apparatus according to a further embodiment of the present
invention, wherein the audio reproduction apparatus corrects only the back channel, although
making corrections to make it recognize that the user is listening in the optimal listening space.
FIG. It is a flow chart for explaining the sound reproduction method concerning the present
invention.
Explanation of sign
[0060]
DESCRIPTION OF SYMBOLS 100 sound reproduction apparatus 110 head related transfer
function database 120 head related transfer function application part 130 1st synthetic ¦
combination part 140 1st band pass filter 150 database of real listening environment
characteristic function 160 2nd band pass filter 170 real listening space characteristic correction
part 180 Second synthesis unit
09-05-2019
20
1/--страниц
Пожаловаться на содержимое документа