close

Вход

Забыли?

вход по аккаунту

?

JPH1132397

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 JPH1132397
[0001]
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a
space sound reproducing apparatus, and more particularly to a space sound reproducing
apparatus for faithfully reproducing a set space sound environment.
[0002]
2. Description of the Related Art The principle of virtual sound image localization will be
described with reference to FIG. FIG. 2 (a) is a diagram for explaining reproduction by a target
position setting sound source, and FIG. 2 (b) is a diagram for explaining reproduction by a real
sound source. Consider simulating the sound stimulation generated by the sound source installed
at a certain target position in each ear of the listener. In this case, the sound can be localized at
the target position for the listener. It is also possible to make the position of the actual sound
source that actually generates sound different from the position to be localized (see “3DSOUND” DR Begault, AP Professional). For example, as shown in FIG. 2 (b), it is assumed that the
positions of two real sound sources are respectively disposed in front of the listener. In order to
simulate the sound x (t) * hL (t), x (t) * hR (t) by the sound source placed at the target position as
shown in FIG. 2A, the coefficients gL (t), gR A convolution operation of (t) with the sound source
signal x (t) is performed. However, gL (t) = (hRR (t) * hL (t) -hRL (t) * hR (t)) / (hLL (t) * hRR (t) hLR (t) * hRL (t)) gR (t) = (hLL (t) * hR (t) -hLR (t) * hL (t)) / (hLL (t) * hRR (t) -hLR (t) * hRL (t))
where The symbol * indicates a convolution operation and / indicates a deconvolution operation.
Also, coefficients gL (t) and gR (t) respectively indicate the acoustic transfer functions hL (t) and
hR (t) from the sound source located at the target position to each ear of the listener from the
10-05-2019
1
actual sound source to the listener The acoustic transfer functions hLL (t), hRL (t), hLR (t) and
hRR (t) to the ears are shown. Included among them are crosstalks hRL (t) and hLR (t) which are
components that propagate from space from each real sound source to reach the ears on
different sides. In order to simulate the sound faithfully, it has been proposed to adopt a circuit
called a crosstalk canceller (Schroeder, MR and Atal, BS (1963). “Computer
simulation of sound transmission in rooms,”in
IEEE International Convention Record(7). New York:
See IEEE Press).
[0003]
However, in general, the acoustic transfer function depends on the position of the listener and
the sound source. Therefore, in the prior art, the limited positional relationship between the
listener and the actual sound source is a precondition for use. That is, in order to achieve desired
sound image localization, the posture and behavior of the listener are also restricted. In order to
address this problem, attempts have also been made to expand the listening position to achieve
the desired sound image localization. However, the expansion range is limited to a narrow area
equal to or less than the wavelength of the sound wave, and within several cm for high frequency
components of several thousand Hz or more. The expansion range is limited to such a narrow
area because high frequency components with short wavelengths also contribute to sound image
localization.
[0004]
On the other hand, it has been proposed to provide a device for detecting the position and
orientation of the listener in a binaural hearing device that uses sound generation sources close
to the listener's ears as headphones as a real sound source. Here, the input sound is processed
using an acoustic transfer function sequentially synthesized with the detected position and
orientation as a clue. However, this device does not reproduce the sound from the target position.
This is because the acoustic transfer function also differs depending on the individual listener.
[0005]
From the above, it has been difficult to realize desired sound image localization. While
stereotactic control in the lateral direction is robust regardless of the listener, a large number of
10-05-2019
2
listeners often perceive front and rear misjudgement between the in-head position and the target
and localized positions. In order to solve the above problems, it has been proposed to use a
speaker and a headphone in combination as an actual sound source (see Japanese Patent LaidOpen No. 126795). For example, if a speaker is arranged in front of the listener, sound image
localization to the front is ensured. Here, when a plurality of speakers are used as actual sound
sources, the sound image localization position can be changed between the speakers by changing
the intensity ratio of the sound emitted from each speaker. However, it lacks a configuration that
changes sound in response to changes in the position, orientation, and individual differences of
the listener. In this case, the sound image localization to the target position is not guaranteed
because the sound reaching the ears from the target position is not simulated.
[0006]
SUMMARY OF THE INVENTION The present invention provides a spatial sound reproducing
apparatus which solves the above-mentioned problems of faithfully reproducing a preset acoustic
environment on each ear of a listener.
