close

Вход

Забыли?

вход по аккаунту

?

DESCRIPTION JPH0715786

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 JPH0715786
[0001]
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an
automatic volume control method in an audio system, and in particular, a microphone provided
in a sound reproduction space detects a sound, and a music signal in the microphone detection
signal with respect to a music signal output from an audio source. The present invention relates
to an automatic volume control method for calculating an estimated magnification of to control
the volume of an audio system.
[0002]
2. Description of the Related Art Unlike inside a house, the interior of a car is always filled with
noises such as engine noise and wind noise. For this reason, when listening to the reproduction
sound of a radio broadcast, a cassette tape, a compact disc or the like in a car, volume setting is
performed according to the size of the noise. That is, when the noise is large, the volume is set to
be larger than in the quiet running state. However, the magnitude of the noise constantly
changes and increases or decreases. Therefore, it is necessary for the driver to adjust the volume
according to the magnitude of the noise, which causes a problem that the operation is
troublesome and it is not preferable for safe driving. For this reason, conventionally, an audio
system provided with an auto volume function has been proposed which automatically adjusts
the volume by estimating the amount of noise or the like.
[0003]
10-04-2019
1
The first method is to detect the low frequency level of the music signal in consideration of low
level, detect the low frequency level as noise, and adjust the volume automatically based on the
low frequency level. That is, a low frequency component of 50 Hz or less in the sound
reproduction space is detected, and if the level of the low frequency component is large, it is
judged that the noise is large, the volume correction amount is made large and the volume is
made large. If the level is low, it is determined that the noise is small, and the volume correction
amount is reduced. Although this method is simple, the low frequency band below 50 Hz is not a
frequency band that human beings feel loud. In other words, the first method has a problem of
lack of certainty because the volume control is not performed for the midrange level where the
noise is most annoying.
[0004]
The second method is to make the car travel at a predetermined speed with music stopped, and
measure and store the low, middle and high frequency characteristics of noise in the passenger
compartment at that time. When traveling while listening to music in such a state, the
microphone detects the sound in the vehicle compartment to obtain the low, mid and high levels
of the vehicle interior. Then, the low-pass level is compared with the low-pass level of noise
previously stored to determine the magnification η, and the noise level of the mid-pass and highpass noise levels stored in the same way is acoustically Estimated as the mid and high frequency
noise levels in the playback space. Further, the mid-range and high-range levels of the music
signal in the vehicle compartment are estimated by subtracting the mid-range and high-range
levels estimated from the mid-range and high-range levels detected by the microphone. After
that, the ratio between the estimated noise level and the estimated music signal level in the
middle and high bands is calculated, and the volume of the reproduced sound is controlled in
consideration of each ratio (it may be considered only in the ratio of the middle band). According
to this method, the volume control can be performed in consideration of the midrange level
where the noise is most annoying. However, it is necessary to measure in advance the frequency
characteristics of noise in the passenger compartment, and the type of car in the car changes,
and the frequency characteristic of noise in the room changes due to the number of passengers
in the car, the seating position of the passenger, the opening and closing of the window, etc. If
you do, there is a problem that you can not follow in real time.
[0005]
10-04-2019
2
Therefore, a third method has been proposed to solve the problems of the first and second
methods (Japanese Patent Application No. 4-124863). In this third method, the microphone
provided in the sound reproduction space detects the sound in the sound reproduction space, the
microphone detection signal is X (t), and the music signal output from the audio source is S (t), S
(t) Assuming that the estimated magnification of the music signal in the microphone detection
signal X (t) with respect to t) is β and μ is a predetermined constant, the following equation βi
+ 1 = βi + μ {dX (t) / dt−βidS (t) / dt} DS (t) / dt where i = 0, 1, 2... Β 0 is the music in the
microphone detection signal by repeating the calculation of the recurrence formula of β
previously obtained until β converges to a constant value The signal component .beta.X (t) is
determined, and X (t)-. Beta.X (t) is calculated using the .beta.X (t) to determine the noise
component N (t) in the microphone detection signal. Then, volume control is performed based on
β and the noise level N (t).
[0006]
According to such an automatic volume control method, it is not necessary to measure in
advance the frequency characteristics of noise in the sound reproduction space, and moreover, it
is possible to change the frequency characteristics of the noise in the sound reproduction space.
It becomes possible to control the volume in real time. However, since the calculation amount of
the recurrence formula is large and the method of determining the volume control amount based
on β after convergence and the noise level N (t) is also complicated, the burden of control
processing becomes large and complete real time Realizing control was difficult. From the above,
it is an object of the present invention to provide an automatic volume control method capable of
complete real-time processing with less burden on control processing.
