close

Вход

Забыли?

вход по аккаунту

?

JP2007174190

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 JP2007174190
The present invention provides an audio system capable of detecting the position of a
microphone in real time only by signal processing without using an additional device. An analysis
control unit refers to peaks of filter coefficients of an adaptive filter. The filter coefficient
simulates an audio propagation path from the speaker 18 to the microphone 11. The position of
this peak (signal corresponding to the direct sound) on the time axis indicates the distance
between the speaker 18 and the microphone 11. Thereby, the positional relationship between
the speakers 18L and 18R and the microphone 11 can be detected. [Selected figure] Figure 1
オーディオシステム
[0001]
The present invention relates to an audio system which detects the positional relationship
between a speaker and a microphone in real time.
[0002]
Some karaoke apparatuses provided with a speaker and a microphone provide a threedimensional sound effect by a stereo effect by a plurality of speakers.
In order to obtain a three-dimensional effect, the positional relationship between the speaker and
the microphone is important. That is, the sound effect is applied so as to obtain the most threedimensional effect at the position of the singer. However, when a singer moves, the problem that
04-05-2019
1
an optimal acoustic effect can not be provided generate | occur | produces.
[0003]
Therefore, a method has been proposed in which an audio signal (ultrasound) of an inaudible
frequency is output from the singing microphone and the distance is measured by the ultrasonic
receiving microphone installed in the karaoke apparatus body (see, for example, Patent
Document 1) ). JP 2000-59880 A
[0004]
However, according to the configuration of Patent Document 1, an ultrasonic wave generator is
required for the microphone, and a microphone for ultrasonic wave reception also needs to be
installed in the karaoke apparatus body. These ultrasonic wave generators and receivers do not
relate to the original functions of the karaoke apparatus (speech voice amplification, karaoke
musical tone reproduction, etc.), and are used only for position detection. Therefore, the
additional device is required only for position detection, and there is a problem that the cost is
increased. There is also a problem that in order to detect a position in real time, it is necessary to
continue emitting ultrasonic waves.
[0005]
Similarly, in the case where the configuration of Patent Document 1 is applied to other fields (for
example, a communication conference apparatus), the function is essentially the same (a voice
collected by a microphone is transmitted and a received voice is emitted by a speaker). There is a
problem that additional equipment which is not relevant is required, which is costly.
[0006]
An object of the present invention is to provide an audio system capable of detecting the position
of a microphone in real time only by signal processing without using an additional device.
[0007]
According to the audio system of the present invention, a microphone that picks up voice and
outputs a pick-up signal, a speaker that emits the output voice signal as voice, and a filter
coefficient that simulates a voice propagation path from the speaker to the microphone A
04-05-2019
2
coefficient estimation unit to calculate, a filter that sets the filter coefficient, filters the output
voice signal, and outputs a simulation signal of a feedback voice signal that is fed back to the
voice propagation path, and the simulation signal from the collected sound signal An adaptive
filter comprising: a subtraction unit that outputs a residual signal from which a feedback speech
signal component in a collected signal is removed by subtraction; an output speech signal
generation unit that generates an output speech signal including the residual signal; Microphone
position estimating means for estimating the distance between the speaker and the microphone
based on the peak position of the filter coefficient on the time axis And wherein the door.
[0008]
In the present invention, the voice collected by the microphone has its feedback voice signal
component removed by the adaptive filter.
The adaptive filter estimates a transfer function of a transfer system (sound transfer system)
from the speaker to the microphone, and generates a simulation signal simulating a feedback
sound (loop signal) from an output sound signal input to the speaker.
The adaptive filter subtracts this simulated signal from the collected signal collected by the
microphone.
The microphone position estimation means refers to the estimated transfer function (filter
coefficient) of this adaptive filter to detect the position of the time axis showing a peak equal to
or higher than a predetermined level. Usually, among the voices from the speaker to the
microphone, the signal with the highest level is the direct sound, so the peak of the filter
coefficient corresponds to the direct sound from the speaker to the microphone. Therefore, the
microphone position estimation means can detect the arrival time of the direct sound by
detecting the position of the peak of the filter coefficient. By multiplying the time of sound by the
speed of sound, the distance between the speaker and the microphone is estimated.
