JP5698279B2 - Sound field recording / reproducing apparatus, method, and program - Google Patents

Description

  The present invention relates to a wave field synthesis method (Ambisonics) technology that collects a sound signal with a microphone installed in a certain sound field, and reproduces the sound field with a speaker using the sound signal. About.

  The wavefront synthesis method and ambisonics are technologies that virtually reproduce a sound field in a remote place using a plurality of microphones and speakers. As such a technique, for example, a technique described in Non-Patent Document 1 is known. In applications such as remote communication systems, real-time sound collection and reproduction is required, so the sound pressure collected by a general microphone array is unique to a sound field reproduction signal for output by a general speaker array. It must be convertible to

Oyama, Furuya, Hiwazaki, Haneda, Suzuki, “Wavefront reconstruction filter for cylindrical microphone / speaker array”, September 2012, Proceedings of Autumn Meeting of Acoustical Society of Japan, pp.605-608

  In the technique described in Non-Patent Document 1, a filter for conversion is derived on the assumption that a cylindrical microphone array / speaker array is used. Therefore, if this filter is applied to a circular microphone array / speaker array, the sound field cannot be reproduced with high accuracy, and a sound field that sounds from all directions around the listener can be reproduced with high accuracy. There was a possibility that could not.

  An object of the present invention is to reproduce a sound field with higher accuracy than in the past when using a circular microphone array / speaker array, and to enhance a sound field where sound comes from all directions around the listener. It is an object to provide a sound field sound collecting / reproducing apparatus, method, and program that can be accurately reproduced.

In order to solve the above problems, a sound field sound collecting and reproducing apparatus according to an aspect of the present invention is a first device in which the circumferential direction of the baffle is a circumferential direction around the axis of a cylindrical rigid baffle having a radius _Rb . Assume that at least two microphones are arranged on the circumference of the radius R _m of the space, R _m ≧ R _b , the circumferential direction of the baffle is φ direction, j is an imaginary unit, ω is frequency, and c is Sound speed, k = ω / c, m is the order in the φ direction, H ⁽¹⁾ _m (・) is the m-th order first-class Hankel function, J _m (・) is the m-order Bessel function, H ⁽¹⁾ _m '(・) is the derivative of the m-th order first-class Hankel function H ⁽¹⁾ _m (・), and J _m ' (・) is the derivative of the m-th order Bessel function J _m (・). , W _m is a weight determined based on m, A (ω) is a complex number determined based on ω, and at least two speakers are circles of a virtual cylinder having a radius R _{s in} a second space different from the first space. Arranged around And the speaker is a circle the secondary source circles are arranged, as the distance to R _ref from the center of the secondary source a circle to a point on the circumference to match the amplitude, generated based on signals collected by the microphone A transform filter unit that generates a filtered signal D ~ _m (ω) by applying a filter F ~ _m (ω) defined by the following equation to the spatiotemporal frequency domain signal P ~ _m (ω) ,

A spatial frequency inverse transform unit that transforms the filtered signal D to _m (ω) into a frequency domain signal by inverse Fourier transform of space, and a frequency domain signal that is transformed into a time domain signal by inverse Fourier transform, and the transformed time And a frequency inverse transform unit that outputs the region signal to the speaker.

Other sound field sound collecting and reproducing apparatus according to an aspect of the invention, a circle of radius R _m of the first space to the circumferential direction of the baffle around the axis of the rigid baffle cylindrical radius R _b and a circumferential direction Suppose that at least two microphones are arranged in the circumference, R _m ≧ R _b , baffle circumferential direction is φ direction, j is an imaginary unit, ω is frequency, c is sound velocity, k = ω / c, m is the order in the φ direction, H ⁽¹⁾ _m (・) is the m-th order Hankel function, J _m (・) is the m-order Bessel function, and H ⁽¹⁾ _m '(・) Is the m-th order Hankel function H ⁽¹⁾ _m (・) derivative, J _m ′ (・) is the m-th order Bessel function J _m (・) derivative, and w _m is based on m And A (ω) is a complex number determined based on ω, and at least two speakers are arranged on the circumference of a virtual cylinder having a radius R _{s in} a second space different from the first space. The speaker is placed Circle and the secondary source a circle, the frequency for converting the R _ref from the center of the secondary source a circle as the distance to a point on the circumference to match the amplitude, the signal picked up by the microphone by a Fourier transform into a frequency domain signal Transformer, spatial frequency transformer that transforms frequency domain signal into spatio-temporal frequency domain signal P ~ _m (ω) by Fourier transform of space, and for spatiotemporal frequency domain signal P ~ _m (ω) Applying a filter F ~ _m (ω) defined by the conversion filter unit to generate a filtered signal D ~ _m (ω),

including.

Other sound field sound collecting and reproducing apparatus according to an aspect of the invention, around the center of the spherical rigid radius R _b baffle, at least two microphones the circumference of a circle of radius R _m of the first space R _m ≧ R _b , the circle where the microphone is placed is the receiving circle, the circumferential direction of the receiving circle is the φ direction, and the plane perpendicular to the receiving circle and the baffle The circumferential direction of the great circle is the θ direction, j is the imaginary unit, ω is the frequency, c is the speed of sound, k = ω / c, m is the order in the φ direction, n is the order in the θ direction, and H ⁽¹⁾ _m (・) is an m-th order first-class Hankel function, J _m (・) is an m-th order Bessel function, and h ⁽¹⁾ _n (・) is an n-th order first-class Hankel function. , J _n (・) is the nth order spherical Bessel function, h ⁽¹⁾ _n ′ (・) is the derivative of the nth order first-order sphere Hankel function h ⁽¹⁾ _n (・), and j _n ′ ( ·) as the derivative of order n of the spherical Bessel functions j _n (·), P ^m _n a (-) And Jandoru陪関number, and the weight determined based on a w _m a m, A and (omega) is a complex number which is determined based on omega, at least two speakers of the radius R _s of the second space different from the first space It is assumed that the circle is placed on the circumference of the great circle of the phantom sphere, the circle where the speaker is placed is the secondary sound source circle, and R _ref is the distance from the center of the secondary sound source circle to the point on the circumference where the amplitude matches As a result of applying the filter F ~ _m (ω) defined by the following equation to the spatio-temporal frequency domain signal P ~ _m (ω) generated based on the signal collected by the microphone A conversion filter unit for generating D ~ _m (ω);

A spatial frequency inverse transform unit that transforms the filtered signal D to _m (ω) into a frequency domain signal by inverse Fourier transform of space, and a frequency domain signal that is transformed into a time domain signal by inverse Fourier transform, and the transformed time And a frequency inverse transform unit that outputs the region signal to the speaker.

