JP2007174189

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 JP2007174189
To provide a sound emission and collection device which has a compact configuration,
suppresses sound coming from a speaker to a microphone, and has an improved S / N ratio. A
plurality of microphones (2A to 2H) are circumferentially arranged on a first surface (10A) of a
substantially short cylindrical casing (1) and point-symmetrically with a center of the circle as a
reference point. The microphones 2A to 2H have directivity, and high sound collection sensitivity
is set in the center direction of the circle. The speaker 3 is disposed on the second surface 10B of
the housing 1 facing the first surface 10A, and emits noise to the outside with nondirectionality.
A part of the sound emitted from the speaker 3 turns to the first surface 10A side via the side
surface 10C of the housing 1. At this time, the microphones 2A to 2H pick up the sound coming
around from the farthest side according to the setting of the directivity, so that the wraparound
sound is sufficiently attenuated before being picked up. [Selected figure] Figure 1
Sound emission device
[0001]
The present invention relates to a sound emission and collection device integrally including a
speaker and a microphone, and more particularly, to a sound emission and collection device
which has a compact configuration and suppresses the sound coming from the speaker to the
microphone.
[0002]
As an audio communication system for performing an audio conference (communication
04-05-2019
1
conference) at a remote place, an audio conference apparatus integrally provided with a speaker
and a microphone has become widespread.
The audio conference apparatus transmits the sound collection signal collected by the
microphone to the connection destination, and emits the sound signal received from the
connection destination from the speaker. When a conference is conducted by a plurality of
people, such an audio conference device is often installed at the center of the conference
participants (the center of the conference desk, etc.). Therefore, it is desirable that such an audio
conference apparatus be miniaturized, and for example, as shown in Patent Document 1, there is
proposed an audio conference apparatus which is miniaturized by omitting a speaker box.
[0003]
Also, since the audio conference apparatus is configured to have a speaker and a microphone in
the same space, when an audio signal received from the connection destination is emitted from
the speaker, this audio is collected as an echo by the microphone, and this echo is included. A
collected signal is transmitted to the connection destination. Therefore, as shown in Patent
Document 2, in an audio conference apparatus having an echo canceller function, an audio
conference apparatus is proposed in which a microphone is accommodated at the end of a
cylindrical elastic body and acoustic coupling between a speaker and the microphone is
suppressed. ing. JP-A-8-204803 JP-A-8-298696
[0004]
However, although the configuration of Patent Document 1 is a compact configuration, the
speaker and the microphone are close to each other, and the sound volume from the speaker to
the microphone is increased. On the other hand, in the configuration of Patent Document 2, the
echo canceller function suppresses the wraparound voice, and the elastic body suppresses the
acoustic coupling inside the housing. However, similar to Patent Document 1, it has a compact
configuration. The speaker and the microphone are close to each other. For this reason, there is
still the problem that the sound emitted from the speaker is likely to be introduced into the
microphone, which places a large processing load on the echo canceller function.
[0005]
SUMMARY OF THE INVENTION It is an object of the present invention to provide a sound
04-05-2019
2
emission and collection device which has a compact configuration but suppresses the sound
coming from a speaker to a microphone and improves the S / N ratio.
[0006]
The sound emission and collection device according to the present invention is a casing having
two opposing faces, and the casing is disposed circumferentially on the first face side of the
casing, and is installed with the directivity axis directed toward the center of the circle. And a
sound emitting surface parallel to a second surface facing the first surface of the housing, the
center of the sound emitting surface and the center of the circle being the first surface, and And a
speaker which emits the sound from the second surface to the outside of the casing on the same
perpendicular line to the second surface.
[0007]
In this configuration, the sound emitted from the speaker installed on the second surface side
serving as the sound emission surface is reflected on the top surface of the desk on which the
sound emission and collection device is installed and the like to the side surface side of the
housing It is propagated.
At this time, the sound strengths transmitted to the respective parts of the side face become
substantially equal.
Then, part of the sound emitted from the speaker wraps around the side surface to the first
surface side where the plurality of unidirectional microphones are installed.
[0008]
Each of the plurality of unidirectional microphones is circumferentially arranged on the first
surface side. At this time, the center of the circle and the center of the sound emitting surface are
on the same perpendicular line with respect to the first and second surfaces, and the center
direction of the circle is the axial direction of directivity, that is, the sound pickup sensitivity is
high. It is installed in the direction.
