close

Вход

Забыли?

вход по аккаунту

?

JP2005198251

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 JP2005198251
PROBLEM TO BE SOLVED: To provide a sphere-based three-dimensional audio signal processing
system and method for simplifying the shape of a human head into a sphere and arranging a
microphone on such a sphere to acquire a three-dimensional audio signal. It is to do. SOLUTION:
In a three-dimensional audio signal processing system using a sphere, acquisition means of a
three-dimensional audio signal for acquiring an audio signal using a predetermined number of
microphones arranged on the sphere, and the three-dimensional audio Post-processing means of
a three-dimensional audio signal for converting an audio signal acquired by the signal acquiring
means to be reproduced in different reproduction environments such as headphones, stereo,
stereo dipole, 4-channel, 5-channel reproduction environment And [Selected figure] Figure 1
Three-dimensional audio signal processing system using sphere and method thereof
[0001]
The present invention relates to a sphere-based three-dimensional audio signal processing
system and method for acquiring a three-dimensional audio signal using a microphone disposed
on a sphere and reproducing the audio signal in various environments.
[0002]
The conventional three-dimensional audio signal acquisition system is mainly based on Binaural
technology, in which a microphone is placed on the ear of a dummy head in the form of a human
to acquire an audio signal, which is then reproduced through headphones. .
[0003]
10-05-2019
1
Such a technique places a microphone in the ear of a dummy head that is actually in the form of
a human to acquire an audio signal, and gives a feeling of being in the place where the sound is
acquired when listening through headphones.
[0004]
However, when reproducing a binaural signal acquired using the dummy head with a speaker,
unlike headphone reproduction, the output of the left speaker can be heard in the right ear and
the output of the right speaker can be heard in the left ear. A talk phenomenon occurs.
In order to remove such crosstalk, various methods have been proposed to design an inverse
filter for crosstalk removal.
[0005]
Recently, a human-shaped dummy head has been simplified into a sphere, and research has been
actively conducted on a system for acquiring a three-dimensional audio signal. However, due to
the characteristics of the sphere, this technique predicts the signal form. As it is possible, it is a
technology that uses a sphere to acquire a three-dimensional audio signal and passes the signal
processing to give it a dummy head-like feeling.
[0006]
Conventional techniques for acquiring a three-dimensional audio signal using a dummy head can
obtain a fairly natural sound because the audio signal is acquired using a dummy head in the
form of a real human.
However, since such conventional techniques slightly vary in head size and shape from one
person to another, an audio signal acquired using a dummy head of a specific shape can not give
satisfactory results to all people.
[0007]
Also, in the prior art as described above, when a binaural signal is reproduced through the
speaker, the signal acquired using the microphone placed on the dummy head's ear is
10-05-2019
2
transmitted through the listener's ear, so the degree of distortion of the signal It gets worse.
[0008]
In addition, the conventional human-shaped dummy head has a problem that recording in a
public place is accompanied with many restrictions because of its size and shape.
[0009]
On the other hand, when judging the direction, humans turn their head little by little left / right.
However, the signal recorded using the dummy head is difficult to determine due to the position
of the fixed ear of the dummy head, and the front signal is judged as the rear signal or the rear
signal is judged as the front signal. Front / back confusion occurs.
[0010]
In addition, since the output of the dummy head is basically a two-channel signal, there is a
problem that it is difficult to expand to multi-channel.
Patent Document 1: JP-A-2000-152372 Patent Document 2: JP-A-2000-354300 Patent
Document 2: U.S. Patent No. 5,778,083 Korea Patent Document No. 2002-087593
[0011]
The present invention has been made in view of the problems of the prior art described above,
and its object is to simplify the shape of the human head into a sphere, and to arrange the
microphone on such a sphere. A sphere-based three-dimensional audio signal processing system
and method for obtaining a two-dimensional audio signal.