[0007]
SUMMARY OF THE INVENTION A spatial acoustic system having a speaker 12 to which acoustic
signals consisting of left and right two channels L and R are supplied as shown in FIG. 1 and a
sound source 11L and 11R disposed close to each ear of a listener. In the playback apparatus, a
spatial sound playback apparatus is configured that includes an antiphase signal successive
estimation unit that sequentially estimates an antiphase signal that is a contribution of the
speaker 12 among the sound in each ear of the listener.
[0008]
In the above-described spatial sound reproducing apparatus, the antiphase signal sequential
estimation unit acts on the input signal to output adaptive filters 24L and 24R for outputting
signals to the sound generation sources 11L and 11R arranged close to each ear of the listener
and the sound generation. The output of the adaptive filters 24L and 24R from the signal
obtained by subtracting 15L and 15R the signal corresponding to the sound presented by the
sound generation source from the signal recorded by the microphones having the microphones
13L and 13R respectively disposed in the vicinity of the source A spatial sound reproducing
apparatus was configured using the signals obtained by subtracting the signals 26L and 26R as
error signals for operating the adaptive filters 24L and 24R.
[0009]
10-05-2019
3
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A space sound reproducing
apparatus is used as a real sound source in combination with a speaker and headphones and
other sound generation sources arranged close to the listener's ears.
In this case, a speaker is installed in front of or behind the listener.
Here, there is provided a configuration for sequentially estimating the antiphase signal of the
contribution of the speaker in the sound in each ear of the listener.
[0010]
Furthermore, as a configuration for estimating the antiphase signal of the contribution by the
speaker, it has an adaptive filter that acts on the input signal and outputs a signal to a sound
source placed close to each ear of the listener.
Then, microphones are arranged in the vicinity of each ear of the listener. Therefore, this
microphone is placed near the sound source placed close to each ear of the listener. The error
signal for operating the above-mentioned adaptive filter is obtained by further subtracting the
output signal of the above-mentioned adaptive filter from the signal obtained by subtracting the
signal corresponding to the sound presented by the sound source from the signal recorded by
the microphone. Use the signal that has been
[0011]
Here, consider a spatial sound reproducing apparatus that reproduces sound using a single
speaker installed in front of the listener and a pair of headphones as a sound source arranged in
the vicinity of the listener. The process until one input signal consisting of left and right two
channels is supplied to one speaker and one pair of headphones will be described. For the sake of
simplicity, the invention will be described for estimating the anti-phase signal of the loudspeaker
acoustic contribution in the left ear.
[0012]
10-05-2019
4
An input signal of one system is branched into input signals of left and right channels, and is
input to the left intensity adjusting unit or the right intensity adjusting unit, respectively. The
input signal of the left channel input to the left intensity adjustment unit is branched into a
speaker signal and a headphone signal, and supplied as a speaker or a headphone. And, the
present invention is provided with a sequential estimation circuit that sequentially estimates the
antiphase signal of the contribution by the speaker among the sounds presented to the left ear.
The contribution of the sound by the speaker output from the successive estimation circuit is
subtracted prior to the output to the headphones. By subtraction, the signal supplied to the
headphones is canceled. Therefore, the left ear is presented with the same sound as in the case
where the sound is presented only from the headphones. The input signal is cut off on the right
side. Therefore, no sound is presented from the headphones to the right ear, but only from the
speaker.
[0013]
Here, the operation of sequentially estimating the antiphase signal of the contribution of the
speaker among the sounds presented to the listener will be described. However, the case of
estimating the antiphase signal of the contribution by the speaker among the sounds presented
to the left ear will be described. The branched speaker signal is supplied to the above-mentioned
adaptive filter. On the other hand, the speaker signal is subtracted from the headphone signal
and input to a separate filter. The signal passing through this filter is supplied to the headphones.
Also, this signal subtracts a microphone signal which is a signal recorded by a microphone close
to the listener's left ear. In this process, the contribution of the sound presented from the nearby
headphone sound source is removed. Therefore, the contribution by the speaker remains. The
output of the above-mentioned adaptive filter is subtracted from the contribution by this speaker
to obtain an error signal for operating the adaptive filter. Therefore, a value that is the reverse
phase of the transfer characteristic from the speaker to the microphone is estimated as the
coefficient of the above-mentioned adaptive filter. Therefore, an antiphase signal of the
contribution by the speaker at the microphone position is predicted as the output of the adaptive
filter for the input signal.
