JP2009100372A - Call device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a call device capable of suppressing howling and improving the sound quality at the same time. <P>SOLUTION: The call device is equipped with a call receiver 1 for transducing into voices an audio signal which is an electric signal made by transducing voices, a call transmitter 2 for transducing voices into an audio signal, and an input/output portion 3 for amplifying the audio signal output by the call transmitter 2 and outputting the amplified signal to the outside while amplifying an audio signal input from the outside and inputting the amplified signal to the call receiver 1. The call transmitter 2 comprises a sound receiving portion 5 for outputting sound pressures of incoming voices, their time differential values, and spatial differential values of the sound pressures in respective axis directions of a two-dimensional Cartesian coordinate system and an audio signal generating portion 6 for performing a preset dead-point formation processing by a spatio-temporal gradient method by using the sound pressures, time differential values, and spatial differential values output by the sound receiving portion 5, creating an audio signal to form a sensitivity-minimizing dead point in the position of the call receiver 1, and outputting the signal to the input/output portion 3. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a call device.

  Conventionally, a receiving unit that converts an audio signal, which is an electrical signal into which voice is converted, into a voice, a transmitting unit that converts voice into a voice signal, and an audio signal output from the transmitting unit is amplified and output to the outside. In addition, a call device including an input / output unit that amplifies an audio signal input from the outside and inputs the signal to a receiving unit is used in, for example, an intercom system.

  In this type of communication device, the sound output from the receiver unit is input to the transmitter unit, amplified as an audio signal at the input / output unit, and output again as speech from the receiver unit. It is necessary to suppress a so-called howling phenomenon.

  Then, in order to suppress the occurrence of howling, as shown in FIG. 7, a first microphone M1 that is disposed in the vicinity of the reception unit 1 composed of a speaker and to which the sound mainly output from the reception unit 1 is input, The second microphone M2 is arranged at a position farther from the receiver 1 than the microphone of the first microphone, and the voice of the speaker is input together with the sound from the receiver 1, and the output of the first microphone from the output M2 of the second microphone. There has been proposed a communication device in which a transmitting unit is configured with a voice processing unit (not shown) that generates a voice signal in which the contribution of the output of the receiving unit 1 is reduced by reducing components common to M1 (for example, , See Patent Document 1).

According to the above communication device, since the component corresponding to the voice output from the receiver 1 is removed from the voice signal, the occurrence of howling is suppressed.
JP 2007-151135 A

  However, since the output of the first microphone M1 includes the voice of the speaker that should be included in the voice signal, the above-described call device suppresses not only the sound generated by the receiver 1 but also the voice of the speaker. As a result, there is a problem that the sound quality of the call is deteriorated.

  Further, according to the experiment of the present inventor, it has been found that the degree to which the component corresponding to the sound from the receiver 1 is suppressed in the audio signal by the above method depends on the frequency of the sound.

  The present invention has been made in view of the above reasons, and an object of the present invention is to provide a communication device capable of improving sound quality while suppressing howling.

  According to the first aspect of the present invention, there is provided a receiving unit that converts an audio signal, which is an electric signal obtained by converting a voice, into a voice, a transmitting unit that converts voice into a voice signal, and amplifies the voice signal output from the transmitting unit. A communication device comprising: an input / output unit for amplifying an audio signal input from the outside and inputting the amplified signal to the receiving unit; and a housing in which the receiving unit and the transmitting unit are respectively fixed. The speech unit includes a sound receiving unit that outputs a sound pressure of the incident sound, a time differential value of the sound pressure, and a spatial differential value of the sound pressure with respect to each axial direction of the two-dimensional orthogonal coordinate system, and a sound receiving unit By using the sound pressure, time differential value, and spatial differential value output from, a predetermined dead point formation process by the spatiotemporal gradient method is performed, so that a dead point with the lowest sensitivity is formed at the position of the receiver. Transmission signal generating means for generating a voice signal and outputting it to the input / output unit; Characterized in that it has.

  According to the present invention, it is possible to improve sound quality as compared with the case where the spatiotemporal gradient method is not used while suppressing howling.

  According to a second aspect of the present invention, in the first aspect of the present invention, the transmitter is provided with position input means for inputting a relative position of the receiver relative to the position of the sound receiver of the transmitter, and the position input means. When the voice signal is input, the transmission signal generating means calculates the parameter to be used for the dead point forming process so that the position of the dead point formed in the audio signal is the position input to the position input means. And a parameter calculation unit that updates a parameter used by the transmission signal generation unit for the dead point forming process to the parameter obtained by the calculation.

  According to the present invention, by inputting the position to the position input means, it is possible to share the transmitter with a plurality of types of communication devices having different positional relationships between the transmitter and receiver.

  According to a third aspect of the present invention, in the first or second aspect of the present invention, the transmitter is configured as a single semiconductor chip.

  According to the present invention, it is possible to reduce the size as compared with the case where the sound receiving means and the transmission signal generating means constituting the transmitting section are configured by individual components.

  According to a fourth aspect of the present invention, in any one of the first to third aspects of the present invention, at least the sound receiving section and the receiving section of the transmitting section are respectively fixed and provided with a case fixed to the housing. .

  According to the present invention, at the time of manufacture, at least the sound receiving unit and the sound receiving unit of the transmitting unit are each fixed to the case as a single call module, so that the sound receiving unit and the sound receiving unit of the transmitting unit are used. With a plurality of types of communication devices having the same positional relationship, the components constituting the communication module and the manufacturing equipment for manufacturing the communication module can be shared, and the manufacturing cost can be reduced.

  According to a fifth aspect of the present invention, in any one of the first to fourth aspects of the present invention, the sound receiving means includes a supported portion that is fixed directly or indirectly to the housing of the communication device and a gimbal with respect to the supported portion. The vibration received through the sound pressure of the sound incident on the diaphragm, which is provided at different locations of the diaphragm, and is provided with a diaphragm that is supported so as to be swingable through the structure and that receives the sound pressure. A plurality of sound pressure detectors for converting sound into an electrical signal, and a spatiotemporal gradient measurement processor for obtaining sound pressure, spatial differential values, and temporal differential values using outputs of the plurality of sound pressure detectors It is characterized by.