04-05-2019
3
[0009]
For this reason, the voice that has rolled around on the first surface side is not a unidirectional
microphone arranged at a position closest to the side position where the voice has looped in, but
is opposite to the side position where the voice has looped in The sound is mainly picked up by a
unidirectional microphone placed at the farthest position of. As a result, the propagation path
(echo path) of the looped speech becomes long, and significant attenuation of the looped speech
can be obtained before it is picked up by the unidirectional microphone.
[0010]
Further, the arrangement positions of the plurality of unidirectional microphones in the sound
emission and collection device of the present invention are characterized in that they are set
point-symmetrically with the center of the circle as a reference point.
[0011]
In this configuration, since the arrangement positions of the plurality of unidirectional
microphones are point-symmetrical, wraparound voices to the respective unidirectional
microphones become substantially equal.
[0012]
Further, according to the sound emission and collection device of the present invention, from the
sound collection signals of the respective unidirectional microphones arranged pointsymmetrically, the unidirectional microphones arranged at the point-symmetrical positions with
respect to each unidirectional microphone and the center of the circle. The present invention is
characterized in that it comprises difference calculation means for calculating a difference
between a sound collection signal with a sex microphone and generating a difference correction
sound collection signal.
[0013]
In this configuration, as described above, the wraparound sound collected by each unidirectional
microphone hardly changes, and in particular, the wraparound voices of the unidirectional
directivity microphones arranged point-symmetrically become substantially equal. By subtracting
the collected sound signals of these unidirectional microphones, a differentially corrected
collected sound signal from which the signal component due to the wraparound speech is
removed can be obtained.
04-05-2019
4
[0014]
Further, the sound emission and collection device according to the present invention detects the
sound source direction based on the signal strengths of the collected sound signals of the
plurality of unidirectional microphones, and collects the directional microphone whose directivity
axis is directed to the sound source direction. It is characterized in that it comprises signal
processing means for outputting a sound signal to the subsequent stage.
[0015]
In this configuration, the sound pickup signal of a unidirectional microphone whose directivity
axis is oriented in the sound source direction is higher in signal strength than the sound pickup
signals of other unidirectional microphones, that is, the sound pressure level is high. Among the
sound collection signals of each unidirectional microphone, the sound collection signal having
the highest signal strength is selectively output.
As a result, the signal strength of the collected signal from the sound source direction becomes
relatively high, and a collected signal having a high S / N ratio can be obtained.
[0016]
Further, the sound emission and collection device according to the present invention includes
signal processing means for detecting the sound source direction based on the signal strength of
the difference correction sound collection signal and outputting the difference correction sound
collection signal corresponding to the sound source direction to the subsequent stage. It is
characterized by
[0017]
In this configuration, the difference obtained by subtracting the sound collection signal of the
unidirectional microphone with the directivity axis facing in the opposite direction from the
sound collection signal of the unidirectional microphone with the directivity axis facing the
sound source direction The corrected sound pickup signal uses the fact that the wraparound
speech component is suppressed and the sound pickup signal from the sound source direction is
further enhanced, and the signal strength is the highest among the respective difference
corrected sound pickup signals, that is, the sound pressure level By selectively outputting the
high differential corrected sound pickup signal, a sound pickup signal having a higher S / N ratio
can be obtained.
04-05-2019
5
[0018]
According to the present invention, a plurality of unidirectional microphones are
circumferentially arranged on one surface of the housing, set with high directivity of directivity
in the central direction of the circle, and the speaker is By being disposed on the other side, it is
possible to effectively increase the propagation distance of the wraparound voice from the
speaker to the microphone.
As a result, it is possible to suppress the sound coming from the speaker to the microphone and
to improve the S / N ratio, while having a compact configuration.
[0019]
A sound emission and collection device according to an embodiment of the present invention will
be described with reference to the drawings.
FIG. 1 is a view showing the configuration of the main part of the sound emission and collection
device 100 of the present embodiment, (A) is a plan view, and (B) is a cross-sectional view taken
along A-A 'in (A).
In FIGS. 1A and 1B, the right side is the + X direction toward the paper surface, and the left side
is the −X direction toward the paper surface.
Further, in FIG. 1A, the upper side toward the paper surface is the + Y direction, and the lower
side toward the paper surface is the −Y direction.
Further, in FIG. 1B, the upper side toward the paper surface is the + Z direction, and the lower
side toward the paper surface is the −Z direction.