Other sound field sound collecting and reproducing apparatus according to an aspect of the invention, around the center of the spherical rigid radius R _b baffle, at least two microphones the circumference of a circle of radius R _m of the first space R _m ≧ R _b , the circle where the microphone is placed is the receiving circle, the circumferential direction of the receiving circle is the φ direction, and the plane perpendicular to the receiving circle and the baffle The circumferential direction of the great circle is the θ direction, j is the imaginary unit, ω is the frequency, c is the speed of sound, k = ω / c, m is the order in the φ direction, n is the order in the θ direction, and H ⁽¹⁾ _m (・) is an m-th order first-class Hankel function, J _m (・) is an m-th order Bessel function, and h ⁽¹⁾ _n (・) is an n-th order first-class Hankel function. , J _n (・) is the nth order spherical Bessel function, h ⁽¹⁾ _n ′ (・) is the derivative of the nth order first-order sphere Hankel function h ⁽¹⁾ _n (・), and j _n ′ ( ·) as the derivative of order n of the spherical Bessel functions j _n (·), P ^m _n a (-) And Jandoru陪関number, and the weight determined based on a w _m a m, A and (omega) is a complex number which is determined based on omega, at least two speakers of the radius R _s of the second space different from the first space It is assumed that the circle is placed on the circumference of the great circle of the phantom sphere, the circle where the speaker is placed is the secondary sound source circle, and R _ref is the distance from the center of the secondary sound source circle to the point on the circumference where the amplitude matches A frequency converter that converts a signal collected by a microphone into a frequency domain signal by Fourier transform, and a spatial frequency that converts a frequency domain signal to a spatio-temporal frequency domain signal P ~ _m (ω) by Fourier transform of space Transformer and transform filter that generates filtered signal D ~ _m (ω) by applying filter F ~ _m (ω) defined by the following equation to spatio-temporal frequency domain signal P ~ _m (ω) And

including.

  Using a filter for a circular microphone array / speaker array, it is possible to convert the collected sound signal of the circular microphone array into a drive signal of the circular speaker array and reproduce the sound field. It is possible to accurately reproduce a sound field where sound comes from all directions.

The figure for demonstrating the example of arrangement | positioning of a microphone and a speaker. The functional block diagram which shows the example of a sound field sound collection reproducing | regenerating apparatus. The flowchart which shows the example of the sound field sound collection reproduction | regeneration method. The figure for demonstrating the example of arrangement | positioning of a microphone and a speaker.

[First embodiment]
Embodiments of the present invention will be described below with reference to the drawings. In the following explanation, the symbols “~”, “ ⁻ ”, etc. used in the text should be described immediately above the character immediately before, but are described immediately after the character due to restrictions on text notation. To do. In the formula, these symbols are written in their original positions. Further, the processing performed for each element of a vector or matrix is applied to all elements of the vector or matrix unless otherwise specified.

<Disposition of microphone array and speaker array>
As shown in FIG. 1, the sound field sound collecting and reproducing apparatus and method according to the first embodiment is centered on the axis of a cylindrical rigid baffle having a radius _Rb of the first space, and the circumferential direction of the baffle is a circle. A microphone array composed of N _ch microphones M1, M2,..., MN _ch arranged on the circumference of a circle with a radius R _m as a direction, and a virtual cylinder circle with a radius R _s of the second space N _ch number of speakers S1, S2 that are arranged in the circumferential, ..., using a speaker array comprised of SN _ch, the first space formed by the sound generated by the sound source S of the first space Reproduce the sound field in the second space. Note that the circumference of the virtual cylinder means the circumference of a circle that is a common part of the plane perpendicular to the axis of the virtual cylinder and the virtual cylinder. The first space and the second space are different from each other. In FIG. 1, the sound source S reproduced in the second space is expressed as a sound source S ′. The circle formed by the microphone array is also called a sound receiving circle, and the circle formed by the speaker array is also called a secondary sound source circle. In other words, the circle where the microphone array is arranged is called a sound receiving circle, and the circle where the speaker array is arranged is called a secondary sound source circle. The axial direction of the baffle is the z direction, and the circumferential direction of the baffle having the radius R _b is the φ direction. In other words, the φ direction is the circumferential direction of the sound receiving circle. The number of microphones arranged in the first space and the number of speakers arranged in the second space may be different. If the number of microphones is larger than the number of speakers arranged in the second space, the reproduction signal may be thinned out. On the other hand, when the number of microphones is smaller than the number of speakers arranged in the second space, the reproduction signal may be interpolated by taking an average between channels. As a method for performing the interpolation, for example, linear interpolation, sinc interpolation, or the like can be applied.

The microphone array forms a circle using two or more microphones. For example, as shown in FIG. 1, about the axis of the rigid baffle cylindrical, disposed N _ch number of microphones equally spaced around the circumference of a circle of radius R _m to the circumferential direction of the baffle and the circumferential direction Is done. N _ch is a predetermined integer of 2 or more.

The radius R _b of the cylindrical rigid baffle and the distance R _m between the baffle axis and the microphone may be any value as long as R _m ≧ R _b . In the case of R _m = R _b , the microphone is arranged on the surface of the peripheral surface of a cylindrical baffle having a radius R _b (= R _m ). In the case of R _m > R _b , as shown in FIG. 1, the microphone is disposed at a position separated from the axis of the cylindrical baffle having a radius R _b by R _m . In other words, the microphone is disposed at a position away from the surface of the peripheral surface of the baffle (R _m -R _b ). For example, the microphone is arranged by being supported by a thin rod-like member protruding vertically from the peripheral surface of the baffle.