  According to a sixth aspect of the present invention, in any one of the first to fourth aspects of the present invention, the sound receiving means includes four microphones that are arranged in a rectangular apex and convert sound pressures of incident sounds into electric signals. And a spatiotemporal gradient measurement processing unit that obtains a sound pressure, a spatial differential value, and a temporal differential value by calculation using the output of each microphone.

  The invention according to claim 7 is the invention according to any one of claims 1 to 6, wherein the transmitter has an instruction input means for inputting an instruction for switching the operation mode, and the transmission signal generating means is an instruction input. In response to an instruction input to the means, in addition to the dead point forming mode for forming a dead point at the position of the receiving part, an audio signal based only on the sound pressure output by the sound receiving means without performing the dead point forming process is generated. The operation mode can also be switched to the omnidirectional mode to be generated.

  According to the present invention, when there is no possibility of occurrence of howling, by switching the operation mode to the omnidirectional mode, all sounds input to the sound receiving means can be output regardless of the position of the sound source. , The loss can be reduced and the sound quality can be improved.

  The invention of claim 8 is the invention according to any one of claims 1 to 7, wherein the transmitter has an instruction input means for inputting an instruction to switch the operation mode, and the transmission signal generating means is an instruction input. A space-time using sound pressure, time differential value and space differential value output by the sound receiving means in addition to the dead point forming mode for forming the dead point at the position of the receiving part according to the instruction input to the means Using an audio signal generated so as to form a dead point at a target position set in advance by a gradient method and an audio signal based only on the sound pressure output by the sound receiving means, a sound source near the target position is used. The operation mode can also be switched to a sound collection area forming mode for generating and outputting a sound signal that selectively reflects sound pressure due to sound.

  According to the present invention, the influence of noise around the speaker on the voice signal can be reduced by operating in the dead point formation mode with the position of the speaker as the target position.

  According to the first aspect of the present invention, the transmitting unit transmits the sound pressure of the incident sound, the time differential value of the sound pressure, and the spatial differential value of the sound pressure in each axial direction of the two-dimensional orthogonal coordinate system. There is a dead point where the sensitivity is minimized by performing a predetermined dead point forming process by a spatiotemporal gradient method using the sound receiving means to output, the sound pressure output by the sound receiving means, the time differential value, and the spatial differential value. Compared to the case where the spatio-temporal gradient method is not used while suppressing howling, since it has a transmission signal generation means for generating an audio signal formed at the position of the receiver and outputting it to the input / output unit. Sound quality can be improved.

  According to the invention of claim 2, the transmitter is configured to input the relative position of the receiver with respect to the position of the receiver of the transmitter, and when the position is input to the position input unit In addition, the transmission signal generation means calculates parameters to be used for the dead point formation processing so that the position of the dead point formed in the voice signal is the position input to the position input means, and the transmission signal generation means Has parameter calculation means for updating the parameter used for the dead point formation processing to the parameter obtained by the calculation, so that the positional relationship between the transmission part and the reception part is obtained by inputting the position to the position input means. The transmitter can be shared by a plurality of different types of communication devices.

  According to the third aspect of the present invention, since the transmitter section is constituted by one semiconductor chip, compared to the case where the sound receiving means and the transmission signal generating means constituting the transmitter section are constituted by individual parts. And downsizing is possible.

  According to the invention of claim 4, at the time of manufacture, at least the sound receiving unit and the receiving unit of the transmitting unit are each fixed to the case as one call module, so that the receiving unit and the transmitting unit are connected. A plurality of types of communication devices having a common positional relationship with the sound receiving unit can share parts constituting the communication module and manufacturing equipment for manufacturing the communication module, thereby reducing manufacturing costs.

  According to the seventh aspect of the present invention, the operation mode can be switched to the omnidirectional mode in which the voice signal is generated only based on the sound pressure output from the sound receiving means without performing the dead point forming process. If there is no possibility of occurrence of noise, switching the operation mode to the omnidirectional mode enables output of all audio input to the sound receiving means regardless of the position of the sound source, and also reduces sound quality by reducing loss. Can be improved.

  According to invention of Claim 8, it is produced | generated so that a dead point may be formed in the target position preset by the spatiotemporal gradient method using the sound pressure which the sound receiving means output, the time differential value, and the space differential value. A sound signal that selectively reflects the sound pressure due to the sound from the sound source in the vicinity of the target position using the sound signal that is based on only the sound pressure output from the sound receiving means and the sound signal output from the sound receiving means. Since the operation mode can be switched to the sound area formation mode, the influence of noise around the speaker on the audio signal can be reduced by operating in the dead point formation mode with the speaker position as the target position. Can do.

  The best mode for carrying out the present invention will be described below with reference to the drawings.

  In the present embodiment, as shown in FIG. 1, a receiving unit 1 that converts an audio signal, which is an electrical signal obtained by converting a voice, into a voice, a transmitting unit 2 that converts a voice into a voice signal, and a transmitting unit 2 And an input / output unit 3 that amplifies the audio signal input from the outside and inputs the amplified audio signal to the receiver unit 1. The receiver 1 and transmitter 2 are each fixed to the housing 4 so that the positional relationship between them does not change.

  The receiver 1 is composed of, for example, a known speaker.