[0020]
04-05-2019
6
The sound emission and collection device 100 of the present embodiment includes a housing 1, a
plurality of microphones 2A to 2H, a speaker 3, and a signal processing function unit shown in
FIG.
[0021]
The housing 1 has a substantially cylindrical external shape, and has substantially the same size
and planar circular first surfaces 10A and second surfaces 10B, and is connected by the edges of
the first surfaces 10A and the second surfaces 10B. And the side surface 10C of the
circumferential surface shape which arrange | positions these at predetermined intervals.
In the vicinity of the edge portion of the second surface 10B, four foot portions 4 arranged at
intervals of approximately 90 degrees are installed.
[0022]
A recess 11 whose plane shape is a circle is formed in the first surface 10A of the housing 1, and
the center of the circle when the first plane 10A is viewed in plan matches the center of the circle
when the recess 11 is viewed in plan Do.
Hereinafter, this center point is referred to as "center point O".
[0023]
The microphones 2A to 2H are unidirectional microphones, and are disposed at pointsymmetrical positions with the center point O as a reference point. Furthermore, each of the
microphones 2A to 2H is installed at a predetermined distance or more from the center point O,
and more preferably, at a position near the edge portion of the first plane 10A.
[0024]
04-05-2019
7
Specifically, as shown in FIG. 1 (A), the microphones 2A to 2H each have a center point O as a
reference point along the inner peripheral wall surface 12 which is the same distance from the
center point O, and the microphone 2A is + X. The microphones 2E are arranged in the -X
direction. Similarly, with the central point O as a reference point, the microphones 2B are
disposed in the + X direction and 45 degrees in the + Y direction, and the microphones 2F are
disposed in 45 degrees in the -X direction and -Y direction. Further, with the central point O as a
reference point, the microphone 2C is disposed in the + Y direction, and the microphone 2G is
disposed in the -Y direction. Further, with the central point O as a reference point, the
microphone 2D is disposed in 45 degrees of −X direction and + Y direction, and the microphone
2H is disposed in 45 degrees of + X direction, −Y direction.
[0025]
Each of the microphones 2A to 2H is installed such that the direction of directivity is in the
direction toward the center point O. Thus, in each microphone, the sound collection sensitivity in
the center point O direction is set higher than the sound collection sensitivity in the other
direction.
[0026]
In the second surface 10B of the housing 1, the second surface 10B and the sound emission
surface are in substantially the same relationship, and the sound emission direction is from the
second surface 10B to the outside direction of the housing 1, A speaker 3 is disposed. The
speaker 3 is a nondirectional speaker including a cone type speaker unit, a horn type speaker
unit, and the like, and the center of the sound emitting surface of the speaker 3 passes through
the center point O of the first plane 10A and a line perpendicular to the first surface 10A. It is
arranged to be located in.
[0027]
Further, although not shown in FIG. 1, a signal processing function unit to be described later is
installed in an empty space other than the arrangement positions of the microphones 2A to 2H
and the speaker 3 in the housing 1. The input / output connector 26 is installed, for example, on
the side surface 10C of the housing 1.
04-05-2019
8
[0028]
Such a sound emission and collection device 100 is arranged and used as shown in FIG.
[0029]
FIG. 2 is a diagram showing a case where two users 201 and 202 use the sound emission and
collection device 100 of the present embodiment, (A) is a plan view, and (B) is a side view.
Also in FIGS. 2A and 2B, the right side is the + X direction toward the paper surface, and the left
side is the −X direction toward the paper surface. Further, in FIG. 2A, the upper side toward the
paper surface is the + Y direction, and the lower side toward the paper surface is the −Y
direction. Further, in FIG. 2B, the upper side toward the paper surface is the + Z direction, and the
lower side toward the paper surface is the −Z direction.
[0030]
The sound emission and collection device 100 is disposed on the top surface substantially at the
center of the top surface of the desk 200. At this time, the plurality of feet 4 are placed in contact
with the top surface of the desk to arrange the housing 1 at a predetermined distance from the
top surface.
[0031]
Further, although not shown, this sound emission and collection device 100 is connected to the
LAN via the aforementioned input / output connector 26, and it is separated from another
location, for example, a different place from the room in which the device is installed. , Is
connected to another sound emitting and collecting device.
[0032]
Users 201 and 202 face each other on both sides of the desk 200 on which the sound emission
and collection device 100 is disposed.
