close

Вход

Забыли?

вход по аккаунту

?

JPH10257582

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 JPH10257582
[0001]
The present invention relates to a method for recording or listening to a sound signal such as a
voice or sound signal generated in a space where there is a reverberation due to a wall reflection
sound etc. in high quality. The present invention relates to a reverberation distortion reduction
microphone device that reduces reverberation distortion included in a received signal such as an
acoustic signal and prevents degradation of clarity.
[0002]
2. Description of the Related Art As a technique for reducing reverberation distortion contained
in speech and the like, there is a technique based on inverse filter processing which cancels the
action of a transfer function from a sound source such as a speaker to a microphone. An example
of such a microphone device of FIG. 2 is shown. The sound signal v (n) (n: discrete time) emitted
from the sound source 1 is received by the two microphones 2a and 2b, subjected to processing
by the inverse filters 3a and 3b, and then added and output by the adder 4 Ru. Let the transfer
functions from sound source 1 to microphones 2a and 2b be z-dg1 (z) and z-dg2 (z), and the
transfer functions of inverse filters 3a and 3b be h1 (z) and h2 (z), respectively. It is known that
h1 (z) and h2 (z) exist which satisfy the following equation.
[0003]
g1 (z) h1 (z) + g2 (z) h2 (z) = z-d (1) where z-d is a delay time common to the above transfer
function, after the sound signal emits a sound source, Represents the time required to reach the
03-05-2019
1
microphone closer to.
[0004]
That is, the output signals z-dg1 (z) v (n), z-dg2 (z) v (n) of the microphones 2a and 2b are
respectively inverse filters 3a and 3b having transfer functions h1 (z) and h2 (z). By processing
and adding after, the reverberation-removed sound signal z-dv (n) is obtained.
[0005]
In the conventional reverberation distortion reduction technique, it is necessary to measure each
transfer function between a sound source and a microphone in order to obtain an inverse filter.
For the measurement, a method such as using a known pseudo-random signal that covers the
frequency band of the sound signal to reduce reverberation distortion is generally used.
However, it is difficult to measure the transfer function when the sound source can not be driven
by the pseudo random signal or when an appropriate speaker can not be installed at the position
of the sound source. In addition, even when the transfer function can be measured and the
inverse filter can be calculated, there is a disadvantage that the reverberation distortion
reduction effect is lost if the sound source position changes.
[0006]
The present invention has been made in view of the above, and the purpose thereof is to be
included in the sound reception signal using only the sound reception signal of each microphone
without measuring the transfer function between the sound source and the microphone. An
object of the present invention is to provide a reverberation reduction microphone device
capable of reducing reverberation distortion.
[0007]
SUMMARY OF THE INVENTION In order to achieve the above object, the present invention
according to claim 1 comprises first, second and third microphones for receiving sound signals
respectively, and these second and third microphones. First and second delay units respectively
providing appropriate delays to respective sound reception signals of the first and second filters;
first and second filters electrically processing respective output signals of the first and second
03-05-2019
2
delay units; An adder for adding the output signals of the first and second filters; a subtractor for
subtracting the output signal of the adder from the sound reception signal of the first
microphone; an output signal of the subtractor; And a filter calculation unit for obtaining a
transfer function that minimizes the root mean square value, which is the power of the output
signal of the subtractor, from the output signals of the second delay and the second delay and
setting the transfer function to the first and second filters. The gist of the door.
[0008]
In the present invention according to claim 1, an appropriate delay is given to each sound
reception signal of the second and third microphones to make a signal delayed from the sound
reception signal of the first microphone, and these respective delay signals are The respective
filters process each of the delay signals so as to minimize the power (square mean value) of the
signal obtained by filtering and adding, and subtracting this sum signal from the sound reception
signal of the first microphone. It defines the transfer function.
[0009]
Thus, the sum signal obtained by adding each delay signal processed by each filter is later than
the direct sound (or the direct reflection sound and the early reflection sound arriving slightly
behind the direct sound) constituting the sound reception signal of the first microphone. It
becomes a replica of the subsequent reverberation.
That is, by subtracting the sum signal from the sound reception signal of the first microphone, it
is possible to obtain a sound signal with reduced reverberation distortion included in the sound
reception signal of the first microphone.
[0010]
BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be
described below with reference to the drawings.
[0011]
FIG. 1 is a block diagram showing the configuration of a reverberation distortion reduction
microphone device according to an embodiment of the present invention.
03-05-2019
3