[0012]
In addition, another object of the present invention is to simplify the shape of human head to a
sphere, arrange a microphone on such a sphere to acquire a three-dimensional audio signal, and
obtain a variety of acquired three-dimensional audio signals. It is an object of the present
10-05-2019
3
invention to provide a sphere-based three-dimensional audio signal processing system and
method thereof for application to various reproduction systems.
[0013]
To achieve the above object, according to the present invention, in a three-dimensional audio
signal processing system using a sphere, a three-dimensional audio signal for acquiring an audio
signal using a predetermined number of microphones arranged on the sphere Conversion means
for converting an audio signal acquired by the acquisition means of the third aspect and the
audio signal acquired by the acquisition means of the three-dimensional audio signal in different
reproduction environments such as headphones, stereo, stereo dipole, four channels, five
channels reproduction environment And a post-processing unit for the three-dimensional audio
signal.
[0014]
Also, three-dimensional audio signal reproduction for reproducing the audio signal converted by
the post-processing means of the three-dimensional audio signal in various different
reproduction environments such as headphones, stereo, stereo dipole, four channels, five
channels reproduction environment Characterized by further comprising means.
[0015]
On the other hand, the method of the present invention is a three-dimensional audio signal
processing method using a sphere, comprising the steps of acquiring a three-dimensional audio
signal for acquiring an audio signal using a predetermined number of microphones disposed on
the sphere After the 3D audio signal converting audio signal to be reproduced in different
reproduction environments such as headphones, stereo, stereo dipole, 4-channel, 5-channel
reproduction environment, the audio signal acquired in the acquisition step of the audio signal
And processing steps.
[0016]
Also, the method of the present invention reproduces the audio signal converted in the postprocessing step of the three-dimensional audio signal in various different reproduction
environments such as headphones, stereo, stereo dipole, four channels, five channels
reproduction environment. The method further comprises the step of reproducing a threedimensional audio signal.
[0017]
10-05-2019
4
According to the present invention, three microphones arranged on a sphere can be used to
acquire a three-dimensional audio signal, and post-processing through various channels such as
five channels, four channels, headphones, stereo, stereo dipole, etc. Unlike the conventional
dummy head, the microphone arranged in the sphere is very used to the shape, so that it can be
used for acquiring three-dimensional audio signals in public places such as music concerts.
[0018]
Hereinafter, the most preferable embodiment of the present invention will be described with
reference to the attached drawings.
[0019]
FIG. 1 is a block diagram of an embodiment of a three-dimensional audio signal processing
system using a sphere according to the present invention.
[0020]
First of all, three-dimensional audio signal acquisition technology using a microphone placed at
90 ° on both sides of a conventional sphere can, to some extent, delineate the time difference
and level difference between two ears that humans use to sense direction Therefore, threedimensional audio effects can be produced.
However, due to the characteristics of the sphere, the signals incident from the same angle on the
front and back surfaces have the same characteristics, so it is quite difficult to distinguish
between the signals coming from the front and the signals coming from the back. .
[0021]
In the present invention, in order to reduce the anteroposterior confusion phenomenon of the
signal acquired using such a sphere, several microphones are arranged on the sphere to make a
difference between the anteroposterior signal and reduce the anteroposterior confusion
phenomenon. In addition, a system and method are presented for reproducing signals acquired
using the several microphones in various reproduction environments such as five channels, four
channels, headphones, stereo, stereo dipole reproduction environment.
[0022]
10-05-2019
5
As shown in FIG. 1, the 3D audio signal processing system according to the present invention is a
3D audio signal processing system for acquiring audio signals using several (e.g. 5) microphones
arranged on a sphere. Three-dimensional audio for converting an audio signal acquired by the
acquisition unit 110 and the three-dimensional audio signal acquisition unit 110 into various
reproduction environments such as headphones, stereo, stereo dipole, four channels, and five
channels reproduction environment A signal post-processing unit 120 is provided.