[0007]
According to the present invention, in the present invention, a microphone for detecting sound is
provided in an audio reproduction space, a microphone detection signal is X (t), and a music
signal output from an audio source is S. (t), the estimated magnification of the music signal in the
microphone detection signal X (t) with respect to S (t) is β, μ is a predetermined constant, and
the following equation using the estimated magnification β previously obtained, β + μ {dX (t ) /
dt-βdS (t) / dt} · dS (t) / dt to calculate a new estimated magnification β, and the difference
between the previously calculated estimated magnification β and the newly calculated estimated
magnification β This is achieved by providing a means to increase or decrease the level of the
audio signal based on the volume control.
[0008]
10-04-2019
3
According to the present invention, the microphone provided in the sound reproduction space
detects the sound in the sound reproduction space, the microphone detection signal is X (t), and
the music signal output from the audio source is S (t), The estimated magnification of the music
signal in the microphone detection signal X (t) with respect to S (t) is β, μ is a predetermined
constant, and the estimated magnification β obtained in the previous time is used to obtain the
following equation β + μ {dX (t) / dt−βdS (t) / dt} · dS (t) / dt is calculated to obtain a new
estimated magnification β, and the level of the audio signal is calculated based on the difference
between the estimated magnification β obtained last time and the estimated magnification β
newly obtained this time Increase or decrease volume control.
In this way, it is possible to obtain a new estimated magnification β in a single calculation, and
to increase or decrease the level of the audio signal based on the difference between the previous
and current estimated magnification β to perform volume control. The processing load can be
greatly reduced, and automatic volume control can be performed completely in real time in
response to changes in noise.
[0009]
1 is a block diagram of a car audio system embodying an automatic volume control method
according to the present invention. 11 is an audio source such as a CD player, 12 is an electronic
volume that electronically controls the volume according to a volume value V determined by a
volume control unit described later, 13 is an amplifier that amplifies the music signal after
volume adjustment with an electronic volume A speaker 14 is driven by an amplifier and outputs
sound, 15 is a car room as an example of a sound reproduction space, 16 is a microphone for
detecting a sound (including music and noise) in the car interior, 17 is a constant microphone
detection signal A level converter that amplifies at a gain and performs full-wave rectification. 18
is a delay circuit that delays the music signal output from the audio source by a fixed time τ. 19
amplifies the delayed music signal at a fixed gain. A level conversion unit for full wave
rectification, 20 receives the microphone detection signal X (t) from the level conversion unit 17
and the music signal S (t) from the level conversion unit 19 Change the estimated magnification
β of the music signal in the microphone detection signal X (t) with respect to S (t), determine the
volume value by correcting the volume value based on the change, and determine the electronic
volume It is a sound volume control unit which performs sound volume control by outputting to
12. The delay time τ in the delay circuit 18 is set to the time required for the music signal to
pass through the path of the electronic volume 12, the amplifier 13, the speaker 14, the vehicle
compartment 15, and the microphone 16.
10-04-2019
4
[0010]
Next, the calculation method of the estimated magnification by the volume control unit 20 and
the determination method of the volume correction amount will be described. Calculation method
of the estimated magnification β The signal detected by the microphone 16 includes not only
music components but also noise components generated in the vehicle compartment 15 and
noise components generated outside the vehicle and entering the vehicle compartment 15 . The
music signal (signal after level conversion) detected by the microphone 16 is αS (t) (α is a
positive constant) in consideration of the characteristics of the amplifier 13 and the speaker 14
and the acoustic characteristics of the passenger compartment 15. Therefore, assuming that the
noise signal is N (t), the microphone detection signal X (t) can be expressed as the following
equation X (t) =. Alpha.S (t) + N (t) (1).
[0011]
The music signal (signal after delay and level conversion) S (t) output from the audio source 11 is
multiplied by a coefficient β to generate a signal βS (t), and this signal βS (t) and the
microphone detection signal X are generated. If (β) closer to α can be obtained further than (β)
by using (t), then βS (t) becomes closer to the music signal component in the microphone
detection signal X (t), and β is S The estimated magnification of the music signal component in
the microphone detection signal X (t) with respect to (t). α is a component of the tangent slope
of the microphone detection signal X (t), and β is a component of the tangent slope of the signal
βS (t) (referred to as an estimated music signal). Therefore, if the inclination of the latter tangent
is made closer to the inclination of the former tangent, β will be closer to α, and as a result, the
estimated music signal βS (t) can be made closer to the music signal component in the
microphone detection signal . Therefore, the differential value of the microphone detection signal
X (t) and the differential value of the estimated music signal βS (t) based on β so far are
obtained, and β is corrected so that the difference therebetween becomes small.