04-05-2019
9
In the example of FIG. 2, the user 201 is in the −X direction with respect to the sound emission
and collection device 100, and the user 202 is in the + X direction with respect to the sound
emission and collection device 100.
[0033]
(1) Sounds uttered from the users 201 and 202 These users 201 and 202 utter to the sound
emitting and collecting apparatus 100 when speaking to the other user in the room of another
sound emitting and collecting apparatus.
[0034]
When the user 201 utters, the uttered sound 301 reaches the microphones 2A to 2H of the
sound emission and collection device 100 while diffusing and attenuating.
Here, as described above, the microphone 2A is set to have directivity with high sound collection
sensitivity in the direction of the center point O of the housing 1 with respect to the microphone
2A, that is, in the -X direction in which the user 201 exists. . For this reason, the microphone 2A
is present at a position farthest from the user 201 as compared with the other microphones 2B
to 2H, but can pick up the utterance sound 301 with high sensitivity. On the other hand, the
microphone 2E which is in point symmetry with the microphone 2A is present at a position
closest to the user 201 as compared with the other microphones 2A to 2D and 2F to 2H.
However, since the microphone 2E has high sound collection sensitivity in the + X direction and
directivity with almost no sound collection sensitivity in the -X direction, the vocal sound 301 is
hardly collected.
[0035]
Further, when the user 202 utters, the utterance sound 302 reaches the microphones 2A to 2H
of the sound emission and collection device 100 while diffusing and attenuating. Here, as
described above, the microphone 2E is set to have directivity with high sound collection
sensitivity in the direction of the center point O of the housing 1 with respect to the microphone
2E, that is, in the + X direction in which the user 202 exists. For this reason, the microphone 2E
is present at a position farthest from the user 202 as compared with the other microphones 2A
to 2D and 2F to 2H, but can pick up the vocal sound 302 with high sensitivity. On the other
04-05-2019
10
hand, the microphone 2A which is in point symmetry with the microphone 2E is present at a
position closest to the user 202 as compared with the other microphones 2B to 2H. However,
since the microphone 2A has high sound collection sensitivity in the -X direction and directivity
with almost no sound collection sensitivity in the + X direction, the vocal sound 302 is hardly
collected.
[0036]
Thus, the user's voice is mainly picked up by the microphone disposed at the opposite position
extending from the side where the user is located through the center point O.
[0037]
By the way, although the invention of the above-mentioned patent document 2 is a sound
emitting and collecting apparatus in which a speaker is disposed on the upper surface and a
microphone is disposed on the side surface, as shown in FIGS. 3 (B) and 3 (D) described later It is
also conceivable to place 93 on the lower side.
In this case, the microphone 92A disposed on the side surface of the housing 91 is set such that
the directivity is directed in the outward direction of the side surface of the housing 91, and the
sound generated by the user 201 closest to the microphone 92A is collected. . In the following
description, the sound emission and collection device having the configuration shown in FIGS. 3B
and 3D is taken as a representative example of a conventional sound emission and collection
device as a comparison object of the present embodiment.
[0038]
FIG. 3A is a conceptual diagram showing the transmission distance Lv1 of the voiced sound to
the main microphones in the sound emission and collection device 100 of the present
embodiment, and FIG. 3B is a side view of the housing. FIG. 8 is a conceptual diagram showing a
transfer distance Lv0 of a vocal sound to a main sound collecting microphone in a sound emitting
and collecting device in which the microphones are arranged. FIGS. 3A and 3B show the case
where the sound generated by the user 201 is picked up by the microphone 2A and the
microphone 92A, respectively.
04-05-2019
11
[0039]
The transmission distance Lv1 of the voiced sound in the sound emission and collection device of
the present embodiment shown in FIG. 3A is longer than the transmission distance Lv0 of the
generated sound in the conventional sound emission and collection device shown in FIG. 3B. .
However, compared with the distance from the user 201 to the sound emission and collection
device (corresponding to the distance from the user 201 to the microphone 92A), the difference
in distance from the side surface on the user 201 side to the microphone 2A is extremely short.
The increase in the amount of attenuation is of the extent that it is hardly necessary to consider.
Therefore, the sound emission and collection device of the present embodiment can collect the
voice with the sensitivity, that is, the sound pressure level substantially equal to the conventional
one.