In addition, the audio signal converted by the post-processing unit 120 of the three-dimensional
audio signal may be reproduced by the three-dimensional audio signal reproducing unit 130 in
various reproduction environments such as headphones, stereo, stereo dipole, four channels, five
channels reproduction environment. Further equipped.
[0023]
The three-dimensional audio signal acquisition unit 110 arranges five microphones on a sphere
with a simplified human head to acquire a three-dimensional audio signal.
The microphone arrangement on the sphere consists of a central microphone for increasing the
front image and two left and right side microphones for compensating for the movement of the
human head.
[0024]
The post-processing unit 120 of the three-dimensional audio signal reproduces the threedimensional audio signal acquired using the five microphones arranged on the sphere by the
three-dimensional audio signal acquisition unit 110 in various reproduction environments. Postprocess.
Here, post-processing is 5 × 5 crosstalk removal filtering for reproduction using five channels
excluding LFE channels in the conventional 5.1 channel reproduction system, left side excluding
center channel among five channels 4x4 crosstalk removal filtering for playback through right /
left and left surround / right surround speakers, filtering to convert multi-channel signals to 2-
10-05-2019
6
channel signals for headphone playback, and 2-channel signals for headphone playback Includes
2 × 2 crosstalk cancellation filtering for playback in stereo and stereo dipole playback
environments.
[0025]
The three-dimensional audio signal reproduction unit 130 may convert the three-dimensional
audio signal converted to be suitable for the reproduction environment by the post-processing
unit 120 of the three-dimensional audio signal, such as headphones, stereo, stereo dipole, four
channels, five channels reproduction environment Playback of three-dimensional audio signals in
various playback environments.
[0026]
The above-described three-dimensional audio signal processing system according to the present
invention will be described in more detail with reference to FIGS. 2 to 10E below.
[0027]
FIG. 2 is an explanatory diagram of an embodiment for explaining a microphone arrangement of
the three-dimensional audio signal processing system according to the present invention.
[0028]
The three-dimensional audio signal acquisition unit 110 arranges five microphones on the
horizontal plane of the sphere as shown in FIG. 2 to acquire an audio signal.
[0029]
As shown in FIG. 2, the first microphone is located at the center of the sphere to obtain an audio
signal from the front, and the side microphone shakes the head left / right when the human
judges the direction. In order to compensate, it arranges 2 each at an angle of 15 ° each on the
left / right side front / rear.
[0030]
The microphones on the front are the first, the microphones on the left are the second and
fourth, and the microphones on the right are the third and fifth, and the audio signals obtained
using the respective microphones are u1, u2, u3, u4, u5 and so on. Do.
10-05-2019
7
[0031]
The post-processing unit 120 of the three-dimensional audio signal is a post-processing for
reproducing the five microphone output signals u1, u2, u3, u4, u5 from the three-dimensional
audio signal acquisition unit 110 in various reproduction systems. I do.
[0032]
FIG. 3 is a detailed block diagram of an embodiment of a post-processing unit of a threedimensional audio signal in the three-dimensional audio signal processing system according to
the present invention.
[0033]
The operation of the post-processing unit 120 of the three-dimensional audio signal is as follows.
[0034]
First, the speaker input signal of the 5-channel reproduction system eliminates crosstalk between
the output signals u1, u2, u3, u4, and u5 of the three-dimensional audio signal acquisition unit
110 and five speakers and five target points. To generate a convolution operation of the 5 × 5
inverse filter 310.
Here, the center speaker input signal, the left speaker input signal, the right speaker input signal,
the left surround speaker input signal, and the right surround speaker input signal.
[0035]
Five target points mean five horizontal points on the sphere as shown in FIG.
[0036]
FIG. 4 is an explanatory diagram of an embodiment showing a target on a sphere at the time of
five-channel reproduction in the three-dimensional audio signal processing system according to
the present invention.
[0037]
10-05-2019
8
In the case of 5-channel reproduction, the output signal of the center speaker is the first target
point, the output signal of the left speaker is the second target point, the output signal of the
right speaker is the third target point, and the output signal of the left surround speaker is the
fourth target As the output signal of the point, right surround speaker is observed only at the
fifth target point, an inverse filter is used to remove crosstalk between each speaker and the
target point.