[0012]
From the above, the calculation method of the estimated magnification β is as follows. If the
product of the microphone detection signal after level conversion is X (t) and the product of the
estimated music signal βS (t) is Y (t), the slopes x (t) and y (t) of each function at any time are x
10-04-2019
5
(t) = dX (t) / dty (t) = dY (t) / dt. The difference between the two is expressed as Z (t) = x (t) -y (t) E
(t) = Z2 (t) = x2 (t) -2x (t) y (t) + y2 (t) .. (2) Then, E (t) is a quadratic function of x (t) and y (t).
Making the level of the estimated music signal Y (t) (= βS (t) close to the level of the music signal
(= αS (t)) detected by the microphone 16 means making Z (t) close to zero, This is nothing but
changing the parameter so that the value of E (t) becomes smaller.
[0013]
Since only β can be changed by the parameter related to E (t) and β is a parameter related to y
(t), the amount of change of E (t) with respect to the amount of change of β is dE (t) / dβ = ( dE
(t) / dy (t). (dy (t) /d.beta.)... (3). E (t) is a quadratic function with β as a parameter, and dE (t) /
dβ represents the slope of E (t) at any β. To reduce the difference between the level of the
estimated music signal (= βS (t)) and the level of the music signal (= αS (t)) actually detected by
the microphone means that E (t) becomes smaller Since it is a requirement, it is sufficient to
change β in the direction in which dE (t) / dβ becomes zero. That is, assuming that the previous
β value is βi and the new β value is βi + 1, the β value is changed by βi + 1 = βi−μ0 (dE
(t) / dβ) (4) We can obtain β, which makes E (t) smaller. In equation (4), μ 0 is a positive
constant.
[0014]
Now, dE (t) / dy (t) and dy (t) / dβ of the right side of the equation (3) are dE (t) / dy (t) = − 2x
(t) +2 y (t) = − 2 { dX (t) / dt- [beta] dS (t) / dt} dy (t) / d [beta] = dS (t) / dt. Therefore, dE (t) / dβ
= -2 {dX (t) / dt-βdS (t) / dt} · (dS (t) / dt) (5). Substituting the equation (5) into the equation (4),
β i + 1 = β i +2 μ 0 {dX (t) / dt−β i dS (t) / dt} · (dS (t) / dt). If 2μ0 is replaced by μ, then βi
+ 1 = βi + μ {dX (t) / dt−βidS (t) / dt} · (dS (t) / dt) (6). The updated estimated magnification
can be obtained by updating the β value by the equation).
[0015]
Determination of Volume Correction Amount The change of the estimation magnification β is
obtained by the following equation: Ri + 1 = βi + 1−βi (7) When the estimated magnification
increases, since βi + 1> βi, Ri + 1> 0. Conversely, when the estimated magnification decreases,
since βi + 1 <βi, Ri + 1 <0. The estimated magnification obtained by the equation (6) becomes
maximum when there is no noise, and decreases as the amount of noise increases. Since the
10-04-2019
6
volume value in the electronic volume 12 should increase or decrease in proportion to the
increase or decrease of noise, when Ri + 1> 0, the correction amount of the volume value is
−ΔV, conversely, Ri + 1 When <0, the correction amount of the volume value is ΔV (where ΔV
is a constant positive value). If Ri + 1 = 0, then the correction amount = zero. Once the correction
amount has been determined in this manner, the volume value V previously set by the electronic
volume 12 is corrected and reset to the electronic volume 12 so that the automatic volume
control following the change in noise is performed. It becomes possible.
[0016]
Overall Operation FIG. 2 is a flow chart showing control processing of the volume control unit
20, which will be described below with reference to FIG. After power on, the volume control unit
20 initializes the volume value VLT, which has been set for the electronic volume 12 immediately
before power off last time, to the electronic volume 12 as an initial volume value V (step 101 in
FIG. 2). The music signal output from the audio source 11 is delayed by τ by the delay circuit 18
and then level-converted by the level converter 19 and output to the volume controller 20 as the
music signal S (t) while the electronic volume 12 → amplifier 13 → output from the speaker 14
into the passenger compartment 15 The sound in the passenger compartment 15 is picked up by
the microphone 16 and then level-converted by the level conversion unit 17 and output to the
volume control unit 20 as a microphone detection signal X (t).