  The input / output unit 3 amplifies the voice signal output from the communication unit 31 connected to the device to be talked via, for example, a signal line (not shown) and the transmission unit 2, and is externally connected via the communication unit 31. And a call processing unit 32 that amplifies an audio signal input from the outside to the communication unit 31 and inputs the amplified signal to the receiver unit 1. When this embodiment is used as a slave unit of an interphone system, a device to be connected to which the communication unit 31 is connected is a master unit of the interphone system. The call processing unit 32 includes a first receiver-side amplifier 32a that amplifies a voice signal input from the communication unit 31, a receiver-side attenuator ILR that attenuates a voice signal output from the first receiver-side amplifier 32a, and a receiver. A second receiver amplifier 32b that amplifies the voice signal output from the side attenuator ILR and inputs the amplified signal to the receiver unit 1, and a first transmitter amplifier 32c that amplifies the voice signal input from the transmitter unit 2; The transmission side attenuator ILT for attenuating the audio signal output from the first transmission side amplifier 32c and the audio signal output from the transmission side attenuator ILT are amplified and output to the outside via the communication unit 31. It comprises a second transmitter amplifier 32d and an attenuation control unit 32e for controlling the attenuation rate of each attenuator ILR, ILT of the call processing unit 32. The attenuation control unit 32e includes the intensity of the voice signal passing through the receiving side attenuator ILR (hereinafter referred to as “receiving signal”) and the voice signal passing through the transmitting side attenuator ILT (hereinafter referred to as “transmission signal”). And the attenuation rate of one of the attenuators ILR and ILT through which the lower-intensity audio signal passes is determined from the attenuation rate of the other attenuator ILR and ILT. Also make it high. That is, the call processing unit 32 reduces the smaller one of the volume of the audio signal input to the receiver 1 (received volume) and the volume of the audio signal output from the transmitter 2 (transmitted volume). Thus, the loop gain is lowered to suppress howling, and the call processing unit 32 is a so-called voice switch as a whole.

  As shown in FIG. 2, the transmitter 2 outputs the sound pressure of the incident sound, the time differential value of the sound pressure, and the spatial differential value of the sound pressure for each axial direction of the two-dimensional orthogonal coordinate system. There is a dead point at which the sensitivity is minimized by performing the predetermined dead point forming process by the spatiotemporal gradient method using the sound receiving unit 5 and the sound pressure, the time differential value, and the spatial differential value output by the sound receiving unit 5. The voice signal generation unit 6 as a transmission signal generation unit that generates a voice signal formed at the position of the reception unit 1 and outputs the voice signal to the call processing unit 32 of the input / output unit 3.

More specifically, the sound receiving unit 5 is provided with an omnidirectional arrangement in which square vertices are arranged so that two rows are arranged in a direction parallel to the x-axis and a direction parallel to the y-axis in the three-dimensional orthogonal coordinate system. Of four microphones 50A, 50B, 50C, 50D and the output signals f _A (t), f _B (t), f _C (t), f _D (t) of the microphones 50A-50D And a spatiotemporal gradient measurement processing unit 51 for performing measurement processing. In the spatiotemporal gradient measurement processing unit 51, the in-phase component M (t) of the sound pressure is determined from the output signals f _A (t), f _B (t), f _C (t), and f _D (t) of the microphones 50A to 50D. , Time differential value (time gradient component) M _t (t), x-axis direction spatial differential value (x-axis direction spatial gradient component) M _x (t), y-axis direction spatial differential value (y-axis direction spatial gradient component) M Each _y (t) is obtained from the following equation.

M (t) = f _A (t) + f _B (t) + f _C (t) + f _D (t)
M _t (t) = df _A (t) / dt + df _B (t) / dt + df _C (t) / dt + df _D (t) / dt
M _x (t) = f _A (t) + f _B (t) −f _C (t) −f _D (t)
_{M y (t) = f A} (t) -f B (t) + f C (t) -f D (t)
The audio signal generation unit 6 outputs the in-phase component M (t), the time differential value M _t (t), the x-axis direction spatial differential value M _x (t), and the y-axis direction spatial differential value M output from the sound receiving unit 5. _A dead point is formed by applying the spatiotemporal gradient method using _y (t).

  Here, in describing the dead point formation processing by the audio signal generation unit 6, the space-time gradient method will be described in detail first.

  The spatiotemporal gradient method was originally proposed as one of the methods for determining an optical flow that is an apparent velocity field in a moving image (see Reference 1). Assumption (f (x, y, t) = f (x + δx, y +) that the image function f (x, y, t) representing the characteristics of the light and shade pattern in the moving image is kept unchanged during the movement. δy, t + δt)), the analysis method is based on an equation that relates the velocity of the optical flow at a certain point (x, y) to the spatial gradient and temporal gradient of the gray-scale distribution of the moving image. The method will be explained in detail below.

At time t + δt, when the density pattern f (x + δx, y + δy, t + δt) at coordinates (x + δx, y + δy) is tailored around (x, y, t),
f (x + δx, y + δy, t + δt) = f (x, y, t) + f x δx + f y δy + f t δt + O (δx + δy + δt) ... (1)
It becomes. Here, O (δx + δy + δt) is a second-order or higher term of δx, δy, δt, but is ignored in the following because it is a minute amount. At this time, the shading pattern at the coordinates (x, y) at time t is
When the density value distribution is moved to the coordinates (x + Δx, y + Δy) after the time Δt has elapsed, the following equation is established from the correspondence.

f (x, y, t) = f (x + δx, y + δy, t + δt)
= f (x, y, t ) + f x δx + f y δy + f t δt ... (2)
_{f x δx + f y δy +} f t δt = 0 ... (3)
Dividing both sides of equation (3) by δt,
_{f x δx / δt + f y} δy / δt + f t = 0 ... (4)
Get. Here, assuming that δt is infinitesimal, assuming that δt → 0, the following equation is obtained.

_{f x dx / dt + f y} dy / dt + f t = 0 ... (5)
Using optical flow velocity v = (u, v) = (dx / dt, dy / dt), equation (5) becomes
_{uf x + vf y + f t} = 0 ... (6)
Equation (6) is an equation relating the gradient of time and space of the gray value of the moving image to the optical flow velocity v.