Note that the filters F to _m (ω) used in the conversion filter unit 4 are derived on the assumption that the baffle cylinder has an infinite length (for details, refer to <Filter Derivation> described later). Therefore, (1) it is preferable that the cylinder of the baffle is long, and (2) the position of the microphone is preferably at the center of the length of the cylinder of the baffle. In this embodiment, the length of the baffle and the installation position of the microphone are not limited to (1) and (2), but in the cases of (1) and (2), the wavefront closest to the sound collection field is synthesized. can do.

N _ch number of microphones arranged in the circumference of a circle of radius R _m is the phi _c as a predetermined angle are located at intervals of phi _c radian respectively. Note that the microphones do not have to be arranged at exactly equal intervals as long as they are arranged at approximately equal intervals. That is, φ _c that is the distance between adjacent microphones does not need to be exactly the same value, and may be almost the same value. Further, the microphones may be arranged at any interval. That is, φ _c that is the interval between adjacent microphones can take an arbitrary value. However, it is possible to reproduce the sound field with high accuracy by arranging the microphones at almost equal intervals, that is, by setting φ _c that is the interval between adjacent microphones to be substantially the same value.

The speaker is also arranged in the same manner as the microphone. In other words, the speaker array forms a circle using two or more speakers. For example, as shown in FIG. 1, _Nch speakers are arranged at equal intervals around the circumference of the virtual cylinder. N _ch is a predetermined integer of 2 or more.

N _ch number of speakers arranged on the circumference of the virtual cylinder, a phi _c as a predetermined angle are located at intervals of phi _c radian respectively. Note that the speakers do not need to be arranged at exactly equal intervals as long as they are arranged at approximately equal intervals. That is, φ _c that is an interval between adjacent speakers does not need to be exactly the same value, and may be almost the same value. Further, the speakers may be arranged at any interval. That is, φ _c that is the distance between adjacent speakers can take an arbitrary value. However, the sound field can be reproduced with high accuracy by arranging the speakers at substantially equal intervals, that is, by setting φ _c that is the interval between adjacent speakers to be substantially the same value.

The positional relationship of the N _ch microphones on the circumference of the receiving circle is preferably the same as the position of the corresponding N _ch speakers in the secondary sound source circle, but may be different. If this position is the same, the sound field can be reproduced more faithfully.

The radius R _{m of the} sound receiving circle is, for example, about 0.2 m. Further, the radius R _s of the secondary sound source circle is, for example, about 1.5 m. In this embodiment, it is assumed that R _s ≧ R _m , but the same processing can be applied even if this is not satisfied (that is, even when R _s <R _m ). However, the accuracy of sound field reproduction is improved when R _s ≧ R _m . The larger the value of the radius R _{m of} the circle formed by the microphone array and the radius R _{s of the} virtual cylinder, the larger the area can be reproduced, but more microphones and speakers are required. It is desirable to experimentally set the radius R _{m of} the circle formed by the microphone array and the radius R _{s of the} virtual cylinder in consideration of the frequency of sound collection and reproduction. Further, the microphone is arranged outside the sound receiving circle, and the speaker is arranged inside the secondary sound source circle. The sound collection range targeted by the microphone array is outside the sound receiving circle, and the reproduction range targeted by the speaker array is inside the secondary sound source circle.

  The speaker may be disposed in the air in the second space in an acoustically transparent state, or may be disposed in the second space in a state that is not acoustically transparent. The acoustically transparent state is a state in which the same transmission characteristic as that of the second space where no speaker is arranged is maintained. For example, if you want to place the speaker in the air in the second space in an acoustically transparent state, the speaker is placed in the air in the second space by hanging it with a thread or fixing it with a thin rod. Is done.

Microphone Mi (i = 1,2, ..., N ch) of the first space r the position of a cylindrical coordinate system ^{_{- m, i = (R m}} , φ m, i, 0) and expressed. The position of the speaker Si (i = 1, 2,..., N _ch ) in the second space is expressed as r ⁻ _{s, i} = (R _s , φ _{s, i} , 0) in the cylindrical coordinate system.

<Sound field recording and playback device>
The sound field sound collecting and reproducing apparatus according to the first embodiment includes, for example, a frequency conversion unit 1, a spatial frequency conversion unit 3, a conversion filter unit 4, a spatial frequency reverse conversion unit 5, and a frequency reverse conversion unit 6 as shown in FIG. The processing of each step illustrated in FIG. 3 is performed.

The microphones M1, M2,..., MN _ch arranged in the first space pick up the sound emitted from the sound source S in the first space and generate a time domain signal. The generated signal is sent to the frequency converter 1. _{^{r - m, i = (R}} m, φ m, i, 0) of the signal collected by the microphone Mi located to time t in the time domain is denoted as p _i (t).

<Frequency converter 1>
The frequency converter 1 converts the signal p _i (t) collected by the microphone Mi into a frequency domain signal P _i (ω) by Fourier transform (step S1). The generated frequency domain signal P _i (ω) is sent to the spatial frequency converter 3. ω is a frequency. A value k = ω / c obtained by dividing ω by the speed of sound c is defined as a wave number. The wave number is a so-called spatial frequency or angular spectrum. For example, the frequency domain signal P _i (ω) is generated by short-time discrete Fourier transform. Of course, the frequency domain signal P _i (ω) may be generated by other existing methods. Further, the frequency domain signal P _i (ω) may be generated using a method such as overlap add. When the input signal is long or when the signal is continuously input as in real time processing, the processing is performed for each frame such as every 10 ms. The frequency domain signal P _i (ω) is defined, for example, as in Expression (1). J in the argument of the function exp is an imaginary unit.