[0009]
The present invention is further characterized by comprising a plurality of the speakers, wherein
the microphone position estimation means detects a plurality of peaks from the filter coefficient
and estimates a distance between each speaker and the microphone based on each peak position.
I assume.
04-05-2019
3
[0010]
In the present invention, when there are a plurality of speakers and the distance between each of
the speakers and the microphone is different, a plurality of peaks are detected in the filter
coefficient.
The microphone position estimation means correlates the plurality of peaks with the sound
output from each speaker, and estimates the distance between each speaker and the microphone.
[0011]
The present invention further includes an output sound signal delay unit for delaying an output
sound signal input to a specific speaker, wherein the microphone position estimation means
changes the delay time of the delay unit and changes the delay time. The peak corresponding to
the specific speaker is determined by detecting the movement of the peak of the filter coefficient
according to
[0012]
In the present invention, a delay time is added to the output sound signal input to any one of the
speakers.
The microphone position estimation means controls this delay time and detects the time change
of the filter coefficient peak. That is, when there are a plurality of speakers and the distance
between each speaker and microphone is different, a plurality of peaks are detected in the filter
coefficient, but the delay time of the output audio signal input to any of the speakers is changed
Thus, the peak of the filter coefficient corresponding to the speaker can be determined. Thereby,
the positional relationship between the speaker and the microphone can be detected in more
detail.
[0013]
According to the present invention, the output sound signal generation unit is provided
04-05-2019
4
independently for each of the plurality of speakers, and a residual signal delay unit for delaying
the residual signal and inputting it to each output sound signal generation unit is provided. The
microphone position estimation means changes the delay time of the residual signal delay unit
and detects the movement of the peak of the filter coefficient due to the change of the delay time,
and is further provided independently for each output speech signal generation unit. The peak
corresponding to each said speaker is characterized by this.
[0014]
In the present invention, a delay time is added to the output sound signal input to each of the
plurality of speakers.
The microphone position estimation means respectively controls this delay time, and detects the
time change of the peak of the filter coefficient. By changing the delay time of each output sound
signal, it is possible to find a speaker corresponding to each peak of the filter coefficient.
Thereby, the positional relationship between the speaker and the microphone can be detected in
more detail.
[0015]
Further, according to the present invention, the microphone position estimation means sets a
delay time of the residual signal delay unit such that peaks corresponding to the respective
speakers of the filter coefficient coincide on a time axis.
[0016]
In the present invention, the delay amount of the delay circuit is further changed, and the moved
peak among the plurality of peaks is made to coincide with one of the peaks on the time axis.
That is, the audio signal input to any one of the speakers is delayed to virtually equalize the
distance between each of the speakers and the microphone. As a result, among the sounds
emitted from the respective speakers, with regard to the sound collected by the microphone, the
position of the singer is simultaneously reached, and the singing voice can be concentrated on
the singer.
04-05-2019
5
[0017]
According to the present invention, by detecting the peak position on the time axis of the filter
coefficient of the adaptive filter, the arrival time of the direct sound from the speaker to the
microphone can be estimated, so that the distance between the speaker and the microphone can
be detected. The position of the microphone can be detected in real time only by signal
processing without using an additional device.
[0018]
Embodiments of the present invention will be described with reference to the drawings.
The present invention is applicable to most systems as long as it is a system using a microphone
and a speaker, but a karaoke apparatus will be described here. FIG. 1 is a block diagram showing
the main part of the karaoke apparatus according to the embodiment. As shown in the figure, the
karaoke apparatus includes a microphone 11, an A / D converter 12, an adaptive filter 13, an
analysis control unit 14, an adder 15, a D / A converter 16, an amplifier 17, and speakers 18L
and 18R. ing.
[0019]
The microphone 11 picks up the voice of the space where the karaoke apparatus is installed
together with the singing voice of the singer and outputs a voice signal corresponding to the
picked-up voice to the A / D converter 12. The collected sound includes the sound emitted from
the speakers 18L and 18R (round sound). The microphone 11 generally uses a dynamic
microphone unit, but other formats such as a condenser microphone unit may be used. Also, the
microphone 11 may be a unidirectional microphone or a nondirectional microphone.