  Next, an assumption is made that “the velocity field can be approximated to be almost constant in the vicinity region Γ of a certain point of interest”. At this time, Equation (6) must be established everywhere in the region Γ. Therefore, evaluation is performed using the square integral on the left side of equation (6) (the following equation (7)), and the velocity field is obtained by the method of least squares.

Differentiating equation (7) with respect to u and v and setting it to 0,
uS _xx + vS _xy + S _xt = 0, uS _xy + vS _yy + S _yt = 0… (8)

Is obtained. Solving equation (8) gives the velocity vector (u, v)
u = (S _yt S _xy -S _xt S _yy ) / (S _xx S _yy -S ² _xy ), v = (S _xt S _xy -S _yt S _xx ) / (S _xx S _yy -S ² _xy )… (Ten)
It is required as follows.

  Next, by applying the spatiotemporal gradient method for obtaining the optical flow velocity in the moving image described above, based on the linear relationship established between the sound pressure at one point in the sound field created by the sound source in the space and the spatiotemporal gradient. A method for localizing the sound source position will be described (see Reference 2).

As shown in FIG. 3, it is assumed that a three-dimensional orthogonal coordinate system having an observation point as an origin is taken, and there are a plurality of point sources that are uncorrelated with each other in front (z> 0). The speed of sound is c, the coordinates of the i-th sound source are (x _i , y _i , z _i ), the distance between the sound source and the observation point is R _i = (x _i ² + y _i ² + z _i ² ) ^1/2 , the source sound g ⁱ (t), as the sum of the spherical waves of the sound field in which each sound source is formed at the observation point when the f ⁱ (t), the composite sound field f that is formed at the observation point from these,

It is expressed. By performing partial differentiation, the x, y differentiation and time differentiation of the sound field at the observation point are expressed by the following equations (12), (13), (14).

here,
ξ ⁱ _x = x _i / R _i ² , ξ ⁱ _y = y _i / R _i ² (15)
Is called the intensity gradient,
τ ⁱ _x = x _i / cR _i , τ ⁱ _y = y _i / cR _i (16)
Is called the time gradient in the x and y directions.

Next, for the sake of simplicity, a sound source localization method in the case of one sound source will be described. For one sound source, equations (12) and (13) are
f _x = -ξ _x f-τ _x f _t , f _y = -ξ _y f-τ _y f _t (17)
Thus, τ _x , τ _y , ξ _x , ξ _y are obtained by applying the method of least squares in the same manner as in equation (1). The evaluation function in the short time window Γ
_{J = ∫ Γ {(f x} + ξ x f + τ x f t) 2 + (f y + ξ y f + τ y f t) 2} dt ... (18)
And When the equation (18) is partially differentiated with respect to τ _x , τ _y , ξ _x , ξ _y and set to 0, the following equation is obtained.

∂J / ∂τ _x = ∫ _Γ 2 (f _x + ξ _x f + τ _x f _t ) ・ f _t dt = 0, ∂J / ∂τ _y = ∫ _Γ 2 (f _y + ξ _y f + τ _y f _t ) ・ f _t dt = 0
… (19)
∂J / ∂ξ _x = ∫ _Γ 2 (f _x + ξ _x f + τ _x f _t ) ・ f _t dt = 0, ∂J / ∂ξ _y = ∫ _Γ 2 (f _y + ξ _y f + τ _y f _t ) ・ f _t dt = 0
… (20)
Where the covariance matrix estimated from the observation window Γ is

Then, equations (19) and (20) are
S _xt + ξ _x S _t + τ _x S _tt = 0, S _yt + ξ _y S _t + τ _y S _tt = 0… (22)
S _x + ξ _x S + τ _x S _t = 0, S _y + ξ _y S + τ _y S _t = 0… (23)
Rewritten. By solving Expressions (22) and (23), τx, τy, ξx, and ξy are obtained as follows.

τ _x = (S _x S _t -SS _xt ) / (SS _tt -S _t ² ) 、 τy = (S _y S _t -SS _yt ) / (SS _tt -S _t ² )… （24）
_{ξ x = (S xt S t} -S x S tt) / (SS tt -S t 2), ξy = (S yt S t -S y S tt) / (SS tt -S t 2) ... (25)
The azimuth angle (x / R, y / R) = (cτ _x , cτ _y ) of the sound source can be obtained from equations (21) and (24). The distance R to the sound source can be obtained by applying the least square method from the equations (15) and (16). Evaluation function

And this is a partial differential with 1 / R and set to 0

It becomes. Solving this
R = c (τ _x ² + τ _y ² ) / (τ _x ξ _x + τ _y ξ _y )… (28)
The distance to the sound source is required.

  Next, a method for directivity control using the spatiotemporal gradient of the sound field will be described (see References 3 to 5). Assuming the case of one sound source, the spatial gradient in the x and y directions of the sound pressure signal f (t) at the observation point is given by equations (12) and (13).

It becomes. When this equation is rewritten using the vector r = (x, y, z) from the sound source to the observation point,

It becomes. Then _{f (t), f t (} t), when ∇f (t) is observed, these weighted sum is

It is expressed. Here, u and u _t are real constants, and w = (w _x , w _y , 0) is a unit vector with the observation point as the origin and pointing in an arbitrary direction. Substituting equation (30) into equation (31),

It becomes. Therefore, the load sum of the spatiotemporal gradient is expressed as the sum of filters having different directivity characteristics H (r) and H _t (r) for f (t) and f _t (t), respectively. When H (r) = α, equation (33) becomes

And can be transformed. Here, if the angle formed by the two vectors a and b is θ, the following formula holds.

Using the formula of equation (38), equation (36) can be rewritten as

Here, from | w | = 1,

It is expressed by the sphere equation. When u + α = 0, equation (35) becomes r · w = 0 (42)
It becomes. Also, when H _t (r) = α, equation (34) becomes

Therefore, if the angle between vectors r and w is θ (r), | w | = 1

It becomes. Therefore, equation (43) becomes

It becomes.