[0038]
In order to design such a 5 × 5 inverse filter, place a sphere in the center of five speakers,
generate impulses on each of the five speakers, and measure the responses at five target points
on the sphere. Find the impulse response between the speakers and the five target points.
[0039]
The inverse function of the impulse response is a 5 × 5 inverse filter that removes crosstalk
between the 5 channel regeneration system and the 5 target points.
[0040]
A speaker input signal of a 5-channel reproduction system is generated by convolution operation
of the 5 × 5 inverse filter 310 and the output signals u1, u2, u3, u4, u5 of the three-dimensional
audio signal acquisition unit 110.
[0041]
On the other hand, in order to generate a four-channel reproduction signal, four of the five
output signals u1, u2, u3, u4, and u5 of the three-dimensional audio signal acquisition unit 110
excluding the first microphone output signal u1 The microphone output signals u 2, u 3, u 4 and
u 5 are used to generate four speaker input signals excluding the LFE channel and the center
channel in the 5.1 channel speaker configuration.
[0042]
The speaker input signal of the four-channel reproduction system is for removing crosstalk
between the output signals u2, u3, u4, u5 of the three-dimensional audio signal acquisition unit
110 and four speakers and four target points. It is generated by the convolution operation of the
4 × 4 inverse filter 320.
10-05-2019
9
Here, the left speaker input signal, the right speaker input signal, the left surround speaker input
signal, and the right surround speaker input signal.
[0043]
The four target points mean four points on the horizontal plane on the sphere as shown in FIG.
[0044]
FIG. 5 is an explanatory diagram of an embodiment showing a target on a sphere at the time of
four-channel reproduction in the three-dimensional audio signal processing system according to
the present invention.
[0045]
In the case of 4-channel reproduction, the output signal of the left speaker is the second target
point, the output signal of the right speaker is the third target point, the output signal of the left
surround speaker is the fourth target point, and the output signal of the right surround speaker
is the fifth Use an inverse filter to eliminate crosstalk between each speaker and the target point,
as observed at the target point only.
[0046]
To design such a 4 × 4 inverse filter, place a sphere in the center of the four speakers, generate
impulses on each of the four speakers, and then measure the response at the four target points
on the sphere, Determine the impulse response between the four speakers and the four target
points.
[0047]
The inverse function of the impulse response is a 4 × 4 inverse filter that eliminates crosstalk
between the 4 channel regeneration system and the 4 target points.
[0048]
A speaker input signal of the 4-channel reproduction system is generated by convolution
operation of the 4 × 4 inverse filter 320 and the output signals u 2, u 3, u 4 and u 5 of the 3D
audio signal acquisition unit 110.
10-05-2019
10
[0049]
On the other hand, the headphone reproduction signal is generated by two methods.
[0050]
The first method places the sphere in the center of the 5-channel reproduction system as shown
in FIG. 6, and uses 5 speakers and impulse response to the left / right 90 ° position of the
sphere to input the 5-channel loudspeaker The signal is converted into a two-channel headphone
reproduction signal by a 5 × 2 filter A330.
[0051]
FIG. 6 is an explanatory diagram of an embodiment showing a speaker and a sphere for
generating a headphone reproduction signal in the three-dimensional audio signal processing
system according to the present invention.
[0052]
In FIG. 6, SIR means the impulse response of the sphere, “LT” indicates the 90 ° point on the
left side of the sphere, and “RT” indicates the 90 ° point on the right side of the sphere.
That is, SIRC-LT means an impulse response from the center speaker to LT.
[0053]
After finding the transfer function to LT and RT at the left / right 90 ° position of the sphere
located at the center of five speakers, this transfer function and the signal for 5-channel
reproduction are given by the following [Equation 1] Generate left / right headphone playback
signal using simple convolution operation.