[0040]
(2) Voice from the other party in another room These users 201 and 202 should listen to the
output voice from the speaker 3 of the sound emitting and collecting apparatus 100 when
listening to the voice from the other user in the room of another sound emitting and collecting
apparatus. It becomes.
[0041]
The speaker 3 is disposed on the second surface 10B (lower surface) of the housing 1, that is, the
surface facing the top surface of the desk 200, and emits the voice from the other user.
The emitted sound 300 is reflected on the top surface of the desk 200 and propagated while
being circumferentially diffused in the horizontal direction, and is diffused from the area of the
second surface 10B to the outside including the upward direction. , Uniformly transmitted to the
space including the users 201 and 202. At this time, part of the sound 300 is transmitted to the
first surface 10A side of the housing 1 via the side surface 10C of the housing 1. This voice is
hereinafter referred to as wraparound voice.
[0042]
Here, as in the case of the above-mentioned vocal sound, each of the microphones 2A to 2H picks
04-05-2019
12
up a wraparound sound transmitted from the end in the direction extending through the central
point O, that is, the side surface 10C at the farthest position, It hardly picks up the wraparound
sound transmitted from the side surface 10C at a position closest to each of the microphones 2A
to 2H. That is, it picks up the wraparound speech with the longest propagation path.
[0043]
FIG. 3C is a conceptual diagram showing the transfer distance Ls1 of the wraparound sound to
the microphone in the sound collection and collection device 100 of the present embodiment,
and FIG. 3D has the same configuration as FIG. 3B. FIG. 16 is a conceptual diagram showing a
transfer distance Ls0 of a wraparound sound to a microphone in a conventional sound emitting
and collecting device.
[0044]
The transmission distance Ls1 of the wraparound speech of the present embodiment shown in
FIG. 3C is longer than the transmission distance Ls0 of the conventional wraparound speech
shown in FIG. 3D.
This is because the conventional transmission distance Ls0 substantially matches the length from
the speaker 93 to the side surface 10C on which the microphone 92A whose directionality is the
outside of the housing 91 is installed. On the other hand, the transmission distance Ls1 of the
present embodiment is approximately the length from the speaker 3 to the side surface 10C, the
height of the side surface 10C, and the length from the position of the side surface 10C to the
microphone 2A disposed farthest at this position. Match the combined distance. As a result, the
transmission distance Ls1 of the wraparound speech in the present embodiment is twice or more
as long as the conventional transmission distance Ls0. As a result, the sound emission and
collection device of the present embodiment can significantly attenuate the incoming sound that
is collected as compared to the conventional sound emission and collection device.
[0045]
Furthermore, conventionally, the propagation direction only changes by 90 degrees from the
second surface 10B to the side surface 10C, but in the configuration of the present embodiment,
the propagation direction changes by 90 degrees from the side surface 10C to the first surface
10A. That is, in the present embodiment, there is one more change of 90 degrees in the
04-05-2019
13
propagation direction than in the prior art. Here, such a change in the propagation direction of
the wraparound speech is such that a reflection wall or the like exists ahead of the propagation
direction, and the change by the reflection by the wall surface is not forced to change, but is
naturally wraparound. Depending on the result, significant attenuation can be obtained.
Therefore, the sound emission and collection device of the present embodiment can attenuate the
wraparound sound very significantly as compared with the conventional sound emission and
collection device.
[0046]
As described above, by using the configuration of the present embodiment, voices and the like
from the user, which are necessary voices, are picked up with high sensitivity, and while the case
is kept small, the voices coming around from the speaker to the microphone are greatly reduced.
Can be attenuated. Thereby, high S / N ratio can be realized.
[0047]
Next, a signal processing functional unit that processes the collected sound signal collected as
described above will be described.
[0048]
FIG. 4 is a block diagram showing the configuration of the sound emission and collection device
of the present embodiment.
The sound emission and collection device of the present embodiment includes the abovedescribed input / output connector 26 in addition to the above-described microphones 2A to 2H
and the speaker 3, and further, as a signal processing function unit, input amplifiers 21A to 21H,
A / D converter 22A to 22H, a microphone signal processing circuit 23, an echo canceller 24, an
input / output interface 25, a D / A converter 31, and an output amplifier 32.
[0049]
The input / output interface 25 supplies an input voice signal input from the input / output
04-05-2019
14
connector 26 to the D / A converter 31 via the echo canceller 24. The D / A converter 31
converts an input audio signal into an analog signal and supplies it to the output amplifier 32,
and the output amplifier 32 amplifies the input audio signal and outputs it to the speaker 3. The
speaker 3 converts the sound of the input sound signal and emits the sound.