[0017]
The volume control unit 20 first sets the estimated magnification β to an appropriate initial
value β 0 according to the volume value VLT (step 102 in FIG. 2), and after i = 0 (step 103), the
level conversion unit 17 , And the microphone detection signal X (t) and the music signal S (t)
from 19 (step 104), the calculation of the equation (6) is performed once to obtain the first
estimated magnification β1, and the built-in memory (not shown) ) (Step 105). Next, the change
R1 of the estimated magnification .beta.1 obtained this time with respect to the estimated
magnification .beta.0 is calculated using the equation (7) (step 106). Then, the sign of the change
R1 is determined (step 107), and if it is positive, the correction amount is set to -.DELTA.V
assuming that the noise input to the microphone 16 is relatively small, and the set volume up to
that V-.DELTA.V obtained by adding -.DELTA.V to the value V is set as a new volume value V (step
108), and reset to the electronic volume 12 (step 109). As a result, the level of the music signal in
the passenger compartment 15 is reduced.
10-04-2019
7
[0018]
Thereafter, the volume control unit 20 increments i to 1 (step 110), and inputs the microphone
detection signal X (t) and the music signal S (t) from the level conversion units 17 and 19 (step
104). The second estimated magnification β2 is calculated based on the equation (6) and stored
(step 105). Next, the change R2 of the estimated magnification β2 determined this time with
respect to the estimated magnification β1 up to that point is calculated using the equation (7)
(step 106). Then, the sign of the change R2 is determined (step 107). If the sign is still positive,
the correction amount is set to -.DELTA.V assuming that the noise input to the microphone 16 is
still small, and the set volume value V V-.DELTA.V obtained by adding -.DELTA.V to the new
volume value V is set (step 108), and the electronic volume 12 is reset (step 109). As a result, the
level of the music signal in the passenger compartment 15 further decreases.
[0019]
Thereafter, the volume control unit 20 increments i to 2 (step 110), and inputs the microphone
detection signal X (t) and the music signal S (t) from the level conversion units 17 and 19 (step
104). The third estimated magnification β3 is calculated based on the equation (6) and stored
(step 105). Next, the change R3 of the estimated magnification β3 determined this time with
respect to the estimated magnification β2 so far is calculated using the equation (7) (step 106).
Then, the sign of the variation R3 is determined (step 107). Here, if R3 becomes zero, the music
signal in the passenger compartment 15 has a magnitude commensurate with the magnitude of
the noise, so the volume value is not corrected.
[0020]
Thereafter, when the noise in the vehicle compartment 15 becomes large, the newly obtained
estimated magnification βi + 1 becomes smaller than the previously obtained estimated
magnification βi, and the sign of the variation Ri + 1 becomes negative. At this time, assuming
that the noise input to the microphone 16 has increased, the correction amount is ΔV, and V +
ΔV obtained by adding ΔV to the previously set volume value V is set as a new volume value V
(step 111). Are set again (step 109). As a result, the level of the music signal in the passenger
compartment 15 is increased to a level commensurate with the magnitude of the noise. Similarly,
if the sign of the change in estimated magnification is negative, it is determined that noise is
increased, and volume value correction is performed to increase the volume, and conversely,
change in estimated magnification If the minute is positive, it is determined that the noise is
10-04-2019
8
decreasing, and the volume value is corrected to decrease. By reducing the volume, it is possible
to automatically control the volume while completely following in real time the noise change in
the passenger compartment 15.
[0021]
Although the volume correction amount is determined in accordance with the sign of the change
in the estimated magnification in the above-described embodiment, the volume correction
amount may be determined in accordance with the sign and the size of the change. Also,
although a car audio system has been taken as an example, it can be applied to other audio
systems, such as a home audio system, in the same way.
[0022]
As described above, according to the present invention, the microphone provided in the sound
reproduction space detects the sound in the sound reproduction space, the microphone detection
signal is X (t), and the music signal output from the audio source is S ( t), the estimated
magnification of the music signal in the microphone detection signal X (t) with respect to S (t) is
β, μ is a predetermined constant, and the estimated magnification β previously obtained is
dt−βdS (t) / dt} · dS (t) / dt is calculated to obtain a new estimated magnification β, and an
audio signal is calculated based on the difference between the estimated magnification β
obtained last time and the estimated magnification β newly obtained this time Since the volume
control is performed to perform the volume control, it is possible to obtain a new estimated
magnification β by one calculation, and furthermore, the level of the audio signal based on the
difference between the previous and the present estimated magnification β. Volume control,
which greatly reduces the burden of control processing, and Automatic volume control that
completely follows the real-time to changes in the size becomes possible.
10-04-2019
9
Документ
Категория
Без категории
Просмотров
0
Размер файла
20 Кб
Теги
description, jph0715786
1/--страниц
Пожаловаться на содержимое документа