Here, the left headphone signal is the right headphone signal, and “conv” indicates the
convolution operation.
[0054]
10-05-2019
11
Next, the second method of 2-channel signal generation for headphone reproduction is to use a 5
× 2 filter B 340 converted from the impulse response of a sphere.
[0055]
FIG. 7 is an explanatory view of an embodiment showing a filter for generating headphone
signals in the three-dimensional audio signal processing system according to the present
invention, and FIG. 8 is a headphone signal in the three-dimensional audio signal processing
system according to the present invention It is explanatory drawing of embodiment for
demonstrating the production | generation process of.
[0056]
The impulse response of the sphere is measured by generating an impulse while changing the
direction of the speaker by 5 ° in the horizontal plane after placing a microphone at a position
of 0 ° in the horizontal plane of the sphere as shown in FIG.
[0057]
Of the measured impulse responses, an inverse function of the 0 ° response in which the
microphone and the speaker are parallel is determined, and convolution is performed with each
impulse response as in the following [Equation 2], and the determined filter is used. Generate
headphone playback signal.
[0058]
Here, denotes an inverse function of the 0 ° impulse response, denotes an impulse response of
each angular sphere, and “conv” denotes a convolution operation.
[0059]
The output signals u 1, u 2, u 3, u 4 and u 5 of the filter and 3D audio signal acquisition unit 110
obtained by such a method are subjected to convolution operation as shown in the following
[Equation 3] to generate a headphone reproduction signal.
[0060]
Meanwhile, to generate left and right speaker input signals for stereo reproduction, transfer
functions between the stereo speakers shown in FIG. 10D and the left and right 90 ° LT and RT
of the sphere as shown in FIG. 9 are used. Then, crosstalk must be removed from the 2 × 2 filter
350.
10-05-2019
12
[0061]
FIG. 9 is an explanatory diagram of an embodiment showing a target on a sphere at the time of
two-channel reproduction in the three-dimensional audio signal processing system according to
the present invention.
[0062]
After impulses are generated by the left and right speakers of the stereo reproduction system of
FIG. 10D, the measured values of the impulses by the left and right 90 of the sphere located at
the center are LT and RT. Impulse response between
[0063]
The inverse function of the impulse response is an inverse filter that removes crosstalk between
the stereo speaker and the target point (LT and RT) of the sphere.
[0064]
One of the 2-channel headphone reproduction signals A and B is selected, and convolution
operation with the 2 × 2 inverse filter 350 is performed to generate left and right speaker input
signals of the stereo reproduction system.
[0065]
Transfer functions between the stereo dipole reproduction system shown in FIG. 10E and the left
90 ° LT, RT of the sphere as in FIG. 9 to generate left and right speaker input signals for stereo
dipole reproduction. To eliminate crosstalk.
[0066]
After impulses are generated by the left and right speakers of the stereo dipole reproduction
system of FIG. 10E, the values of impulses measured by the left and right 90 of the sphere
located at the center are the loudspeaker and the sphere of the stereo dipole reproduction
system. Impulse response between LT and RT.
[0067]
The inverse function of the impulse response is an inverse filter that removes crosstalk between
the loudspeakers of the stereo dipole reproduction system and the target points (LT and RT) of
10-05-2019
13
the sphere.
[0068]
One of the 2-channel headphone reproduction signals A and B is selected, and convolution
operation with the 2 × 2 inverse filter 360 is performed to generate left and right speaker input
signals of the stereo dipole reproduction system.
[0069]
10A to 10E are detailed configuration diagrams of an embodiment of a three-dimensional audio
signal reproduction unit in the three-dimensional audio signal processing system according to
the present invention.
[0070]
The three-dimensional audio signal reproduction unit 130 reproduces, in the respective
reproduction environments, signals converted through the conversion filter suitable for the
reproduction environment by the post-processing unit 120 of the three-dimensional audio signal.