<Spatial frequency converter 3>
The spatial frequency conversion unit 3 converts the frequency domain signal P _i (ω) into the spatio-temporal frequency domain signals P _m (ω) by Fourier transform of space (step S3). The spatio-temporal frequency domain signal P _m (ω) is calculated for each frequency ω. The converted spatio-temporal frequency domain signal P _m (ω) is sent to the conversion filter unit 4. The spatial frequency conversion unit 3 calculates P _m (ω) defined by Equation (2), for example.

m is the order in the φ direction. Here, −M ≦ m ≦ M, and m is an integer. Expression (2) is an example of conversion to the spatio-temporal frequency domain, and spatial Fourier transform may be performed by other methods. Also, a method of performing two-dimensional DFT by combining equations (1) and (2) may be used.

<Conversion filter unit 4>
The transform filter unit 4 applies a predetermined filter F to _m (ω) to the spatio-temporal frequency domain signal P to _m (ω) by the following equation to generate a filtered signal D to _m (ω) ( Step S4). The filtered signal D _m (ω) is transmitted to the spatial frequency inverse transform unit 5.

R _ref is a distance from the center of the secondary sound source circle to a point on the circumference where the amplitudes coincide (exist on the same plane as the secondary sound source circle). Therefore, when R _ref = 0, the amplitudes match at the center of the secondary sound source circle, and when R _ref ≠ 0, the amplitudes match on the circumference of the concentric circle of the secondary sound source circle of radius R _ref . When R _ref = 0, the set of points whose amplitudes match from the center of the secondary sound source circle does not form a circle but becomes a point (center of the secondary sound source circle). It is included in the expression “circumference in which the amplitudes coincide from the center of the circle”. R _ref may be paraphrased as the distance from the center of the secondary sound source circle to the point on the same plane as the secondary sound source circle that matches the amplitude. In the case of equation (4), the amplitude coincides with the target sound field on the circumference of the concentric circle of the secondary sound source circle determined by R _ref , whereas in the case of equation (5), there is a certain point (in the secondary sound source circle) The amplitude is the same at the center. A (ω) is a complex number determined based on ω (an arbitrary complex number depending on ω), and is a value for performing an equalizing operation such as, for example, lowering the gain in the high frequency band. In the case of formula (5), specifically, the amplitude is the same height as the speaker array, and the amplitude at the center of the secondary sound source circle or the circumference of the concentric circle of the secondary sound source circle of radius R _ref. Match. w _m is a weight function and is defined as follows, for example.

Γ (z) and Y _v (z) represent a gamma function and a Neumann function, respectively. The conversion filter is obtained by numerically calculating the above equation. The integral value of equation (4) or the like may be approximately calculated by discretization, for example, as in the following equation.

Here, i is the discretization index, Δkz is the discretization interval, and I is the total range. Equation (11) is an example of a method of discretizing and obtaining an integral value approximately, and may be discretized by other methods. It can be discretized using various approximation formulas used in the definition of the piecewise quadrature method.

When R _b = R _m , that is, when the microphone is arranged on the surface of the circumferential surface of the cylindrical baffle having the radius R _b (= R _m ), the equations (4) and (5) are respectively It can be simplified as follows.

Reproduced in the second space by applying the filter F _m (ω) defined by the equation (4) or the equation (5) or the equation (4-2) or the equation (5-2). Since the amplitude of the signal can be matched on a predetermined circumference, the amplitude of the signal reproduced in a wider range than before is matched.

<Spatial frequency inverse transform unit 5>
The spatial frequency inverse transform unit 5 transforms the filtered signal _D˜m (ω) into the frequency domain signal D _i (ω) by inverse spatial Fourier transform (step S5). The converted frequency domain signal D _i (ω) is sent to the frequency inverse transform unit 6. The spatial frequency inverse transform unit 5 calculates the frequency domain signal D _i (ω) defined by the equation (12), for example.

<Inverse frequency converter 6>
The frequency inverse transform unit 6 transforms the frequency domain signal D _i (ω) into the time domain signal d _i (t) by inverse Fourier transform (step S6). The time domain signal d _i (t) obtained for each frame by the inverse Fourier transform is appropriately shifted to obtain a linear sum to be a continuous time domain signal. For the inverse Fourier transform, an existing method such as a short-time discrete inverse Fourier transform may be used. The time domain signal d _i (t) is sent to the speakers S1, S2 _,.

The speaker Si reproduces sound based on the time domain signal d _i (t). For example, i = 1, ..., a ^{_{N ch, r - s, i}} = (R s, φ s, i, 0) speakers positioned on Si reproduces sound based on the time-domain signal d _i (t) . Thereby, the sound field of the first space can be reproduced in the second space.

As described above, when the number of microphones is larger than the number of speakers, the time domain signal d _i (t) may be thinned out. On the other hand, when the number of microphones is smaller than the number of speakers, interpolation may be performed by taking an average of the time domain signals d _i (t). As a method for performing the interpolation, for example, linear interpolation, sinc interpolation, or the like can be applied.

<Effect>
Using a filter for a circular microphone array / speaker array, it is possible to convert the collected sound signal of the circular microphone array into a drive signal of the circular speaker array and reproduce the sound field. It is possible to accurately reproduce a sound field where sound comes from all directions. Therefore, the sound field can be reproduced with higher accuracy than in the past.

<Derivation of filter>
Hereinafter, the reason why the filters F to _m (ω) are expressed as in Expression (4) and Expression (5) will be described.

Here, the cylindrical coordinate system r ⁻ = (r, φ, z) is considered. The sound field on the xy plane is reproduced using the secondary sound source circle arranged in the reproduction area from the sound pressure distribution acquired in the sound receiving circle arranged in the sound collection area.

Composite sound due to secondary sound sources successively arranged on the circumference field P _syn (r ^-, ω), the secondary source drive signals D ^- and (r _s, omega), the secondary source and the target sound field Using the transfer function G (r ⁻ −r ⁻ _s , ω), the following is expressed.

Since this can be regarded as a convolution with respect to φ of D (•) and G (•), it can be expressed in the form of a product in the circular harmonic spectrum region as follows.