[0020]
The A / D converter 12 converts the audio signal output from the microphone 11 into a digital
signal and outputs the digital signal to the adaptive filter 13.
[0021]
The adaptive filter 13 includes a digital filter such as an FIR filter, and removes the wraparound
04-05-2019
6
sound from the collected sound signal of the microphone 11 converted to a digital signal by the
A / D converter 12 and outputs the resultant to the adder 15.
[0022]
A detailed block diagram of the adaptive filter 13 is shown in FIG.
The adaptive filter 13 includes an FIR filter 131, an adder 132, and a coefficient estimation unit
133.
The coefficient estimation unit 133 estimates a transfer function of an acoustic transfer system
(sound propagation path from the microphone 11 to the speaker 18), and calculates and sets
filter coefficients of the FIR filter 131 so as to simulate the estimated transfer function. The
estimation of the transfer function and the calculation of the filter coefficient are based on the
signal (output audio signal input to the speaker 18) input from the adder 15 using the residual
signal that is the signal output from the adder 132 as a reference signal. And using an adaptive
algorithm. The adaptive algorithm is an algorithm that calculates filter coefficients so that the
residual signal is as small as possible.
[0023]
As a result, a signal simulating a wraparound signal (a sound signal from the speaker 18 to the
microphone 11) of the sound transmission system is generated in the FIR filter 131, and a
wraparound signal is subtracted from the output signal of the microphone 11 in the adder 132.
Only the signal can be attenuated efficiently. As a result, the adaptive filter 13 can prevent
howling and echo caused by the loop phenomenon of the loop signal.
[0024]
The adder 15 adds the singing voice signal, which is an output signal of the adaptive filter, and
the musical tone signal (monaural) output from the musical tone reproduction unit (not shown)
of the karaoke apparatus, and outputs the result to the D / A converter 16.
[0025]
04-05-2019
7
The D / A converter 16 converts the output signal of the adder 15 into an analog audio signal
and outputs the analog audio signal to the amplifier 17.
[0026]
The amplifier 17 is a so-called power amplifier, which amplifies the audio signal output from the
D / A converter 16, and branches and outputs the amplified audio signal to the speaker 18L and
the speaker 18R.
In this example, the monaural musical tone signal and the singing voice signal are added by the
adder 15, and output equally to the speakers 18L and 18R (in center localization).
[0027]
The speaker 18 </ b> L and the speaker 18 </ b> R respectively emit sound based on the
amplified sound signal output from the amplifier 17.
Although the speaker 18L and the speaker 18R generally use a cone type speaker unit, other
types such as a horn type speaker unit may be used.
[0028]
The analysis control unit 14 is connected to the coefficient estimation unit 133 of the adaptive
filter 13 as shown in FIGS. 1 and 2, and reads out the filter coefficient to be set in the FIR filter
131. Then, based on the peak position of the filter coefficient, the distance between the
microphone 11 and the speakers 18L and 18R is detected.
[0029]
Here, a method in which the analysis control unit 14 detects a distance will be described in detail.
The time-axis component of the filter coefficient of the adaptive filter 13 is shown in FIG. The
04-05-2019
8
horizontal axis of the graph shown in the figure represents the tap number of the FIR filter, ie,
time, and the vertical axis represents the gain of each tap, ie, the level of the wraparound sound.
FIG. 6A shows an example of filter coefficients in the case of one speaker and one microphone.
The adaptive filter 13 is a digital filter, and the filter coefficients are represented as discrete
signals, but in the figure, they are represented as continuous signals in order to facilitate the
description.
[0030]
As described above, the adaptive filter 13 estimates the transfer function of the acoustic transfer
system based on the output sound signal input to the speaker using the residual signal as a
reference signal, and adjusts the filter coefficients in accordance with the estimated transfer
function. Calculate Therefore, the time axis component of the filter coefficient shown in the
figure corresponds to the feedback signal from the speaker to the microphone. When the level of
the feedback signal from the speaker to the microphone is high, the level of the filter coefficient
is increased to cancel this.