[0014]
By the way, in human perception, a phenomenon called an advance sound effect is known. That
is, when sounds having correlation with each other are emitted simultaneously from a plurality of
10-05-2019
5
sound sources, the listener localizes the sound in the direction of the ear to which the sound
wave arrives first. In the previous example, delaying the presentation of the sound through the
headphones to the left ear allows the sound to the right ear to be presented first. On top of that,
if the presentation of the sound from the speakers, clues to the localization in the direction of the
speakers placed forward can be obtained regardless of the difference in the listeners. As a result,
the listener can be localized to the right front (preceded sound effect: also referred to as the law
of the first wave front, the hearth effect). "Spatial sound" by Brauert, Morimoto, ed., Goto,
Kashima Publishing Co., Ltd.).
[0015]
Embodiments of the present invention will be specifically described with reference to the
drawings. FIG. 3 is a diagram for explaining the positional relationship between the listener and
the perceived position. FIG. 3 (a) is a view for explaining control based on the difference between
the two ears, and FIG. 3 (b) is a view for explaining the relationship between listener rotation and
perceived position. As shown in FIG. 3 (a), it is known that the position of the perceived sound
image is biased when there is a difference in the intensity of sound stimulation between the left
and right ears ("Spatial sound" by Brauert, edited by Morimoto, Goto; See Kashima Publishing
Association). The degree of displacement of the perceived position is controlled on the basis of
the difference in intensity between the left and right ears or the arrival time difference of sound
waves between the left and right ears. Therefore, the localization ratio can be controlled using
the intensity ratio or the time difference of the sound emitted from the speaker and the
headphone. Here, it is possible to reduce the intensity of the sound presented from the left side
of the headphone to the left ear, or to shift the perceived sound image position to the right as the
delay time to the left ear is increased. If the correspondence between the sound image position,
the intensity ratio of the headphone signal and the speaker signal, and the time difference are
known in advance, sound image position control to the target position can be quantitatively
performed.
[0016]
In the present invention, by presenting the sound from the speaker whose position is fixed, it is
possible to compensate the sound image position to move in the direction opposite to the
movement or rotation of the listener relative to the listener. This will be described with reference
to FIG. 3 (b). Consider the case where the perceived sound image position is on the right side of
the speaker position. When the listener rotates counterclockwise, the speaker whose position is
fixed rotates relatively clockwise with respect to the listener. Here, the intensity or arrival time of
10-05-2019
6
the sound presented from the headphones to the left ear is reduced or delayed compared to the
sound presented from the speaker to the right ear. As well known, the sound image position is
moved to the right more than the speaker position, considering that the sound can be always
localized to the speaker position by presenting the sound to the listener's both ears only from the
speaker. be able to. That is, the compensation of the perceived sound image position
accompanying the movement or rotation of the listener itself is not limited to the case where the
target position is the speaker position.
[0017]
In FIG. 3B, the sound can be localized in a symmetrical direction by exchanging the right and left
in the above-described operation. In addition, the sound image can be localized in the front-rear
symmetrical direction by exchanging the front and back with respect to the listener position at
the speaker position. That is, according to the present invention, sound image localization to a
target position can be realized for the listener in all directions in the horizontal plane.
[0018]
An embodiment of the present invention will be more specifically described with reference to
FIG. In this embodiment, it is assumed that one real sound source speaker 12 is used and signals
of one system of two channels on the left and right are inputted. The embodiment of the space
sound reproducing apparatus shown in FIG. 1 includes a left headphone sound source 11L, a
right headphone sound source 11R, a speaker 12, a left microphone 13L, a right microphone
13R, a left inverse characteristic simulation filter 14L, and a right inverse characteristic
simulation filter 14R. Left subtraction unit 15L, right subtraction unit 15R, left delay unit 16L,
right delay unit 16R, left intensity adjustment unit 21L, right intensity adjustment unit 21R,
addition unit 22, left variable filter 23L, right variable filter 23R, left adaptive filter 24L, right
adaptive filter 24R, left subtraction unit 25L, right subtraction unit 25R, left subtraction unit 26L,
right subtraction unit 26R, left delay unit 27L, right delay unit 27R, left headphone signal 31L,
right headphone signal 31R, left A speaker signal 32L, a right speaker signal 32R, and a target
position control signal 33 are provided as components.