[0050]
Each of the microphones 2A to 2H picks up external voice, converts it into a pick-up signal, and
outputs it to the input amplifiers 21A to 21H. Each of the input amplifiers 21A to 21H amplifies
the collected sound signal and outputs it to the A / D converters 22A to 22H. The A / D
converters 22 </ b> A to 22 </ b> H convert the respective collected signals into digital signals,
and output the converted signals to the microphone signal processing circuit 23. Hereinafter,
sound collection signals collected by the microphones 2A to 2H and output from the A / D
converters 22A to 22H will be simply referred to as signals A to H, respectively.
[0051]
FIG. 5 is a detailed block diagram of the microphone signal processing circuit 23. The
microphone signal processing circuit 23 includes adders (subtractors) 231A to 231H, a select /
mixing circuit 232, and a maximum signal strength detection circuit 233.
[0052]
The signal A output from the A / D converter 22A is input to the adder 231A, and the signal E
output from the A / D converter 22E is input to the adder 231A. The adder 231A subtracts the
signal E from the signal A and outputs a correction signal A. Here, the signal A is a sound pickup
signal by the microphone 2A, and the signal E is a sound pickup signal by the microphone E. As
described above, since the microphones 2A and 2E are disposed at point-symmetrical positions
with respect to the center point O, the wraparound voices collected by each are substantially
equivalent. As a result, by subtracting the signal E from the signal A, it is possible to reduce this
wraparound speech component.
[0053]
Similarly, the correction signal B is generated by subtracting the signal F from the signal B in the
04-05-2019
15
adder 231B, the correction signal C is generated by subtracting the signal G from the signal C in
the adder 231C, and the correction signal D is added It is generated by subtracting the signal H
from the signal D in the unit 231D.
[0054]
The signal E output from the A / D converter 22E is input to the adder 231E, and the signal A
output from the A / D converter 22A is input to the adder 231E.
The adder 231 E subtracts the signal A from the signal E and outputs a correction signal E.
Similarly, the correction signal F is generated by subtracting the signal B from the signal F in the
adder 231F, the correction signal G is generated by subtracting the signal C from the signal G in
the adder 231G, and the correction signal H is added It is generated by subtracting the signal D
from the signal H in the unit 231H.
[0055]
As a result, each of the correction signals A to H can reduce the wraparound speech component.
[0056]
The generated correction signals A to H are input to the select / mixing circuit 232 and the
maximum signal strength detection circuit 233.
The maximum signal strength detection circuit 233 compares the signal strengths of the
correction signals A to H, that is, the sound pressure levels, selects the correction signal of the
highest signal strength, and selects the correction signal of the highest signal strength. To the
select / mixing circuit 232. The select / mixing circuit 232 selects a corresponding correction
signal from the input correction signals A to H based on the selection information supplied from
the maximum signal strength detection circuit 233 and outputs the selected correction signal to
the echo canceller 24. The maximum signal strength detection circuit 233 detects the correction
signal of the highest signal strength, selects the correction signal of the maximum signal
strength, and a plurality of correction signals adjacent to the correction signal, and selects /
mixes the signal. You may give to 232. Furthermore, in view of the case where there are a
plurality of sound sources in different directions, a plurality of correction signals may be selected
04-05-2019
16
in order from the correction signal with the highest signal strength and supplied to the selection
/ mixing circuit 232. In these cases, the select / mixing circuit 232 selects a plurality of
corresponding correction signals based on the selection information, mixes them, and outputs the
mixed signals to the echo canceller 24.
[0057]
By performing such selection processing, a correction signal with a low signal strength that is
unlikely to be considered as an utterance from the user is deleted, so that the S / N can be further
improved.
[0058]
FIG. 6 is a detailed block diagram of the echo canceller 24.