[0031]
As shown in the figure, the filter coefficient has one peak higher than a predetermined level. The
peak refers to the coefficient having the largest level among components above a predetermined
level (threshold). Here, among the sounds from the speaker to the microphone, the signal with
the highest level is the direct arrival sound, so the peak of the filter coefficient corresponds to the
direct sound from the speaker to the microphone. Therefore, the time component of the peak of
the filter coefficient indicates the arrival time of the direct voice from the speaker to the
microphone. Therefore, the analysis control unit 14 connected to the adaptive filter 13 detects
the time component of this peak, and multiplies the time component by the speed of sound to
calculate the distance between the speaker and the microphone. Since the speaker is fixed, the
distance can detect the position of the microphone.
[0032]
Here, in this example, a plurality of speakers 18L and 18R are provided, and in fact, as shown in
FIG. 6B, the filter coefficient has two peaks of a predetermined level or more. Since the distance
between each speaker and the microphone is different, the arrival time of the direct voice from
04-05-2019
9
each speaker to the microphone is different, and the filter coefficient has two different peaks of
time axis components. The small peak of the time axis component (short time) is the peak
corresponding to the speaker close to the microphone, and the large peak of the time axis
component (long time) is the peak corresponding to the speaker far from the microphone is
there. The analysis control unit 14 determines in advance to which side of the speaker 18L and
the speaker 18R the microphone 11 is located (speaker nearer and speaker farther) so that each
peak of the filter coefficient corresponds to the output sound of which speaker. It can be
determined whether it is a corresponding one. In order to determine which side the microphone
11 is located, it may be set in advance in anticipation of a situation in which the microphone is
used, or may be set by the user using the operation unit or the like of the karaoke apparatus. It is
also good.
[0033]
For example, when the speaker 18L is close to the microphone 11, the small peak of the time
axis component shown in FIG. 3B corresponds to the output sound of the speaker 18L. On the
other hand, the large peak of the time axis component corresponds to the output sound of the
speaker 18R. Therefore, the distance between each speaker 18 and the microphone 11 can be
detected. The position of the microphone 11 can be detected by detecting the distance between
each speaker 18 and the microphone 11. That is, when the installation position of the speaker is
determined and the distance between the speaker 18L and the speaker 18R is known, the
position of the microphone 11 can be determined by triangulation.
[0034]
The distance information of the microphone detected as described above is output to, for
example, a sound control unit or the like of the karaoke apparatus, and is used to give a sound
effect in the sound control unit. The sound control unit controls, for example, the volume of the
speaker 18 based on the distance information between the speaker 18 and the microphone 11 so
that an optimum sound effect is given at the position of the singer (microphone). It may also be
used to give other effects such as spotlighting the position of the singer.
[0035]
Further, in order to facilitate identification of the peak due to the wraparound sound from the left
04-05-2019
10
and right speakers, the following modification is possible in the present invention. FIG. 4 is a
block diagram showing the main part of a karaoke apparatus according to a modification. The
same reference numerals as in the block diagram of the main part of the karaoke apparatus
shown in FIG. 1 denote the same parts, and a description thereof will be omitted.
[0036]
The karaoke apparatus includes a microphone 11, an A / D converter 12, an adaptive filter 13,
an analysis control unit 14, adders 21L and 21R, a mixer 22, a delay 23, a D / A converter 16, an
amplifier 17, and speakers 18L and 18R. Have.
[0037]
In this karaoke apparatus, the adaptive filter 13 is connected to the adders 21L and 21R and the
mixer 22.
The adder 21L is connected to the delay 23, and the D / A converter 16 is connected to the adder
21R and the delay 23. The analysis control unit 14 is also connected to the adaptive filter 13 and
the delay 23.
[0038]
The output signal of the adaptive filter 13 is input to the adders 21L and 21R. The adders 21L
and 21R respectively add the output signal (singing voice signal) of the adaptive filter 13 and the
stereo musical tone signals (L channel and R channel) output from the musical tone reproducing
unit (not shown) of the karaoke apparatus. The output signal of the adder 21R is input to the D /
A converter 16, and the output signal of the adder 21L is input to the delay 23. The output
signals of the adders 21L and 21R are branched and input to the mixer 22, respectively. The
mixer 22 mixes these signals and outputs the mixed signal to the adaptive filter 13 as a monaural
signal. The adaptive filter 13 updates the filter coefficients using the above-mentioned residual
signal as a reference signal based on this monaural signal.