The reverberation distortion reduction microphone device shown in the figure receives each of
the first, second and third microphones 21, 22, 23, and the second and third microphones 22, 23
that receive the sound signal generated from the sound source 1. First and second filters 31 and
32 for electrically processing the output signals of the first and second delay units 51 and 52
and the first and second delay units 51 and 52 for providing appropriate delay to the sound
signal. , An adder 41 for adding the output signals of the first and second filters 31 and 32, a
subtractor 61 for subtracting the output signal of the adder 41 from the sound reception signal
of the first microphone, and an output of the subtractor 61 From the signal and the output
signals of the first and second delay units 51 and 52, a transfer function that minimizes the
power (square mean value) of the output signal of the subtractor 61 is determined and set in the
first and second filters 31 and 32. Do And a filter calculation unit 71.
[0012]
In the reverberation distortion reduction microphone device configured as described above, the
sound signal v (n) such as voice or sound signal generated from the sound source 1 is received
by the first, second, and third microphones 21, 22, 23. Be done.
The sound reception signals of the second and third microphones 22 and 23 are delayed by τ
sample time by the delay units 51 and 52 respectively, subjected to the processing by the filters
31 and 32, and added by the adder 41. Are subtracted from the sound reception signal of the
first microphone 21.
The filter calculation unit 71 uses the difference signal w (n) output from the subtractor 61 and
the output signals from the delay units 51 and 52 to minimize the power of the difference signal
w (n) h1 (z), h2 (z) is sequentially calculated and set in the filters 31 and 32.
[0013]
Next, the process in the filter calculation unit 71 will be described in detail. Assuming that the
transfer functions from the sound source 1 to the first, second and third microphones are z-d0
g0 (z), z-dg1 (z) and z-dg2 (z), respectively, the above difference signal w (n) Can be expressed as
follows. w (n) = {z-d0 g0 (z)-{z-(d + τ) g1 (z) h1 (z) + z-(d + τ) g2 (z) h2 (z)}} v (n ) (2) where z-d0
03-05-2019
4
represents the time required for the sound signal from the sound source 1 to reach the first
microphone, and z-d represents the time between the sound source 1 and the second microphone
and the sound source 1 and the second The delay time common to the transfer function between
the three microphones represents the time required for the sound signal to generate a sound
source and to reach the microphone closer to the sound source.
[0014]
By setting the delay time τ so as to satisfy d0 <d + τ, equation (2) can be modified as follows.
[0015]
Similar to the equation (1), the relational expression in {} of the second term on the right side of
the equation (3) can be expressed as g0 ′ ′ (z) − {g1 (z) h1 (z) + g2 (z) h2 (z)} = 0 Transfer
functions h1 (z) and h2 (z), which are (5), exist, and (a good approximation of these transfer
functions) can be calculated by minimizing the following equation (6).
[0016]
The first term of the equation (6) represents the power of the error signal w (n), and the second
term is a device for reducing distortion caused by autocorrelation or the like of the sound signal
v (n).
[0017]
The reverberation distortion (second term of the equation (3)) included in the difference signal w
(n) is reduced by the above processing.
That is, w (n) becomes a sound signal in which the reverberation distortion is reduced compared
to the sound reception signal of the first microphone.
[0018]
As described above, according to the present invention, each delayed signal obtained by delaying
each of the sound receiving signals of the second and third microphones is subjected to filter
processing and then added to the sum signal of the first microphone. This sum signal is
03-05-2019
5
subtracted from the sound reception signal of the first microphone so as to be a replica of the
reverberation sound that follows the direct sound (or the direct sound and the initial reflection
sound that arrives slightly behind the direct sound) constituting the sound reception signal. This
makes it possible to obtain a sound signal with reduced reverberation distortion included in the
sound reception signal of the first microphone, so there is no need to measure each transfer
function between the sound source and the microphone as in the prior art. Only the reception
signal can be used to reduce the reverberation distortion included in the sound signal.
Therefore, even when the sound source can not be driven by a known pseudo random signal or
the like, or when an appropriate speaker can not be installed at the sound source position, a
highly clear sound reception signal with reduced reverberation distortion is obtained even when
the sound source moves. be able to.
[0019]
Brief description of the drawings
[0020]
1 is a block diagram showing a configuration of a reverberation distortion reduction microphone
device according to an embodiment of the present invention.
[0021]
2 is a block diagram showing a configuration of a conventional reverberation distortion
reduction microphone device.
[0022]
Explanation of sign
[0023]
DESCRIPTION OF SYMBOLS 1 sound source 21, 22, 23, 23 1st, 2nd, 3rd microphone 31, 32 1st,
2nd filter 41 Adder 51, 52 1st, 2nd delay device 61 Subtractor 71 Filter calculation part
03-05-2019
6
Документ
Категория
Без категории
Просмотров
0
Размер файла
14 Кб
Теги
jph10257582
1/--страниц
Пожаловаться на содержимое документа