The echo canceller 24 includes an adaptive filter 241 and an adder 242. The adaptive filter 241
includes a digital filter such as an FIR filter, estimates a transfer function of an acoustic
propagation path from the speaker 3 to the microphones 2A to 2H, and filter coefficients of the
FIR filter so as to simulate the estimated transfer function. Calculate The adaptive filter 241
generates a pseudo-regression sound signal using the estimated filter coefficient, and outputs the
pseudo-regression sound signal to the adder 242. The adder 242 subtracts the pseudoregression sound signal from the output signal of the microphone signal processing circuit 23,
and outputs the result as an output sound signal to the input / output interface 25. Here, the
estimation of the transfer function and the calculation of the filter coefficient are fed back to the
adaptive filter 241 as a reference signal using the residual signal which is the signal output from
the adder 242, based on the input audio signal supplied to the speaker 3. It is done iteratively
using an adaptive algorithm. This optimizes the estimation of the transfer function and the
setting of the filter coefficients.
[0059]
By performing such processing, the wraparound speech component is further suppressed, and
the audio signal output to the input / output interface 25 further improves the S / N ratio.
[0060]
04-05-2019
17
As described above, in the sound emission and collection device of the present embodiment, the
wraparound sound can be mechanically reduced by setting the positional relationship between
the speaker and the microphone as described above.
Further, by setting the microphone installation pattern as described above, it is possible to
effectively suppress the wraparound speech component included in the collected sound signal of
each microphone, and by performing the echo canceling, the wraparound speech component Can
be further suppressed. This makes it possible to realize a very good S / N ratio for the output
audio signal.
[0061]
In the present embodiment, the recess 11 of the first surface 10A of the housing 1 is formed, and
the microphones 2A to 2H are disposed on the inner peripheral surface 12 of the recess 11.
However, as shown in FIG. The microphones 2A to 2H may be arranged in a structure.
[0062]
FIG. 7 is a diagram showing the configuration of the main part of the sound emission and
collection device of this embodiment of the sound emission and collection device of another
configuration of this embodiment, (A) is a plan view, (B) is (A) It is an AA 'sectional view in.
In the sound emission and collection device shown in FIG. 7, the microphones 2A to 2H are
disposed on the first surface 10A, and the microphones 2A to 2H are covered with a mesh-like
cover 13. The other configuration is the same. Even if it is such composition, the abovementioned effect can be produced.
[0063]
Further, in the present embodiment, the short cylindrical casing 1 has been described as an
example, but it may be an elliptic cylindrical shape whose plane cross section is an ellipse, or
may be a rectangular parallelepiped shape.
[0064]
Moreover, although the example which makes the 2nd surface 10B side which has the speaker 3
facing the top surface of the desk 200 and arrange | positions in this embodiment was shown,
04-05-2019
18
the 2nd surface 10B side which has the speaker 3 is a room where a user exists. It may be
arranged to be directed to the ceiling and to connect the foot 4 to the ceiling surface.
[0065]
Further, in the present embodiment, the case where eight microphones and one speaker are
shown, but as described above, the microphone and the speaker are disposed on the facing
surface of the housing, and the directivity of the microphone as described above Is set, the
number of microphones and the number of speakers can be set as appropriate.
[0066]
Further, the configuration of the microphone signal processing circuit 23 is not limited to the
example described above.
FIG. 8 is a block diagram showing another configuration of the microphone signal processing
circuit 23.
The microphone signal processing circuit 23 shown in FIG. 8 differs from the microphone signal
processing circuit 23 shown in FIG. 5 only in the signal combining portion.
The signal A output from the A / D converter 22A is input to the adder 231A, and the signal B
output from the A / D converter 22B is input to the adder 231A. The adder 231A adds the signal
A and the signal B and outputs the result. Similarly, the adder 231B adds and outputs the signal B
and the signal C, the adder 231C adds and outputs the signal C and the signal D, and the adder
231D adds the signal D and the signal E. Output. Further, the adder 231E adds and outputs the
signal E and the signal F, the adder 231F adds and outputs the signal F and the signal G, and the
adder 231G adds the signal G and the signal H. The adder 231H adds the signal H and the signal
A and outputs the added signal. As described above, the microphone signal processing circuit 23
illustrated in FIG. 8 adds and outputs the collected sound signals obtained from two adjacent
microphones. As described above, by adding the collected sound signals of the adjacent
microphones, the collected sound signal component from the front direction of the microphone,
that is, the direction in which the high sound collection sensitivity is set is strengthened, and the
collected sound signals from other directions The ingredients are weakened. This provides a
more directional signal.
04-05-2019
19
[0067]
Furthermore, the configuration of the microphone signal processing circuit 23 may be as follows.
FIG. 9 is a block diagram of a signal combining unit in another microphone signal processing
circuit 23. The microphone signal processing circuit 23 shown in FIG. 9 also differs from the
microphone signal processing circuit 23 shown in FIG. 5 only in the signal combining portion.