[0039]
04-05-2019
11
The delay 23 adds a delay time set by the analysis control unit 14 to the output signal of the
adder 21L and outputs it. The output signal of the delay 23 is input to the D / A converter 16.
The D / A converter 16 converts the output signal of the adder 21 R and the output signal of the
delay 23 into an analog audio signal, and outputs the analog audio signal to the amplifier 17. The
amplifier 17 amplifies each analog signal (L channel, R channel) and outputs it to the speakers
18L and 18R. The speaker 18L and the speaker 18R emit sounds in accordance with the signals
of the respective channels.
[0040]
The analysis control unit 14 changes the delay amount of the delay 23. As a result, the distance
of the L channel in a system outside the adaptive filter 13 is artificially increased. The analysis
control unit 14 can determine that the peak of the filter coefficient that moves in response to the
change in the delay amount corresponds to the output sound of the speaker 18L. Therefore, the
position of the microphone 11 (the distance from the speakers 18L and 18R) can be detected.
[0041]
A method of detecting the position of the microphone 11 by the analysis control unit 14 will be
described in detail. FIG. 5 shows time axis components of filter coefficients of the adaptive filter
13 in this karaoke apparatus. The horizontal axis of the graph shown in the figure represents
time, and the vertical axis represents a level. The figure shows an example of filter coefficients
when there are two speakers and one microphone, and the distance between each speaker and
the microphone is different. Note that the filter coefficients are represented as continuous signals
also in the same drawing in order to facilitate the description.
[0042]
The time axis component of the filter coefficient of the adaptive filter 13 corresponds to the
feedback signal from the speaker to the microphone, as described above. When the level of the
feedback signal from the speaker to the microphone is high, the level of the filter coefficient is
increased to cancel this. In the karaoke apparatus shown in FIG. 1, as shown in FIG. 3B, the filter
coefficient usually has two peaks of a predetermined level or more. That is, since the karaoke
apparatus in FIG. 1 has two speakers and one microphone, and the distance between each
speaker and the microphone is different, the filter coefficient has two peaks with different time
04-05-2019
12
axis components. . The small peak of the time axis component (short time) is the peak
corresponding to the speaker close to the microphone, and the large peak of the time axis
component (long time) is the peak corresponding to the speaker far from the microphone is
there.
[0043]
The analysis control unit 14 in FIG. 4 changes the delay amount of the delay 23. Then, as shown
in FIG. 5, the position on the time axis of one of the two peaks of the filter coefficient is moved.
[0044]
That is, since the L-channel output sound signal is given a delay time and output from the
speaker, the adaptive filter 13 adapts the filter coefficient according to the delayed sound signal,
and the position of the peak moves. It can be determined that this moved peak is a peak
corresponding to the L channel direct sound. The analysis control unit 14 can determine whether
the peak corresponds to the audio signal output from the L channel speaker by detecting the
movement of the peak. The amount of delay may be fixed (the analysis control unit 14 only turns
on / off the delay 23) or may be changed. Although FIG. 4 shows an example in which the delay
23 is connected to an L channel signal, it may be an R channel.
[0045]
As described above, the analysis control unit 14 determines which speaker the plurality of peaks
of the filter coefficient correspond to the audio signal output from, respectively, and detects the
distance between each speaker and the microphone. Can. Since each speaker is identified and the
distance to the microphone is detected, the position of the microphone can be detected without
making mistakes in the left and right of the speaker. The detected microphone position
information is output to a sound control unit or the like of the karaoke apparatus, and is used to
apply a sound effect. Of course, microphone position information may be applied to other
applications.
[0046]
04-05-2019
13
Moreover, the following aspects are also considered as another modification. FIG. 6 is a block
diagram showing the main part of a karaoke apparatus according to another modification. The
same reference numerals as in the block diagram of the main part of the karaoke apparatus
shown in FIG. 1 denote the same parts, and a description thereof will be omitted. As shown in the
figure, the karaoke apparatus includes a microphone 11, an A / D converter 12, an adaptive filter
13, an analysis control unit 14, delays 31L and R, adders 21L and R, a D / A converter 16, an
amplifier 17, Speakers 18L and 18R and a mixer 32 are provided.