[0019]
The anti-phase signal successive estimation unit that sequentially estimates the anti-phase signal
due to the speaker contribution of the sound stimulation in each listener's ear in the above
configuration has the left microphone 13L indicated by hatching in FIG. Right microphone 13R,
10-05-2019
7
left reverse characteristic simulation filter 14L, right reverse characteristic simulation filter 14R,
left subtraction unit 15L, right subtraction unit 15R, left delay unit 16L, right delay unit 16R, left
variable filter 23L, right variable filter 23R, left The adaptive filter 24L, the right adaptive filter
24R, the left subtraction unit 25L, the right subtraction unit 25R, the left subtraction unit 26L,
and the right subtraction unit 26R.
[0020]
The flow of signals indicated by solid arrows will be described for the above embodiments.
For convenience of explanation, the description will be given focusing on the left side L in the
configuration. The left headphone signal 31L and the left speaker signal 32L are respectively
input to the left intensity delay adjustment unit 21L. The left headphone signal 31L is input to
the left subtraction unit 25L, while the left speaker signal 32L is supplied to the left subtraction
unit 25L and the addition unit 22. The left subtraction unit 25L subtracts the left speaker signal
32L from the left headphone signal 31L, and passes the output thereof to the left variable filter
23L. The signal that has passed through the left variable filter 23L passes through the left
inverse characteristic simulation filter 14L and is supplied to the left headphone sound source
11L and is also allowed to pass through the left delay unit 16L. The left subtraction unit 15L
subtracts the signal output from the left delay unit 16L from the signal recorded by the left
microphone 13L. By the way, the signal for speaker is added between the LRs in the adding unit
22 and supplied to the left delay unit 27L and the speaker 12. The output of the left delay unit
27L is input to the left adaptive filter 24L. The output signal of the left adaptive filter 24L is
input to the left subtraction unit 26L. The left subtraction unit 26L subtracts the signal of the left
adaptive filter 24L from the signal of the left subtraction unit 15L. The signal from the left
subtraction unit 26L is an error signal for operating the left adaptive filter 24L. However, the
filter coefficient in the left inverse characteristic simulation filter 14L is set to the inverse
characteristic of the acoustic transfer function between the left headphone sound source 11L and
the left microphone 13L. The delay time of the left delay unit 16L is set to a delay time at which
the sound reaching the left microphone 13L and the output of the left delay unit 16L are
synchronized. Further, the delay time of the left delay unit 16L is set to be the same as the delay
time of the left delay unit 27L. At this time, the component derived from the left headphone 11L
sound generation source is removed from the recording signal by the left microphone 13L as the
output signal of the left subtraction unit 15L. That is, the contribution from the speaker 12
remains. Accordingly, a difference obtained by subtracting the output signal from the left
adaptive filter 24L from the contribution from the speaker 12 is obtained as the output signal
from the left subtraction unit 26L. This is used as an error signal for the adaptive operation of
the left adaptive filter 24L to sequentially estimate a value -hL which is the opposite phase of the
10-05-2019
8
acoustic transfer function hL from the speaker 12 to the left microphone 13L. This value -hL acts
on the input signal as a filter coefficient in the left adaptive filter 24L, and as a result, the
antiphase signal of the contribution from the speaker 12 is estimated as the output signal. As the
filter coefficient in the left variable filter 23L, the one estimated by the left adaptive filter 24L is
used, so the same filter coefficient -hL is always applied to the input signal to the left variable
filter 23L and the left adaptive filter 24L through the filter coefficient. Acted.
Since the filter coefficient -hL is sequentially estimated in the left adaptive filter 24L, the
listener's left ear has a contribution from the speaker 12 regardless of the listener's position,
orientation change, and individual differences among the listeners. The removed sound is
presented. For the right side R, both the right headphone signal 31R and the right speaker signal
32R are cut off by the target position control signal 33 in the right intensity adjustment unit 21R.
Since no signal flows in each part on the right side, no sound is presented from the headphones
11R. Therefore, in the right ear of the listener, the sound emitted from the speaker 12 supplied
with the left speaker signal 32L is presented.