From the equations (41), (42), and (45), the following properties are obtained for H (r) and H _t (r).
1) The two directivity characteristics H (r) and H _t (r) have a rotationally symmetric body with w as the axis. 2) When H (r) = 0, the distribution of r has a diameter 1 / u (u ≠ 0). ) To form a spherical surface or plane (u = 0) 3) When H _t (r) = 0, the distribution of r forms a conical surface or plane (u _t = 0) with apex angle 2cu _t (ut ≠ 0) 4) The intersection of the distributions of r when H (r) = 0 and H _t (r) = 0 is a circle or plane.

Get. Therefore, the frequency response T (r, w) from the sound source r to s (t) is
T (r, w) = H (r) + jωH _t (r) (47)
If H (r) and H _t (r) are real numbers, if T (r, w) = 0, H (r) = 0, H _t (r) = 0 (48)
It becomes. Therefore, it can be seen from equation (47) that the zero distribution where S (ω) = 0 does not depend on the frequency ω, but only on the sound source position r. Therefore, the time slope and x of the sound pressure at the observation point, when the spatial gradient in the y-direction is obtained, to form the zero sensitivity region (dead point), at a certain moment f, f _t, f _x, the f _y It is only necessary to take the load sum and perform compensation filter processing (low-pass filter processing).

The audio signal generation unit 6 of the present embodiment includes an in-phase component M (t), a time differential value M _t (t), an x-axis direction spatial differential value M _x (t), and a y-axis direction output from the sound receiving unit 5. coefficient vector defining the vector M = (M (t) M t (t) M x (t) M y (t)) T to the spatial differential value M _y (t) components, the load for these as elements After calculating the load sum with W = (W W _t W _x W _y ) ^T , it has a dead point forming unit 61 that forms a dead point at a predetermined arbitrary position by passing through the low-pass filter 61a. . Specifically, the directivity characteristics H (r) and H _t (r) described above are replaced with H ₁ (r _i ) and H ₂ (r _i ), respectively, and defined as follows (where r _i is a sound source) i's position vectors, n _ix and n _iy are the x and y components of r _i / | r _i |, respectively).

Further, two constraint conditions such as the following formulas (51) and (52) are set for these H ₁ (r _i ) and H ₂ (r _i ).
W ^H H ₁ (r _i ) ＝ p… (51)
W ^H H ₂ (r _i ) ＝ q… (52)
Here, p and q are positive real constants, respectively. Then, the gain due to the load sum using the coefficient vector W does not depend on the coefficient vector W and becomes p + jωq. In order to compensate for this, the dead point forming unit 61 in FIG. A first-order low-pass filter 61a of (p + jωq) ⁻¹ is provided after the stage. As a result, the gain of the dead point forming unit 61 as a whole is 1 with respect to the sound source i at the position r _i . Furthermore, a vector indicating the position of the assumed speaker is set as r ₁ with respect to the position of the sound receiving unit 5 of the transmitting unit 2, and p and q are, for example, p = 1 / | r ₁ |, q = 1, respectively. / c.

  Then, the coefficient vector W is the output power of the dead point forming unit 61 in the observation time interval Γ under the conditions of the equations (51) and (52).

Is obtained by using a Minimum Variance Beamformer (MV method) for minimizing. This solution is expressed by the following equations (54) and (55).

However, Bij (i, j = a, x, y) in equation (55) is expressed by equation (56), and b _a (t), b _x (t), b in equation (56) _y (t) is represented by the equations (57) to (59), respectively.

Instead of using the low-pass filter 61a, a coefficient vector W such that W ^H H ₁ (r _i ) = 0 and W ^H H ₂ (r _i ) = 0 may be selected. In this case, a dead point independent of the frequency can be formed at the position r _i of the sound source i. That is, the output O (t) of the dead point forming unit 61 does not include the sound pressure of the sound emitted from the sound source existing at the dead point.

  In the present embodiment, the center of the range where sound is generated in the receiver 1 as a parameter (hereinafter simply referred to as “parameter”) such as the coefficient vector W used by the dead point forming unit 61 in the above-described dead point forming process. Is used so that a dead point is formed at the position (hereinafter simply referred to as “the position of the receiver 1”), and the audio signal generator 6 outputs the output O (t ) Is input to the call processing unit 32 of the input / output unit 3 as an output of the audio signal generation unit 6, and the operation in the dead point formation mode is possible.

  That is, in the dead point formation mode, a dead point is formed at the position of the receiver 1, thereby reducing the loop gain over all frequencies and suppressing howling. Further, compared to the case where the spatio-temporal gradient method is not used as in the conventional example, since the loss is reduced with respect to the sound from the sound source other than the receiver 1, the sound quality can be improved. Furthermore, unlike the conventional example, since the sound generated in the receiver 1 is removed regardless of the frequency, the howling margin is increased compared to the conventional example.

  In addition, since howling is suppressed by the dead point forming process as described above, in this embodiment, the necessity for suppressing howling due to attenuation by the attenuators ILR and ILT of the call processing unit 32 is reduced. FIG. 4 is a graph in which the ratio of the intensity of the received signal to the intensity of the transmitted signal is taken on the horizontal axis, and the gains of the attenuators ILR and ILT are taken on the vertical axis. The gain of the ILR is shown, and the solid line and the broken line on the lower right side show the gain of the transmitting side attenuator ILT. When the dead point forming unit 61 is not used, the gains of the respective attenuators ILR and ILT need to be set low (increase the attenuation rate) as shown by broken lines, whereas in the present embodiment, as described above. Since the necessity of suppressing howling due to attenuation by the attenuators ILR and ILT of the call processing unit 32 is reduced, the gains of the attenuators ILR and ILT are generally increased (attenuation rate) as shown by the solid line. Therefore, the sound quality can be improved. Further, in the present embodiment, a two-way simultaneous call is possible because the attenuation rate in the call processing unit 32 is suppressed to be relatively low even for a voice signal having a lower intensity among the transmission signal and the reception signal. It has become.