[0047]
In this example, the adaptive filter 13 is connected to the analysis control unit 14, the delays 31
L and 31 R, and the mixer 32. The delay 31L is connected to the adder 21L and the analysis
control unit 14, and the delay 31R is connected to the adder 21R and the analysis control unit
14. The mixer 32 mixes the stereo musical tone signal (L channel, R channel) output from the
musical tone reproduction unit (not shown) of the karaoke apparatus with the output signal of
the adaptive filter 13 and outputs the mixed signal to the adaptive filter 13.
[0048]
The delays 31L and 31R respectively add the delay amount set by the analysis control unit 14 to
the output signal of the adaptive filter and output the same. The output signals of the delays 31L
and R are input to the adders 21L and R, respectively. The adders 21L and 21R mix the output
signals of the delays 31L and 31R with the L channel musical tone signal and the R channel
musical tone signal, respectively, and output the mixed signal to the D / A converter 16. The D /
A converter 16 converts the output signals of the adders 21 L and 21 R into analog audio signals,
and outputs the analog audio signals to the amplifier 17. The amplifier 17 amplifies each analog
signal (L channel, R channel) and outputs it to the speakers 18L and 18R. The speaker 18L and
the speaker 18R emit sounds in accordance with the signals of the respective channels.
[0049]
The analysis control unit 14 changes the delay amount of the delays 31L and 31R, refers to the
filter coefficient of the adaptive filter 13 updated according to the change of the delay amount,
and detects the position of the microphone 11 (distance from the speakers 18L and 18R). ) To
04-05-2019
14
detect. For the method of detecting the position of the microphone 11 by the analysis control
unit 14, as shown in FIG. 5, change the delay time of the delay 31 L (or the delay 31 R) and refer
to the peak of the filter coefficient moved on the time axis. To do. The peak moving by the delay
of the delay 31L is the peak corresponding to the speaker 18L, and the peak moving by the delay
of the delay 31R is the peak corresponding to the speaker 18R. Further, the analysis control unit
14 changes the delay amounts of the delays 31L and 31R, and performs control such that a
plurality of peaks of the filter coefficient coincide with each other on the time axis. FIG. 7 is a
diagram showing time axis components of filter coefficients of the adaptive filter in this
modification. Also in the graph shown in the figure, the horizontal axis represents time, and the
vertical axis represents a level. Also, in the same figure, an example of filter coefficients in the
case where there are two speakers and one microphone and the distances between the respective
speakers and the microphone are different is shown. Note that the filter coefficients are
represented as continuous signals also in the same drawing in order to facilitate the description.
[0050]
As shown in the figure (A), in this example, there are also two speakers and one microphone, and
the distance between each speaker and the microphone is different. Will have one. The small
peak of the time axis component is a peak corresponding to the speaker close to the microphone,
and the large peak of the time axis component is a peak corresponding to the speaker far from
the microphone. The analysis control unit 14 changes the delay amount of the delay 31 (L or R),
and determines which speaker corresponds to the audio signal output from which peak the
moved peak is. This detects the position of the microphone. The detected microphone position
information is output to a sound control unit or the like of the karaoke apparatus, and is used to
apply a sound effect.
[0051]
In this example, the analysis control unit 14 further changes the delay amount of the delay 31 (L
or R), and causes the moved peak to coincide with the other peak on the time axis. That is, the
audio signal input to any one of the speakers is delayed to virtually equalize the distance between
each of the speakers and the microphone. As a result, among the sounds emitted from the
respective speakers, the component of the sound collected by the microphone (a loud sound)
simultaneously reaches the position of the singer.
[0052]
04-05-2019
15
As described above, since the L channel or R channel signal (loud-speaking sound) is delayed and
simultaneously reaches the microphone 11, the sound image of the microphone voice is localized
at the position of the singer, and the singing voice is concentrated on the singer it can. Further,
according to this example, since the karaoke musical tone signal is not delayed (only the signal
collected by the microphone is inputted to the delay 31), the sound image localization of the
karaoke musical tone is compared with the example shown in FIG. It is hard to break the feeling,
and further improvement of the sound quality can be expected.