The microphone signal processing circuit 23 shown in FIG. 9 includes adders 237A to 237H and
delay circuits 234A to 234H and 235A to 235H and 236A to 236H. The signals A to H are input
to the delay circuits 234A to 234H, 235A to 235H, and 236A to 236H, respectively. For
example, the signal A is input to the delay circuits 234A, 235A, 236A, and the other signals B to
H are similarly processed.
[0068]
The delay circuits 234A to 234H, 235A to 235H, 236A to 236H delay the input signals so that
the three signals input to the adders 237A to 237H are in phase.
[0069]
The adder 237A adds the output signal (signal A) of the delay 234A, the output signal (signal B)
of the delay 235B, and the output signal (signal C) of the delay 236C, and outputs the result.
Similarly, the adder 237B adds and outputs the delayed signal B, the signal C, and the signal D,
and the adder 237C adds the delayed signal C, signal D, and signal E to each. The adder 237D
adds and outputs the signal D, the signal E, and the signal F which have been subjected to the
delay processing. Further, the adder 237E adds the delayed signal E, the signal F and the signal G
to each other and outputs it, and the adder 237F adds the delayed signal F, signal G and signal H
to each other and outputs it The adder 237G adds the delayed signal G, the signal H and the
signal A to each other and outputs the added signal. The adder 237H adds the delayed signal H,
the signal A and the signal B to each other and outputs it. Do. As a result, the collected sound
signals from three adjacent microphones are summed in the same phase. As a result, the signal
strength in the specific direction is further increased, the S / N is improved, and the directivity in
the specific direction is further improved. The number of signals to be added is not limited to
three, and the S / N ratio in the specific direction can be improved by adding or subtracting more
signals.
04-05-2019
20
[0070]
Although the microphone signal processing circuit 23 shown in FIGS. 8 and 9 has a configuration
in which the output signals A to H of the A / D converters 22A to 22H are directly processed, the
microphone signal processing circuit 23 is generated using a circuit as shown in FIG. The
corrected signals A to H may be input. This further improves the S / N ratio.
[0071]
It is a figure which shows the structure of the principal part of the sound emission and collection
device of embodiment of this invention. It is the figure which showed the case where two users
201 and 202 use the sound emission and collection apparatus 100 of embodiment of this
invention. It is a conceptual diagram showing transfer distance Lv1 of the vocal sound to the
microphone which performs the main sound collection in the sound emission and collection
device 100 of the embodiment of the present invention, and the vocal sound to the main sound
collecting microphone in the conventional sound emitting and collecting device FIG. 16 is a
conceptual diagram showing the transfer distance Lv0 of the speaker, and is a conceptual
diagram showing the transfer distance Ls1 of the wraparound sound to the microphone, and the
transfer distance Ls0 of the wraparound sound to the microphone in the conventional sound
emitting and collecting apparatus having a speaker on the side of the case. It is a conceptual
diagram showing. It is a block diagram showing composition of a sound emission and collection
device of an embodiment of the present invention. 5 is a detailed block diagram of a microphone
signal processing circuit 23. FIG. It is a detailed block of the echo canceller 24. It is a figure
which shows the structure of the principal part of the sound emission and collection apparatus of
this embodiment of the sound emission and collection apparatus of the other structure of this
embodiment. 15 is a block diagram showing another configuration of the microphone signal
processing circuit 23. FIG. FIG. 12 is a block diagram of a signal synthesis unit in another
microphone signal processing circuit 23;
Explanation of sign
[0072]
100-sound emission and collection device, 200-desk, 201, 202-user, 300-sound emission, 301,
302-vocal sound, 1, 91-case, 10A-first surface of case 1, 10B-case Second surface of body 1,
10C-side surface of housing 1, 2A-2H, 92A-microphone, 3-speaker, 4-foot, 21A-21H-input
04-05-2019
21
amplifier, 22A-22H-A / D converter, 23- Microphone signal processing circuit, 231A-231H,
237A-237H-adder, 232-select / mixing circuit, 233-maximum signal strength detection circuit,
234A-234H, 235A-235H, 236A-236H-delay circuit, 24-echo canceller , 241-adaptive filter, 242adder, 25-input / output interface, 26-input / output connector, 31-D / A converter, 32-output
Pump
04-05-2019
22