  Furthermore, this embodiment can be operated in an operation mode other than the above-described dead point formation mode. More specifically, the audio signal generation unit 6 of the present embodiment uses the output M (t) of the sound receiving unit 5 and the output O (t) of the dead point forming unit 61 to reverse the dead point formation mode. In addition, the sound collection unit 61 extracts and outputs the sound signal S (t) in which the sound from the sound source at the position considered as the dead point is reflected in the dead point forming unit 61 and the influence of the sound from the surrounding sound sources is reduced. The inputs to the area forming unit 62 and the call processing unit 32 of the input / output unit 3 are output O (t) of the dead point forming unit 61 and output S (t) of the sound collection area forming unit 62 according to the operation mode. And a switching unit 63 that selectively switches to either one of the output M (t) of the sound receiving unit 5. That is, the state where the output O (t) of the dead point forming unit 61 is input to the input / output unit 3 is the dead point forming mode. Hereinafter, the state in which the output S (t) of the sound collection area forming unit 62 is input to the input / output unit 3 is referred to as a sound collection area forming mode, and the output M (t) of the sound receiving unit 5 is the input / output unit. The state input to 3 is called an omnidirectional mode. Furthermore, the transmission unit 2 controls the switching unit 63 according to the instruction input unit 81 to which an instruction for switching the operation mode is input, and the instruction input to the instruction input unit 81, and sets parameters to the dead point forming unit 61. And a transmission control unit 7 for instructing. The instruction input unit 81 may have, for example, a push button switch to receive an operation input for instructing switching of an operation mode, or a known multiplexed communication technique from another device (for example, a master unit of an interphone system). It is also possible to receive an electric signal transmitted by being superimposed on the audio signal.

  An operation mode other than the dead point formation mode will be described. First, the omnidirectional mode is a mode in which the audio signal obtained by the sound receiving unit 5 is output to the input / output unit 3 as it is. For example, the sound input to the sound receiving unit 5 is connected to the communication unit 31. For example, when recording in another device, no audio signal is input to the receiver 1 (in other words, there is no received signal), and there is a risk of howling without processing by the dead point forming unit 61. Used when there is no. If such an omnidirectional mode is used, all sounds input to the sound receiving unit 5 can be output regardless of the position of the sound source, and the loss is less than when using other operation modes. There is an advantage that the sound quality can be improved.

  Next, the sound collection area formation mode will be described. The operation in the sound collection area forming mode was proposed by the present inventor in Japanese Patent Application No. 2007-149570. More specifically, in the sound collection area forming mode, the position of the dead point formed by the dead point forming unit 61 is not the position of the receiving unit 1 but the speaker as a sound source to which the sound is to be input to the receiving unit 5. A parameter is designated by the transmission control unit 7 to the dead point forming unit 61 so as to be a predetermined target position assumed as a position.

  The sound collection area forming unit 62 is an audio signal in which both sound from a sound source (speaker) at a target position (hereinafter referred to as “target sound”) and noise from sound sources around the target position are reflected. The in-phase component of the output of the sound receiving unit 5 (hereinafter referred to as “total sound pressure” in the description of the sound collection area forming mode) M (t) and noise from the sound source around the target position are mainly reflected. In contrast to the output of the dead point forming unit 61 (hereinafter referred to as “noise component” in the description of the sound collection area forming mode) O (t), which is a known audio signal, a conventionally known spectral subtraction method (references) 6) is applied to extract the audio signal S (t) mainly reflecting the target sound.

  Hereinafter, the operation of the sound collection area forming unit 62 will be described. First, each output M (t), O (t) to be used is divided for each unit time (frame time) by the frame dividing unit 62a, and each divided output M (t, k), O (t, k) is transformed from the time domain to the frequency domain by a fast Fourier transform unit (FFT) 62b (where k represents a frame number). The average amplitude μ (= E {| O (f, k |}) of the noise component O (f, k) is calculated by the noise average amplitude calculator 62c, and the total sound pressure M (( The average amplitude μ of the noise component O (f, k) is subtracted from the amplitude value | M (f, k) | of the f, k), and the subtracted value (| M (f, k) | −μ) Output S (f, k) that does not contain noise by multiplying the phase (= exp {j∠M (f, k)}) of the total sound pressure M (f, k) calculated by the phase calculation unit 62e. = (| M (f, k) | −μ) · exp {j∠M (f, k)} is extracted, and this output S (f, k) is inversely transformed by the fast Fourier transform to return from the frequency domain to the time domain. Thus, the speech signal S (t) in which the noise around the speaker is suppressed and the target sound is emphasized is obtained.

  As described above, in the sound collection area formation mode, the sound collection area formation unit 62 generates the audio signal S (f, k) in which the target sound emitted from the sound source existing at the target position (dead point) is emphasized. Even if a noise source exists in the direction of the position, only the sound emitted from the sound source existing at the target position can be emphasized and output to the input / output unit 3. FIG. 5 shows the total sound pressure M (f) output from the sound receiving unit 5, the noise component O (f) output from the dead point forming unit 61, and the amplitude of the audio signal S (f) output from the sound collection area forming unit 62. An example of the frequency characteristic of (sound pressure) is shown, and it can be seen that the noise component is sufficiently suppressed in the output S (f) of the sound collection area forming unit 62. That is, if the sound collection area formation mode is used, it is possible to make a call by extracting the voice of the speaker even in an environment with a large ambient noise. Further, in the normal use state, the target position is set to a position sufficiently away from the receiver 1, so that an effect of suppressing howling can be obtained even in the sound collection area forming mode. Furthermore, in the sound collection area formation mode, not only the sound that directly reaches the sound receiving unit 5 from the receiver 1 but also the sound that has passed through the room is suppressed, so that howling due to reverberant sound is also suppressed.