[0053]
Also in this modification, the positional information of the microphone may be output to the
sound control unit of the karaoke apparatus, and may be used for giving a sound effect in the
sound control unit, or may be applied to other applications. .
[0054]
In the above example, the karaoke apparatus has been described, but the application example of
the present invention is not limited to this, and may be used for a PA system (sound output
apparatus).
When used in a loudspeaker system, the adder 15 is omitted in FIG. 1 and no musical tone signal
is input. In this case, the output sound signal generation unit, which is a component of the
present invention, corresponds to, for example, an amplifier.
[0055]
Further, the audio system of the present invention is not limited to the case where it is applied to
the karaoke apparatus (sound amplification apparatus) as described above. For example, as
shown in FIG. 8, the present invention can be applied to a voice input / output device (so-called
communication conference device). FIG. 8 is a block diagram showing the configuration of the
voice input / output device. The same reference numerals as in the block diagram of the main
part of the karaoke apparatus shown in FIG. 1 denote the same parts, and a description thereof
will be omitted. As shown in the figure, this voice input / output device includes a microphone
11, an A / D converter 12, an adaptive filter 13A, an adaptive filter 13B, an analysis control unit
04-05-2019
16
14, a D / A converter 16, an amplifier 17, speakers 18L and 18R, And an input / output interface
51.
[0056]
In this example, the adaptive filter 13A and the adaptive filter 13B are connected to the analysis
control unit 14 and the input / output interface 51. Further, the input / output interface 51 is
connected to the D / A converter 16 and other audio input / output devices (not shown).
[0057]
The input / output interface 51 receives an audio signal (speech signal) from another audio input
/ output device connected via a network or the like. The received signal is emitted as sound from
the speaker 18 through the D / A converter 16 and the amplifier 17. Further, the input / output
interface 51 picks up the sound with the microphone 11, and outputs the audio signal input
through the A / D converter 12 and the adaptive filters 13A and 13 to other audio input / output
devices.
[0058]
In this example, the adaptive filter 13A and the adaptive filter 13B respectively input a plurality
of speech signals (reception signals A and B in FIG. 8) received by the input / output interface,
and the voices collected from the microphone 11 Cancel the feedback component. That is, in this
example, the adaptive filter 13 cancels the component (echo) of the sound returned from the
speaker 18 to the microphone 11. Different speech signals (received signals A and B) are input to
the speakers 18L and 18R, respectively, and different sounds are emitted from the speakers. The
adaptive filter 13A and the adaptive filter 13B cancel feedback components related to the
respective speech signals.
[0059]
The analysis control unit 14 refers to the filter coefficients of the adaptive filter 13A and the
adaptive filter 13B, and detects the arrival time of the direct sound from the speaker 18 to the
04-05-2019
17
microphone 11 as described above. In this example, since the sound emitted from each speaker
is different (the adaptive filter corresponding to each speaker is determined to be one to one), the
analysis control unit 14 may calculate the distance between each speaker and the microphone. it
can. Therefore, in this voice input / output device, the position of the microphone can be
detected (two-dimensionally detected) in detail without giving a delay time to the voice signal.
[0060]
Also in this example, the positional information of the microphone may be output to the sound
control unit of the voice input / output device, and may be used for giving a sound effect in the
sound control unit, or may be output to another sound input / output device Etc. may be applied
to any application.
[0061]
Block diagram showing the detailed configuration of the main part of the karaoke apparatus
Block diagram showing the detailed configuration of the adaptive filter Figure showing the time
axis component of the filter coefficient of the adaptive filter block diagram showing the
configuration of the main part of the karaoke apparatus according to the modification Figure
showing the time-axis component of the filter coefficient of the adaptive filter in a block diagram
showing the configuration of the main part of the karaoke apparatus according to the other
modification example diagram showing the time-axis component of the filter coefficient of the
adaptive filter in the other modification Block diagram showing the configuration of the device
Explanation of sign
[0062]
11-microphone 12-A / D converter 13-adaptive filter 14-analysis control unit 15-adder 16- D / A
converter 17-amplifier 18-speaker
04-05-2019
18
Документ
Категория
Без категории
Просмотров
0
Размер файла
30 Кб
Теги
jp2007174190
1/--страниц
Пожаловаться на содержимое документа