[0071]
The 5-channel reproduction signal of the post-processing unit 120 of the three-dimensional
audio signal is an input of the 5-channel reproduction system shown in FIG. 10A, and the 4channel reproduction signal is an input of the 4-channel reproduction system shown in FIG. 10B.
[0072]
The headphone reproduction signals A and B are input signals to the headphones shown in FIG.
10C.
[0073]
The stereo reproduction signal is an input signal of the stereo reproduction system of FIG. 10D,
and the stereo dipole reproduction signal is an input signal of the stereo dipole reproduction
system of FIG. 10E.
[0074]
FIG. 11 is a flowchart of an embodiment of a three-dimensional audio signal processing method
10-05-2019
14
according to the present invention.
[0075]
As shown in FIG. 11, first, audio signals are acquired using five microphones arranged on the
sphere (1101), and the acquired audio signals are reproduced in five channels, four channels,
stereo, stereo dipole, headphones, etc. Perform post-processing to reproduce at step 1102.
[0076]
Next, the post-processed converted audio signal is reproduced in an actual reproduction
environment (1103).
[0077]
As described above, the method of the present invention is stored in a recording medium (CD
ROM, RAM, ROM, floppy disk, hard disk, magneto-optical disk, etc.) in a form embodied in a
program and readable by a computer.
[0078]
The present invention is not limited to the above-described embodiment, and various
modifications can be made without departing from the technical concept of the present
invention.
[0079]
FIG. 1 is a block diagram of an embodiment of a three-dimensional audio signal processing
system using a sphere according to the present invention.
It is explanatory drawing of embodiment for demonstrating the microphone arrangement |
positioning of the three-dimensional audio signal processing system which concerns on this
invention d.
FIG. 5 is a detailed block diagram of an embodiment of a post-processing unit of a threedimensional audio signal in the three-dimensional audio signal processing system according to
the present invention.
10-05-2019
15
It is explanatory drawing of embodiment which shows the target on a spherical body at the time
of 5 channels reproduction | regeneration in the three-dimensional audio signal processing
system which concerns on this invention.
It is explanatory drawing of embodiment which shows the target on a spherical body at the time
of 4 channels reproduction | regeneration in the three-dimensional audio signal processing
system which concerns on this invention.
It is an explanatory view of an embodiment showing a speaker and a sphere for generation of a
headphone reproduction signal in a three-dimensional audio signal processing system according
to the present invention.
It is explanatory drawing of embodiment which shows the filter for the production | generation
of a headphone signal in the three-dimensional audio signal processing system which concerns
on this invention.
It is an explanatory view of an embodiment for explaining a generation process of a headphone
signal in a three-dimensional audio signal processing system concerning the present invention.
It is explanatory drawing of embodiment which shows the target on a spherical body at the time
of 2 channel reproduction | regeneration in the three-dimensional audio signal processing
system which concerns on this invention.
FIG. 6 is a detailed block diagram of an embodiment of a three-dimensional audio signal
reproduction unit in the three-dimensional audio signal processing system according to the
present invention.
FIG. 6 is a detailed block diagram of an embodiment of a three-dimensional audio signal
reproduction unit in the three-dimensional audio signal processing system according to the
present invention.
10-05-2019
16
FIG. 6 is a detailed block diagram of an embodiment of a three-dimensional audio signal
reproduction unit in the three-dimensional audio signal processing system according to the
present invention.
FIG. 6 is a detailed block diagram of an embodiment of a three-dimensional audio signal
reproduction unit in the three-dimensional audio signal processing system according to the
present invention.
FIG. 6 is a detailed block diagram of an embodiment of a three-dimensional audio signal
reproduction unit in the three-dimensional audio signal processing system according to the
present invention.
It is a flowchart figure of the embodiment to the three-dimensional audio signal processing
method concerning the present invention.
10-05-2019
17
Документ
Категория
Без категории
Просмотров
0
Размер файла
26 Кб
Теги
jp2005198251
1/--страниц
Пожаловаться на содержимое документа