Here, as a method of obtaining the sound signal S (t) in which the noise around the target position is reduced as described above in the sound collection area forming unit 62, a known method is used instead of the spectrum subtraction method as described above. An independent component analysis (ICA) method may be used. Further, instead of making the parameters used by the dead point forming unit 61 constant in the sound collection area forming mode, the transmission control unit 7 performs the dead point forming unit 61 at a predetermined timing during the operation in the sound collecting area forming mode. controls in-phase component M output by the sound receiving unit 5 (t), the time differential value M _t (t), x-axis spatial differential value M _x (t), y-axis spatial differential value M _y (t) The position of the speaker is detected by a well-known sound source localization method using, and the target position is determined by changing the parameters so that the target position (the position of the dead point) becomes the position of the detected speaker. You may make it adjust automatically to the position of.

  In addition, the present embodiment includes a position input unit 82 for inputting the position of the receiver 1 with respect to the position of the sound receiver 5 (strictly, for example, the center position of the range surrounded by the microphones 50A to 50D). When the position of the receiver 1 is input to the position input unit 82, the transmission control unit 7 calculates parameters such that a dead point is formed at the position of the receiver 1 input to the position input unit 82. In the dead point forming mode, the dead point forming unit 61 is instructed to use the parameter. That is, the transmission control unit 7 is parameter calculation means in the claims. The position input unit 82 may be, for example, a push button switch that receives an operation input indicating the position of the receiver 1 or a terminal to which a setting device is connected.

By the way, in the sound receiving unit 5, instead of receiving sound by the four microphones 50A to 50D as described above, sound is received by one diaphragm 52 as shown in FIG. 6A. One microphone 50E as shown in FIG. 6 may be used (see References 7 and 8). The diaphragm 52 is composed of a thin disk as a whole, and is supported by a supported portion 52a fixed to the housing 4 and outward in the radial direction of the supported portion 52a (see FIG. 6A). A first connecting portion 52b projecting from both the upper and lower sides), an intermediate portion 52c having an annular shape and having both ends in the diameter direction of the inner periphery connected to the first connecting portion 52b, A second connecting portion 52d projecting from the portion 52c in a direction perpendicular to the protruding direction of the first connecting portion 52b (on the left and right sides in FIG. 6A), and an annular shape. In addition, both ends of the inner circumference in the diametrical direction each have a vibration part 52e connected to the second connection part 52d. A microphone 50E shown in FIGS. 6 (a) and 6 (b) projects from a package 53 having a housing recess 53a in which the diaphragm 52 is housed and fixed to the housing 4, and a bottom surface of the housing recess 53a of the package 53. And a column 54 fixed to the supported portion 52 of the diaphragm 52 so that the thickness direction of the diaphragm 52 is directed in the depth direction of the housing recess 53a. Further, the connecting portions 52b and 52d of the diaphragm 52 can be elastically deformed so as to be twisted. That is, the vibration part 52e is supported with respect to the supported part 52a via a so-called biaxial orthogonal gimbal structure, and the left and right directions in FIG. The seesaw movement is possible by twisting the first connecting part 52b, and the seesaw movement is possible in all directions in the vertical direction of FIG. 6A by being able to move the seesaw by twisting the second connecting part 52d. (Oscillation) is possible. Further, the vibration part 52e can be elastically deformed such that the peripheral part is displaced in the thickness direction with respect to the central part. If such a vibration plate 52 is used, the force received by the vibration part 52e is converted into an electric signal by the sound pressure of the sound incident on the vibration part 52e (for example, the sound pressure is detected based on the vibration of the vibration part 52e). By providing sound pressure detectors at a plurality of locations of the diaphragm 52, the in-phase component M (t), time differential value M _t (t), x-axis direction spatial differential value M _x (t), y-axis direction spatial differential it is possible to obtain a value M _y (t) from one diaphragm 52. In this case, the directions of the x-axis and the y-axis are directions orthogonal to the thickness direction of the diaphragm 52 that has been elastically restored. As the sound pressure detection part, for example, one electrode is provided on the vibration part 52e of the diaphragm 52, and the other electrode is provided on the bottom surface of the housing recess 53a of the package 53. And a piezoelectric element that detects distortion (that is, converts into an electric signal) generated in the diaphragm 52 can be used. Further, in-phase component from the output of the sound pressure detector M (t), the time differential value M _{t (t),} x-axis spatial differential value M _x (t), obtaining a y-axis direction spatial differential value M _y (t) Since the spatio-temporal gradient measurement processing unit can be realized by a well-known technique, a description thereof will be omitted.

  Each configuration of the transmitter 2 as described above is integrated on one semiconductor chip using a well-known semiconductor process technology, and can be configured as one device called a so-called MEMS (Micro Electro Mechanical Systems). In addition, it is desirable to do so from the viewpoint of miniaturization.