On the other hand, the sound pressure distribution P _rcv (R _m , 0, φ, obtained from the target sound field P _des (r ⁻ , ω) on the circumference of the radius R _m of the sound collection field P _rcv (r ⁻ , ω). It is necessary to express using ω). Around the same position as the sound receiving circle, when the infinitely long rigid cylinder of radius R _b (≦ R _m) is installed, the sound collection field P _rcv (r ^-, ω) is incident as: It can be expressed by the sum of the sound field P _inc (r ⁻ , ω) and the scattered sound field P _sct (r ⁻ , ω).
^{_{P rcv (r -, ω)}} = P inc (r -, ω) + P sct (r -, ω) ... (15)
From the boundary condition on the rigid cylinder,

Holds. Also, P _inc (r ⁻ , ω) and P _sct (r ⁻ , ω) can be written as follows in the helical wave spectrum region.

Therefore,

It becomes. As described above, assuming that the sound collection field is invariant in the z-axis direction, the relational expression between the incident sound field and the sound collection field can be derived as follows.

Since the target sound field is equal to the incident sound field,

It becomes.
Synthetic sound field P _syn (r ^-, ω) and the target sound field P _des (r ^-, ω) from the fact that it is sufficient to match,

For simplicity, it is assumed that G (r ⁻ −r ⁻ _s , ω) has a monopole characteristic.

As described above, the conversion formula of the secondary sound source drive signal is obtained from the sound pressure distribution on the sound receiving circle, but it is necessary to give in advance a circle whose amplitude matches the target sound field as r = R _ref .

  If the secondary sound source is assumed to be a line sound source, the conversion formula can be simplified.

Than

It becomes. However, the actual speaker is close to the monopole characteristic, and correction for approximating the line sound source is required.

Therefore, the following frequency characteristics are corrected for each output signal.

  In the case of Rb = Rm, it can be simplified using the following relational expression.

[Second Embodiment]
A description will be given centering on differences from the first embodiment.

<Disposition of microphone array and speaker array>
Sound field sound collection reproducing apparatus and method of the second embodiment, as shown in FIG. 4, around the center of the spherical rigid radius R _b baffle, the circumference of a circle of radius R _m of the first space N _ch number of microphones M1 disposed in, M2, ..., MN and a microphone array constituted by _ch, N _ch disposed on a great circle of the circumference of the virtual sphere of radius R _s of the second space The sound field of the first space formed by the sound generated by the sound source S in the first space is used in the second space using the speaker array composed of the speakers S1, S2 _,. Reproduce. The circle formed by the microphone array is also called a sound receiving circle, and the great circle of the virtual sphere formed by the speaker array is also called a secondary sound source circle. The direction perpendicular to the sound receiving circle and the secondary sound source circle is the z direction, the circumferential direction of the sound receiving circle and the secondary sound source circle is the φ direction, and a great circle of a plane perpendicular to the sound receiving circle and the baffle The circumferential direction of the great circle between the plane perpendicular to the secondary sound source circle and the phantom sphere is the θ direction.

In the first embodiment, microphones are respectively provided on a circle of radius R _m and a circumference of a virtual cylinder of radius R _s centered on the axis of a cylindrical rigid baffle having a radius R _b and the circumferential direction of the baffle being the circumferential direction. And a speaker. On the other hand, in the second embodiment, a microphone and a circle of a radius R _{m in} the first space and a great circle of a virtual sphere of radius R _s around the center of a spherical rigid body baffle having a radius R _b are respectively provided A speaker is arranged.

Microphone Mi (i = 1,2, ..., N ch) of the first space r a position in a spherical coordinate system ^{_{- m, i = (R m}} , π / 2, φ m, i) and expressed. The position of the speaker Si (i = 1, 2,..., N _ch ) in the second space is expressed as r ⁻ _{s, i} = (R _s , π / 2, φ _{s, i} ) in a spherical coordinate system.

<Conversion filter unit 4>
The transform filter unit 4 applies a predetermined filter F to _m (ω) to the spatio-temporal frequency domain signal P to _m (ω) by the following equation to generate a filtered signal D to _m (ω) ( Step S4). The filtered signal D _m (ω) is transmitted to the spatial frequency inverse transform unit 5.

R _ref is a distance from the center of the secondary sound source circle to a point on the circumference where the amplitudes coincide (exist on the same plane as the secondary sound source circle). n is the order in the θ direction. h ⁽¹⁾ _n (•), j _n (•), h ⁽¹⁾ _n '(•), j _n ' (•), and P ^m _n (•) are n-order first-class spherical Hankel functions, respectively. , N-th order spherical Bessel function, n-th order first-class spherical Hankel function h ⁽¹⁾ _n (・) derivative, n-order spherical Bessel function j _n (・) derivative, Legendre 陪 function Is defined as

P _n (·) represents a Legendre polynomial. The conversion filter is obtained by numerically calculating the above equation.

Further, when R _b = R _m , that is, when the microphone is arranged on the surface of the circumferential surface of the cylindrical baffle having the radius R _b (= R _m ), the equations (31) and (32) are respectively It can be simplified as follows.

By applying the filter F _m (ω) defined by the expression (31) or the expression (32), or the expression (31-2) or the expression (32-2), it is reproduced in the second space. Since the amplitude of the signal can be matched on a predetermined circumference, the amplitude of the signal reproduced in a wider range than before is matched.

  With such a configuration, the same effect as that of the first embodiment can be obtained.

<Derivation of filter>
Hereinafter, the reason why the filters F to _m (ω) are expressed as in Expression (31) and Expression (32) will be described.

Here, the spherical coordinate system r ⁻ = (r, θ, φ) is considered. The sound field on the xy plane is reproduced using the secondary sound source circle arranged in the reproduction area from the sound pressure distribution acquired in the sound receiving circle arranged in the sound collection area.

The synthesized sound field P _syn (r ⁻ , ω) by the secondary sound source continuously arranged on the circumference can be written as follows in the circular harmonic spectrum region, as in the first embodiment.