[0021]
In this case, a clue to control the position of the perceived sound to the right of the position of
the speaker 12 can be obtained. This is because the sound of the speaker presented to the right
ear arrives earlier than the sound of the headphones presented to the left ear. Here, if the
intensity of the right headphone audio signal is increased by delaying or weakening the left
headphone signal 31L than the left speaker signal 32L, the perceived sound image position is
from the front to the right. It can move. Here, the position of the perceived sound image is set by
controlling the intensity of the left headphone signal 31L by the left intensity adjustment unit
21L so that the intensity ratio associated with the target position is given according to the target
position control signal 33. be able to.
[0022]
In the above embodiment, sound image localization to a symmetrical azimuth can be realized by
exchanging the left and right actions. Further, if the supply of the left and right headphone
signals 31L and 31R is cut off for both left and right, only the sound from the speaker 12 is
emitted. In this case, the direction of the localized sound is the speaker position. In addition,
sound image localization to a symmetrical direction can be realized by exchanging the front and
back of the speaker position presenting the left and right sound. As a result, sound image
10-05-2019
9
localization control can be performed over the entire circumference.
[0023]
In general, when a sound is presented only from headphones, the position of the listener, even if
the orientation changes, the perceived sound image position does not change relatively to the
listener. However, in this embodiment, if the position of the speaker is fixed, the movement of the
absolute position of the perceived sound image is suppressed even if the position and orientation
of the listener change. For example, as shown in FIG. 3 (b), the amount of movement of the
perceived absolute position is less than the amount of change of the listener's orientation and
thus the absolute target position by the headphones. That is, the position of the perceived sound
image moves relative to the listener's position and orientation in the direction opposite to the
listener's change, and the acoustic environment in each listener's ears for changes in the
listener's position and orientation is automatic. To be compensated.
[0024]
In the above configuration, if the configuration in which the reverse phase signal of the
contribution by the speaker among the sound stimulation in each ear of the listener is
sequentially estimated and the headphones and the microphone are made to function in parallel
for each listener, The above-described effects of sound image localization can be realized
independently for the listener.
[0025]
The above-described spatial sound reproducing apparatus according to the present invention
estimates the reverse phase signal of the contribution by the speaker in the presented sound
regardless of the position and orientation of the listener and the individual difference of the
listener. .
Thereby, the sound of the speaker alone can be presented to the ear near the speaker, and the
contribution of the sound of the speaker can be sequentially removed from the ear on the
opposite side. Therefore, by positioning the speaker in front of or behind the listener, localization
to a desired target position can be realized regardless of the individual difference of the listener.
The position of the sound image can be compensated and changed relatively to the change of the
position and direction of the listener relative to the position of the listener. That is, when sound
10-05-2019
10
image localization is performed to a desired target position, the restriction on the feasible
listening position is relaxed. In addition, by sequentially estimating the antiphase signal of the
contribution by the speaker among the sound stimulation in each ear of the listener and
providing a sound source for proximity to the listener for each listener, a plurality of listeners
having similar effects Can be realized.
[0026]
In summary, according to the present invention, sound image localization to a desired target
position can be easily realized regardless of changes in the position and orientation of the
listener and individual differences in the listener. In particular, it is possible to avoid front-rear
misjudgment and in-head localization that often occur when binaural hearing.
[0027]
Brief description of the drawings
[0028]
1 is a diagram for explaining an embodiment.
[0029]
2 is a diagram for explaining the principle of virtual sound image localization.
[0030]
3 is a diagram for explaining the positional relationship between the listener and the perceived
position.
[0031]
Explanation of sign
[0032]
11L left headphone sound source 11R right headphone sound source 12 speaker 13L left
microphone 13R right microphone 14L left inverse characteristic simulation filter 14R right
inverse characteristic simulation filter 15L left subtraction portion 15R right subtraction portion
10-05-2019
11
16L left delay portion 16R right delay portion 21L left intensity adjustment Unit 21R Right
intensity adjustment unit 22 Addition unit 23L Left variable filter 23R Right variable filter 24R
Right adaptive filter 25L Left subtraction unit 25R Right subtraction unit 26L Left subtraction
unit 26R Right subtraction unit 27R Left delay unit 27R Right delay unit 31L Signal for left
headphone 31R Signal for right headphone 32L Signal for left speaker 32R Signal for right
speaker 33 Target position control signal
10-05-2019
12
Документ
Категория
Без категории
Просмотров
0
Размер файла
24 Кб
Теги
jph1132397
1/--страниц
Пожаловаться на содержимое документа