Further, the receiver 1 and at least the sound receiver 5 of the transmitter 2 are fixed to a common case (not shown), and the receiver 1 and transmitter 2 are integrated through the case at the time of manufacture. If the sound receiving unit 5 is housed in the housing 4 as a single call module as a whole, a plurality of types of call devices in which the positional relationship between the sound receiving unit 1 and the sound receiving unit 5 of the transmitter unit 2 is common at the time of manufacture. Therefore, it is desirable that the communication module parts and manufacturing equipment can be shared, and the manufacturing cost can be reduced.
<List of references>
Reference 1: Shigeru Ando “Velocity vector distribution measurement system using spatio-temporal differential calculation of images” Transactions of the Society of Instrument and Control Engineers 22-12, 1330/1336 (1986)
Reference 2: Shigeru Ando, Hiroyuki Shinoda, Katsuya Ogawa, Satoshi Mitsuyama "Three-dimensional sound source localization sensor system based on spatiotemporal gradient method" Vol. 29, No. 5, p520-528, 1993
Reference 3: N. Ono, T. Arita, Y. Senjo, and S. Ando, “Directivity steeringprinciple for biomimicry silicon microphone”, Proc. Int. Conf. Solid State Sensors,
Actuators, and Microsystems (Transducers'05), pp. 792-795, 2005.
Reference 4: Ono, Ando, “Measurement of sound field and directivity control, 22nd Sensing Forum document, pp. 305-310, 2005.
Reference 5: Ono, Arita, Chiaki, Ando, “Theory of Directional Control and Sound Source Separation Based on Spatiotemporal Gradient Measurement, Proc. Of the Spring Meeting of the Acoustical Society of Japan 2005, 2-6-13, pp. 607-608, 2005.
Reference 6: SFBoll "Suppression of Acoustic Noise in Speech. UsingSpectral Subtraction" IEEE Trans.on. Acoustics, Speech and Signal Processing Vol. ASSP-27, No. 2, pp. 113-1, 1979
Reference 7: Junji Ono, Akihito Saito, Shigeru Ando “Theory and Experiments of Localization Sensors Simulating Ultra-Small Sound Sources (2nd Report)”, 19th Sensing Forum, pp.379-382,2002
Reference 8: Junji Ono, Akito Saito, Shigeru Ando, "Theory and Experiment of Differential Detection Type Sound Source Localization Sensor Imitating Drosophila", Auditory Society, pp.187-192,2002

It is a block diagram which shows embodiment of this invention. It is a block diagram which shows a transmission part same as the above. It is explanatory drawing for demonstrating the spatiotemporal gradient method in the same as the above. It is explanatory drawing which shows an effect same as the above. It is explanatory drawing which shows the effect of a sound collection area formation mode same as the above. (A) (b) shows the microphone in another form same as the above, (a) is a top view of a diaphragm, (b) is sectional drawing. It is a perspective view which shows a prior art example.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Reception part 2 Transmission part 3 Input / output part 4 Housing 5 Sound reception part 6 Voice signal generation part (Transmission signal generation means in Claim)
7 Transmission control unit (parameter calculation means in claims)
50A-50D Microphone 51 Spatiotemporal gradient measurement processing unit 52 Diaphragm 52a Supported portion 52e Vibrating unit 81 Instruction input unit 82 Position input unit

Claims (8)

A voice receiving unit that converts a voice signal, which is an electrical signal converted from voice, into a voice, a voice transmitting unit that converts voice into a voice signal, and amplifies the voice signal output from the voice transmitting unit and outputs the amplified signal to the outside A communication device comprising: an input / output unit that amplifies a voice signal input from the input unit and inputs it to the receiver unit; and a housing in which the receiver unit and the transmitter unit are fixed,
A transmitter unit that outputs a sound pressure of an incident sound, a time differential value of the sound pressure, and a spatial differential value of the sound pressure in each axial direction of a two-dimensional orthogonal coordinate system; By using a sound pressure, a temporal differential value, and a spatial differential value output by the means to perform a predetermined dead point forming process by a spatiotemporal gradient method, a dead point where sensitivity is minimized is formed at the position of the receiver. And a transmission signal generating means for generating a simple voice signal and outputting it to the input / output unit. The transmitter is a position input means for inputting the relative position of the receiver relative to the position of the receiver of the transmitter;
Parameters that the transmission signal generation means should use for dead point formation processing so that when the position is input to the position input means, the position of the dead point formed in the audio signal becomes the position input to the position input means. The communication apparatus according to claim 1, further comprising: a parameter calculation unit that calculates a parameter used by the transmission signal generation unit for the dead point forming process to a parameter obtained by the calculation.   3. The communication device according to claim 1 or 2, wherein the transmitter is configured on one semiconductor chip.   The communication device according to any one of claims 1 to 3, further comprising a case in which at least the sound receiving unit and the receiving unit of the transmitting unit are fixed to the housing, respectively.   The sound receiving means includes a supported portion that is directly or indirectly fixed to the housing of the communication device and a vibrating portion that is supported so as to be swingable with respect to the supported portion via a gimbal structure and receives sound pressure. A plurality of sound pressure detectors that convert the force received by the sound pressure of the sound incident on the diaphragm, the vibration portions of the diaphragm being incident on the diaphragm, and a plurality of sound 5. The communication device according to claim 1, further comprising: a spatiotemporal gradient measurement processing unit that obtains a sound pressure, a spatial differential value, and a temporal differential value by using an output of the pressure detection unit. .   The sound receiving means is provided with a rectangular apex arrangement, each of the four microphones for converting the sound pressure of the incident sound into an electric signal, and the sound pressure, the spatial differential value, and the time differential by calculation using the output of each microphone. The communication apparatus according to claim 1, further comprising a spatiotemporal gradient measurement processing unit that obtains a value. The transmitter has an instruction input means for inputting an instruction for switching the operation mode,
In response to the instruction input to the instruction input unit, the transmission signal generation unit outputs the dead point forming process without performing the dead point formation process in addition to the dead point formation mode for forming the dead point at the position of the receiving unit. The call device according to any one of claims 1 to 6, wherein the operation mode can be switched to an omnidirectional mode that generates an audio signal based only on sound pressure. The transmitter has an instruction input means for inputting an instruction for switching the operation mode,
In response to the instruction input to the instruction input means, the transmission signal generating means, in addition to the dead point forming mode for forming a dead point at the position of the receiving part, the sound pressure and time differential value output by the sound receiving means Using a sound signal generated so as to form a dead point at a preset target position by a spatiotemporal gradient method using a spatial differential value and a sound signal based only on the sound pressure output by the sound receiving means 2. The operation mode can be switched to a sound collection area forming mode for generating and outputting an audio signal that selectively reflects sound pressure generated by sound from a sound source near the target position. The communication device according to any one of? 7.

Images