On the other hand, the sound pressure distribution P (R _m , π / 2, φ obtained from the target sound field P _des (r ⁻ , ω) on the circumference of the radius R _m of the sound collection field P _rcv (r ⁻ , ω). , ω). As the same position as the center and the sound receiving circle, when the rigid body ball of radius R _b (≦ R _m) is installed, the sound collection field P _rcv (r ^-, ω) incident sound field P _inc (r ^-, ω) and the scattered sound field P _sct (r ⁻ , ω).
^{_{P rcv (r -, ω)}} = P inc (r -, ω) + P sct (r -, ω) (38)
From the boundary condition on a hard sphere,

Holds. In addition, P _inc (r ⁻ , ω) and P _sct (r ⁻ , ω) can be written in the spherical harmonic spectrum region as follows.

Here, Y ^m _n (θ, φ) represents a spherical harmonic function. Therefore,

It becomes. From the above, assuming that the sound collection field is universal in the z-axis direction, the relational expression between the incident sound field and the sound collection field can be derived as follows.

Furthermore,

And

Because

It becomes.
Synthetic sound field P _sys (r ^-, ω) and the target sound field P _des (r ^-, ω) from the fact that it is sufficient to match,

For simplicity, it is assumed that G (r ⁻ −r ⁻ _s , ω) has a monopole characteristic.

As described above, the conversion formula of the secondary sound source driving signal is obtained from the sound pressure distribution on the sound receiving circle. However, it is necessary to give in advance a circle whose amplitude matches the target sound field as r = R _ref .

  If the secondary sound source is assumed to be a line sound source, the conversion formula can be simplified.

Than

It becomes. However, the actual speaker is close to the monopole characteristic, and correction for approximating the line sound source is required. From the equation (26), the following frequency characteristics are corrected for each output signal.

When R _b = R _m , it can be derived using the following relational expression.

[Modifications, etc.]
Each unit constituting the sound field sound collecting / reproducing device may be provided in either the sound collecting device arranged in the first space or the reproducing device arranged in the second space. In other words, each processing of the frequency conversion unit 1, the spatial frequency conversion unit 3, the conversion filter unit 4, the spatial frequency reverse conversion unit 5, and the frequency reverse conversion unit 6 is performed by a sound collection device arranged in the first space. It may be executed or may be executed by a playback device arranged in the second space. The signal generated by the sound collection device is transmitted to the reproduction device.

  The positions of the first space and the second space are not limited to those shown in FIGS. The first space and the second space may be adjacent to each other or separated from each other. Also, the orientation of the first space and the second space may be any.

  As long as the sound field sound collecting / reproducing apparatus includes the conversion filter unit 4, the sound field collecting / reproducing device may not include other units. For example, the sound field sound collecting / reproducing apparatus may include a transform filter unit 4, a spatial frequency inverse transform unit 5, and a frequency inverse transform unit 6. Further, the sound field sound collecting / reproducing apparatus may include a frequency conversion unit 1, a spatial frequency conversion unit 3, and a conversion filter unit 4.

  You may perform the process of the frequency converter 1 and the process of the spatial frequency converter 3 simultaneously. Similarly, the process of the spatial frequency inverse transform unit 5 and the process of the frequency inverse transform unit 6 may be performed simultaneously. Further, the spatial frequency conversion unit 3 and the spatial frequency inverse conversion unit 5 may be interchanged.

  The sound field sound collecting / reproducing apparatus can be realized by a computer. In this case, the processing content of each part of this apparatus is described by a program. Then, by executing this program on a computer, each unit in this apparatus is realized on the computer.

  The program describing the processing contents can be recorded on a computer-readable recording medium. In this embodiment, these apparatuses are configured by executing a predetermined program on a computer. However, at least a part of these processing contents may be realized by hardware.

  The present invention is not limited to the above-described embodiment, and can be modified as appropriate without departing from the spirit of the present invention.

DESCRIPTION OF SYMBOLS 1 Frequency conversion part 3 Spatial frequency conversion part 4 Conversion filter part 5 Spatial frequency reverse conversion part 6 Frequency reverse conversion part

Claims (9)

And at least two microphones are arranged in the circumferential direction of the center axis of the rigid baffle cylindrical radius R _b above baffle circumference of radius R _m of the first space to the circumferential direction, R _m ≧ R _b , the circumferential direction of the baffle as φ direction, j as imaginary unit, ω as frequency, c as sound velocity, k = ω / c, m as order in φ direction, H ^{(1 )} _m (・) is the m-th order first-class Hankel function, J _m (・) is the m-th order Bessel function, and H ⁽¹⁾ _m '(・) is the m-th order first-class Hankel function H ^{(1 )} _m (・), J _m ′ (・) is the m-th order Bessel function J _m (・), w _m is a weight determined based on m, and A (ω) is based on ω It is assumed that the number is a complex number, and at least two speakers are arranged on the circumference of a virtual cylinder having a radius R _{s in} a second space different from the first space, and the circle in which the speakers are arranged is a secondary sound source circle. and then, above the R _ref As the distance from the center of the secondary source a circle to a point on the circumference to match the amplitude,
Spatial frequency domain signal P ~ _m by applying a filter F ~ _m (ω) defined by the following equation with respect to (omega) filtering after signal D when generated based on the collected sound signal by the microphone ~ conversion filter unit that generates _m (ω);

A spatial frequency inverse transform unit that transforms the filtered signal D to _m (ω) into a frequency domain signal by inverse Fourier transform of space;
A frequency inverse transform unit that transforms the frequency domain signal into a time domain signal by inverse Fourier transform and outputs the transformed time domain signal to the speaker;
Sound field collection and playback device including And at least two microphones are arranged in the circumferential direction of the center axis of the rigid baffle cylindrical radius R _b above baffle circumference of radius R _m of the first space to the circumferential direction, R _m ≧ R _b , the circumferential direction of the baffle as φ direction, j as imaginary unit, ω as frequency, c as sound velocity, k = ω / c, m as order in φ direction, H ^{(1 )} _m (・) is the m-th order first-class Hankel function, J _m (・) is the m-th order Bessel function, and H ⁽¹⁾ _m '(・) is the m-th order first-class Hankel function H ^{(1 )} _m (・), J _m ′ (・) is the m-th order Bessel function J _m (・), w _m is a weight determined based on m, and A (ω) is based on ω It is assumed that the number is a complex number, and at least two speakers are arranged on the circumference of a virtual cylinder having a radius R _{s in} a second space different from the first space, and the circle in which the speakers are arranged is a secondary sound source circle. and then, above the R _ref As the distance from the center of the secondary source a circle to a point on the circumference to match the amplitude,
A frequency converter that converts the signal collected by the microphone into a frequency domain signal by Fourier transform;
A spatial frequency converter that converts the frequency domain signal into a spatio-temporal frequency domain signal P _m (ω) by Fourier transform of the space;
A transform filter unit that generates a filtered signal D to _m (ω) by applying a filter F to _m (ω) defined by the following equation to the spatio-temporal frequency domain signal P to _m (ω):

Sound field collection and playback device including Around the center of the spherical rigid baffle having a radius R _b, and at least two microphones are arranged in the circumference of a circle of radius R _m of the first space, and R _m ≧ R _b, the microphone Is the receiving circle, the circumferential direction of the receiving circle is the φ direction, the circumferential direction of the great circle between the plane perpendicular to the receiving circle and the baffle is the θ direction, and j is The imaginary unit, ω is the frequency, c is the speed of sound, k = ω / c, m is the order in the φ direction, n is the order in the θ direction, and H ⁽¹⁾ _m ( It is a kind of Hankel function, J _m (・) is an m-th order Bessel function, h ⁽¹⁾ _n (・) is an n-th order first-class sphere Hankel function, and j _n (・) is an n-th order sphere Bessel H ⁽¹⁾ _n ′ (・) is the derivative of the nth-order first-order sphere Hankel function h ⁽¹⁾ _n (・), and j _n ′ (・) is the nth-order spherical Bessel function j _n ( ·) of the differential, and P ^m _n a (·) and the associated Legendre, a w _m to m And the weight determined in Zui, a complex number determined based A a (omega) to omega, at least two speakers are in the great circle of the circumference of the virtual sphere of radius R _s of the second space different from the first space Suppose that the circle where the speaker is placed is a secondary sound source circle, and R _ref is the distance from the center of the secondary sound source circle to a point on the circumference where the amplitudes match,
Spatial frequency domain signal P ~ _m by applying a filter F ~ _m (ω) defined by the following equation with respect to (omega) filtering after signal D when generated based on the collected sound signal by the microphone ~ conversion filter unit that generates _m (ω);

A spatial frequency inverse transform unit that transforms the filtered signal D to _m (ω) into a frequency domain signal by inverse Fourier transform of space;
A frequency inverse transform unit that transforms the frequency domain signal into a time domain signal by inverse Fourier transform and outputs the transformed time domain signal to the speaker;
Sound field collection and playback device including Around the center of the spherical rigid baffle having a radius R _b, and at least two microphones are arranged in the circumference of a circle of radius R _m of the first space, and R _m ≧ R _b, the microphone Is the receiving circle, the circumferential direction of the receiving circle is the φ direction, the circumferential direction of the great circle between the plane perpendicular to the receiving circle and the baffle is the θ direction, and j is The imaginary unit, ω is the frequency, c is the speed of sound, k = ω / c, m is the order in the φ direction, n is the order in the θ direction, and H ⁽¹⁾ _m ( It is a kind of Hankel function, J _m (・) is an m-th order Bessel function, h ⁽¹⁾ _n (・) is an n-th order first-class sphere Hankel function, and j _n (・) is an n-th order sphere Bessel H ⁽¹⁾ _n ′ (・) is the derivative of the nth-order first-order sphere Hankel function h ⁽¹⁾ _n (・), and j _n ′ (・) is the nth-order spherical Bessel function j _n ( ·) of the differential, and P ^m _n a (·) and the associated Legendre, a w _m to m And the weight determined in Zui, a complex number determined based A a (omega) to omega, at least two speakers are in the great circle of the circumference of the virtual sphere of radius R _s of the second space different from the first space Suppose that the circle where the speaker is placed is a secondary sound source circle, and R _ref is the distance from the center of the secondary sound source circle to a point on the circumference where the amplitudes match,
A frequency converter that converts the signal collected by the microphone into a frequency domain signal by Fourier transform;
A spatial frequency converter that converts the frequency domain signal into a spatio-temporal frequency domain signal P _m (ω) by Fourier transform of the space;
A transform filter unit that generates a filtered signal D to _m (ω) by applying a filter F to _m (ω) defined by the following equation to the spatio-temporal frequency domain signal P to _m (ω):

Sound field collection and playback device including And at least two microphones are arranged in the circumferential direction of the center axis of the rigid baffle cylindrical radius R _b above baffle circumference of radius R _m of the first space to the circumferential direction, R _m ≧ R _b , the circumferential direction of the baffle as φ direction, j as imaginary unit, ω as frequency, c as sound velocity, k = ω / c, m as order in φ direction, H ^{(1 )} _m (・) is the m-th order first-class Hankel function, J _m (・) is the m-th order Bessel function, and H ⁽¹⁾ _m '(・) is the m-th order first-class Hankel function H ^{(1 )} _m (・), J _m ′ (・) is the m-th order Bessel function J _m (・), w _m is a weight determined based on m, and A (ω) is based on ω It is assumed that the number is a complex number, and at least two speakers are arranged on the circumference of a virtual cylinder having a radius R _{s in} a second space different from the first space, and the circle in which the speakers are arranged is a secondary sound source circle. and then, above the R _ref As the distance from the center of the secondary source a circle to a point on the circumference to match the amplitude,
The conversion filter unit applies a filter F to _m (ω) defined by the following equation to the spatio-temporal frequency domain signal P to _m (ω) generated based on the signal collected by the microphone. A transform filter step for generating a filtered signal D ~ _m (ω);

A spatial frequency inverse transform unit transforms the filtered signal D to _m (ω) into a frequency domain signal by inverse Fourier transform of the space,
A frequency inverse transform unit that transforms the frequency domain signal into a time domain signal by inverse Fourier transform, and outputs the transformed time domain signal to the speaker;
Sound field collection and playback method including And at least two microphones are arranged in the circumferential direction of the center axis of the rigid baffle cylindrical radius R _b above baffle circumference of radius R _m of the first space to the circumferential direction, R _m ≧ R _b , the circumferential direction of the baffle as φ direction, j as imaginary unit, ω as frequency, c as sound velocity, k = ω / c, m as order in φ direction, H ^{(1 )} _m (・) is the m-th order first-class Hankel function, J _m (・) is the m-th order Bessel function, and H ⁽¹⁾ _m '(・) is the m-th order first-class Hankel function H ^{(1 )} _m (・), J _m ′ (・) is the m-th order Bessel function J _m (・), w _m is a weight determined based on m, and A (ω) is based on ω It is assumed that the number is a complex number, and at least two speakers are arranged on the circumference of a virtual cylinder having a radius R _{s in} a second space different from the first space, and the circle in which the speakers are arranged is a secondary sound source circle. and then, above the R _ref As the distance from the center of the secondary source a circle to a point on the circumference to match the amplitude,
A frequency conversion step in which the frequency conversion unit converts the signal collected by the microphone into a frequency domain signal by Fourier transform;
A spatial frequency transforming step, wherein the spatial frequency transforming unit transforms the frequency domain signal into a spatiotemporal frequency domain signal P to _m (ω) by a Fourier transform of the space;
Conversion filter portion, by applying a filter F ~ _m (ω) defined by the following equation to generate a filtered signal after D ~ _m (ω) with respect to the space-time frequency domain signal P ~ _m (ω) A transform filter step;

Sound field collection and playback method including Around the center of the spherical rigid baffle having a radius R _b, and at least two microphones are arranged in the circumference of a circle of radius R _m of the first space, and R _m ≧ R _b, the microphone Is the receiving circle, the circumferential direction of the receiving circle is the φ direction, the circumferential direction of the great circle between the plane perpendicular to the receiving circle and the baffle is the θ direction, and j is The imaginary unit, ω is the frequency, c is the speed of sound, k = ω / c, m is the order in the φ direction, n is the order in the θ direction, and H ⁽¹⁾ _m ( It is a kind of Hankel function, J _m (・) is an m-th order Bessel function, h ⁽¹⁾ _n (・) is an n-th order first-class sphere Hankel function, and j _n ( H ⁽¹⁾ _n ′ (・) is the derivative of the nth-order first-order sphere Hankel function h ⁽¹⁾ _n (・), and j _n ′ (・) is the nth-order spherical Bessel function j _n ( ·) of the differential, and P ^m _n a (·) and the associated Legendre, a w _m to m And the weight determined in Zui, a complex number determined based A a (omega) to omega, at least two speakers are in the great circle of the circumference of the virtual sphere of radius R _s of the second space different from the first space Suppose that the circle where the speaker is placed is a secondary sound source circle, and R _ref is the distance from the center of the secondary sound source circle to a point on the circumference where the amplitudes match,
The conversion filter unit applies a filter F to _m (ω) defined by the following equation to the spatio-temporal frequency domain signal P to _m (ω) generated based on the signal collected by the microphone. A transform filter step for generating a filtered signal D ~ _m (ω);

A spatial frequency inverse transform unit transforms the filtered signal D to _m (ω) into a frequency domain signal by inverse Fourier transform of the space,
A frequency inverse transform unit that transforms the frequency domain signal into a time domain signal by inverse Fourier transform, and outputs the transformed time domain signal to the speaker;
Sound field collection and playback method including Around the center of the spherical rigid baffle having a radius R _b, and at least two microphones are arranged in the circumference of a circle of radius R _m of the first space, and R _m ≧ R _b, the microphone Is the receiving circle, the circumferential direction of the receiving circle is the φ direction, the circumferential direction of the great circle between the plane perpendicular to the receiving circle and the baffle is the θ direction, and j is The imaginary unit, ω is the frequency, c is the speed of sound, k = ω / c, m is the order in the φ direction, n is the order in the θ direction, and H ⁽¹⁾ _m ( It is a kind of Hankel function, J _m (・) is an m-th order Bessel function, h ⁽¹⁾ _n (・) is an n-th order first-class sphere Hankel function, and j _n ( H ⁽¹⁾ _n ′ (・) is the derivative of the nth-order first-order sphere Hankel function h ⁽¹⁾ _n (・), and j _n ′ (・) is the nth-order spherical Bessel function j _n ( ·) of the differential, and P ^m _n a (·) and the associated Legendre, a w _m to m And the weight determined in Zui, a complex number determined based A a (omega) to omega, at least two speakers are in the great circle of the circumference of the virtual sphere of radius R _s of the second space different from the first space Suppose that the circle where the speaker is placed is a secondary sound source circle, and R _ref is the distance from the center of the secondary sound source circle to a point on the circumference where the amplitudes match,
A frequency conversion step in which the frequency conversion unit converts the signal collected by the microphone into a frequency domain signal by Fourier transform;
A spatial frequency transforming step, wherein the spatial frequency transforming unit transforms the frequency domain signal into a spatiotemporal frequency domain signal P to _m (ω) by a Fourier transform of the space;
Conversion filter portion, by applying a filter F ~ _m (ω) defined by the following equation to generate a filtered signal after D ~ _m (ω) with respect to the space-time frequency domain signal P ~ _m (ω) A transform filter step;

Sound field collection and playback method including   A sound field recording / reproducing program for causing a computer to function as each part of the sound field sound collecting / reproducing apparatus according to claim 1.

Images